reciprocity hypothesis psychology

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 26 June 2018

Reciprocity of social influence

• Ali Mahmoodi   ORCID: orcid.org/0000-0002-8133-2546 1 , 2 ,
• Bahador Bahrami   ORCID: orcid.org/0000-0003-0802-5328 3 , 4   na1 &
• Carsten Mehring   ORCID: orcid.org/0000-0001-8125-5205 1 , 2   na1  

Nature Communications volume  9 , Article number:  2474 ( 2018 ) Cite this article

14k Accesses

42 Citations

50 Altmetric

Metrics details

• Human behaviour
• Neuroscience
• Social behaviour

Humans seek advice, via social interaction, to improve their decisions. While social interaction is often reciprocal, the role of reciprocity in social influence is unknown. Here, we tested the hypothesis that our influence on others affects how much we are influenced by them. Participants first made a visual perceptual estimate and then shared their estimate with an alleged partner. Then, in alternating trials, the participant either revised their decisions or observed how the partner revised theirs. We systematically manipulated the partner’s susceptibility to influence from the participant. We show that participants reciprocated influence with their partner by gravitating toward the susceptible (but not insusceptible) partner’s opinion. In further experiments, we showed that reciprocity is both a dynamic process and is abolished when people believed that they interacted with a computer. Reciprocal social influence is a signaling medium for human-to-human communication that goes beyond aggregation of evidence for decision improvement.

Similar content being viewed by others

Language is primarily a tool for communication rather than thought

Using games to understand the mind

Principal component analysis

Introduction.

When we are uncertain, we look for a second opinion and those opinions often change our decisions and preferences 1 , 2 , 3 , 4 , 5 . Moreover, social influence is not restricted to difficult or critical decisions: evaluations of the comments in the news media are affected by previous scores of the content 6 and risk preferences alter after observing other people’s choices 7 . Social influence extends to perceptual judgment 8 , 9 , 10 and long-term memory 11 . Social information can help improve decision accuracy 12 , outcome value 13 , and evaluative judgment 14 , 15 , 16 . On the other hand, social influence can also lead to catastrophic outcomes. Social influence causes information cascades 17 urging individuals to ignore their own accurate information in favor of the cascaded falsehoods. In some cases, groups are less biased when their individuals resist social influence 18 . Social influence can undermine group diversity 19 leading to disastrous phenomena such as market bubble 20 , rich-get-richer dynamics 21 , and zealotry 22 .

Humans tend to reciprocate in social interaction 23 , 24 . We react to respect with respect and hostility with hostility 23 . Smiling staff get more tips 25 . Violators of trust in Trust Games 26 , and free riders (i.e., those who do not contribute) in Public Good Games are punished 27 . Since social influence is, by definition, mediated via social interaction, one may wonder if reciprocity extends to social influence itself. However, despite a wealth of research on reciprocal behavior, to our knowledge, no study has examined the existence of reciprocity in how we receive and inflict influence on others 28 , 29 .

Several studies in social decision-making in humans have shown a sensible correspondence between the reliability of social information (e.g., advice) and the extent to which that advice is assimilated in decisions and preferences. For example, human agents integrate their own choice with that of an advisor by optimally tracking the trustworthiness of the advisor 30 and are able to track the expertise of several agents concurrently 31 . Social information is integrated into value and confidence judgment based on its reliability 32 or credibility 33 . People can integrate information from themselves and others based on their confidence 9 . These studies offer strong evidence for what previous works in social psychology have called “informational conformity” 34 , which is based on the assumption that others can have access to information that will help the agent achieve better accuracy. The prediction drawn from this informational account is that social influence should hinge on the reliability of the source and quality of the advice but not on conventions and norms such as reciprocity. If we could benefit from others’ (critical) opinion and the accuracy of our opinion is our only concern, then we should welcome reliable advice irrespective of the advisor’s attitude toward our opinion.

On the other hand, numerous studies indicate that people perform less than ideally when using social information 10 , 35 , 36 , 37 . When aggregating individual and social information about a perceptual decision, humans follow a simplifying heuristic, dubbed equality bias: the tendency to allow everyone equal say in a collective decision irrespective of their differential accuracy or expertise 36 , 37 . The universal prevalence of the equality bias is important because it shows that even though humans do have the cognitive and computational capacity to track the trustworthiness 30 , reliability 31 , and credibility 33 of others, they still choose to employ a simple heuristic. Other studies have shown that in a different set of experimental conditions, people show an egocentric bias by relying on their own individual information more than they should 38 , 39 . These findings suggest that normative concerns such as equality and maintaining a good self-image may also play an important role in interactive decision-making. These factors are not consistent with the Bayesian theory of social information aggregation 30 , 32 , 33 , which requires that the influence that people take from others should not depend on norms and conventions. An empirical observation of reciprocity in advice-taking would be inconsistent with the Bayesian theory of social influence 30 , 32 , 33 .

To summarize, aligning with other people’s choices by taking their advice could be motivated informationally 40 to increase accuracy, or normatively 41 to affiliate with others and maintain positive self-esteem. Following other people’s advice often leads to more accurate decisions 9 , 42 , 43 . Alignment can contribute to a positive self-image 44 and is used as a compensatory tool among minorities 45 . Being ignored in a virtual game can damage one’s self-image 46 . Social exclusion has negative consequences on the excluded 47 . We hypothesized that people would reciprocate influence with others because reciprocity is a pervasive social norm 48 . If one breaks the norm of reciprocity, one should expect to be punished 49 , for example, by being ignored. Therefore, participants would reciprocate with a reciprocating partner in order to maintain influence over them and avoid the negative experience of losing influence 47 . On the other hand, for a non-reciprocating partner who violates the norm, participants may have a desire to punish the partner by ignoring their opinion. We tested these hypotheses by investigating if participants in a social decision-making task take less advice from insusceptible partners, i.e., those who do not take advice from the participant. Conversely, we also tested if participants take more advice from susceptible partners, i.e., those who are influenced by the participant’s suggestions.

We adopted and modified an experimental perceptual task inspired by recent work on aggregation of social and individual information 10 . Participants estimated the location of a visual target on a computer screen. Then they saw the estimate of their partner about the location of the same target. The participant did not know that this partner’s opinion was, in reality, generated by sampling randomly from a distribution centered on the correct answer. After the two initial estimates were disclosed, the participant or the partner was allowed to revise their estimate. An algorithm generated the partner’s revised estimate simulating susceptible or insusceptible partners. Experiment 1 showed that, participants took more advice from the partner who took more advice from the participant. Experiment 2A and 2B investigated the dynamics of reciprocity and whether reciprocity depends critically on whether we believe it changes the partner’s state of mind. Participants thought they worked with either a human or a computer partner and reciprocity disappeared when subjects believed they were working with a computer partner. Finally, our results also showed that reciprocity had a profound impact on participants’ evaluation of their own performance, which was lower when working with a non-reciprocating partner.

Experiment 1

In experiment 1, 20 participants were recruited one at a time and told that they would cooperate with three partners who were participating in the same experiment simultaneously in other laboratory rooms connected via internet. In reality, each participant was coupled with a computer algorithm. The algorithm generated three distinct behavioral profiles, corresponding to three experimental conditions (see below), which were administered in a block design in counterbalanced order. Participants were not informed about this arrangement. In each trial, the participant made a perceptual estimate about the location of a target on the screen (Fig.  1 ). After stating her initial estimate, the participant saw the opinion of the partner about the same stimulus. Next, the participant either revised her estimate or observed the partners revise theirs. Participants were required to put their second estimate between their own first estimate and that of their partner’s. The acceptable range included staying on their first estimate or moving all the way to their partner’s first estimate. Using this constraint, we assured that the amount of change made in the second stage is solely due to observing the partner’s choice and not because of a change of mind 50 . The three partners differed in their susceptibility to taking influence from the participant. In the baseline condition, the participant always made the second estimate and the partner never contributed a second estimate. Hence, participants were not able to observe the susceptibility of the baseline partner. In the susceptible condition, the partner was influenced strongly by the participant and revised her initial estimate by conspicuously gravitating toward the participant’s estimate. Vice versa, in the insusceptible condition, the partner more or less ignored the participant’s opinion. The partner’s initial estimate was generated identically in all three conditions by sampling randomly from a distribution centered on the correct answer.

Experimental task. a Participants first observed a series of dots on the screen. Participants were required to indicate where they saw the very first dot (yellow dot) and then declare their numerical confidence. After making their individual estimates, they were presented with the estimate of a partner (red dot) concerning the same stimulus. After observing the partner’s choice, in some trials the participants and in other trials the partner was given a second chance to revise their initial estimate. Afterwards, they were briefly presented with their initial choices and the second choice. In experiment 1, they did the task with three different alleged human partners, which only varied in the second choice strategy in different blocks: in the baseline blocks, the participant made all second choices. In the susceptible blocks, the partner was very influenced by the participant’s first choice, however in the insusceptible blocks, the partner was much less influenced by the participant’s first choice compared to susceptible blocks. b Influence was computed as the angular displacement toward the peer’s choice divided by their initial distance from each other’s choice

We computed the influence that participants took from their partner as the ratio of the angular displacement (in radians) between their initial and final estimate toward their partner’s estimate divided by their initial angular distance from their partner (Fig.  1b ). Overall, participants were influenced by their partners’ opinions (mean influence ± std. dev. = 0.36 ± 0.11; Wilcoxon sign rank test vs zero, Z  = 3.62, p  = 0.0002). Consistent with previous studies 30 , 33 , 36 , this indicates that our participants did use social information (the partners’ choices) to improve their decisions (mean ± std. dev. error after first estimate 67 ± 7 radians and after second estimate 64 ± 6 radians, Wilcoxon sign rank test, Z  = 3, p  = 0.002). To test our main hypothesis, we asked whether revised opinions were more influenced by the susceptible than the insusceptible (Fig.  2a ) and baseline (Fig.  2b ) partners. The difference between the average influence in the susceptible and the insusceptible condition, which is a measure of reciprocity, was significantly larger than zero (Fig.  2a–c , Wilcoxon sign rank test, Z = 3.33, p  = 0.002 after Bonferroni correction). Similarly, the influence from the susceptible partner was larger than the influence in the baseline condition (Fig.  2b, c , Wilcoxon sign rank test, Z  = 2.34, p  = 0.03 after Bonferroni correction). The difference between insusceptible and baseline was not significant (Wilcoxon sign rank test, Z  = 1.11, p  = 0.26).

Results of experiment 1. a Reciprocity, computed as influence in the susceptible condition minus influence in the insusceptible condition, plotted across participants. b Reciprocity, computed as influence in the susceptible condition minus influence in the baseline condition, is plotted across participants. c Average reciprocity across participants in insusceptible and baseline conditions compared to the susceptible condition. Error bars indicate the standard errors and were computed across the participants’ average reciprocities. * p  < 0.05, ** p  < 0.005; Wilcoxon sign rank test. d Performance ratings for self and all partners as reported at the end of the experiment

The three different partners’ initial estimates were produced from an identical generative process using exactly the same distribution and therefore ensuring that the partners’ accuracies were perfectly controlled across conditions. However, one might argue that the observed result may be due to the difference in perceived accuracy of the partner. Indeed, we may think more highly of those who confirm our decisions more often and, subsequently, take our (misguided) assessment of their competence as grounds for integrating their estimate into our own revised opinion. To test this hypothesis directly, at the end of each experiment, we asked the participants to rate the precision of the partners they interacted within the experiment, how much they liked different partners, and their own performance as well. A mixed effect model showed that perceived precision of the partners, the actual precision of the partners, the participants’ actual precision in different conditions, and the liking of the partner did not have any effect on reciprocity (Supplementary Note  1 ). The performance ratings of the partner did not differ across conditions (Fig.  2d , repeated measures analysis of variance (ANOVA), F (2.49, 39) = 1.07, p  = 0.36). In each trial, after participants registered their estimates, they were required to report their confidence about their estimates using a scale from 1 to 6. Employing mixed effect models, we showed that condition (baseline, susceptible, or insusceptible) had a significant effect on influence even if confidence was included as a potential confound (Supplementary Note  1 ).

At the end of the experiment, we asked participants to rate how much they liked their three partners on a scale of 1–10. The mean score ± std. dev. was 7.64 ± 0.7 for the susceptible partner, 5.94 ± 2.53 for the insusceptible partner, and 7.52 ± 1.41 for the baseline partner. Our data shows that people liked the susceptible partner over the insusceptible one (Wilcoxon sign rank test, Z  = 2.62, p  = 0.008). There was no difference between susceptible and baseline partners (Wilcoxon sign rank test, Z  = −0.34, p  = 0.72) and the p -value for the difference between the baseline and the insusceptible partner was only around the threshold (Wilcoxon sign rank test, Z  = −1.91, p  = 0.056).

Experiment 2A and 2B

The results of our first experiment show that human participants were more influenced by partners which were reciprocally more influenced by the participants. We next asked whether human participants change their advice-taking strategy in response to a change in the advice-taking strategy of a partner over time. To answer this question, we carried out experiment 2A using the same paradigm as in experiment 1 but with an important modification. The participants were told that they are working with the same partner during the entire experiment. The partner’s strategy changed across time: in one part of the experiment, the partner was susceptible, in the other part, she was not. Between the two conditions of the experiment, there was a smooth transition (from susceptible to insusceptible or vice versa) and the order of the conditions was counterbalanced across participants (Supplementary Figure  1 ).

We calculated participants’ trial-by-trial influence in the two conditions. Replicating experiment 1, reciprocity, again defined as influence in the susceptible condition minus influence in the insusceptible condition, was significantly larger than zero (Fig.  3a , Wilcoxon sign rank test, Z  = 3.54, p  = 0.0003). A mixed effect model showed that condition (susceptible or insusceptible) had a significant effect on influence even if confidence was included as a potential confound (Supplementary Note  1 ). Again, using a mixed effect model, we showed that the change of influence across conditions cannot be explained by a change in perceived performance of self or partner (Supplementary Note  1 ). An interesting question is whether reciprocity is affected by the condition (susceptible or insusceptible) with which the participant started the experiment. To answer this question, we compared the reciprocity for participants who were first exposed to the susceptible partner to participants who started with the insusceptible partner. Our result showed no difference in reciprocity between these two groups (mean ± std. dev. for those who started with reciprocal partner 0.049 ± 0.1 and for those who started with non-reciprocal partner 0.08 ± 0.07, Wilcoxon rank-sum test, z  = −1.07, p  = 0.28).

Results of experiments 2A and 2B. a Reciprocity when participants believe they interact with a human partner. b Reciprocity when participants believe they interact with a computer partner. c Influence for alleged human and computer partners across susceptible and insusceptible conditions. Dots indicate each participant. Black dots depict the mean influence across participants while error bars depict the standard error of the mean. d Average reciprocity across participants when participants think they interact with a human or a computer partner. Error bars depict the standard errors. e Difference in participants’ performance rating for self, insusceptible minus susceptible, plotted for each participant. The inset shows the mean and standard error across participants. ** p  < 0.005, ns not significant; Wilcoxon sign rank test

The present results demonstrate that if a participant observes any change in the amount of influence she has over her partner, she will in return modify the amount of advice she takes from her partner. We, therefore, hypothesized that our participants exploit reciprocity as a social signal to communicate with their partner. We predicted that participants would not show reciprocity when working with a computer. In other words, reciprocity depends critically on whether we believe it changes the partner’s state of mind. To test this prediction, we conducted experiment 2B in which participants were told that they were working with a computer. All other aspects of the experiment remained as in experiment 2A. As predicted, we did not observe reciprocity when participants believed they were working with a computer (Fig.  3b–d , Wilcoxon sign rank test, Z  = −1.57, p  = 0.11). In fact, a majority of participants showed the opposite pattern observed in experiment 2A (Fig.  3b ). A two-way ANOVA with factors condition (susceptible or insusceptible as within subjects factor) and type of partner (believed to be human or computer as between subjects factor) and influence as the dependent variable showed a significant interaction of type of partner and condition ( F (1, 58) = 14.8, p  = 0.00001). The effect of condition alone was not significant ( F (1, 58) = 0.49, p  = 0.51), but there was also a significant between-subject effect of type of partner (Fig.  3c , F (1, 58) = 4.16, p  = 0.04). A post hoc analysis between reciprocity in human and computer condition confirmed a significant difference in reciprocity between these two conditions (Fig.  3c, d , Wilcoxon rank-sum test, Z  = 2.97, p  = 0.003). Taken together, these results demonstrate that participants were more influenced when they thought their partner is a computer (mean influence ± std. dev.: 42 ± 0.16) compared to when they thought their partner is a human (mean influence ± std. dev.: 0.34 ± 0.13).

We then investigated whether the distance between the participants’ and the partners’ initial estimates affected the influence that participants took from their partner or the strength of reciprocity (see Supplementary Note  1 for details). We did not find a significant effect of distance (mixed ANOVA, F (2.6, 151) = 1.17, p  = 0.31) nor significant interactions between distance and condition (mixed ANOVA, F (2.6,151) = 1.47, p  = 0.22) or between distance, condition, and experiment ( F (2.6, 151) = 1, p  = 0.38). The interaction between distance and experiment was only around the significance threshold ( F (2.6, 151) = 2.7, p  = 0.05). We also did not find a significant effect of distance on reciprocity in experiment 2A (repeated measures ANOVA, F (2.61, 75) = 1.63, p  = 0.16). Taken together, these results show that the distance between the initial estimates did not affect the influence that participants took from their partner nor the strength of reciprocity.

Finally, we asked how being in the susceptible and insusceptible conditions changed the participants’ ratings of their own and their partner’s performance. To answer this question, participants were asked to rate their own and their partners’ performance on a 1–10 scale at the end of each condition, i.e., twice in each experiment 2A and 2B. As there was not any difference in performance rating between experiment 2A and 2B, neither for self nor partner (Supplementary Figure  2 ), we aggregated the data of the two experiments. Participants’ rating of their own performance was significantly lower in the insusceptible than in the susceptible condition (Fig.  3e , Wilcoxon sign rank test, Z  = 3.04, p  = 0.002), while participants’ rating of their partners’ performance remained unaffected by the partners’ susceptibility (mean rating ± std. dev.: 6.63 ± 1.58 for susceptible condition, and 6.81 ± 1.59 for insusceptible condition; Wilcoxon sign rank test, Z  = −0.9, p  = 0.36).

An important question in human social interaction is how people weigh others’ opinion 51 . Bayesian theories recommend that different opinions should be weighted by their reliability in order for the group to benefit from putting the opinions together 52 . Indeed, some empirical evidence has supported this view 9 , 30 , 32 , 33 while others have shown other decision aggregation strategies in human social interaction 10 , 36 .

We developed an experimental paradigm inspired by previous work on social information aggregation 10 . We quantified how people weighted their peer’s opinion in the context of a visual perceptual task. We tested if this weighting depends on the weight their peers assigned to the participants’ opinion. In experiment 1, participants worked with three different alleged human partners in separate blocks. Our participants were more influenced by the susceptible partner compared to the baseline and insusceptible partners. In experiment 2A, the behavior of a single partner changed dynamically within the same experiment from susceptible to insusceptible or vice versa. We replicated the result of experiment 1 by showing that participants were more influenced in the susceptible condition than in the insusceptible condition. In experiment 2B, we showed that participants did not reciprocate when their peer was a computer even though they took greater influence from the computer’s advice.

When combining opinions in an optimal Bayesian way to maximize accuracy, each source of information should be weighted based on its reliability 29 . Consequently, reciprocity of social influence, i.e., weighting others’ opinion by the weight they give to our opinion is not consistent with Bayesian reliability-based information aggregation. In our experiment, we systematically controlled the accuracy of the participants’ partners such that they were identical across experimental conditions. Participants rated their own performance lower in the insusceptible condition than in the susceptible condition but did not distinguish between the partners’ accuracies. With such judgment of their performance and that of their partners, a hypothetical Bayesian participant would have taken more influence from the insusceptible partner. This is actually what participants did when working with a computer partner. However, when working with a human partner, they followed the opposite strategy and took more influence from the susceptible partner.

Why do people go against an information integration strategy that is more likely to maximize their accuracy? We propose that reciprocity is a pervasive social norm 48 , and abiding by norms is sometimes rewarding in itself and could hence become a goal 53 . As a consequence, individuals may be ready to pay a cost (in terms of reduced accuracy) to adhere to these norms 54 . This explanation is supported by the finding that participants did not reciprocate with a computer partner as participants did not expect the computer to comply with the reciprocity norm.

In the susceptible condition, participants may reciprocate with their reciprocating partners in order to keep their influence over them and avoid the pain of being ignored 47 . As such, showing reciprocity toward a susceptible partner may be driven by a form of loss aversion. Why would people be aversive to losing influence? Recent works have suggested that influence over others may be inherently valuable both behaviorally 55 and neurobiologically 56 . There is now compelling evidence that others’ agreement with our opinion is a strong driver of human brain’s reward network 1 , 57 . In the insusceptible condition, on the other hand, ignoring the insusceptible partner may be motivated by wishing to punish someone who does not comply with the norm of reciprocity.

Following the norm of reciprocity might also improve people’s self-efficacy. It is possible that in the insusceptible condition players perceive the experiment as a status competition. In this view, ignoring the insusceptible partners in response to being ignored by them could serve as a signal from the participant that she/he is not willing to accept an inferior position 58 , 59 . Several studies in behavioral economics (ultimatum game in particular) have shown that one reason why people reject unfair offers is because they want to send the signal that they will not be easily dominated and thereby refuse to accept an inferior social status compared to their peer 60 . Indeed, multiple studies have shown that people do not like to be in an inferior position where their choices are less selected than others’ and they use various strategies to compete with their peers in having more influence 56 , 61 , 62 . However, people do not engage in a status competition with a computer that is consistent with the difference in reciprocity between human and computer experiments (Fig.  3d ). In addition to the above, ignoring the non-reciprocating partner may also serve to protect the participant’s “wounded pride” 63 and maintain their self-esteem. Our finding that participants rate their own performance higher when playing with the susceptible than with the insusceptible partner (Fig.  3e ) is consistent with the hypothesis that having influence over others improves self-efficacy. It should be noted, however, that, there is no one-to-one correspondence between the perception of self-efficacy and reciprocity: while the difference in self-efficacy between the susceptible and insusceptible conditions is the same for experiment 2A and 2B (Supplementary Figure  2 ), the difference in influence is not: in experiment 2A, the influence in the insusceptible condition is less than in the susceptible condition but in experiment 2B, the influence in the susceptible condition is identical to the insusceptible condition (Fig.  3c, d ). Hence, reciprocity cannot be entirely explained by changes in self-efficacy.

Reciprocating influence is consistent with cognitive balance theory 64 , which posits that humans change their preference to be similar to those they like and dissimilar to those they do not. In experiment 1, participants liked the susceptible partner more than the insusceptible one and were more influenced by the partner they liked more. However, in our experiment, the participants were not allowed to change their estimate away from their peers (participants were instructed to make their second choice between their and their partner’s initial estimates). This restriction makes it difficult to directly address the relationship between cognitive balance theory and reciprocity observed in the current study.

In the insusceptible condition, participants’ perceived performance of themselves dropped significantly (Fig.  3e ). Previous studies show that humans are good at tracking their accuracy even in the absence of any external feedback 9 , 65 . It is been argued that people are able to get insight into their accuracy through past experience 66 . However, in social contexts, their judgment could be affected by the environment 67 depending on whether they compete or cooperate with a peer 68 . Our performance rating results confirm the effects of social context on human performance monitoring. This finding shows that being ignored exerts a devastating impact on self-efficacy. One possibility is that being repeatedly ignored in the insusceptible condition may induce a negative emotional impression on the participant that impairs the participant’s self-evaluation. Another possibility is that participants may interpret the partner’s revised estimate as the correct position of the target. By definition, the insusceptible partner’s revised estimates would fall further from those of the participant. The inevitable conclusion for the ignored participants would be that their opinion must have been less precise in the insusceptible blocks. Future studies could investigate each of these potential explanations.

Previously, we showed that when working together in a dyad 36 , people tend to operate by an “equality bias” giving equal weight to their own and their partner’s decision. Participants fulfilled this goal either by adjusting the weight they assign to each other’s opinions 36 or by matching their confidence to the confidence of the other people they worked with 61 . Hence, in both cases, people mutually adapted to each other’s behavior when required to make decisions together. Similarly, participants exhibited mutual adaptation of social influence in the present study.

Our results imply that humans do not only consider others’ reliability to compute the weight that they assign to others’ opinion, but instead they take into account other factors like reciprocity as well. We conclude that reciprocity plays a significant role in human advice-taking and social influence, which violates the optimal account of human information integration. Reciprocity as a social norm helps people to fulfill objectives of social interaction including maintaining a positive self-image.

A total of 80 healthy adult participants (39 females, mean age ± std. dev.: 25 ± 2.9) participated in three experiments after having given written informed consent. Each participant participated in only one of the experiments. Participants were students at the University of Freiburg, Germany. The experimental procedures were approved by the ethics committee of the University of Freiburg. All experiments were performed using Psychophysics Toolbox 69 implemented in MATLAB (Mathworks). The data were analyzed using MATLAB and SPSS.

Experimental task 1

This experiment was designed to investigate whether participants were more influenced by whom they influenced more. Participants first made a perceptual estimate about the location of a target on the screen. Afterwards, they were presented with the estimate from their partner regarding the same stimulus. This was followed by making a second choice about the location of the target or observing a second choice of their partner.

In more detail, the experiment went on as follows: participants ( N  = 20, 9 females, mean age ± std. dev.: 25 ± 2.8) were presented with a sequence of 91 visual stimuli consisting of small circular Gaussian blobs ( r  = 5 mm) in rapid serial visual presentation on the screen (resolution = 2560 × 1440 Dell U2713HM 27″). The first item was presented for 30 ms and every other stimulus was presented for 15 ms each. Participants’ task was to identify the location of the first stimulus. Participants were required to wait until the presentation of all stimuli were finished, and then indicate the location of the target stimulus using the computer mouse (Fig.  1 ). The reported location was marked by a yellow dot. After participants reported their initial estimate, they were required to report their confidence about their estimate on a numerical scale from 1 (low confidence) to 6 (high confidence). Afterwards, participants were shown the choice of their partners about the same stimulus (see below, Constructing partners section for further details) by a small dot on the screen. Then, either the participant revised her estimate or observed the partner revise theirs. After the second estimate was made, all estimates were presented to the participant for 3 s. In this stage, the first choice was shown by a hexagon to be distinguished from the second choice, which was shown by a circle (Fig.  1 ). There was not any time pressure on participants in any stage of the experiment and the experiment did not move to next stage until the participants had registered their responses (Fig.  1 ). Participants were told that their payoff will be calculated based on their first and second estimates. However, everyone was given a fixed amount at the end of the experiment.

During the course of the experiment, participants were exposed to three different partners. Partners varied in their susceptibility to the participants’ estimates (i.e., the amount of influence the participants’ first estimate has on the partner’s second estimate). In the baseline blocks, the participant always made the second estimate and the partner never contributed a second estimate. In the susceptible and insusceptible blocks, the partner and the participants made the second estimate in the odd and even trials, respectively. In the susceptible block, the partner was influenced strongly by the participant and vice versa in the insusceptible block (see below for details in the section “Constructing partners”). In each block, the participants worked with only one partner and each block contained 30 trials. Participants worked with each partner for five blocks. For example, they worked with baseline partner in block 1, then with the susceptible partner in block 2, then with insusceptible partner in block 3, then again with the baseline partner in block 4, and so on. Participants completed 15 blocks in total. The order of the partners was randomized across participants. The three partners were shown by blue, red, and turquoise markers, which were randomly assigned to the different partners at the beginning of each subject’s experiment. After finishing the experiment, the participants were required to estimate their own and the three partners’ performance on a numerical scale from 1 to 10. They were instructed to only consider the first choice to assess their partners’ performance. At the end, we asked them whether they thought they interacted with real people or with a computer algorithm. All participants indicated that they believed they were interacting with real human partners.

Three participants were excluded from the final analyses of this experiment. One participant did not notice she played against three different partners. The other two participants were excluded because they resampled in the second stage, meaning that their second estimates were not between their own and their partner’s initial estimates (contrary to the task instruction). In the Supplementary Figure  3 , we show that our findings remain valid and statistically significant when these three subjects were not excluded from the analysis.

Experimental task 2A

This experiment was designed to test whether reciprocity is a dynamic process. The experiment used the same paradigm as experiment 1 but participants ( N  = 30, 15 females, mean age ± std. dev.: 25 ± 2.8) were told that they do the task with only one partner, which is the same gender as themselves. The partner changed its susceptibility during the course of the experiment, either from susceptible to insusceptible or vice versa. The experiment consisted of 11 blocks in total. Half of the participants were first probed in the susceptible condition, which lasted five blocks and then with a transition block in between, they switched to five insusceptible blocks. The other half completed the opposite order. The average advice/influence that the partner took from our participants is depicted in Supplementary Figure  1 . The transition block was designed in order to avoid a sudden change of the partner’s behavior. During the transition block, the partner’s advice-taking strategy linearly (see below) switched from susceptible to insusceptible or vice versa.

After each session of the experiment, all participants were debriefed to assess to what extent they believed the cover story. We interviewed them with indirect questions about the cover story and all participants stated that they believed they were working with other human participants in neighboring experimental rooms.

Experimental task 2B

This experiment differed from experiment 2A in one respect: participants ( N  = 30, 15 females, mean age ± std. dev.: 24 ± 3.1) were told that their partner in the experiment is a computer. Any other aspects of the experiment were identical to experiment 2A and they received exactly the same task instructions as in experiment 2A except that the human partner was replaced by a computer.

Performance rating

In experiment 1, participants rated their own and the three different partners’ performance once at the end of the experiment. Note that a different color identified each partner during the experiment. In experiment 2A and 2B, participants rated their own and their partner’s performance at the end of each block. This way, we obtained a pair of performance ratings for the self and the susceptible partner and another pair for the self and the insusceptible partner.

Constructing partners

The error distribution of all partners’ first choices was modeled from participants’ actual estimation errors during a pilot experiment. Ten participants performed an experiment identical to experiment 1 of the current study. We aggregated errors of all participants ( N  = 10) and fitted the concentration parameter kappa of a von Mises distribution centered to the target, yielding the value kappa = 7.4. Then, in each trial, we drew the first choice of the partner from this distribution. We speculated that participants’ assessment of their partners’ performance may be strongly influenced by the few trials with high confidence (confidence level of 5 or 6). To avoid this potential problem, the partner’s first choice was not taken from the von Mises distribution in high confidence trials but randomly drawn from a uniform distribution centered on the participants’ choice with a width of ±20°.

The second choice of the partner was computed differently for susceptible and insusceptible partners. For experiment 1, the influence that the insusceptible partner took from the participants in each trial was chosen with a probability of 0.65 from a uniform distribution on the interval [0, 0.2], with a probability of 0.2 randomly from a uniform distribution on the interval [0.3, 0.7], and with a probability of 0.15 randomly from a uniform distribution on the interval [0.7, 0.9]. For experiment 2, the influence that the insusceptible partner took from the participants was chosen randomly from a uniform distribution on the interval [0, 0.2]. For the susceptible partner, in all experiments, the influence was chosen with a probability of 0.5 randomly from a uniform distribution on the interval [0.7, 1], with a probability of 0.2 randomly from a uniform distribution on the interval [0.3, 0.7], and with a probability of 0.3 randomly from a uniform distribution on the interval [0, 0.3]. In the transition block, the influence of the partner was a linear interpolation between the susceptible and the insusceptible partner:

where inf s and inf ins were the influences of the susceptible and insusceptible partners, respectively (as explained above). λ gradually increased with time from 0 at the beginning to 1 at the end of the transition block for a transition from the susceptible to the insusceptible condition. For the transition from the insusceptible to the susceptible condition, λ decreased gradually from 1 to 0.

In experiment 1, on average, the advice that the partners took from the participants was 0.3 and 0.55 in the insusceptible and susceptible conditions, respectively. In experiment 2A, on average, the advice that the partner took from the participants was 0.07 and 0.5 in the insusceptible and susceptible conditions, respectively. The second choice of the partners in experiment 2B was designed exactly the same as in experiment 2A and the average advice that the partner took in the insusceptible and the susceptible condition was identical to experiment 2A.

Computation of error bars

All error bars in the figures depict standard errors. Standard errors were computed across subjects; in situations where we had multiple measurements from each participant, we first computed the mean value for each participant, and then computed the standard error across the participants’ mean.

Data availability

The behavioral data that support the findings of this study and the code that was used to generate the findings and to conduct the experiments of this study will be provided to all readers upon request.

Campbell-Meiklejohn, D. K., Bach, D. R., Roepstorff, A., Dolan, R. J. & Frith, C. D. How the opinion of others affects our valuation of objects. Curr. Biol. 20 , 1165–1170 (2010).

Article   PubMed   PubMed Central   CAS   Google Scholar  

Berns, G. S., Capra, C. M., Moore, S. & Noussair, C. Neural mechanisms of the influence of popularity on adolescent ratings of music. Neuroimage 49 , 2687–2696 (2010).

Article   PubMed   Google Scholar  

Zaki, J., Schirmer, J. & Mitchell, J. P. Social influence modulates the neural computation of value. Psychol. Sci. 22 , 894–900 (2011).

Klucharev, V., Hytönen, K., Rijpkema, M., Smidts, A. & Fernández, G. Reinforcement learning signal predicts social conformity. Neuron 61 , 140–151 (2009).

Article   PubMed   CAS   Google Scholar  

Garvert, M. M., Moutoussis, M., Kurth-Nelson, Z., Behrens, T. E. & Dolan, R. J. Learning-induced plasticity in medial prefrontal cortex predicts preference malleability. Neuron 85 , 418–428 (2015).

Muchnik, L., Aral, S. & Taylor, S. J. Social influence bias: a randomized experiment. Science 341 , 647–651 (2013).

Article   ADS   PubMed   CAS   Google Scholar  

Suzuki, S., Jensen, E. L., Bossaerts, P. & O’Doherty, J. P. Behavioral contagion during learning about another agent’s risk-preferences acts on the neural representation of decision-risk. Proc. Natl. Acad. Sci. USA 113 , 3755–3760 (2016).

Article   ADS   PubMed   PubMed Central   CAS   Google Scholar  

Asch, S. E. & Guetzkow, H. in Groups, Leadership and Men; Research in Human Relations 222–236 (Carnegie Press, Oxford, 1951).

Bahrami, B. et al. Optimally interacting minds. Science 329 , 1081–1085 (2010).

Toelch, U., Bach, D. R. & Dolan, R. J. The neural underpinnings of an optimal exploitation of social information under uncertainty. Soc. Cogn. Affect. Neurosci. 9 , 1746–1753 (2013).

Article   PubMed   PubMed Central   Google Scholar  

Edelson, M., Sharot, T., Dolan, R. J. & Dudai, Y. Following the crowd: brain substrates of long-term memory conformity. Science 333 , 108–111 (2011).

Yaniv, I. Receiving other people’s advice: influence and benefit. Organ. Behav. Hum. Decis. Process. 93 , 1–13 (2004).

Article   Google Scholar  

Farrell, S. Social influence benefits the wisdom of individuals in the crowd. Proc. Natl. Acad. Sci. USA 108 , E625–E625 (2011).

Article   ADS   PubMed   PubMed Central   Google Scholar  

Harvey, N. & Fischer, I. Taking advice: accepting help, improving judgment, and sharing responsibility. Organ. Behav. Hum. Decis. Process. 70 , 117–133 (1997).

Kennedy, J., Kleinmuntz, D. N. & Peecher, M. E. Determinants of the justifiability of performance in ill-structured audit tasks. J. Account. Res. 35 , 105–123 (1997).

Hartley, P. & Trout, R. in Marketing Dynamism & Sustainability: Things Change, Things Stay the Same 71–74 (Springer, New Orleans, 2015).

Raafat, R. M., Chater, N. & Frith, C. Herding in humans. Trends Cogn. Sci. 13 , 420–428 (2009).

Madirolas, G. & de Polavieja, G. G. Improving collective estimations using resistance to social influence. PLoS Comput. Biol. 11 , e1004594 (2015).

Lorenz, J., Rauhut, H., Schweitzer, F. & Helbing, D. How social influence can undermine the wisdom of crowd effect. Proc. Natl. Acad. Sci. USA 108 , 9020–9025 (2011).

Chari, V. V. & Kehoe, P. J. Financial crises as herds: overturning the critiques. J. Econ. Theory 119 , 128–150 (2004).

Article   MATH   Google Scholar  

Salganik, M. J., Dodds, P. S. & Watts, D. J. Experimental study of inequality and unpredictability in an artificial cultural market. Science 311 , 854–856 (2006).

Festinger, L., Riecken, H. & Schachter, S. When Prophecy Fails: A Social and Psychological Study of a Modern Group that Predicted the Destruction of the World (Lulu Press, Minneapolis, 2017).

Fehr, E. & Gächter, S. Fairness and retaliation: the economics of reciprocity. J. Econ. Perspect. 14 , 159–181 (2000).

Sanfey, A. G. Social decision-making: insights from game theory and neuroscience. Science 318 , 598–602 (2007).

Tidd, K. L. & Lockard, J. S. Monetary significance of the affiliative smile: a case for reciprocal altruism. Bull. Psychon. Soc. 11 , 344–346 (1978).

King-Casas, B. et al. Getting to know you: reputation and trust in a two-person economic exchange. Science 308 , 78–83 (2005).

Fehr, E. & Gächter, S. Altruistic punishment in humans. Nature 415 , 137–140 (2002).

Frith, C. D. & Frith, U. Mechanisms of social cognition. Annu. Rev. Psychol. 63 , 287–313 (2012).

Toelch, U. & Dolan, R. J. Informational and normative influences in conformity from a neurocomputational perspective. Trends Cogn. Sci. 19 , 579–589 (2015).

Behrens, T. E., Hunt, L. T., Woolrich, M. W. & Rushworth, M. F. Associative learning of social value. Nature 456 , 245–249 (2008).

Boorman, E. D., O’Doherty, J. P., Adolphs, R. & Rangel, A. The behavioral and neural mechanisms underlying the tracking of expertise. Neuron 80 , 1558–1571 (2013).

De Martino, B., Bobadilla-Suarez, S., Nouguchi, T., Sharot, T. & Love, B. C. Social information is integrated into value and confidence judgments according to its reliability. J. Neurosci. 37 , 6066–6074 (2017).

Park, S. A., Goïame, S., O’Connor, D. A. & Dreher, J.-C. Integration of individual and social information for decision-making in groups of different sizes. PLoS Biol. 15 , e2001958 (2017).

Myers, D. in Social Psychology 11th edn, 586–610 (McGraw Hill, New York, 2013).

Soll, J. B. & Larrick, R. P. Strategies for revising judgment: how (and how well) people use others’ opinions. J. Exp. Psychol. Learn. Mem. Cogn. 35 , 780 (2009).

Mahmoodi, A. et al. Equality bias impairs collective decision-making across cultures. Proc. Natl. Acad. Sci. USA 112 , 3835–3840 (2015).

ADS   PubMed   PubMed Central   CAS   Google Scholar  

Hertz, U., Romand-Monnier, M., Kyriakopoulou, K. & Bahrami, B. Social influence protects collective decision making from equality bias. J. Exp. Psychol. Hum. Percept. Perform. 42 , 164 (2016).

Heyes, C. What’s social about social learning? J. Comp. Psychol. 126 , 193 (2012).

Yaniv, I. & Kleinberger, E. Advice taking in decision making: egocentric discounting and reputation formation. Organ. Behav. Hum. Decis. Process. 83 , 260–281 (2000).

Deutsch, M. & Gerard, H. B. A study of normative and informational social influences upon individual judgment. J. Abnorm. Soc. Psychol. 51 , 629 (1955).

Article   CAS   Google Scholar  

Cialdini, R. B. & Goldstein, N. J. Social influence: compliance and conformity. Annu. Rev. Psychol. 55 , 591–621 (2004).

Mackie, D. M. Systematic and nonsystematic processing of majority and minority persuasive communications. J. Pers. Soc. Psychol. 53 , 41 (1987).

Cialdini, R. B. Influence: Science and practice (Allyn and Bacon, Boston, 2001).

Arndt, J., Schimel, J., Greenberg, J. & Pyszczynski, T. The intrinsic self and defensiveness: evidence that activating the intrinsic self reduces self-handicapping and conformity. Pers. Soc. Psychol. Bull. 28 , 671–683 (2002).

Tafarodi, R. W., Kang, S.-J. & Milne, A. B. When different becomes similar: compensatory conformity in bicultural visible minorities. Pers. Soc. Psychol. Bull. 28 , 1131–1142 (2002).

Williams, K. D., Cheung, C. K. & Choi, W. Cyberostracism: effects of being ignored over the Internet. J. Pers. Soc. Psychol. 79 , 748 (2000).

Eisenberger, N. I., Lieberman, M. D. & Williams, K. D. Does rejection hurt? An fMRI study of social exclusion. Science 302 , 290–292 (2003).

Gouldner, A. W. The norm of reciprocity: a preliminary statement. Am. Sociol. Rev . 25 , 161–178 (1960).

Fehr, E. & Fischbacher, U. Third-party punishment and social norms. Evol. Hum. Behav. 25 , 63–87 (2004).

Van den Berg, R. et al. A common mechanism underlies changes of mind about decisions and confidence. Elife 5 , e12192 (2016).

Bates, J. M. & Granger, C. W. The combination of forecasts. Ops Res. Q. 20 , 451–468 (1969).

Nitzan, S. & Paroush, J. Optimal decision rules in uncertain dichotomous choice situations. Int. Econ. Rev . 23 , 289–297 (1982).

Gintis, H. The hitchhiker’s guide to altruism: gene-culture coevolution, and the internalization of norms. J. Theor. Biol. 220 , 407–418 (2003).

Article   MathSciNet   PubMed   Google Scholar  

Atran, S. & Ginges, J. Religious and sacred imperatives in human conflict. Science 336 , 855–857 (2012).

Hertz, U. & Bahrami, B. Intrinsic value of social influence over others. Preprint at PsyArXiv doi:10.17605/OSF.IO/6JM7T (2018).

Hertz, U. et al. Neural computations underpinning the strategic management of influence in advice giving. Nat. Commun. 8 , 2191 (2017).

Izuma, K. & Adolphs, R. Social manipulation of preference in the human brain. Neuron 78 , 563–573 (2013).

Yamagishi, T. et al. Rejection of unfair offers in the ultimatum game is no evidence of strong reciprocity. Proc. Natl. Acad. Sci. USA 109 , 20364–20368 (2012).

Xiao, E. & Houser, D. Emotion expression in human punishment behavior. Proc. Natl. Acad. Sci. USA 102 , 7398–7401 (2005).

Yamagishi, T. et al. The private rejection of unfair offers and emotional commitment. Proc. Natl. Acad. Sci. USA 106 , 11520–11523 (2009).

Bang, D. et al. Confidence matching in group decision-making. Nat. Hum. Behav. 1 , 0117 (2017).

Mahmoodi, A., Bang, D., Ahmadabadi, M. N. & Bahrami, B. Learning to make collective decisions: the impact of confidence escalation. PLoS ONE 8 , e81195 (2013).

Straub, P. G. & Murnighan, J. K. An experimental investigation of ultimatum games: information, fairness, expectations, and lowest acceptable offers. J. Econ. Behav. Organ. 27 , 345–364 (1995).

Heider, F. The Psychology of Interpersonal Relations (Psychology Press, New York, 2013).

Fleming, S. M., Weil, R. S., Nagy, Z., Dolan, R. J. & Rees, G. Relating introspective accuracy to individual differences in brain structure. Science 329 , 1541–1543 (2010).

Bandura, A. Self-efficacy: toward a unifying theory of behavioral change. Psychol. Rev. 84 , 191 (1977).

Allport, F. H. The group fallacy in relation to social science. Am. J. Sociol. 29 , 688–706 (1924).

Wittmann, M. K. et al. Self-other mergence in the frontal cortex during cooperation and competition. Neuron 91 , 482–493 (2016).

Brainard, D. H. The psychophysics toolbox. Spat. Vis. 10 , 433–436 (1997).

Download references

Acknowledgements

This work was supported by a PhD scholarship (A.M.) from the Graduate School Scholarship Program of the German Academic Exchange Service (DAAD), a European Research Council Starting Grant “NeuroCoDec #309865” (B.B.), the German Research Foundation (DFG, grant no INST 39/1014-1 FUGG) (C.M.) and the “Struktur -und Innovationsfonds Baden-Württemberg (SI-BW)” of the state of Baden-Württemberg (C.M.). We thank Ulf Toelch for helping to implement the experimental task and Helena Gavrilova, Tobias Pistohl, and Luke Bashford for helping to collect data.

Author information

These authors contributed equally: Bahador Bahrami, Carsten Mehring.

Authors and Affiliations

Bernstein Centre Freiburg, University of Freiburg, Hansastrasse 9a, 79104, Freiburg, Germany

Ali Mahmoodi & Carsten Mehring

Faculty of Biology, University of Freiburg, Schänzlestraße 1, 79104, Freiburg, Germany

Institute of Cognitive Neuroscience, University College London, 17 Queen Square London, London, WC1N 3AR, UK

Bahador Bahrami

Faculty of Psychology and Educational Sciences, Ludwig Maximilian University, Leopoldstrasse 13, 80802, Munich, Germany

You can also search for this author in PubMed   Google Scholar

Contributions

A.M., B.B., and C.M. designed the experiments. A.M. collected the data. A.M. carried out the data analysis. A.M., B.B., and C.M. interpreted the results and wrote the manuscript.

Corresponding author

Correspondence to Ali Mahmoodi .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Supplementary information, peer review file, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Mahmoodi, A., Bahrami, B. & Mehring, C. Reciprocity of social influence. Nat Commun 9 , 2474 (2018). https://doi.org/10.1038/s41467-018-04925-y

Download citation

Received : 05 February 2018

Accepted : 05 June 2018

Published : 26 June 2018

DOI : https://doi.org/10.1038/s41467-018-04925-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Disentangling the contribution of individual and social learning processes in human advice-taking behavior.

• Maayan Pereg
• Nitzan Shahar

npj Science of Learning (2024)

Brain Systems Underlying Fundamental Motivations of Human Social Conformity

• Xinling Chen
• Chunliang Feng

Neuroscience Bulletin (2023)

Social Influence Under Uncertainty in Interaction with Peers, Robots and Computers

• Joshua Zonca
• Alessandra Sciutti

International Journal of Social Robotics (2023)

Making brand activism successful: How advice-giving can boost support behavior and reap benefits for the brand

• Carina Thürridl
• Frauke Mattison Thompson

Marketing Letters (2023)

Information aggregation and collective intelligence beyond the wisdom of crowds

• Tatsuya Kameda
• Wataru Toyokawa
• R. Scott Tindale

Nature Reviews Psychology (2022)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

Chapter 4: Attraction/Relationships

4 – Attraction and Relationships

Introduction.

Think about your social world….friends, family, classmates, and intimate relationships in your life. How have these relationships shaped you into the person you are today? How do social relationships and social support influence us? What about dating—how do we tend to choose a mate? Do opposites indeed attract or are we more attracted to those who are similar to ourselves? In this section, we will review the importance of social relationships and social support and examine the various factors that influence social relationships and attraction. After completing the readings in this section, you will be able to do the following:

Learning Objectives

• Understand the importance of friendships and social support on our overall health and well being
• Understand and describe various factors that contribute to attraction
• Examine the impact that our physical state may have on our feelings of social connectivity
• Understand what attracts us to others.
• Review research that suggests that friendships are important for our health and well-being.
• Examine the influence of the Internet on friendship and developing relationships.
• Understand what happens to our brains when we are in love.
• Consider the complexity of love.
• Examine the construct and components of social support.

Friendship and love, and more broadly, the relationships that people cultivate in their lives, are some of the most valuable treasures a person can own. This module explores ways in which we try to understand how friendships form, what attracts one person to another, and how love develops. It also explores how the Internet influences how we meet people and develop deep relationships. Finally, this module will examine social support and how this can help many through the hardest times and help make the best times even better.

The importance of relationships has been examined by researchers for decades. Many researchers point to sociologist Émile Durkheim’s classic study of suicide and social ties ( 1951 ) as a starting point for this work. Durkheim argued that being socially connected is imperative to achieving personal well-being. In fact, he argued that a person who has no close relationships is likely a person who is at risk for suicide. It is those relationships that give a person meaning in their life. In other words, suicide tends to be higher among those who become disconnected from society. What is interesting about that notion is when people are asked to describe the basic necessities for life—people will most often say food, water, and shelter, but seldom do people list “close relationships” in the top three. Yet time and time again, research has demonstrated that we are social creatures and we need others to survive and thrive. Another way of thinking about it is that close relationships are the psychological equivalent of food and water; in other words, these relationships are necessary for survival. Baumeister and Leary ( 1995 ) maintain that humans have basic needs and one of them is the need to belong; these needs are what makes us human and give a sense of purpose and identity to our lives ( Brissette, Cohen, & Seeman, 2000 ; Ryff, 1989 ).

Given that close relationships are so vital to well-being, it is important to ask how interpersonal relationships begin. What makes us like or love one person but not another? Why is it that when bad things happen, we frequently want to talk to our friends or family about the situation? Though these are difficult questions to answer because relationships are complicated and unique, this module will examine how relationships begin; the impact of technology on relationships; and why coworkers, acquaintances, friends, family, and intimate partners are so important in our lives.

Attraction: The Start of Friendship and Love

Why do some people hit it off immediately? Or decide that the friend of a friend was not likable? Using scientific methods, psychologists have investigated factors influencing attraction and have identified a number of variables, such as similarity, proximity (physical or functional), familiarity, and reciprocity, that influence with whom we develop relationships.

Often we “stumble upon” friends or romantic partners; this happens partly due to how close in proximity we are to those people. Specifically, proximity or physical nearness has been found to be a significant factor in the development of relationships. For example, when college students go away to a new school, they will make friends consisting of classmates, roommates, and teammates (i.e., people close in proximity). Proximity allows people the opportunity to get to know one other and discover their similarities—all of which can result in a friendship or intimate relationship. Proximity is not just about geographic distance, but rather functional distance , or the frequency with which we cross paths with others. For example, college students are more likely to become closer and develop relationships with people on their dorm-room floors because they see them (i.e., cross paths) more often than they see people on a different floor. How does the notion of proximity apply in terms of online relationships? Deb Levine ( 2000 ) argues that in terms of developing online relationships and attraction, functional distance refers to being at the same place at the same time in a virtual world (i.e., a chat room or Internet forum)—crossing virtual paths.

Familiarity

One of the reasons why proximity matters to attraction is that it breeds familiarity ; people are more attracted to that which is familiar. Just being around someone or being repeatedly exposed to them increases the likelihood that we will be attracted to them. We also tend to feel safe with familiar people, as it is likely we know what to expect from them. Dr. Robert Zajonc ( 1968 ) labeled this phenomenon the mere-exposure effect . More specifically, he argued that the more often we are exposed to a stimulus (e.g., sound, person) the more likely we are to view that stimulus positively. Moreland and Beach ( 1992 ) demonstrated this by exposing a college class to four women (similar in appearance and age) who attended different numbers of classes, revealing that the more classes a woman attended, the more familiar, similar, and attractive she was considered by the other students.

There is a certain comfort in knowing what to expect from others; consequently research suggests that we like what is familiar. While this is often on a subconscious level, research has found this to be one of the most basic principles of attraction ( Zajonc, 1980 ). For example, a young man growing up with an overbearing mother may be attracted to other overbearing women not because he likes being dominated but rather because it is what he considers normal (i.e., familiar).

When you hear about couples such as Sandra Bullock and Jesse James, or Kim Kardashian and Kanye West, do you shake your head thinking “this won’t last”? It is probably because they seem so different. While many make the argument that opposites attract, research has found that is generally not true; s imilarity is key. Sure, there are times when couples can appear fairly different, but overall we like others who are like us. Ingram and Morris ( 2007 ) examined this phenomenon by inviting business executives to a cocktail mixer, 95% of whom reported that they wanted to meet new people. Using electronic name tag tracking, researchers revealed that the executives did not mingle or meet new people; instead, they only spoke with those they already knew well (i.e., people who were similar).

When it comes to marriage, research has found that couples tend to be very similar, particularly when it comes to age, social class, race, education, physical attractiveness, values, and attitudes ( McCann Hamilton, 2007 ; Taylor, Fiore, Mendelsohn, & Cheshire, 2011 ). This phenomenon is known as the matching hypothesis ( Feingold, 1988 ; Mckillip & Redel, 1983 ). We like others who validate our points of view and who are similar in thoughts, desires, and attitudes.

Reciprocity

Another key component in attraction is reciprocity ; this principle is based on the notion that we are more likely to like someone if they feel the same way toward us. In other words, it is hard to be friends with someone who is not friendly in return. Another way to think of it is that relationships are built on give and take; if one side is not reciprocating, then the relationship is doomed. Basically, we feel obliged to give what we get and to maintain equity in relationships. Researchers have found that this is true across cultures ( Gouldner, 1960 ).

“In poverty and other misfortunes of life, true friends are a sure refuge. They keep the young out of mischief; they comfort and aid the old in their weakness, and they incite those in the prime of life to noble deeds.”— Aristotle

Research has found that close friendships can protect our mental and physical health when times get tough. For example, Adams, Santo, and Bukowski ( 2011 ) asked fifth- and sixth-graders to record their experiences and self-worth, and to provide saliva samples for 4 days. Children whose best friend was present during or shortly after a negative experience had significantly lower levels of the stress hormone cortisol in their saliva compared to those who did not have a best friend present. Having a best friend also seemed to protect their feelings of self-worth. Children who did not identify a best friend or did not have an available best friend during distress experienced a drop in self-esteem over the course of the study.

Workplace friendships

Friendships often take root in the workplace, due to the fact that people are spending as much, or more, time at work than they are with their family and friends ( Kaufman & Hotchkiss, 2003 ). Often, it is through these relationships that people receive mentoring and obtain social support and resources, but they can also experience conflicts and the potential for misinterpretation when sexual attraction is an issue. Indeed, Elsesser and Peplau ( 2006 ) found that many workers reported that friendships grew out of collaborative work projects, and these friendships made their days more pleasant.

In addition to those benefits, Riordan and Griffeth ( 1995 ) found that people who worked in an environment where friendships could develop and be maintained were more likely to report higher levels of job satisfaction, job involvement, and organizational commitment, and they were less likely to leave that job. Similarly, a Gallup poll revealed that employees who had “close friends” at work were almost 50% more satisfied with their jobs than those who did not ( Armour, 2007 ).

Internet friendships

What influence does the Internet have on friendships? It is not surprising that people use the Internet with the goal of meeting and making new friends ( Fehr, 2008 ; McKenna, 2008 ). Researchers have wondered if the issue of not being face-to-face reduces the authenticity of relationships, or if the Internet really allows people to develop deep, meaningful connections. Interestingly, research has demonstrated that virtual relationships are often as intimate as in-person relationships; in fact, Bargh and colleagues found that online relationships are sometimes more intimate ( Bargh et al., 2002 ). This can be especially true for those individuals who are more socially anxious and lonely—such individuals who are more likely to turn to the Internet to find new and meaningful relationships ( McKenna, Green, & Gleason, 2002 ). McKenna et al. ( 2002 ) suggest that for people who have a hard time meeting and maintaining relationships, due to shyness, anxiety, or lack of face-to-face social skills, the Internet provides a safe, nonthreatening place to develop and maintain relationships. Similarly, Penny Benford ( 2008 ) found that for high-functioning autistic individuals, the Internet facilitated communication and relationship development with others, which would have been more difficult in face-to-face contexts, leading to the conclusion that Internet communication could be empowering for those who feel frustrated when communicating face to face.

Is all love the same? Are there different types of love? Examining these questions more closely, Robert Sternberg’s ( 2004 ; 2007 ) work has focused on the notion that all types of love are comprised of three distinct areas: intimacy, passion, and commitment. Intimacy includes caring, closeness, and emotional support. The passion component of love is comprised of physiological and emotional arousal; these can include physical attraction, emotional responses that promote physiological changes, and sexual arousal. Lastly, commitment refers to the cognitive process and decision to commit to love another person and the willingness to work to keep that love over the course of your life. The elements involved in intimacy (caring, closeness, and emotional support) are generally found in all types of close relationships—for example, a mother’s love for a child or the love that friends share. Interestingly, this is not true for passion. Passion is unique to romantic love, differentiating friends from lovers. In sum, depending on the type of love and the stage of the relationship (i.e., newly in love), different combinations of these elements are present.

Taking this theory a step further, anthropologist Helen Fisher explained that she scanned the brains (using fMRI) of people who had just fallen in love and observed that their brain chemistry was “going crazy,” similar to the brain of an addict on a drug high ( Cohen, 2007 ). Specifically, serotonin production increased by as much as 40% in newly in-love individuals. Further, those newly in love tended to show obsessive-compulsive tendencies. Conversely, when a person experiences a breakup, the brain processes it in a similar way to quitting a heroin habit ( Fisher, Brown, Aron, Strong, & Mashek, 2009 ). Thus, those who believe that breakups are physically painful are correct! Another interesting point is that long-term love and sexual desire activate different areas of the brain. More specifically, sexual needs activate the part of the brain that is particularly sensitive to innately pleasurable things such as food, sex, and drugs (i.e., the striatum—a rather simplistic reward system), whereas love requires conditioning—it is more like a habit. When sexual needs are rewarded consistently, then love can develop. In other words, love grows out of positive rewards, expectancies, and habit ( Cacioppo, Bianchi-Demicheli, Hatfield & Rapson, 2012 ).

Love and the Internet

The ways people are finding love has changed with the advent of the Internet. In a poll, 49% of all American adults reported that either themselves or someone they knew had dated a person they met online ( Madden & Lenhart, 2006 ). As Finkel and colleagues ( 2007 ) found, social networking sites, and the Internet generally, perform three important tasks. Specifically, sites provide individuals with access to a database of other individuals who are interested in meeting someone. Dating sites generally reduce issues of proximity, as individuals do not have to be close in proximity to meet. Also, they provide a medium in which individuals can communicate with others. Finally, some Internet dating websites advertise special matching strategies, based on factors such as personality, hobbies, and interests, to identify the “perfect match” for people looking for love online. In general, scientific questions about the effectiveness of Internet matching or online dating compared to face-to-face dating remain to be answered.

It is important to note that social networking sites have opened the doors for many to meet people that they might not have ever had the opportunity to meet; unfortunately, it now appears that the social networking sites can be forums for unsuspecting people to be duped. In 2010 a documentary, Catfish , focused on the personal experience of a man who met a woman online and carried on an emotional relationship with this person for months. As he later came to discover, though, the person he thought he was talking and writing with did not exist. As Dr. Aaron Ben-Zeév stated, online relationships leave room for deception; thus, people have to be cautious.

Social Support

When bad things happen, it is important for people to know that others care about them and can help them out. Unsurprisingly, research has found that this is a common thread across cultures ( Markus & Kitayma, 1991 ; Triandis, 1995 ) and over time ( Reis, Sheldon, Gable, Roscoe, & Ryan, 2000 ); in other words, social support is the active ingredient that makes our relationships particularly beneficial. But what is social support? One way of thinking about social support is that it consists of three discrete conceptual components.

Perceived Social Support

Have you ever thought that when things go wrong, you know you have friends/family members that are there to help you? This is what psychologists call perceived social support or “a psychological sense of support” ( Gottlieb, 1985 ). How powerful is this belief that others will be available in times of need? To examine this question, Dr. Arnberg and colleagues asked 4,600 survivors of the tragic 2004 Indian Ocean (or Boxing Day) Tsunami about their perception of social support provided by friends and family after the event. Those who experienced the most amount of stress found the most benefit from just knowing others were available if they needed anything (i.e., perceived support). In other words, the magnitude of the benefits depended on the extent of the stress, but the bottom line was that for these survivors, knowing that they had people around to support them if they needed it helped them all to some degree.

Perceived support has also been linked to well-being. Brannan and colleagues ( 2012 ) found that perceived support predicted each component of well-being (high positive affect, low negative affect, high satisfaction with life) among college students in Iran, Jordan, and the United States. Similarly, Cohen and McKay ( 1984 ) found that a high level of perceived support can serve as a buffer against stress. Interestingly enough, Dr. Cohen found that those with higher levels of social support were less likely to catch the common cold. The research is clear—perceived social support increases happiness and well-being and makes our live better in general ( Diener & Seligman, 2002 ; Emmons & Colby, 1995 ).

Received Social Support

Received support is the actual receipt of support or helping behaviors from others ( Cohen & Wills, 1985 ). Interestingly, unlike perceived support, the benefits of received support have been beset with mixed findings ( Stroebe & Stroebe, 1996 ). Similar to perceived support, receiving support can buffer people from stress and positively influence some individuals—however, others might not want support or think they need it. For example, dating advice from a friend may be considered more helpful than such advice from your mom! Interestingly, research has indicated that regardless of the support-provider’s intentions, the support may not be considered as helpful to the person receiving the support if it is unwanted ( Dunkel-Schetter, Blasband, Feinstein, & Herbert, 1992 ; Cutrona, 1986 ). Indeed, mentor support was viewed negatively by novice ESOL teachers (those teaching English as a second language in other countries; Brannan & Bleistein, 2012 ). Yet received support from family was perceived as very positive—the teachers said that their family members cared enough to ask about their jobs and told them how proud they were. Conversely, received mentor support did not meet teachers’ needs, instead making them feel afraid and embarrassed to receive mentor support.

Quality or Quantity?

With so many mixed findings, psychologists have asked whether it is the quality of social support that matters or the quantity (e.g., more people in my support network ). Interestingly, research by Friedman and Martin ( 2011 ) examining 1,500 Californians over 8 decades found that while quality does matter, individuals with larger social networks lived significantly longer than those with smaller networks. This research suggests we should count the number of our friends / family members—the more, the better, right? Not necessarily: Dunbar ( 1992 ; 1993 ) argued that we have a cognitive limit with regard to how many people with whom we can maintain social relationships. The general consensus is about 150—we can only “really” know (maintain contact and relate to) about 150 people. Finally, research shows that diversity also matters in terms of one’s network, such that individuals with more diverse social networks (i.e., different types of relationships including friends, parents, neighbors, and classmates) were less likely to get the common cold compared to those with fewer and less diverse networks ( Cohen, Doyle, Turner, Alper, & Skoner, 2003 ). In sum, it is important to have quality relationships as well as quantity—and as the Beatles said, “all you need is love—love is all you need.”

Test Your Knowledge

When choosing someone to date, do you find yourself drawn to others with similar opinions and views or do you tend to be drawn to those opposite of yourself? What factors lead to strong relationships? Is similarity important? Or do our differences complement one another? In the following articles, Johnson (2018) refutes the old adage that opposites attract, while Sommers (2011) addresses the idea of the matching hypothesis with regards to dating and relationships. Read the article “Opposites do not attract”

Read the article “The science of small talk”

Does our physical state influence our social relationships? Inagaki & Eisenberger (2013) examined this premise, investigating if physical warmth could impact our feelings of social connectivity. What are some ways that we can use this data to intervene with clients who may be experiencing depression or loneliness?

Read the article “The heartwarming nature of social bonds”

In the following article, Bazzini et. al (2010) review the “beauty is good” stereotype and how this stereotype may be reflected in Disney movies. As you review the article, think about the implications this stereotype may have on child development. Are there other examples of this stereotype being portrayed in other children’s shows? What about on social media sites or video platforms?

Read the article “Do Animated Disney Characters Portray and Promote the Beauty–Goodness Stereotype?”

Attraction and Relationships Resources

Brannan, D. & Mohr, C. D. (2020). Love, friendship, and social support. In R. Biswas-Diener & E. Diener (Eds), Noba textbook series: Psychology. Champaign, IL: DEF publishers. Retrieved from Love, Friendship, and Social Support

Outside Resources

• Adams, R. E., Santo, J., & Bukowski, W. M. (2011). The presence of a best friend buffers the effects of negative experiences. Developmental Psychology, 47(6) , 1786–1791. doi:10.1037/a0025401
• Aristotle. (n.d.). In poverty and other misfortunes of life…. BrainyQuote.com. Retrieved July 25, 2013, from Aristotle – In poverty and other misfortunes of life, true…
• Armour, S. (2007, August 2). Friendships and work: A good or bad partnership? USA Today . Retrieved from Friendship and work: A good or bad partnership? – USATODAY.com
• Bargh, J. A., McKenna, K. Y. A, & Fitsimons, G. G. (2002). Can you see the real me? Activation and expression of the true self on the Internet. Journal of Social Issues, 58 , 33–48.
• Baumeister, R. & Leary, M. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin , 117(3), 497–529.
• Benford, P. (2008). The use of Internet-based communication by people with autism (Doctoral dissertation, University of Nottingham).
• Brannan, D., & Bleisten, T. (2012). Novice ESOL teachers’ perceptions of social support and self-efficacy. TESOL Quarterly, 46 , 519–541.
• Brannan, D., Biswas-Diener, R., Mohr, C. D., Mortazavi, S., & Stein, N. (2012). Friends and family, a cross-cultural investigation of social support and subjective well-being. Journal of Positive Psychology, 8(1), 65–75.
• Brissette, I., Cohen, S., & Seeman, T. E. (2000). Measuring social integration and social networks. In S. Cohen, L. Underwood, & B. Gottlieb (Eds.), Measuring and intervening in social support, (pp. 53–85), New York, NY: Oxford University Press.
• Cacioppo, S., Bianchi-Demicheli, F., Hatfield, E., & Rapson, R. L. (2012). Social neuroscience of love. Clinical Neuropsychiatry, 9(1) , 3–13.
• Cohen, E. (2007, February 15). Loving with all your … brain. CNN.com . Retrieved July 25th, 2013, from Loving with all your… brain .
• Cohen, S., & McKay, G. (1984). Social support, stress, and the buffering hypothesis: A theoretical analysis. In A. Baum, J. E. Singer, & S. E. Taylor (Eds.), Handbook of psychology and health (pp. 253–267), Volume IV. Hillsdale, NJ: Erlbaum.
• Cohen, S., & Wills, T. A. (1985). Stress, social support and the buffering hypothesis. Psychological Bulletin, 98 , 310–357.
• Cohen, S., Doyle, W. J., Turner, R. B., Alper, C. M., & Skoner, D. P. (2003). Sociability and susceptibility to the common cold. Psychological Science, 14 , 389–395.
• Cutrona, C. (1986). Behavioral manifestations of social support: A microanalytic investigation. Journal of Personality and Social Psychology, 51(1) , 201–208.
• Diener, E. & Seligman, M. E. P. (2002). Very happy people. Psychological Science, 13 , 81–84.
• Dunbar, R. I. M. (1993). Coevolution of neocortical size, group size and language in humans. Behavioral and Brain Sciences, 16 , 681–735.
• Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in primates. Journal of Human Evolution, 22 , 469–493. doi:10.1016/0047-2484(92)90081-J
• Dunkel-Schetter, C., Blasband, D., Feinstein, L., & Herbert, T. (1992). Elements of supportive interactions: When are attempts to help effective? In Spacapan, S. & Oskamp, S. (Eds.) Helping and being helped: Naturalistic studies . (pp. 83–114). Thousand Oaks, CA, US: Sage Publications, Inc.
• Durkheim, E. (1951). Suicide: A study in sociology . Ornstein, R. & Swencionis, C. (Eds). New York, NY: Free Press.
• Elsesser, L., & Peplau, L. A. (2006). The glass partition: Obstacles to cross-sex friendships at work. Human Relations, 59(8) , 1077–1100.
• Emmons, R. A. & Colby, P. M. (1995). Emotional conflict and well-being relation to perceived availability, daily utilization, and observer reports of social support. Journal of Personality and Social Psychology, 68 , 947–959.
• Fehr, B. (2008). Friendship formation. In S. Sprecher, A. Wenzel, & J. Harvey (Eds.), Handbook of Relationship Initiation (pp. 29–54). New York, NY: Psychology Press.
• Feingold, Alan (1988). Matching for attractiveness in romantic partners and same-sex friends: A meta-analysis and theoretical critique. Psychological Bulletin 104 , 226–235.
• Finkel, E. J., Burnette J. L., & Scissors L. E. (2007). Vengefully ever after: Destiny beliefs, state attachment anxiety, and forgiveness. Journal of Personality and Social Psychology, 92 , 871–886.
• Fisher, H. E., Brown, L. L., Aron, A., Strong, G., & Mashek, D. (2009). Reward, addiction, and emotion regulation systems associated with rejection in love. Journal of Neurophysiology, 104 , 51–60.
• Friedman, H. S. & Martin, L. R. (2011). The Longevity Project: Surprising Discoveries for Health and Long Life from the Landmark Eight-Decade Study . New York, NY: Hudson Street Press.
• Gottlieb, B. H. (1985). Social support and community mental health. In S. Cohen & S. Syme (Eds.), Social Support and Health (pp. 303–326). Orlando, FL: Academic Press.
• Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement. American Sociological Review, 25 , 169–186.
• Ingram, P., & Morris, M. W. (2007). Do people mix at mixers? Structure, homophily, and the “life of the party.” Administrative Science Quarterly, 52 , 558–585.
• Kaufman, B. E., & Hotchkiss, J. L. 2003. The economics of labor markets (6th ed.). Mason, OH: Thomson South-Western.
• Levine, D. (2000). Virtual attraction: What rocks your boat. Cyberpsychology & Behavior, 3(4) , 565–573. doi:10.1089/109493100420179
• Madden, M. & Lenhart, A. (2006). Americans who are seeking romance use the Internet to help them in their search, but there is still widespread public concern about safety of online dating. Pew/Internet and American Life Project. Retrieved from
• Markus, H., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98 , 224–253.
• McCann Hamilton, V. (2007) Human relations: The art and science of building effective relationships . Upper Saddle River, NJ: Pearson Prentice Hall.
• McKenna, K. A. (2008) MySpace or your place: Relationship initiation and development in the wired and wireless world. In S. Sprecher, A. Wenzel, & J. Harvey (Eds.), Handbook of relationship initiation (pp. 235–247). New York, NY: Psychology Press.
• McKenna, K. Y. A., Green, A. S., & Gleason, M. E. J. (2002). Relationship formation on the Internet: What’s the big attraction? Journal of Social Issues, 58 , 9–31.
• Mckillip, J., & Redel, S. L. (1983). External validity of matching on physical attractiveness for same- and opposite-sex couples. Journal of Applied Social Psychology, 13 , 328–337.
• Moreland, R. L., & Beach, S. R. (1992). Exposure effects in the classroom: The development of affinity among students. Journal of Experimental Social Psychology, 28 , 255–276.
• Reis, H. T., Sheldon, K. M., Gable, S. L., Roscoe, R., & Ryan, R. (2000). Daily well being: The role of autonomy, competence, and relatedness. Personality and Social Psychology Bulletin, 26 , 419–435.
• Riordan, C. M., & Griffeth, R. W. (1995). The opportunity for friendship in the workplace: An underexplored construct. Journal of Business and Psychology, 10 , 141–154.
• Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 57(6) , 1069–1081.
• Sternberg, R. J. (2007). Triangulating Love. In Oord, T. J. The Altruism Reader: Selections from Writings on Love, Religion, and Science (pp 331-347). West Conshohocken, PA: Templeton Foundation.
• Sternberg, R. J. (2004). A Triangular Theory of Love. In Reis, H. T.; Rusbult, C. E. Close Relationships (pp: 528-276). New York, NY: Psychology Press.
• Stroebe, W., & Stroebe, M. (1996). The social psychology of social support. In Social psychology: Handbook of basic principles (pp. 597–621). New York, NY: Guilford Press.
• Taylor, L. S., Fiore, A. T., Mendelsohn, G. A., & Cheshire, C. (2011). “Out of my league”: A real-world test of the matching hypothesis. Personality and Social Psychology Bulletin, 37 , 942–955.
• Triandis, H. C. (1995). Individualism and collectivism . Boulder, CO: Westview.
• Triangular Theory of Love. (n.d.). In Wikipedia. Retrieved April 3, 2013, from Wikipedia, Triangular theory of love
• Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. American Psychologist, 35(2) , 151–175.
• Zajonc, R. B. (1968) Attitudinal effects of mere exposure. Journal of Personality and Social Psychology, 9 , 1–27.
• Bazzini, D., Curtin, L., Joslin, S., Regan, S. & Martz, D. (2010). Do animated Disney characters portray and promote the beauty-goodness stereotype? Journal of Applied Social Psychology, 40, 10, 2687-2709. Do Animated Disney Characters Portray and Promote the Beauty–Goodness Stereotype?
• Inagaki, T.K., & Eisenberger, N.I. (2013, November 5). Shared Neural Mechanisms Underlying Social Warmth and Physical Warmth. Psychological Science. The ‘Heartwarming’ Nature of Social Bonds
• Johnson, M.D. (2018, February 12). No, opposites do not attract. The Conversation. The Conversation, No, opposites do not attract
• Sommers, S. (2011, July 6). Who do you love? The science of falling in love. Psychology Today. Psychology Today, Who Do You Love?

The relative closeness or distance from a given comparison standard. The further from the standard a person is, the less important he or she considers the standard. When a person is closer to the standard he/she is more likely to be competitive.

The frequency with which we cross paths with others.

The notion that people like people/places/things merely because they are familiar with them.

The actual act of receiving support (e.g., informational, functional).

The actual act of receiving support (e.g., informational, functional)

The people who care about and support a person.

Social Psychology Copyright © by Jennifer Croyle is licensed under a Creative Commons Attribution 4.0 International License , except where otherwise noted.

Share This Book

• Bipolar Disorder
• Therapy Center
• When To See a Therapist
• Types of Therapy
• Best Online Therapy
• Best Couples Therapy
• Best Family Therapy
• Managing Stress
• Sleep and Dreaming
• Understanding Emotions
• Self-Improvement
• Healthy Relationships
• Student Resources
• Personality Types
• Guided Meditations
• Verywell Mind Insights
• 2024 Verywell Mind 25
• Mental Health in the Classroom
• Editorial Process
• Meet Our Review Board
• Crisis Support

What Is Reciprocity?

Kendra Cherry, MS, is a psychosocial rehabilitation specialist, psychology educator, and author of the "Everything Psychology Book."

Sean is a fact-checker and researcher with experience in sociology, field research, and data analytics.

Natalie Racioppa / Getty Images

• Development

Reciprocity is a process of exchanging things with other people to gain a mutual benefit. The norm of reciprocity (sometimes referred to as the rule of reciprocity) is a social norm where, if someone does something for you, you then feel obligated to return the favor.

If someone talks about something being a two-way street or give-and-take, these are other words and phrases for reciprocity. Learn how reciprocity develops, types of reciprocity, how it's used, and more.

How Reciprocity Develops

The socialization process plays an important role in developing the need to reciprocate. Through experience, children learn to share with others, take turns, and engage in reciprocal actions. Reciprocity plays an important role in the development and continuation of relationships. It also plays an important role in persuading others to adopt certain beliefs or behaviors.

If you have ever felt obligated to do something for someone because they first did something for you, then you were likely responding to the norm of reciprocity. This is an example of just one type of social norm that can have a powerful influence on our behavior .

The reciprocity norm operates on a simple principle: People tend to feel obligated to return favors after people do favors for them.

When your new neighbors bring over a plate of cookies to welcome you to the neighborhood, for instance, you might feel obligated to return the favor when they ask you to take care of their dog while they are on vacation.

Types of Reciprocity

There are three main types of reciprocity:

• Generalized reciprocity : This form often involves exchanges within families or friends. There is no expectation of a returned favor; instead, people simply do something for another person based on the assumption that the other person would do the same thing for them. This type of reciprocity is related to altruism.
• Balanced reciprocity : This type involves a calculation of the value of the exchange and an expectation that the favor will be returned within a specified time frame. For example, someone might exchange something they have, whether it is a skill or tangible item, for something of perceived equal value.
• Negative reciprocity : This form of reciprocity happens when one party involved in the exchange is trying to get more about it than the other person. Selling a much-needed item at an inflated price is one example of negative reciprocity.

Uses for Reciprocity

One area where this norm is commonly employed is in the field of marketing. Marketers utilize a broad range of strategies to convince consumers to make purchases. Some are straightforward such as sales, coupons, and special promotions. Others are far more subtle and make use of principles of human psychology of which many people are not even aware.

Charities also sometimes use reciprocity in an attempt to increase their donations. They might send you free greeting cards or an ink pen, for example, in the hopes that you will reciprocate by donating money to their organization.

Research indicates that, while reciprocity may initially cause people to make a charitable donation, this response reduces over time.

Examples of Reciprocity

Examples of reciprocity in business include:

• A salesperson giving a freebie to a potential customer, hoping that it will lead them to return the favor by purchasing something
• A leader offering attention and mentorship to followers in exchange for loyalty
• Offering customers some valuable information in exchange for signing up for future marketing offers

In relationships, reciprocity often looks like supporting one another in different situations. For example, you might comfort your partner when something doesn't go their way. In return, they will provide comfort and support when you are having a bad day.

Impact of Reciprocity

Reciprocity has a few obvious benefits. For one thing, taking care of others helps the survival of the species.  

By reciprocating, we ensure that other people receive help when they need it and that we receive assistance when we need it.

Reciprocity also allows people to get things done that they would not be able to do on their own. By working together or exchanging services, people can accomplish more than they would individually.

One seminal experiment showed how powerful reciprocity could be in the real world. In 1974, sociologist Phillip Kunz mailed out handwritten Christmas cards with a note and photograph of him and his family to approximately 600 randomly selected people. All of the recipients of the cards were complete strangers. Shortly after mailing the cards, responses began trickling in.

Kunz received nearly 200 replies. Why would so many people reply to a complete stranger? This is the rule of reciprocity at work. Since Kunz had done something for them (sent a thoughtful note during the holiday season), many recipients felt obligated to return the favor.

Reciprocity and Persuasion

There are also a number of persuasion techniques that employ the tactic of reciprocity. These strategies are used by people who are trying to persuade you to take action or conform with a request, such as salespeople or politicians.

One of these is known as the "that's-not-all" technique. Let's say you're shopping for a new mobile phone. The salesperson shows your phone and tells you the price, but you're still not quite sure. If the salesperson offers to add a phone case at no additional charge, you might feel like they're doing you a favor, which in turn might make you feel obligated to buy the phone.

Reciprocity in Relationships

Reciprocity is a critical component of a healthy relationship . It involves a mutual exchange of support, emotional investment, care, and love. Reciprocity in a relationship is characterized by:

• Each partner feeling able to share their needs
• A willingness to meet the needs of the other person
• Open and honest communication
• Interdependence, in which partners support one another while maintaining a clear sense of self
• Emotional reciprocity, which involves showing empathy and support for another person and the return of that same empathy and support when you need it

It is not a transactional exchange where each person keeps score. Instead, relationship reciprocity focuses on a balanced give and take where people strive to communicate their needs, respond to their partner, and note when each person's needs change.

In relationships, reciprocity involves a mutually beneficial exchange of support that makes each person feel cared for and loved. It is marked by sharing needs, caring for each other, empathy, and interdependence. Because each person provides emotional support that is then reciprocated, both people in the relationship get the care that they need to thrive.

Tips for Navigating Reciprocity

In many cases, the reciprocity norm is actually a good thing. It helps people behave in socially acceptable ways and allows them to engage in a social give-and-take with others. But what should you do if you are trying to overcome the urge to reciprocate, such as trying to avoid the need to purchase an item after receiving a freebie?

Some tips that can help:

• Give it some time. Experts suggest that the urge to reciprocate is strongest immediately after the initial exchange.   If you can wait, you will probably feel less pressure to return the favor.
• Evaluate the exchange. Think about whether the favor measures up to the expected return. In many cases, the initial gift or favor is much smaller than the requested return favor.

Understanding how the reciprocity norm influences behavior may help you better evaluate persuasive messages and requests.

Potential Pitfalls of Reciprocity

Reciprocity is not always an even exchange, which opens up the potential for imbalance or even abuse. Research has shown that people are often willing to perform a proportionately larger favor after someone has done something small for them.

Engaging in that first reciprocal exchange can make it more likely that you'll respond to other, often bigger, requests in the future. In marketing, this is often called the "foot-in-the-door" technique. Someone starts off by making a small request, and once you agree to it, they then make a much bigger request.

Another approach known as the "door-in-the-face" technique can also take advantage of reciprocity. The persuaded starts by asking for a large favor they know you will reject. They then appear to concede by asking for a much smaller favor, which you might feel obligated to fulfill.

In reality, the small favor was the intent all along, but by appearing to do you a favor by making a smaller request, you feel compelled to return the favor by saying yes to the smaller request.

American Psychological Association. Reciprocity .

Chuan A, Kessler JB, Milkman KL. Field study of charitable giving reveals that reciprocity decays over time . PNAS . 2018;115(8):1766-1771. doi:10.1073/pnas.1708293115

Nohe C, Hertel G. Transformational leadership and organizational citizenship behavior: A meta-analytic test of underlying mechanisms . Front Psychol . 2017;8:1364. doi:10.3389/fpsyg.2017.01364

Molm LD. The structure of reciprocity . Social Psychology Quarterly . 2010;(73)2:119-131. doi:10.1177/0190272510369079

Kunz PR, Woolcott M. Season's greetings: From my status to yours . Social Science Research . 1976;5(3):269-278. doi:10.1016/0049-089X(76)90003-X

Cialdini RB. Pre-Suasion: A Revolutionary Way to Influence and Persuade . Simon & Schuster . 2016.

Goyal N, Miller JG. The importance of timing in reciprocity: An investigation of reciprocity norms among Indians and Americans . Journal of Cross-Cultural Society . 2017;(49)3;381-403. doi:10.1177/0022022117746239

Comello ML, Myrick JG, Raphiou AL. A health fundraising experiment using the "foot-in-the-door" technique .  Health Mark Q . 2016;33(3):206‐220. doi:10.1080/07359683.2016.1199209

Guéguen N. Door-in-the-face technique and delay to fulfill the final request: An evaluation with a request to give blood .  J Psychol . 2014;148(5):569‐576. doi:10.1080/00223980.2013.817963

Molm, L. The structure of reciprocity . Social Psychology Quarterly. Published April 2010

Zimbardo PG, Leippe MR. The Psychology of Attitude Change and Social Influence. New York: McGraw-Hill. 1991.

By Kendra Cherry, MSEd Kendra Cherry, MS, is a psychosocial rehabilitation specialist, psychology educator, and author of the "Everything Psychology Book."

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List

Reciprocity of social influence

Ali mahmoodi.

1 Bernstein Centre Freiburg, University of Freiburg, Hansastrasse 9a, 79104 Freiburg, Germany

2 Faculty of Biology, University of Freiburg, Schänzlestraße 1, 79104 Freiburg, Germany

Bahador Bahrami

3 Institute of Cognitive Neuroscience, University College London, 17 Queen Square London, London, WC1N 3AR UK

4 Faculty of Psychology and Educational Sciences, Ludwig Maximilian University, Leopoldstrasse 13, 80802 Munich, Germany

Carsten Mehring

Associated data.

The behavioral data that support the findings of this study and the code that was used to generate the findings and to conduct the experiments of this study will be provided to all readers upon request.

Humans seek advice, via social interaction, to improve their decisions. While social interaction is often reciprocal, the role of reciprocity in social influence is unknown. Here, we tested the hypothesis that our influence on others affects how much we are influenced by them. Participants first made a visual perceptual estimate and then shared their estimate with an alleged partner. Then, in alternating trials, the participant either revised their decisions or observed how the partner revised theirs. We systematically manipulated the partner’s susceptibility to influence from the participant. We show that participants reciprocated influence with their partner by gravitating toward the susceptible (but not insusceptible) partner’s opinion. In further experiments, we showed that reciprocity is both a dynamic process and is abolished when people believed that they interacted with a computer. Reciprocal social influence is a signaling medium for human-to-human communication that goes beyond aggregation of evidence for decision improvement.

Humans give and receive social influence—e.g. advice—in many situations, but it is not known whether social influence is a reciprocal process, like trade. Here, the authors show that people are more likely to follow a partner's advice if that partner has previously complied with their advice.

Introduction

When we are uncertain, we look for a second opinion and those opinions often change our decisions and preferences 1 – 5 . Moreover, social influence is not restricted to difficult or critical decisions: evaluations of the comments in the news media are affected by previous scores of the content 6 and risk preferences alter after observing other people’s choices 7 . Social influence extends to perceptual judgment 8 – 10 and long-term memory 11 . Social information can help improve decision accuracy 12 , outcome value 13 , and evaluative judgment 14 – 16 . On the other hand, social influence can also lead to catastrophic outcomes. Social influence causes information cascades 17 urging individuals to ignore their own accurate information in favor of the cascaded falsehoods. In some cases, groups are less biased when their individuals resist social influence 18 . Social influence can undermine group diversity 19 leading to disastrous phenomena such as market bubble 20 , rich-get-richer dynamics 21 , and zealotry 22 .

Humans tend to reciprocate in social interaction 23 , 24 . We react to respect with respect and hostility with hostility 23 . Smiling staff get more tips 25 . Violators of trust in Trust Games 26 , and free riders (i.e., those who do not contribute) in Public Good Games are punished 27 . Since social influence is, by definition, mediated via social interaction, one may wonder if reciprocity extends to social influence itself. However, despite a wealth of research on reciprocal behavior, to our knowledge, no study has examined the existence of reciprocity in how we receive and inflict influence on others 28 , 29 .

Several studies in social decision-making in humans have shown a sensible correspondence between the reliability of social information (e.g., advice) and the extent to which that advice is assimilated in decisions and preferences. For example, human agents integrate their own choice with that of an advisor by optimally tracking the trustworthiness of the advisor 30 and are able to track the expertise of several agents concurrently 31 . Social information is integrated into value and confidence judgment based on its reliability 32 or credibility 33 . People can integrate information from themselves and others based on their confidence 9 . These studies offer strong evidence for what previous works in social psychology have called “informational conformity” 34 , which is based on the assumption that others can have access to information that will help the agent achieve better accuracy. The prediction drawn from this informational account is that social influence should hinge on the reliability of the source and quality of the advice but not on conventions and norms such as reciprocity. If we could benefit from others’ (critical) opinion and the accuracy of our opinion is our only concern, then we should welcome reliable advice irrespective of the advisor’s attitude toward our opinion.

On the other hand, numerous studies indicate that people perform less than ideally when using social information 10 , 35 – 37 . When aggregating individual and social information about a perceptual decision, humans follow a simplifying heuristic, dubbed equality bias: the tendency to allow everyone equal say in a collective decision irrespective of their differential accuracy or expertise 36 , 37 . The universal prevalence of the equality bias is important because it shows that even though humans do have the cognitive and computational capacity to track the trustworthiness 30 , reliability 31 , and credibility 33 of others, they still choose to employ a simple heuristic. Other studies have shown that in a different set of experimental conditions, people show an egocentric bias by relying on their own individual information more than they should 38 , 39 . These findings suggest that normative concerns such as equality and maintaining a good self-image may also play an important role in interactive decision-making. These factors are not consistent with the Bayesian theory of social information aggregation 30 , 32 , 33 , which requires that the influence that people take from others should not depend on norms and conventions. An empirical observation of reciprocity in advice-taking would be inconsistent with the Bayesian theory of social influence 30 , 32 , 33 .

To summarize, aligning with other people’s choices by taking their advice could be motivated informationally 40 to increase accuracy, or normatively 41 to affiliate with others and maintain positive self-esteem. Following other people’s advice often leads to more accurate decisions 9 , 42 , 43 . Alignment can contribute to a positive self-image 44 and is used as a compensatory tool among minorities 45 . Being ignored in a virtual game can damage one’s self-image 46 . Social exclusion has negative consequences on the excluded 47 . We hypothesized that people would reciprocate influence with others because reciprocity is a pervasive social norm 48 . If one breaks the norm of reciprocity, one should expect to be punished 49 , for example, by being ignored. Therefore, participants would reciprocate with a reciprocating partner in order to maintain influence over them and avoid the negative experience of losing influence 47 . On the other hand, for a non-reciprocating partner who violates the norm, participants may have a desire to punish the partner by ignoring their opinion. We tested these hypotheses by investigating if participants in a social decision-making task take less advice from insusceptible partners, i.e., those who do not take advice from the participant. Conversely, we also tested if participants take more advice from susceptible partners, i.e., those who are influenced by the participant’s suggestions.

We adopted and modified an experimental perceptual task inspired by recent work on aggregation of social and individual information 10 . Participants estimated the location of a visual target on a computer screen. Then they saw the estimate of their partner about the location of the same target. The participant did not know that this partner’s opinion was, in reality, generated by sampling randomly from a distribution centered on the correct answer. After the two initial estimates were disclosed, the participant or the partner was allowed to revise their estimate. An algorithm generated the partner’s revised estimate simulating susceptible or insusceptible partners. Experiment 1 showed that, participants took more advice from the partner who took more advice from the participant. Experiment 2A and 2B investigated the dynamics of reciprocity and whether reciprocity depends critically on whether we believe it changes the partner’s state of mind. Participants thought they worked with either a human or a computer partner and reciprocity disappeared when subjects believed they were working with a computer partner. Finally, our results also showed that reciprocity had a profound impact on participants’ evaluation of their own performance, which was lower when working with a non-reciprocating partner.

Experiment 1

In experiment 1, 20 participants were recruited one at a time and told that they would cooperate with three partners who were participating in the same experiment simultaneously in other laboratory rooms connected via internet. In reality, each participant was coupled with a computer algorithm. The algorithm generated three distinct behavioral profiles, corresponding to three experimental conditions (see below), which were administered in a block design in counterbalanced order. Participants were not informed about this arrangement. In each trial, the participant made a perceptual estimate about the location of a target on the screen (Fig.  1 ). After stating her initial estimate, the participant saw the opinion of the partner about the same stimulus. Next, the participant either revised her estimate or observed the partners revise theirs. Participants were required to put their second estimate between their own first estimate and that of their partner’s. The acceptable range included staying on their first estimate or moving all the way to their partner’s first estimate. Using this constraint, we assured that the amount of change made in the second stage is solely due to observing the partner’s choice and not because of a change of mind 50 . The three partners differed in their susceptibility to taking influence from the participant. In the baseline condition, the participant always made the second estimate and the partner never contributed a second estimate. Hence, participants were not able to observe the susceptibility of the baseline partner. In the susceptible condition, the partner was influenced strongly by the participant and revised her initial estimate by conspicuously gravitating toward the participant’s estimate. Vice versa, in the insusceptible condition, the partner more or less ignored the participant’s opinion. The partner’s initial estimate was generated identically in all three conditions by sampling randomly from a distribution centered on the correct answer.

Experimental task. a Participants first observed a series of dots on the screen. Participants were required to indicate where they saw the very first dot (yellow dot) and then declare their numerical confidence. After making their individual estimates, they were presented with the estimate of a partner (red dot) concerning the same stimulus. After observing the partner’s choice, in some trials the participants and in other trials the partner was given a second chance to revise their initial estimate. Afterwards, they were briefly presented with their initial choices and the second choice. In experiment 1, they did the task with three different alleged human partners, which only varied in the second choice strategy in different blocks: in the baseline blocks, the participant made all second choices. In the susceptible blocks, the partner was very influenced by the participant’s first choice, however in the insusceptible blocks, the partner was much less influenced by the participant’s first choice compared to susceptible blocks. b Influence was computed as the angular displacement toward the peer’s choice divided by their initial distance from each other’s choice

We computed the influence that participants took from their partner as the ratio of the angular displacement (in radians) between their initial and final estimate toward their partner’s estimate divided by their initial angular distance from their partner (Fig.  1b ). Overall, participants were influenced by their partners’ opinions (mean influence ± std. dev. = 0.36 ± 0.11; Wilcoxon sign rank test vs zero, Z  = 3.62, p  = 0.0002). Consistent with previous studies 30 , 33 , 36 , this indicates that our participants did use social information (the partners’ choices) to improve their decisions (mean ± std. dev. error after first estimate 67 ± 7 radians and after second estimate 64 ± 6 radians, Wilcoxon sign rank test, Z  = 3, p  = 0.002). To test our main hypothesis, we asked whether revised opinions were more influenced by the susceptible than the insusceptible (Fig.  2a ) and baseline (Fig.  2b ) partners. The difference between the average influence in the susceptible and the insusceptible condition, which is a measure of reciprocity, was significantly larger than zero (Fig.  2a–c , Wilcoxon sign rank test, Z = 3.33, p  = 0.002 after Bonferroni correction). Similarly, the influence from the susceptible partner was larger than the influence in the baseline condition (Fig.  2b, c , Wilcoxon sign rank test, Z  = 2.34, p  = 0.03 after Bonferroni correction). The difference between insusceptible and baseline was not significant (Wilcoxon sign rank test, Z  = 1.11, p  = 0.26).

Results of experiment 1. a Reciprocity, computed as influence in the susceptible condition minus influence in the insusceptible condition, plotted across participants. b Reciprocity, computed as influence in the susceptible condition minus influence in the baseline condition, is plotted across participants. c Average reciprocity across participants in insusceptible and baseline conditions compared to the susceptible condition. Error bars indicate the standard errors and were computed across the participants’ average reciprocities. * p  < 0.05, ** p  < 0.005; Wilcoxon sign rank test. d Performance ratings for self and all partners as reported at the end of the experiment

The three different partners’ initial estimates were produced from an identical generative process using exactly the same distribution and therefore ensuring that the partners’ accuracies were perfectly controlled across conditions. However, one might argue that the observed result may be due to the difference in perceived accuracy of the partner. Indeed, we may think more highly of those who confirm our decisions more often and, subsequently, take our (misguided) assessment of their competence as grounds for integrating their estimate into our own revised opinion. To test this hypothesis directly, at the end of each experiment, we asked the participants to rate the precision of the partners they interacted within the experiment, how much they liked different partners, and their own performance as well. A mixed effect model showed that perceived precision of the partners, the actual precision of the partners, the participants’ actual precision in different conditions, and the liking of the partner did not have any effect on reciprocity (Supplementary Note  1 ). The performance ratings of the partner did not differ across conditions (Fig.  2d , repeated measures analysis of variance (ANOVA), F (2.49, 39) = 1.07, p  = 0.36). In each trial, after participants registered their estimates, they were required to report their confidence about their estimates using a scale from 1 to 6. Employing mixed effect models, we showed that condition (baseline, susceptible, or insusceptible) had a significant effect on influence even if confidence was included as a potential confound (Supplementary Note  1 ).

At the end of the experiment, we asked participants to rate how much they liked their three partners on a scale of 1–10. The mean score ± std. dev. was 7.64 ± 0.7 for the susceptible partner, 5.94 ± 2.53 for the insusceptible partner, and 7.52 ± 1.41 for the baseline partner. Our data shows that people liked the susceptible partner over the insusceptible one (Wilcoxon sign rank test, Z  = 2.62, p  = 0.008). There was no difference between susceptible and baseline partners (Wilcoxon sign rank test, Z  = −0.34, p  = 0.72) and the p -value for the difference between the baseline and the insusceptible partner was only around the threshold (Wilcoxon sign rank test, Z  = −1.91, p  = 0.056).

Experiment 2A and 2B

The results of our first experiment show that human participants were more influenced by partners which were reciprocally more influenced by the participants. We next asked whether human participants change their advice-taking strategy in response to a change in the advice-taking strategy of a partner over time. To answer this question, we carried out experiment 2A using the same paradigm as in experiment 1 but with an important modification. The participants were told that they are working with the same partner during the entire experiment. The partner’s strategy changed across time: in one part of the experiment, the partner was susceptible, in the other part, she was not. Between the two conditions of the experiment, there was a smooth transition (from susceptible to insusceptible or vice versa) and the order of the conditions was counterbalanced across participants (Supplementary Figure  1 ).

We calculated participants’ trial-by-trial influence in the two conditions. Replicating experiment 1, reciprocity, again defined as influence in the susceptible condition minus influence in the insusceptible condition, was significantly larger than zero (Fig.  3a , Wilcoxon sign rank test, Z  = 3.54, p  = 0.0003). A mixed effect model showed that condition (susceptible or insusceptible) had a significant effect on influence even if confidence was included as a potential confound (Supplementary Note  1 ). Again, using a mixed effect model, we showed that the change of influence across conditions cannot be explained by a change in perceived performance of self or partner (Supplementary Note  1 ). An interesting question is whether reciprocity is affected by the condition (susceptible or insusceptible) with which the participant started the experiment. To answer this question, we compared the reciprocity for participants who were first exposed to the susceptible partner to participants who started with the insusceptible partner. Our result showed no difference in reciprocity between these two groups (mean ± std. dev. for those who started with reciprocal partner 0.049 ± 0.1 and for those who started with non-reciprocal partner 0.08 ± 0.07, Wilcoxon rank-sum test, z  = −1.07, p  = 0.28).

Results of experiments 2A and 2B. a Reciprocity when participants believe they interact with a human partner. b Reciprocity when participants believe they interact with a computer partner. c Influence for alleged human and computer partners across susceptible and insusceptible conditions. Dots indicate each participant. Black dots depict the mean influence across participants while error bars depict the standard error of the mean. d Average reciprocity across participants when participants think they interact with a human or a computer partner. Error bars depict the standard errors. e Difference in participants’ performance rating for self, insusceptible minus susceptible, plotted for each participant. The inset shows the mean and standard error across participants. ** p  < 0.005, ns not significant; Wilcoxon sign rank test

The present results demonstrate that if a participant observes any change in the amount of influence she has over her partner, she will in return modify the amount of advice she takes from her partner. We, therefore, hypothesized that our participants exploit reciprocity as a social signal to communicate with their partner. We predicted that participants would not show reciprocity when working with a computer. In other words, reciprocity depends critically on whether we believe it changes the partner’s state of mind. To test this prediction, we conducted experiment 2B in which participants were told that they were working with a computer. All other aspects of the experiment remained as in experiment 2A. As predicted, we did not observe reciprocity when participants believed they were working with a computer (Fig.  3b–d , Wilcoxon sign rank test, Z  = −1.57, p  = 0.11). In fact, a majority of participants showed the opposite pattern observed in experiment 2A (Fig.  3b ). A two-way ANOVA with factors condition (susceptible or insusceptible as within subjects factor) and type of partner (believed to be human or computer as between subjects factor) and influence as the dependent variable showed a significant interaction of type of partner and condition ( F (1, 58) = 14.8, p  = 0.00001). The effect of condition alone was not significant ( F (1, 58) = 0.49, p  = 0.51), but there was also a significant between-subject effect of type of partner (Fig.  3c , F (1, 58) = 4.16, p  = 0.04). A post hoc analysis between reciprocity in human and computer condition confirmed a significant difference in reciprocity between these two conditions (Fig.  3c, d , Wilcoxon rank-sum test, Z  = 2.97, p  = 0.003). Taken together, these results demonstrate that participants were more influenced when they thought their partner is a computer (mean influence ± std. dev.: 42 ± 0.16) compared to when they thought their partner is a human (mean influence ± std. dev.: 0.34 ± 0.13).

We then investigated whether the distance between the participants’ and the partners’ initial estimates affected the influence that participants took from their partner or the strength of reciprocity (see Supplementary Note  1 for details). We did not find a significant effect of distance (mixed ANOVA, F (2.6, 151) = 1.17, p  = 0.31) nor significant interactions between distance and condition (mixed ANOVA, F (2.6,151) = 1.47, p  = 0.22) or between distance, condition, and experiment ( F (2.6, 151) = 1, p  = 0.38). The interaction between distance and experiment was only around the significance threshold ( F (2.6, 151) = 2.7, p  = 0.05). We also did not find a significant effect of distance on reciprocity in experiment 2A (repeated measures ANOVA, F (2.61, 75) = 1.63, p  = 0.16). Taken together, these results show that the distance between the initial estimates did not affect the influence that participants took from their partner nor the strength of reciprocity.

Finally, we asked how being in the susceptible and insusceptible conditions changed the participants’ ratings of their own and their partner’s performance. To answer this question, participants were asked to rate their own and their partners’ performance on a 1–10 scale at the end of each condition, i.e., twice in each experiment 2A and 2B. As there was not any difference in performance rating between experiment 2A and 2B, neither for self nor partner (Supplementary Figure  2 ), we aggregated the data of the two experiments. Participants’ rating of their own performance was significantly lower in the insusceptible than in the susceptible condition (Fig.  3e , Wilcoxon sign rank test, Z  = 3.04, p  = 0.002), while participants’ rating of their partners’ performance remained unaffected by the partners’ susceptibility (mean rating ± std. dev.: 6.63 ± 1.58 for susceptible condition, and 6.81 ± 1.59 for insusceptible condition; Wilcoxon sign rank test, Z  = −0.9, p  = 0.36).

An important question in human social interaction is how people weigh others’ opinion 51 . Bayesian theories recommend that different opinions should be weighted by their reliability in order for the group to benefit from putting the opinions together 52 . Indeed, some empirical evidence has supported this view 9 , 30 , 32 , 33 while others have shown other decision aggregation strategies in human social interaction 10 , 36 .

We developed an experimental paradigm inspired by previous work on social information aggregation 10 . We quantified how people weighted their peer’s opinion in the context of a visual perceptual task. We tested if this weighting depends on the weight their peers assigned to the participants’ opinion. In experiment 1, participants worked with three different alleged human partners in separate blocks. Our participants were more influenced by the susceptible partner compared to the baseline and insusceptible partners. In experiment 2A, the behavior of a single partner changed dynamically within the same experiment from susceptible to insusceptible or vice versa. We replicated the result of experiment 1 by showing that participants were more influenced in the susceptible condition than in the insusceptible condition. In experiment 2B, we showed that participants did not reciprocate when their peer was a computer even though they took greater influence from the computer’s advice.

When combining opinions in an optimal Bayesian way to maximize accuracy, each source of information should be weighted based on its reliability 29 . Consequently, reciprocity of social influence, i.e., weighting others’ opinion by the weight they give to our opinion is not consistent with Bayesian reliability-based information aggregation. In our experiment, we systematically controlled the accuracy of the participants’ partners such that they were identical across experimental conditions. Participants rated their own performance lower in the insusceptible condition than in the susceptible condition but did not distinguish between the partners’ accuracies. With such judgment of their performance and that of their partners, a hypothetical Bayesian participant would have taken more influence from the insusceptible partner. This is actually what participants did when working with a computer partner. However, when working with a human partner, they followed the opposite strategy and took more influence from the susceptible partner.

Why do people go against an information integration strategy that is more likely to maximize their accuracy? We propose that reciprocity is a pervasive social norm 48 , and abiding by norms is sometimes rewarding in itself and could hence become a goal 53 . As a consequence, individuals may be ready to pay a cost (in terms of reduced accuracy) to adhere to these norms 54 . This explanation is supported by the finding that participants did not reciprocate with a computer partner as participants did not expect the computer to comply with the reciprocity norm.

In the susceptible condition, participants may reciprocate with their reciprocating partners in order to keep their influence over them and avoid the pain of being ignored 47 . As such, showing reciprocity toward a susceptible partner may be driven by a form of loss aversion. Why would people be aversive to losing influence? Recent works have suggested that influence over others may be inherently valuable both behaviorally 55 and neurobiologically 56 . There is now compelling evidence that others’ agreement with our opinion is a strong driver of human brain’s reward network 1 , 57 . In the insusceptible condition, on the other hand, ignoring the insusceptible partner may be motivated by wishing to punish someone who does not comply with the norm of reciprocity.

Following the norm of reciprocity might also improve people’s self-efficacy. It is possible that in the insusceptible condition players perceive the experiment as a status competition. In this view, ignoring the insusceptible partners in response to being ignored by them could serve as a signal from the participant that she/he is not willing to accept an inferior position 58 , 59 . Several studies in behavioral economics (ultimatum game in particular) have shown that one reason why people reject unfair offers is because they want to send the signal that they will not be easily dominated and thereby refuse to accept an inferior social status compared to their peer 60 . Indeed, multiple studies have shown that people do not like to be in an inferior position where their choices are less selected than others’ and they use various strategies to compete with their peers in having more influence 56 , 61 , 62 . However, people do not engage in a status competition with a computer that is consistent with the difference in reciprocity between human and computer experiments (Fig.  3d ). In addition to the above, ignoring the non-reciprocating partner may also serve to protect the participant’s “wounded pride” 63 and maintain their self-esteem. Our finding that participants rate their own performance higher when playing with the susceptible than with the insusceptible partner (Fig.  3e ) is consistent with the hypothesis that having influence over others improves self-efficacy. It should be noted, however, that, there is no one-to-one correspondence between the perception of self-efficacy and reciprocity: while the difference in self-efficacy between the susceptible and insusceptible conditions is the same for experiment 2A and 2B (Supplementary Figure  2 ), the difference in influence is not: in experiment 2A, the influence in the insusceptible condition is less than in the susceptible condition but in experiment 2B, the influence in the susceptible condition is identical to the insusceptible condition (Fig.  3c, d ). Hence, reciprocity cannot be entirely explained by changes in self-efficacy.

Reciprocating influence is consistent with cognitive balance theory 64 , which posits that humans change their preference to be similar to those they like and dissimilar to those they do not. In experiment 1, participants liked the susceptible partner more than the insusceptible one and were more influenced by the partner they liked more. However, in our experiment, the participants were not allowed to change their estimate away from their peers (participants were instructed to make their second choice between their and their partner’s initial estimates). This restriction makes it difficult to directly address the relationship between cognitive balance theory and reciprocity observed in the current study.

In the insusceptible condition, participants’ perceived performance of themselves dropped significantly (Fig.  3e ). Previous studies show that humans are good at tracking their accuracy even in the absence of any external feedback 9 , 65 . It is been argued that people are able to get insight into their accuracy through past experience 66 . However, in social contexts, their judgment could be affected by the environment 67 depending on whether they compete or cooperate with a peer 68 . Our performance rating results confirm the effects of social context on human performance monitoring. This finding shows that being ignored exerts a devastating impact on self-efficacy. One possibility is that being repeatedly ignored in the insusceptible condition may induce a negative emotional impression on the participant that impairs the participant’s self-evaluation. Another possibility is that participants may interpret the partner’s revised estimate as the correct position of the target. By definition, the insusceptible partner’s revised estimates would fall further from those of the participant. The inevitable conclusion for the ignored participants would be that their opinion must have been less precise in the insusceptible blocks. Future studies could investigate each of these potential explanations.

Previously, we showed that when working together in a dyad 36 , people tend to operate by an “equality bias” giving equal weight to their own and their partner’s decision. Participants fulfilled this goal either by adjusting the weight they assign to each other’s opinions 36 or by matching their confidence to the confidence of the other people they worked with 61 . Hence, in both cases, people mutually adapted to each other’s behavior when required to make decisions together. Similarly, participants exhibited mutual adaptation of social influence in the present study.

Our results imply that humans do not only consider others’ reliability to compute the weight that they assign to others’ opinion, but instead they take into account other factors like reciprocity as well. We conclude that reciprocity plays a significant role in human advice-taking and social influence, which violates the optimal account of human information integration. Reciprocity as a social norm helps people to fulfill objectives of social interaction including maintaining a positive self-image.

A total of 80 healthy adult participants (39 females, mean age ± std. dev.: 25 ± 2.9) participated in three experiments after having given written informed consent. Each participant participated in only one of the experiments. Participants were students at the University of Freiburg, Germany. The experimental procedures were approved by the ethics committee of the University of Freiburg. All experiments were performed using Psychophysics Toolbox 69 implemented in MATLAB (Mathworks). The data were analyzed using MATLAB and SPSS.

Experimental task 1

This experiment was designed to investigate whether participants were more influenced by whom they influenced more. Participants first made a perceptual estimate about the location of a target on the screen. Afterwards, they were presented with the estimate from their partner regarding the same stimulus. This was followed by making a second choice about the location of the target or observing a second choice of their partner.

In more detail, the experiment went on as follows: participants ( N  = 20, 9 females, mean age ± std. dev.: 25 ± 2.8) were presented with a sequence of 91 visual stimuli consisting of small circular Gaussian blobs ( r  = 5 mm) in rapid serial visual presentation on the screen (resolution = 2560 × 1440 Dell U2713HM 27″). The first item was presented for 30 ms and every other stimulus was presented for 15 ms each. Participants’ task was to identify the location of the first stimulus. Participants were required to wait until the presentation of all stimuli were finished, and then indicate the location of the target stimulus using the computer mouse (Fig.  1 ). The reported location was marked by a yellow dot. After participants reported their initial estimate, they were required to report their confidence about their estimate on a numerical scale from 1 (low confidence) to 6 (high confidence). Afterwards, participants were shown the choice of their partners about the same stimulus (see below, Constructing partners section for further details) by a small dot on the screen. Then, either the participant revised her estimate or observed the partner revise theirs. After the second estimate was made, all estimates were presented to the participant for 3 s. In this stage, the first choice was shown by a hexagon to be distinguished from the second choice, which was shown by a circle (Fig.  1 ). There was not any time pressure on participants in any stage of the experiment and the experiment did not move to next stage until the participants had registered their responses (Fig.  1 ). Participants were told that their payoff will be calculated based on their first and second estimates. However, everyone was given a fixed amount at the end of the experiment.

During the course of the experiment, participants were exposed to three different partners. Partners varied in their susceptibility to the participants’ estimates (i.e., the amount of influence the participants’ first estimate has on the partner’s second estimate). In the baseline blocks, the participant always made the second estimate and the partner never contributed a second estimate. In the susceptible and insusceptible blocks, the partner and the participants made the second estimate in the odd and even trials, respectively. In the susceptible block, the partner was influenced strongly by the participant and vice versa in the insusceptible block (see below for details in the section “Constructing partners”). In each block, the participants worked with only one partner and each block contained 30 trials. Participants worked with each partner for five blocks. For example, they worked with baseline partner in block 1, then with the susceptible partner in block 2, then with insusceptible partner in block 3, then again with the baseline partner in block 4, and so on. Participants completed 15 blocks in total. The order of the partners was randomized across participants. The three partners were shown by blue, red, and turquoise markers, which were randomly assigned to the different partners at the beginning of each subject’s experiment. After finishing the experiment, the participants were required to estimate their own and the three partners’ performance on a numerical scale from 1 to 10. They were instructed to only consider the first choice to assess their partners’ performance. At the end, we asked them whether they thought they interacted with real people or with a computer algorithm. All participants indicated that they believed they were interacting with real human partners.

Three participants were excluded from the final analyses of this experiment. One participant did not notice she played against three different partners. The other two participants were excluded because they resampled in the second stage, meaning that their second estimates were not between their own and their partner’s initial estimates (contrary to the task instruction). In the Supplementary Figure  3 , we show that our findings remain valid and statistically significant when these three subjects were not excluded from the analysis.

Experimental task 2A

This experiment was designed to test whether reciprocity is a dynamic process. The experiment used the same paradigm as experiment 1 but participants ( N  = 30, 15 females, mean age ± std. dev.: 25 ± 2.8) were told that they do the task with only one partner, which is the same gender as themselves. The partner changed its susceptibility during the course of the experiment, either from susceptible to insusceptible or vice versa. The experiment consisted of 11 blocks in total. Half of the participants were first probed in the susceptible condition, which lasted five blocks and then with a transition block in between, they switched to five insusceptible blocks. The other half completed the opposite order. The average advice/influence that the partner took from our participants is depicted in Supplementary Figure  1 . The transition block was designed in order to avoid a sudden change of the partner’s behavior. During the transition block, the partner’s advice-taking strategy linearly (see below) switched from susceptible to insusceptible or vice versa.

After each session of the experiment, all participants were debriefed to assess to what extent they believed the cover story. We interviewed them with indirect questions about the cover story and all participants stated that they believed they were working with other human participants in neighboring experimental rooms.

Experimental task 2B

This experiment differed from experiment 2A in one respect: participants ( N  = 30, 15 females, mean age ± std. dev.: 24 ± 3.1) were told that their partner in the experiment is a computer. Any other aspects of the experiment were identical to experiment 2A and they received exactly the same task instructions as in experiment 2A except that the human partner was replaced by a computer.

Performance rating

In experiment 1, participants rated their own and the three different partners’ performance once at the end of the experiment. Note that a different color identified each partner during the experiment. In experiment 2A and 2B, participants rated their own and their partner’s performance at the end of each block. This way, we obtained a pair of performance ratings for the self and the susceptible partner and another pair for the self and the insusceptible partner.

Constructing partners

The error distribution of all partners’ first choices was modeled from participants’ actual estimation errors during a pilot experiment. Ten participants performed an experiment identical to experiment 1 of the current study. We aggregated errors of all participants ( N  = 10) and fitted the concentration parameter kappa of a von Mises distribution centered to the target, yielding the value kappa = 7.4. Then, in each trial, we drew the first choice of the partner from this distribution. We speculated that participants’ assessment of their partners’ performance may be strongly influenced by the few trials with high confidence (confidence level of 5 or 6). To avoid this potential problem, the partner’s first choice was not taken from the von Mises distribution in high confidence trials but randomly drawn from a uniform distribution centered on the participants’ choice with a width of ±20°.

The second choice of the partner was computed differently for susceptible and insusceptible partners. For experiment 1, the influence that the insusceptible partner took from the participants in each trial was chosen with a probability of 0.65 from a uniform distribution on the interval [0, 0.2], with a probability of 0.2 randomly from a uniform distribution on the interval [0.3, 0.7], and with a probability of 0.15 randomly from a uniform distribution on the interval [0.7, 0.9]. For experiment 2, the influence that the insusceptible partner took from the participants was chosen randomly from a uniform distribution on the interval [0, 0.2]. For the susceptible partner, in all experiments, the influence was chosen with a probability of 0.5 randomly from a uniform distribution on the interval [0.7, 1], with a probability of 0.2 randomly from a uniform distribution on the interval [0.3, 0.7], and with a probability of 0.3 randomly from a uniform distribution on the interval [0, 0.3]. In the transition block, the influence of the partner was a linear interpolation between the susceptible and the insusceptible partner:

where inf s and inf ins were the influences of the susceptible and insusceptible partners, respectively (as explained above). λ gradually increased with time from 0 at the beginning to 1 at the end of the transition block for a transition from the susceptible to the insusceptible condition. For the transition from the insusceptible to the susceptible condition, λ decreased gradually from 1 to 0.

In experiment 1, on average, the advice that the partners took from the participants was 0.3 and 0.55 in the insusceptible and susceptible conditions, respectively. In experiment 2A, on average, the advice that the partner took from the participants was 0.07 and 0.5 in the insusceptible and susceptible conditions, respectively. The second choice of the partners in experiment 2B was designed exactly the same as in experiment 2A and the average advice that the partner took in the insusceptible and the susceptible condition was identical to experiment 2A.

Computation of error bars

All error bars in the figures depict standard errors. Standard errors were computed across subjects; in situations where we had multiple measurements from each participant, we first computed the mean value for each participant, and then computed the standard error across the participants’ mean.

Data availability

Electronic supplementary material

Acknowledgements

This work was supported by a PhD scholarship (A.M.) from the Graduate School Scholarship Program of the German Academic Exchange Service (DAAD), a European Research Council Starting Grant “NeuroCoDec #309865” (B.B.), the German Research Foundation (DFG, grant no INST 39/1014-1 FUGG) (C.M.) and the “Struktur -und Innovationsfonds Baden-Württemberg (SI-BW)” of the state of Baden-Württemberg (C.M.). We thank Ulf Toelch for helping to implement the experimental task and Helena Gavrilova, Tobias Pistohl, and Luke Bashford for helping to collect data.

Author contributions

A.M., B.B., and C.M. designed the experiments. A.M. collected the data. A.M. carried out the data analysis. A.M., B.B., and C.M. interpreted the results and wrote the manuscript.

Competing interests

The authors declare no competing interests.

These authors contributed equally: Bahador Bahrami, Carsten Mehring.

Supplementary Information accompanies this paper at 10.1038/s41467-018-04925-y.

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

• Subscriber Services
• For Authors
• Publications
• Archaeology
• Art & Architecture
• Bilingual dictionaries
• Classical studies
• Encyclopedias
• English Dictionaries and Thesauri
• Language reference
• Linguistics
• Media studies
• Medicine and health
• Names studies
• Performing arts
• Science and technology
• Social sciences
• Society and culture
• Overview Pages
• Subject Reference
• English Dictionaries
• Bilingual Dictionaries

Recently viewed (0)

• Save Search
• Share This Facebook LinkedIn Twitter

Related Content

Related overviews, more like this.

Show all results sharing these subjects:

• Life Sciences

reciprocity hypothesis

Quick reference.

The proposal that altruism can benefit the altruist because it will be reciprocated by other members of the group. For example, meerkats spend part of their time above ground foraging and part standing upright watching for predators; if a sentinel sees a predator it gives an alarm call that sends the others running for safety. If the sentinel is exposed to greater risk than other members of the group, the time it spends on watch will be worth it if the amount of time it can spend foraging safely while others are on watch exceeds that risk. Observation suggests this is not what happens, however, and that sentinels spend as much time as they can in a safe lookout position close to a burrow, even if they live alone. Some authorities have suggested that food sharing is a form of reciprocity, but this may involve little or no cost to the giver, in which case it is not altruistic.

From:   reciprocity hypothesis   in  A Dictionary of Ecology »

Subjects: Science and technology — Life Sciences

Related content in Oxford Reference

Reference entries.

View all related items in Oxford Reference »

Search for: 'reciprocity hypothesis' in Oxford Reference »

• Oxford University Press

PRINTED FROM OXFORD REFERENCE (www.oxfordreference.com). (c) Copyright Oxford University Press, 2023. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single entry from a reference work in OR for personal use (for details see Privacy Policy and Legal Notice ).

date: 27 June 2024

• Cookie Policy
• Privacy Policy
• Legal Notice
• Accessibility
• [185.147.128.134]
• 185.147.128.134

Character limit 500 /500

Home > ETD > DISSERTATIONS > AAI3099155

The reciprocity hypothesis as an explanation of perception shifts in product judgment

Chung-Chiang Hsiao , Purdue University

Researchers in consumer and social psychology have developed various theories to account for shifts in perceptions of targets caused by contextual stimuli. In this previous work, the perception shift of a target stimulus is observed as it reacts to the presentation of contextual stimuli. Attention in the past research has focused entirely on the target. As a result, no one has investigated possible shifts in perceptions of contextual stimuli. The present research proposes and tests a reciprocity hypothesis to explain what happens to the contextual stimulus and suggests that not only is the perception of the target shifted by the contextual stimulus, but also the perception of the contextual stimulus is shifted in a direction opposite to the perception shift of the target. That is, the perception of the contextual stimulus might be shifted away from the target in a contrast condition, but toward the target in an assimilation condition. Two studies with 390 undergraduates examined the hypothesis that, after priming, the perception of the contextual stimulus might be shifted in a direction opposite to the perception shift of the target. In Study 1, the perception shift of the contextual stimulus was examined when the contextual stimulus was manipulated to produce a contrast effect on the target (the perception of the target was shifted away from the contextual stimulus). The result showed that the perception of the contextual stimulus was shifted away from the target as well. In Study 2, the perception shift of the contextual stimulus was examined when the contextual stimulus was manipulated to produce an assimilation effect on the target (the perception of the target was shifted toward the contextual stimulus). The result showed that the perception of the contextual stimulus was shifted toward the target. That was, results in both studies supported the predictions. Thus, after the priming procedure, the perception of the contextual stimulus might be shifted in a direction opposite to the perception shift of the target (symmetrical perception shift).

Wegener, Purdue University.

Subject Area

Behavioral psychology|Marketing|Social psychology

Off-Campus Purdue Users: To access this dissertation, please log in to our proxy server .

Since November 30, 2005

Advanced Search

• Notify me via email or RSS
• Purdue Libraries
• Purdue University Press Open Access Collections

Links for Authors

• Policies and Help Documentation
• Collections
• Disciplines

Home | About | FAQ | My Account | Accessibility Statement

Privacy Copyright

Attachment Theory In Psychology

Saul Mcleod, PhD

Editor-in-Chief for Simply Psychology

BSc (Hons) Psychology, MRes, PhD, University of Manchester

Saul Mcleod, PhD., is a qualified psychology teacher with over 18 years of experience in further and higher education. He has been published in peer-reviewed journals, including the Journal of Clinical Psychology.

Learn about our Editorial Process

Olivia Guy-Evans, MSc

Associate Editor for Simply Psychology

BSc (Hons) Psychology, MSc Psychology of Education

Olivia Guy-Evans is a writer and associate editor for Simply Psychology. She has previously worked in healthcare and educational sectors.

On This Page:

Attachment theory is a lifespan model of human development emphasizing the central role of caregivers (attachment figures) who provide a sense of safety and security.

Attachment theory hypothesizes that early caregiver relationships establish social–emotional developmental foundations, but change remains possible across the lifespan due to interpersonal relationships during childhood, adolescence, and adulthood.

Attachment can be defined as a deep and enduring emotional bond between two people in which each seeks closeness and feels more secure when in the presence of the attachment figure. 

The initial and perhaps most crucial emotional bond forms between infants and their primary caregivers.

Distinct behaviors characterize attachment in children and adults, such as seeking closeness with the attachment figure when distressed or threatened (Bowlby, 1969).

John Bowlby

Attachment theory in psychology finds its roots in the pioneering work of John Bowlby (1958). During the 1930s, Bowlby was a psychiatrist at a Child Guidance Clinic in London, treating numerous emotionally troubled children.

His experiences there underscored the significance of a child’s relationship with their mother in shaping their social, emotional, and cognitive development.

It molded his understanding of the connection between early separations from the mother and subsequent maladjustment, leading him to develop his attachment theory.

The attachment bond isn’t coincidental. Its primary purpose is to ensure the survival of the vulnerable infant, requiring the constant presence of a caregiver (Bowlby, 1973, 1980).

Viewed from this lens, attachment emerges as an evolutionary concept. The behavior of seeking proximity is universally observed across cultures (Van Ijzendoorn & Sagi-Schwartz, 2008).

Bowlby (1988) contended that the drive for proximity arises from an interconnected set of behavioral systems that collectively shape behavior. These include the attachment, caregiving, and exploratory behavioral systems.

Attachment Behavioral System

The attachment behavioral system concerns the tendency of an individual to seek security during times of stress (Mikulincer & Shaver, 2003), which can be internal (i.e., hunger, fatigue, illness) or from external features of the environment, such as threatening stimuli (Bowlby, 1988). 

The more extreme the stress, the more intense the attachment system activation.  The attachment system is most readily activated during the first five years of life, a period characterized by high levels of vulnerability and dependence. 

Once the attachment system is activated, the infant is motivated to seek proximity to significant others (attachment figures) to protect themselves from physical or emotional harm (Bowlby, 1969). 

If this goal is achieved, the infant develops feelings of safety and security, and their attachment system becomes deactivated.  The infant will call upon a range of attachment behaviors with the goal of attaining proximity to the attachment figure. 

Bowlby (1988) suggests the attachment behavioral system remains important throughout life and will also motivate adults to seek proximity in times of stress during adulthood.

Caregiving System

The attachment figure is viewed as a ‘safe haven’, and their role is to correspondingly alter their level of responsiveness to deactivate the infant’s attachment system by promoting feelings of security. 

George and Solomon (1996) call this reciprocal response of the attachment figure to the infant’s attachment system the ‘caregiving’ system. 

Bowlby (1969) posits that the caregiving system exists to provide protection and support to others in need of assistance, through providing sensitive and responsive care. 

The caregiving system is activated when an individual expresses a need for support or their attachment system is activated, and is deactivated when the care recipient appears to be in a secure state (Shaver & Mikulincer, 2006). 

Once activated, the caregiver may utilize a variety of behavioral strategies intended to improve the other person’s well-being, re-establishing their felt security, and facilitating their coping efforts. 

Caregiving strategies include validating a person’s worries, providing physical closeness and affection, and communicating that a person is loved and valued (Mikulincer & Shaver, 2007). 

In addition to attachment behaviors, the caregiving system can support exploratory behaviors (Feeney, 2004). 

The Importance of Early Emotional Bonds

• Attachment behavior in adults toward the child includes responding sensitively and appropriately to the child’s needs.  Such behavior appears universal across cultures.
• Attachments are most likely to form with those who responded accurately to the baby’s signals, not the person they spent more time with. Schaffer and Emerson called this sensitive responsiveness.
• Reciprocity is the mutual, two-way interaction between an infant and caregiver, where both respond to each other’s signals, such as when a baby’s smile evokes a smile in return. This form of interactional synchrony is vital for a child’s development, establishing their foundational trust and shaping future relationships and learning.

Exploratory Behavioral System

When infants feel safe and secure, and their attachment system is deactivated, their energy can be devoted to what Bowlby (1969) refers to as the exploratory behavioral system. 

The exploratory behavioral system refers to behaviors that drive the organism to interact with the environment in a bid to inspect it, manipulate it, and master it (Mikulincer & Shaver, 2007). 

According to Bowlby (1969), the exploratory system is activated by novelty and is terminated when a person exhibits a sense of competence and familiarity with their environment.   From this perspective, attachment figures can also be seen as a ‘secure base’ which infants use to explore their social world (Ainsworth, Blehar, Waters, & Wall, 1978). 

The more assured the infant is in the availability of their attachment figure in times of stress, the more likely they will interact with others and their environment.  Thus attachment, far from interfering with exploration, is viewed as nurturing exploration. 

Caregivers who provide a secure base allow infants to become autonomous, inquisitive, and experimental.  Children who lack a secure base find their attachment system keeps overriding their attempts to be autonomous and to competently interact with their social environment. 

This, in turn, can impair and harm a child’s social, emotional, and cognitive development (Bowlby, 1980).  Of course, not all attachment figures become a secure base, and this function is based on the responsiveness of their caregiver towards the infant (Ainsworth & Wittig, 1969). 

Ainsworth et al. proposed the interconnecting between attachment and exploratory systems are adaptive as they ensure a balance between protection and exploration of the social and physical environment.

Ainsworth’s Strange Situation

Mary Ainsworth and her colleagues discovered three major patterns that infants attach to their primary caregivers (“mother figures”) from their Strange Situation Procedure (Ainsworth et al., 1978).

The study recruited four different samples of infants at around one year of age, and engaged them in the Strange Situation procedure, roughly described below:

An infant was put into an unfamiliar environment with his or her mother and was free to explore the environment; a stranger entered the room and gradually approached the infant; the mother then left the room, returning after the infant spent some time alone with the stranger.

Ainsworth and colleagues observed how comfortable each infant was physically farther away from the mother in an unfamiliar environment, how each infant interacted with the stranger, and how each infant greeted the mother upon her return.

Based on the observations, they sorted the infants into three groups: secure, anxious, and avoidant.

Attachment Styles

Attachment styles refer to the particular way in which an individual relates to other people. The style of attachment is formed at the very beginning of life, and once established, it is a style that stays with you and plays out today in how you relate in intimate relationships and in how you parent your children.

The concept involves one’s confidence in the availability of the attachment figure for use as a secure base from which one can freely explore the world when not in distress and a safe haven from which one can seek support, protection, and comfort in times of distress.

Secure Attachment

Bowlby (1988) described secure attachment as the capacity to connect well and securely in relationships with others while also having the capacity for autonomous action as situationally appropriate.

Secure attachment is characterized by trust, an adaptive response to being abandoned, and the belief that one is worthy of love.

An infant with a secure attachment is characterized as actively seeking and maintaining proximity with the mother, especially during the reunion episode. The infant may or may not be friendly with the stranger, but always shows more interest in interacting with the mother.

Additionally, during the same situation, the infant tended to be slightly distressed during separation from the mother, but the infant rarely cried.

Ainsworth and colleagues interpreted infants who were securely attached to their mothers, showed less anxiousness and more positive attitudes toward the relationship, and were likely because they believed in their mothers’ responsiveness towards their needs.

Anxious (Ambivalent) Attachment

Anxious attachment (also called ambivalent ) relationships are characterized by a concern that others will not reciprocate one’s desire for intimacy. This is

caused when an infant learns that their caregiver or parent is unreliable and does not consistently provide responsive care towards their needs.

An anxiously attached infant is characterized as being somewhat ambivalent (and resistant) to the mother. The infant often demonstrated signs of resisting interactions with the mother, especially during the strange situation reunion episode.

However, once contact with the mother was gained, the infant also showed strong intentions to maintain such contact. Overall, ambivalent infants often displayed maladaptive behaviors throughout the Strange Situation.

Ainsworth and colleagues found ambivalent infants to be anxious and unconfident about their mothers’ responsiveness, and their mothers were observed to lack “the fine sense of timing” in responding to the infants’ needs.

As adults, those with an anxious preoccupied attachment style are overly concerned with the uncertainty of a relationship. They hold a negative working model of self and a positive working model of others.

Avoidant Attachment

Children with avoidant attachment styles tend to avoid interaction with the caregiver, and show no distress during separation. This may be because the parent has ignored attempts to be intimate, and the child may internalize the belief that they cannot depend on this or any other relationship.

An infant with an avoidant attachment was characterized as displaying little to no tendency to seek proximity with the mother.

The infant often showed no distress during separation from the mother, interacted with the stranger similarly to how he or she would interact with the mother, and showed slight signs of avoidance (turning away, avoiding eye contact, etc.) when reunited with the mother.

Ainsworth and colleagues interpreted infants’ avoidance behaviors as a defensive mechanism against the mothers’ own rejecting behaviors, such as being uncomfortable with physical contact or being more easily angered by the infants.

Disorganized (Fearful) Attachment

Main and Solomon (1986) discovered that a sizable proportion of infants did not fit into secure, anxious, or avoidant, based on their behaviors in the Strange Situation experiment. They categorized these infants as having a disorganized attachment type .

Disorganized attachment is classified by children who display sequences of behaviors that lack readily observable goals or intentions, including obviously contradictory behaviors or stilling/freezing of movements.

Main and Solomon found that the parents of disorganized infants often had unresolved attachment-related traumas, which caused the parents to display either frightened or frightening behaviors, resulting in the disorganized infants being confused or forcing them to rely on someone they were afraid of at the same time.

Stages of Attachment

Rudolph Schaffer and Peggy Emerson (1964) investigated if attachment develops through a series of stages by studying 60 babies at monthly intervals for the first 18 months of life (this is known as a longitudinal study).

The children were all studied in their own homes, and a regular pattern was identified in the development of attachment.

The babies were visited monthly for approximately one year, their interactions with their carers were observed, and carers were interviewed.

A diary was kept by the mother to examine the evidence for the development of attachment. Three measures were recorded:

• Stranger Anxiety – response to arrival of a stranger. • Separation Anxiety – distress level when separated from carer, degree of comfort needed on return. • Social Referencing – degree that child looks at carer to check how they should respond to something new (secure base).

They discovered that baby’s attachments develop in the following sequence:

Asocial (0 – 6 weeks)

Very young infants are asocial in that many kinds of stimuli, both social and non-social, produce a favorable reaction, such as a smile.

Indiscriminate Attachments (6 weeks to 7 months)

Infants indiscriminately enjoy human company; most babies respond equally to any caregiver. They get upset when an individual ceases to interact with them.

From 3 months, infants smile more at familiar faces and can be easily comfortable by a regular caregiver.

Specific Attachment (7 – 9 months)

Special preference for a single attachment figure.  The baby looks to particular people for security, comfort, and protection.  It shows fear of strangers (stranger fear) and unhappiness when separated from a special person ( separation anxiety ).

Some babies show stranger fear and separation anxiety much more frequently and intensely than others; nevertheless, they are seen as evidence that the baby has formed an attachment.  This usually develops at one year of age.

Multiple Attachment (10 months and onwards)

Many of the babies from the Schaffer and Emerson study had multiple attachments by 10 months old, including attachments to mothers, fathers, grandparents, siblings, and neighbors.

The baby becomes increasingly independent and forms several attachments. By 18 months, the majority of infants have formed multiple attachments.

The multiple attachments formed by most infants vary in their strength and importance to the infant. Attachments are often structured in a hierarchy, whereby an infant may have formed three attachments, but one may be stronger than the other two, and one may be the weakest.

The results of the study indicated that attachments were most likely to form with those who responded accurately to the baby’s signals, not the person they spent more time with.  Schaffer and Emerson called this sensitive responsiveness.

Intensely attached infants had mothers who responded quickly to their demands and, interacted with their child. Infants who were weakly attached had mothers who failed to interact.

The Lasting Impact of Early Attachment

According to Bowlby’s theory (1988), when we form our primary attachment, we also make a mental representation of what a relationship is (internal working model), which we then use for all other relationships in the future i.e., friendships, working, and romantic relationships.

The different attachment styles may be viewed as internal working models of “relationships” that evolved from event experiences (Main, Kaplan, & Cassidy, 1985).

This would suggest that early interactions with caregivers could not only shape how an infant understood and behaved in relationships (as exemplified by infant attachment styles), but that such impact could be carried forward into adult attachment .

According to Bowlby (1969) later relationships are likely to be a continuation of early attachment styles (secure and insecure) because the behavior of the infant’s primary attachment figure promotes an internal working model of relationships which leads the infant to expect the same in later relationships.

In other words, there will be continuity between early attachment experiences and later relationships. This is known as the continuity hypothesis.

In humans, attachment does not conclude in infancy, or even childhood, but instead is active throughout the lifespan, with individuals gaining comfort from both physical and mental representations of significant others (Bowlby, 1969).

It is through an individual’s internal working model that childhood patterns of attachment are carried forward across the life cycle into adolescence and adulthood.

The notion of security is still important; however, the growing emergence of autonomy is also significant as the attachment system in adults is less likely to be activated due to them being able to tolerate higher levels of distress compared to children.

During adulthood, new attachment bonds are formed which may become a significant source of support during periods of distress, or during periods of goal achievement and exploration.

Researchers have proposed that working models are interconnected within a complex hierarchical structure (Bowlby, 1980; Bretherton, 1985, 1990; Collins & Read, 1994; Main, Kaplan, & Cassidy, 1985).

For example, the highest level model comprises beliefs and expectations across all types of relationship, and lower level models hold general rules about specific relations, such as romantic or parental, underpinned by models specific to events within a relationship with a single person.

The existence of multiple mental models is supported by evidence which demonstrates considerable within-person variability in the expectations and beliefs that people hold about the self and others (Baldwin & Fehr, 1995).

Furthermore, although specific models of attachment relationships are positively associated with more overarching general working models, the correlations are small to moderate (less than .40), indicating that they comprised distinct beliefs regarding the self and significant others (Cozzarelli, Hoekstra, & Bylsma, 2000).

Likely, general mental models indicate a typical appraisal of the self and others across relationships, and relationship-specific beliefs about the self and one’s partner would plausibly represent only a part of these generalized beliefs.

Key Takeaways

• Attachment is defined as a “lasting psychological connectedness between human beings” (Bowlby, 1969, p. 194), and may be considered interchangeable with concepts such as “affectional bond” and “emotional bond.”
• Attachment is characterized by specific behaviors in children, such as seeking proximity to the attachment figure when upset or threatened (Bowlby, 1969).
• Attachment theory explains how the parent-child relationship emerges and influences subsequent development.
• A person’s first attachment is often established with the primary caregiver during infancy. However, it must be noted that attachment is not unique to infant-caregiver relationships but may also be present in other social relationships.
• Attachments of various kinds are formed through the repeated act of “attachment behaviors” or “attachment transactions,” a continuing process of seeking and maintaining a certain level of proximity to another specified individual (Bowlby, 1969).
• Because caregivers vary in sensitivity and responsiveness, not all infants attach to caregivers in the same way.

Ainsworth, M. D. S., & Bell, S. M. (1970). Attachment, exploration, and separation: Illustrated by the behavior of one-year-olds in a strange situation. Child Development, 41 , 49-67.

Ainsworth, M. D. S. (1973). The development of infant-mother attachment. In B. Cardwell & H. Ricciuti (Eds.), Review of child development research (Vol. 3, pp. 1-94) Chicago: University of Chicago Press.

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation . Lawrence Erlbaum.

Ainsworth, M. D. S. (1991). Attachments and other affectional bonds across the life cycle. In C . M. Parkes, J. Stevenson-Hinde, & P. Marris (Eds.), Attachment across the life cycle (pp. 33-51). London: Routledge.

Ainsworth. M. D. S., & Wittig, B. A. (1969) Attachment and exploratory behaviour of one-year-olds in a strange situation. In: B. M. Foss (Ed.) Determinants of infant behaviour , IV. London: Methuen, p. 111-136.

Bowlby, J. (1958). The nature of the childs tie to his mother. International Journal of Psychoanalysis, 39 , 350-371.

Bowlby J. (1969). Attachment. Attachment and loss:   Vol. 1. Loss. New York: Basic Books.

Bowlby, J., and Robertson, J. (1952). A two-year-old goes to the hospital. Proceedings of the Royal Society of Medicine, 46, 425–427.

Bowlby, J. (1973). Attachment and loss, Vol. 2: Separation . New York: Basic Books.

Bowlby, J. (1980). Attachment and loss, Vol. 3: Loss, sadness and depression . New York: Basic Books.

Bowlby, J. (1988). A secure base: Parent–child attachment and healthy human development . New York: Basic Books.

Dollard, J. & Miller, N.E. (1950). Personality and psychotherapy . New York: McGraw-Hill.

Feeney, B. C. (2004). A secure base: responsive support of goal strivings and exploration in adult intimate relationships.  Journal of Personality and Social Psychology, 87 (5), 631.

George, C., & Solomon, J. (1996). Representational models of relationships: Links between caregiving and attachment.  Infant Mental Health Journal, 17 (3), 198-216.

Harlow, H. F. & Zimmermann, R. R. (1958). The development of affective responsiveness in infant monkeys. Proceedings of the American Philosophical Society, 102, 501 -509.

Mikulincer, M., & Shaver, P. R. (2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. Advances in Experimental Social Psychology, 35 , 53-152.

Mikulincer, M., & Shaver, P. R. (2007).  Attachment in adulthood: Structure, dynamics,  and change . New York: Guilford Press.

Prior, V., & Glaser, D. (2006). Understanding attachment and attachment disorders: Theory, evidence and practice. Jessica Kingsley Publishers.

Schaffer, H. R., & Emerson, P. E. (1964). The development of social attachments in infancy. Monographs of the Society for Research in Child Development, 1-77.

Shaver, P. R., & Mikulincer, M. (2006). Attachment theory, individual psychodynamics, and relationship functioning.  The Cambridge handbook of personal relationships , 251-271.

Van Ijzendoorn, M. H., & Sagi-Schwartz, A. (2008). Cross-cultural patterns of attachment: Universal and contextual dimensions. In J. Cassidy & P. Shaver (Eds) Handbook of Attachment: Theory, Research and Clinical Applications (pp. 880-905). New York: Guildford Press.

What is attachment theory in relationships?

Attachment theory is a psychological theory developed by British psychologist John Bowlby that explains how humans form emotional bonds with others, particularly in the context of close relationships.

The theory suggests that infants and young children have an innate drive to seek proximity to their primary caregivers for safety and security, and that the quality of these early attachments can have long-term effects on social and emotional development.

What are the 4 attachments in a relationship?

Attachment theory suggests that there are four types of attachments people can develop based on their early experiences with caregivers. These four types are secure, anxious-preoccupied, avoidant-dismissive, and disorganized.

People with secure attachments are comfortable with intimacy and have positive views of themselves and others. Those with anxious-preoccupied attachments worry about being rejected and may become overly clingy in relationships.

People with avoidant-dismissive attachments may avoid close relationships and prioritize independence. Those with disorganized attachments may have difficulty regulating their emotions and behavior in close relationships due to past trauma or abuse.

Attachment styles can change over time , but understanding one’s attachment style can provide insight into how one approaches relationships and areas for personal growth.

What do psychologists mean by attachment?

Attachment in psychology refers to the emotional bond between individuals, typically seen in relationships between parents and children. It’s a crucial part of social and emotional development and impacts future relationships. Attachment can be secure or insecure (avoidant, ambivalent, or disorganized).

Related Articles

Child Psychology

Vygotsky vs. Piaget: A Paradigm Shift

Interactional Synchrony

Internal Working Models of Attachment

Learning Theory of Attachment

Stages of Attachment Identified by John Bowlby And Schaffer & Emerson (1964)

Child Psychology , Personality

How Anxious Ambivalent Attachment Develops in Children

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• My Bibliography
• Collections
• Citation manager

Save citation to file

Email citation, add to collections.

• Create a new collection
• Add to an existing collection

Add to My Bibliography

Your saved search, create a file for external citation management software, your rss feed.

• Search in PubMed
• Search in NLM Catalog
• Add to Search

Indirect reciprocity can foster large-scale cooperation

Affiliations.

• 1 Department of Psychology, University of Zurich, Zurich 8050, Switzerland.
• 2 Faculty of Economics and Business, Groningen University, Groningen 9700AB, the Netherlands.
• 3 Department of Psychology, Leiden University, Leiden 2333AK, the Netherlands.
• 4 Faculty of Behavioral and Social Sciences, University of Groningen, Groningen 9712TS, the Netherlands.
• 5 Behavioral Ecology and Sociobiology Unit, German Primate Center, Leibniz Institute for Primate Research, Göttingen 37077, Germany.
• PMID: 38913888
• DOI: 10.1073/pnas.2409894121

PubMed Disclaimer

Conflict of interest statement

Competing interests statement:The authors declare no competing interest.

Similar articles

• Indirect reciprocity undermines indirect reciprocity destabilizing large-scale cooperation. Schnell E, Muthukrishna M. Schnell E, et al. Proc Natl Acad Sci U S A. 2024 May 7;121(19):e2322072121. doi: 10.1073/pnas.2322072121. Epub 2024 Apr 29. Proc Natl Acad Sci U S A. 2024. PMID: 38683991 Free PMC article.
• Direct and indirect reciprocity among individuals and groups. Romano A, Saral AS, Wu J. Romano A, et al. Curr Opin Psychol. 2022 Feb;43:254-259. doi: 10.1016/j.copsyc.2021.08.003. Epub 2021 Aug 13. Curr Opin Psychol. 2022. PMID: 34481332 Review.
• Transient nature of cooperation by pay-it-forward reciprocity. Horita Y, Takezawa M, Kinjo T, Nakawake Y, Masuda N. Horita Y, et al. Sci Rep. 2016 Jan 20;6:19471. doi: 10.1038/srep19471. Sci Rep. 2016. PMID: 26786178 Free PMC article.
• Two distinct neural mechanisms underlying indirect reciprocity. Watanabe T, Takezawa M, Nakawake Y, Kunimatsu A, Yamasue H, Nakamura M, Miyashita Y, Masuda N. Watanabe T, et al. Proc Natl Acad Sci U S A. 2014 Mar 18;111(11):3990-5. doi: 10.1073/pnas.1318570111. Epub 2014 Mar 3. Proc Natl Acad Sci U S A. 2014. PMID: 24591599 Free PMC article.
• Human cooperation. Rand DG, Nowak MA. Rand DG, et al. Trends Cogn Sci. 2013 Aug;17(8):413-25. doi: 10.1016/j.tics.2013.06.003. Epub 2013 Jul 13. Trends Cogn Sci. 2013. PMID: 23856025 Review.

Publication types

• Search in MeSH

LinkOut - more resources

Full text sources.

• Citation Manager

NCBI Literature Resources

MeSH PMC Bookshelf Disclaimer

The PubMed wordmark and PubMed logo are registered trademarks of the U.S. Department of Health and Human Services (HHS). Unauthorized use of these marks is strictly prohibited.

Relationship between shame proneness and cooperation among chinese adolescents: a moderated mediation model

• Published: 26 June 2024

Cite this article

• Huanlin Zhang 1 ,
• Na Hao 1 , 2 ,
• Liying Cui   ORCID: orcid.org/0000-0001-5919-1946 1 &
• Aruna Wu 1  

This study examines the relationship between shame proneness and cooperation in Chinese adolescents, and the role of interpersonal trust and cognitive emotion regulation strategies (CERSs). A total of 783 students (355 boys and 374 girls; M age  = 13.85, SD  = 1.88 years) completed questionnaires measuring shame proneness, interpersonal relationships, CERSs, cooperative tendency, and cooperative behavior. The findings indicated that adolescents’ shame proneness negatively predicted cooperative behavior; however, it negatively predicted cooperative tendency only partially, and only affected the dimension of gregariousness. Adolescents’ shame proneness weakened their interpersonal trust, thus reducing their gregariousness and cooperative behavior. A preference for negative CERSs strengthened this negative relationship, whereas a preference for positive CERSs weakened the relationship. These findings suggest that we should teach adolescents positive emotion regulation strategies and guide them to adopt reasonable adjustment strategies to cope with negative events and improve their emotional regulation and social adaptability.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price includes VAT (Russian Federation)

Instant access to the full article PDF.

Rent this article via DeepDyve

Institutional subscriptions

Data availability

The data are available from the corresponding author on reasonable request.

Anastasopoulos, D. (1997). Shame in psychotherapy with adolescents. Journal of Child Psychotherapy, 23 (1), 103–123. https://doi.org/10.1080/00754179708254531

Article   Google Scholar  

Arnett, J. J. (1999). Adolescent storm and stress reconsidered. American Psychologist, 54 (5), 317–326. https://doi.org/10.1037/0003-066X.54.5.317

Article   PubMed   Google Scholar  

Balliet, D., & Van Lange, P. A. M. (2013). Trust, conflict, and cooperation: A meta-analysis. Psychological Bulletin, 139 (5), 1090–1112. https://doi.org/10.1037/a0030939

Bastin, C., Rakesh, D., Harrison, B. J., Davey, C. G., Allen, N. B., Muller, S., & Whittle, S. (2021). Feelings of shame and guilt are associated with distinct neural activation in youth. Biological Psychology, 159 , 108025. https://doi.org/10.1016/j.biopsycho.2021.108025

Behrendt, H., & Ben-Ari, R. (2012). The positive side of negative emotion: The role of guilt and shame in coping with interpersonal conflict. Journal of Conflict Resolution, 56 (6), 1116–1138. https://doi.org/10.1177/0022002712445746

Bower, G. H. (1981). Mood and memory. American Psychologist, 36 (2), 129–148. https://doi.org/10.1037/0003-066X.36.2.129

Cândea, D. M., & Szentagotai-Tătar, A. (2018). Shame-proneness, guilt-proneness and anxiety symptoms: A meta-analysis. Journal of Anxiety Disorders, 58 , 78–106. https://doi.org/10.1016/j.janxdis.2018.07.005

Cesare, C., Francesco, P., Valentino, Z., Barbara, D., Olivia, R., Gianluca, C.,... & Enrico, M. (2016). Shame proneness and eating disorders: a comparison between clinical and non-clinical samples. Eating and Weight Disorders-Studies on Anorexia, Bulimia and Obesity, 21 (4), 701–707. https://doi.org/10.1007/s40519-016-0328-y

Chen, X. P., Xie, X., & Chang, S. (2011). Cooperative and competitive orientation among Chinese people: Scale development and validation. Management and Organization Review, 7 (2), 353–379. https://doi.org/10.1111/j.1740-8784.2011.00215.x

Cramer, P. (2015). Defense mechanisms: 40 years of empirical research. Journal of Personality Assessment, 97 , 114–122. https://doi.org/10.1080/00223891.2014.947997

Cui, L. Y., Bu, W. W., Gao, Q. L., Wu, Q., Huang, Y., Han, X. G., & Luo, J. L. (2022). The development of discriminatory perception of junior high school students and influence on the cooperative behavior of internal and external groups. Acta Psychologica Sinica, 54 (03), 259–269. https://doi.org/10.3724/SP.J.1041.2022.00259

De Hooge, I. E., Breugelmans, S. M., & Zeelenberg, M. (2008). Not so ugly after all: When shame acts as a commitment device. Journal of Personality and Social Psychology, 95 (4), 933–943. https://doi.org/10.1037/a0011991

De Hooge, I. E., Zeelenberg, M., & Breugelmans, S. M. (2007). Moral sentiments and cooperation. Differential influences of shame and guilt. Cognition and Emotion, 21 (5), 1025–1042. https://doi.org/10.1080/02699930600980874

De Hooge, I. E., Zeelenberg, M., & Breugelmans, S. M. (2010). Restore and protect motivations following shame. Cognition and Emotion, 24 (1), 111–127. https://doi.org/10.1080/02699930802584466

De Rubeis, S., & Hollenstein, T. (2009). Individual differences in shame and depressive symptoms during early adolescence. Personality and Individual Differences, 46 (4), 477–482. https://doi.org/10.1016/j.paid.2008.11.019

Declerck, C. H., Boone, C., & Kiyonari, T. (2014). No place to hide: When shame causes proselfs to cooperate. Journal of Social Psychology, 154 (1), 74–88. https://doi.org/10.1080/00224545.2013.855158

Dempsey, H. L., & Mann, A. (2017). Guilt, shame, sympathy, and prosocial behavior. Journal of Articles in Support of the Null Hypothesis, 14 (1), 59–68.

Google Scholar  

Ding, F., Chao, L. M., & Zhang, C. C. (2013). The effect of shame on cooperative behavior of primary school children. Youth & Juvenile Research, 1 , 19–24. https://doi.org/10.3969/j.issn.1673-8950.2013.01.005

Ding, X. F., Gao, J., Zhang, Z. X., & Qian, M. Y. (2012). Mediation role of shame-focused coping between shame and depression in high school students. Chinese Mental Health Journal, 26 (6), 450–454. https://doi.org/10.3969/j.issn.1000-6729.2012.06.011

Dou, K., Liu, Y. Z., Wang, Y. J., & Nie, Y. G. (2018a). Willingness to cooperate: Emotion enhancement mechanism of perceived social mindfulness on cooperative behaviour. Acta Psychologica Sinica, 50 (1), 101–114. https://doi.org/10.3724/SP.J.1041.2018.00101

Dou, K., Nie, Y. G., Wang, Y. J., & Liu, Y. Z. (2018b). Tust or defence? The enhancing effect of perceived social mindfulness on cooperative behavior during interactive game. Journal of Psychological Science, 41 (2), 390–396. https://doi.org/10.16719/j.cnki.1671-6981.20180221

Drummy, M. L. (2022). Shame in adolescence: Addressing shame’s role in adolescence in psychotherapy (Unpublished doctoral dissertation). Pacifica Graduate Institute.

Dumontheil, I., Apperly, I. A., & Blakemore, S. J. (2010). Online usage of theory of mind continues to develop in late adolescence. Developmental Science, 13 (2), 331–338. https://doi.org/10.1111/j.1467-7687.2009.00888.x

Elison, J., Garofalo, C., & Velotti, P. (2014). Shame and aggression: Theoretical considerations. Aggression and Violent Behavior, 19 (4), 447–453. https://doi.org/10.1016/j.avb.2014.05.002

Fergus, T. A., Valentiner, D. P., McGrath, P. B., & Jencius, S. (2010). Shame-and guilt-proneness: Relationships with anxiety disorder symptoms in a clinical sample. Journal of Anxiety Disorders, 24 (8), 811–815. https://doi.org/10.1016/j.janxdis.2010.06.002

Ferguson, T. J. (2005). Mapping shame and its functions in relationships. Child Maltreatment, 10 (4), 377–386. https://doi.org/10.1177/1077559505281430

Ferrin, D. L., Bligh, M. C., & Kohles, J. C. (2008). It takes two to tango: An interdependence analysis of the spiraling of perceived trustworthiness and cooperation in interpersonal and intergroup relationships. Organizational Behavior and Human Decision Processes, 107 (2), 161–178. https://doi.org/10.1016/j.obhdp.2008.02.012

Gao, J., Qian, M. Y., Wang, W. Y. (2011). Cognitive emotion regulation of shame and general negative emotions. Chinese Journal of Clinical Psychology, 19 (6), 807–809. https://doi.org/10.16128/j.cnki.1005-3611.2011.06.018

Garnefski, N., Kraaij, V., & Spinhoven, P. (2001). Negative life events, cognitive emotion regulation and emotional problems. Personality and Individual Differences, 30 (8), 1311–1327. https://doi.org/10.1016/S0191-8869(00)00113-6

Gausel, N., Vignoles, V. L., & Leach, C. W. (2016). Resolving the paradox of shame: Differentiating among specific appraisal-feeling combinations explains pro-social and self-defensive motivation. Motivation and Emotion, 40 (1), 118–139. https://doi.org/10.1007/s11031-015-9513-y

Govorun, O., Fuegen, K., & Payne, B. K. (2006). Stereotypes focus defensive projection. Personality and Social Psychology Bulletin, 32 , 781–798. https://doi.org/10.1177/0146167205285556

Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach . Guilford Press.

He, M. Z. (2021). Study on the relationship and intervention among rejection sensitivity, emotion regulation strategy and interpersonal disturbance of junior high school students (Unpublished master dissertation). Central China Normal University https://doi.org/10.27159/d.cnki.ghzsu.2021.002400 .

He, Z., Chen, Y. C., & Li, Y. Q. (2017). The effect of social support, cognitive emotion regulation on aggression of junior middle school students. China Journal of Health Psychology, 25 (4), 532–536. https://doi.org/10.13342/j.cnki.cjhp.2017.04.015

Irwin, K., Edwards, K., & Tamburello, J. A. (2015). Gender, trust and cooperation in environmental social dilemmas. Social Science Research, 50 , 328–342. https://doi.org/10.1016/j.ssresearch.2014.09.002

Johnson, R. C., Danko, G. P., Darvill, T. J., Bochner, S., Bowers, J. K., Huang, Y.-H.,... Pennington, D. (1989). Cross-cultural assessment of altruism and its correlates. Personality and Individual Differences, 10 (8), 855–868. https://doi.org/10.1016/0191-8869(89)90021-4

Kaufmann, M., Quirin, M., & Baumann, N. (2022). Blaming others: Individual differences in self-projection. Personality and Individual Differences, 196 , 111721. https://doi.org/10.1016/j.paid.2022.111721

Kim, S., Thibodeau, R., & Jorgensen, R. S. (2011). Shame, guilt, and depressive symptoms: A meta-analytic review. Psychological Bulletin, 137 (1), 68–96. https://doi.org/10.1037/a0021466

Lagattuta, K. H. and Thompson, R. A. (2007). The development of selfconscious emotions and their underlying appraisals. In J. L. R. Tracy, R. W. Robins, & Tangney (Eds), The self-conscious emotions (pp. 91–113). Guilford Press.

Leach, C. W., & Cidam, A. (2015). When is shame linked to constructive approach orientation? A meta-analysis. Journal of Personality and Social Psychology, 109 (6), 983–1002. https://doi.org/10.1037/pspa0000037

Lei, L., & Wang, J. F. (2015). Relationship between medical college students’ interpersonal trust, shame and social support. China Journal of Health Psychology, 23 (2), 246–249. https://doi.org/10.13342/j.cnki.cjhp.2015.02.025

Lewis, M. (1992). Shame: The exposed self . Free Press.

Lewis, M. (1997). The self in self-conscious emotions. Annuals of the New York Academy of Sciences, 818 (1), 118–142.

Li, J., & Zhu, L. Q. (2014). Cooperation in children with high-functioning autism. Acta Psychologica Sinica, 46 (9), 1301–1316. https://doi.org/10.3724/SP.J.1041.2014.01301

Li, P. Y., Jiao, J. L., & Liu, Y. (2022). Relationship between shame and aggressive behavior in middle school students: The chain mediating role of cognitive emotion regulation strategy and anger emotion. China Journal of Health Psychology, 30 (12), 1826–1831. https://doi.org/10.13342/j.cnki.cjhp.2022.12.014

Li, S. Q., & Wang, L. S. (2020). The effect of endogenous shame and group identity on prosocial behavior. Journal of Psychological Science, 43 (4), 957–962. https://doi.org/10.16719/j.cnki.1671-6981.20200428

Liang, Y., & Cui, L. Y. (2019). Effect of narcissism on junior school students’ prosocial behavior: Moderating effect of self-esteem stability. Psychological Research Psychologische Forschung, 12 (2), 187–192. https://doi.org/10.3969/j.issn.2095-1159.2019.02.012

Liu, C. J., Li, Y. M., & Li, S. (2007). Advance in research on social dilemmas: Evidence from experimental social psychology. Advances in Psychological Science, 15 (2), 379–384. https://doi.org/10.3969/j.issn.1671-3710.2007.02.030

Loader, P. (1998). Such a shame—A consideration of shame and shaming mechanisms in families. Child Abuse Review: Journal of the British Association for the Study and Prevention of Child Abuse and Neglect, 7 (1), 44–57. https://doi.org/10.1002/(SICI)1099-0852(199801/02)7:1%3c44::AID-CAR334%3e3.0.CO;2-7

Maciejewski, D. F., van Lier, P. A., Branje, S. J., Meeus, W. H., & Koot, H. M. (2017). A daily diary study on adolescent emotional experiences: Measurement invariance and developmental trajectories. Psychological Assessment, 29 (1), 35–49. https://doi.org/10.1037/pas0000312

Maxwell, S. R., Zepeda, A., & Rzotkiewicz, A. (2018). Peers, social control, or guilt? Self-conscious emotions as explanation for antisocial behavior. Deviant Behavior, 39 (3), 322–331. https://doi.org/10.1080/01639625.2016.1272970

Mills, R. S. (2003). Possible antecedents and developmental implications of shame in young girls. Infant and Child Development: An International Journal of Research and Practice, 12 (4), 329–349. https://doi.org/10.1002/icd.308

Mills, R. S. (2005). Taking stock of the developmental literature on shame. Developmental Review, 25 (1), 26–63. https://doi.org/10.1016/j.dr.2004.08.001

Mills, R. S., Hastings, P. D., Serbin, L. A., Stack, D. M., Abela, J. R., Arbeau, K. A., & Lall, D. I. (2015). Depressogenic thinking and shame proneness in the development of internalizing problems. Child Psychiatry & Human Development, 46 (2), 194–208. https://doi.org/10.1007/s10578-013-0416-4

Muris, P., & Meesters, C. (2014). Small or big in the eyes of the other: On the developmental psychopathology of self-conscious emotions as shame, guilt, and pride. Clinical Child and Family Psychology Review, 17 (1), 19–40. https://doi.org/10.1007/s10567-013-0137-z

Nathanson, D. L. (1992). Shame and pride: Affect, sex, and the birth of the self . Norton.

Nie, Y. G., Li, T., & Li Z. X. (2011). Relationship among adolescent life events, self-consciousness and their mental crisis. China Journal of Health Psychology, 19 (4), 435–437. https://doi.org/10.13342/j.cnki.cjhp.2011.04.008

Nunney, S. J., van der Schalk, J., & Manstead, A. S. (2022). Emotion and intergroup cooperation: How verbal expressions of guilt, shame, and pride influence behavior in a social dilemma. Journal of Behavioral Decision Making, 35 (4), e2273. https://doi.org/10.1002/bdm.2273

Orth, U., Robins, R. W., & Soto, C. J. (2010). Tracking the trajectory of shame, guilt, and pride across the life span. Journal of Personality and Social Psychology, 99 (6), 1061–1071. https://doi.org/10.1037/a0021342

Ozmen, E., Ozmen, D., Cetinkaya, A. C., Taskin, E. O., & Dundar, P. E. (2009). The relationship between gender and depression, self-esteem, hopelessness, submissive acts, guilt, shame and anger in adolescents. In E. P. Lamont (Eds.), Social Psychology: New Research (pp.135–146). Nova Science Publishers.

Passanisi, A., Gervasi, A. M., Madonia, C., Guzzo, G., & Greco, D. (2015). Attachment, self-esteem and shame in emerging adulthood. Procedia-Social and Behavioral Sciences, 191 , 342–346. https://doi.org/10.1016/j.sbspro.2015.04.552

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88 (5), 879–903. https://doi.org/10.1037/0021-9010.88.5.879

Qi, S. H., Zhang, T., Li, F. R., & Li, Z. W. (2008). Development of a middle school students’ shame scale. Chinese Journal of Clinical Psychology, 16 (6), 599–601. https://doi.org/10.16128/j.cnki.1005-3611.2008.06.017

Qiu, F. H., Luo, Y. J., & Jia, S. W. (2016). The influence of individual aggression on categorical perception of angry expression. Acta Psychologica Sinica, 48 (8), 946–956. https://doi.org/10.3724/SP.J.1041.2016.00946

Reimer, M. S. (1996). “Sinking into the ground”: The development and consequences of shame in adolescence. Developmental Review, 16 (4), 321–363. https://doi.org/10.1006/drev.1996.0015

Ren, Y. S. (2019). The relationship between narcissistic personality and negative risk-taking behavior of secondary vocational school students: The moderating effect of cognitive emotion regulation strategy (Unpublished master dissertation). Anhui Normal University.

Rodriguez Mosquera, P. M., Manstead, A. S. R., Fischer, A. H., & Zaalberg, R. (2008). Attack, disapproval, or withdrawal? The role of honour in anger and shame responses to being insulted. Cognition & Emotion, 22 (8), 1471–1498. https://doi.org/10.1080/02699930701822272

Roos, S., Hodges, E., & Salmivalli, C. (2014). Do guilt- and shame-proneness differentially predict prosocial, aggressive, and withdrawn behaviors during early adolescence? Developmental Psychology, 50 (3), 941–946. https://doi.org/10.1037/a0033904

Rotter, J. B. (1967). A new scale for the measurement of interpersonal trust. Journal of Personality, 35 (4), 651–665. https://doi.org/10.1111/j.1467-6494.1967.tb01454.x

Rudolph, K. D., Hammen, C., & Daley, S. E. (2006). Mood disorders. Behavioral and emotional disorders in adolescents: Nature, assessment, and treatment. In D. A. Wolfe, & E. J. Mash (Eds.), Behavioral and emotional disorders in adolescents: Nature, assessment, and treatment (pp. 300–342). Guilford Press.

Schimmenti, A. (2012). Unveiling the hidden self: Developmental trauma and pathological shame. Psychodynamic Practice, 18 (2), 195–211. https://doi.org/10.1080/14753634.2012.664873

Silverman, M. J. (2019). Songwriting to target state shame, guilt, and pride in adults with substance use disorder on a detoxification unit: A cluster-randomized study. Substance Use & Misuse, 54 (8), 1345–1354. https://doi.org/10.1080/10826084.2019.1580742

Simonds, L. M., John, M., Fife-Schaw, C., Willis, S., Taylor, H., Hand, H.,... & Winton, H. (2016). Development and validation of the adolescent shame-proneness scale. Psychological assessment, 28 (5), 549–562. https://doi.org/10.1037/pas0000206

Smiley, P. A., Rasmussen, H. F., Buttitta, K. V., Hecht, H. K., Scharlach, K. M., & Borelli, J. L. (2020). Parent control and child shame: Associations with children’s task persistence and depressive symptoms in middle childhood. Parenting, 20 (4), 311–336. https://doi.org/10.1080/15295192.2019.1694837

Sun, X. L., & Wu, M. Z. (2011). Relationships between self-esteem, rejection sensitivity, interpersonal trust and social anxiety of undergraduates. Chinese Journal of Clinical Psychology, 19 (4), 537–539. https://doi.org/10.16128/j.cnki.1005-3611.2011.04.012

Tangney, J. P., & Tracy, J. L. (2012). Self-conscious emotions. In M. R. Leary & J. P. Tangney (Eds.), Handbook of self and identity (pp. 446–478). Guilford Press.

Tangney, J. P., Stuewig, J., & Mashek, D. J. (2007). Moral emotions and moral behavior. Annual Review of Psychology, 58 (1), 345–372. https://doi.org/10.1146/annurev.psych.56.091103.070145

Article   PubMed   PubMed Central   Google Scholar  

Tangney, J. P., Stuewig, J., & Martinez, A. G. (2014). Two faces of shame: The roles of shame and guilt in predicting recidivism. Psychological Science, 25 (3), 799–805. https://doi.org/10.1177/0956797613508790

Tangney, J. P., Wagner, P., & Gramzow, R. (1992). Proneness to shame, proneness to guilt, and psychopathology. Journal of Abnormal Psychology, 101 (3), 469–478. https://doi.org/10.1037/0021-843X.101.3.469

Tangney, J. P., Wagner, P. E., Hill-Barlow, D., Marschall, D. E., & Gramzow, R. (1996). Relation of shame and guilt to constructive versus destructive responses to anger across the lifespan. Journal of Personality and Social Psychology, 70 (4), 797–809. https://doi.org/10.1037/0022-3514.70.4.797

Taylor, H. M. (2012). Shame in adolescence: Gender differences and links with social anxiety . University of Surrey.

van Eickels, R. L., Tsarpalis-Fragkoulidis, A., & Zemp, M. (2022). Family cohesion, shame-proneness, expressive suppression, and adolescent mental health—A path model approach. Frontiers in Psychology, 13 , 921250. https://doi.org/10.3389/fpsyg.2022.921250

Vani, M. F., Lucibello, K. M., Welsh, T., & Sabiston, C. M. (2023). Body-related shame disrupts attentional focus over time in adolescence. Journal of Adolescence, 95 (7), 1520–1527. https://doi.org/10.1002/jad.12216

Vuilleumier, P. (2005). How brains beware: Neural mechanisms of emotional attention. Trends in Cognitive Sciences, 9 (12), 585–594. https://doi.org/10.1016/j.tics.2005.10.011

Wang, L. S. (2017). The regulation of shame emotion in adolescence . (Unpublished doctor dissertation). East China Normal University.

Wang, X. D., Wang, X. L., & Ma, H. (Eds). (1999). Rating scales for mental health . Chinese Mental Health Journal Publisher.

Wang, Y. L., Chen, H. L., & Yuan, Y. (2020). Effect of social exclusion on adolescents’ self-injury: The mediation effect of shame and the moderating effect of cognitive reappraisal. Journal of Psychological Science, 43 (2), 333–339. https://doi.org/10.16719/j.cnki.1671-6981.20200211

Wang, Y. L., Chen, H. L., Qin, Y. L., & Lin, X. Y. (2019). The self-punishment function of adolescents’ self-injury: From guilt or shame?. Psychological Development and Education, 35 (2), 219–226. https://doi.org/10.16187/j.cnki.issn1001-4918.2019.02.11

Wang, Y., & Li, M. J. (2020). Constructive or destructive? A dual-pathway model of shame. Advances in Psychological Science, 28 (8), 1325–1336. https://doi.org/10.3724/SP.J.1042.2020.01325

Wei, H. Y. (2014). Study on the relationship between adolescent Cooperative tendency and self-esteem, interpersonal trust and perspective taking . (Unpublished master dissertation). Shanghai Normal University.

Wojcik, K. D., Cox, D. W., & Kealy, D. (2019). Adverse childhood experiences and shame-and guilt-proneness: Examining the mediating roles of interpersonal problems in a community sample. Child Abuse & Neglect, 98 , 104233. https://doi.org/10.1016/j.chiabu.2019.104233

Wu, J., Wu, H. C., Hsieh, C. M., & Ramkissoon, H. (2022a). Face consciousness, personal norms, and environmentally responsible behavior of Chinese tourists: Evidence from a lake tourism site. Journal of Hospitality and Tourism Management, 50 , 148–158. https://doi.org/10.1016/j.jhtm.2022.01.010

Wu, Q., Zhang, Y. X., He, W., & Cui, L. Y. (2022b). The relationship between adolescents’ materialism and cooperative propensity: The mediating role of greed and the moderating role of awe. Personality and Individual Differences, 189 , 111484. https://doi.org/10.1016/j.paid.2021.111484

Xavier, A., Pinto Gouveia, J., & Cunha, M. (2016). Non-suicidal self-injury in adolescence: The role of shame, self-criticism and fear of self-compassion. Child & Youth Care Forum, 45 , 571–586. https://doi.org/10.1007/s10566-016-9346-1

Xie, X. F., Yu, Y. Y., Chen, X., & Chen, X. P. (2006). The measurement of cooperative and competitive personality. Acta Psychologica Sinica, 38 (1), 116–125.

Zhu, X. Z., Luo, F. S., & Yao, S. Q. (2007). Reliability and validity of the cognitive emotion regulation questionnaire-chinese version. Chinese Journal of Clinical Psychology, 15 (2), 121–124. https://doi.org/10.3969/j.issn.1005-3611.2007.02.004

Download references

Author information

Authors and affiliations.

Department of Psychology, Shanghai Normal University, Shanghai, 200234, China

Huanlin Zhang, Na Hao, Liying Cui & Aruna Wu

Baoding Vanguard Road Primary School, Baoding, 071000, China

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Liying Cui .

Ethics declarations

Conflict of interest.

We have no know conflict of interest to disclose.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Zhang, H., Hao, N., Cui, L. et al. Relationship between shame proneness and cooperation among chinese adolescents: a moderated mediation model. Curr Psychol (2024). https://doi.org/10.1007/s12144-024-06248-2

Download citation

Accepted : 06 June 2024

Published : 26 June 2024

DOI : https://doi.org/10.1007/s12144-024-06248-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Shame proneness
• Cooperation
• Interpersonal trust
• Cognitive emotion regulation strategies
• Find a journal
• Publish with us
• Track your research

What’s wrong with shame?

Shame gets a bad rep these days. It appears to be some kind of psychological scapegoat. Allegedly, shame can explain why people feel really bad all the time or have bad relationships . It could be a sign of religious fanaticism. The “walk of shame” is viewed as a consequence of sexual promiscuity. Some influencers go as far as to label shame as “dangerous.”

I’ll zag here and make the opposite claim. Shame is useful. It’s beneficial. And it’s necessary for healthy social development.

But what exactly is shame? Shame stems from our moral conscience . It’s a signal that goes off when we’ve done something wrong in social contexts. When people are caught breaking a rule or violating a norm, they may feel shame as a result.

We go to great lengths to avoid feeling ashamed. It’s painful for a reason—it keeps our dark impulses in check. It’s our mind’s way of letting us know when we’ve done something bad, in part because we care about our social reputations and don’t want others to view us unfavorably. Without shame, people would behave like lunatics. When we describe someone as “shameless,” we mean it as a pejorative.

What makes shame different from guilt?

Some might push back at my claim by noting that guilt (not shame) is actually the emotion that occurs after we’ve done something bad and that shame is different because it’s about one’s feelings of self-worth as a whole person. Others might posit that shame is a more “public” emotion that occurs when others judge us, and we internalize this judgment, whereas guilt is a more “private” emotion that occurs when people feel that they’ve violated their own personal standards.

I disagree with both of these framings.

Research suggests that shame can be experienced internally and externally but that in the majority of cases , people feel these emotions publicly. Private feelings of shame or guilt are less common. Mostly, people experience these feelings in the presence of others that they know well (e.g., significant others, family members), although embarrassment tends to be more common in front of acquaintances or strangers.

Studies show that shame and guilt overlap in substantial ways . People generally rate them as intense and unpleasant feelings that lasted a while and occurred in very serious situations. Participants who described their experiences with each emotion said that they felt personally responsible for their (wrong) actions, a sense of responsibility, and a desire to make amends. They also felt other co-occurring emotions, such as anger and disgust, which were directed inward.

That means people felt high arousal negativity as a result of their own actions, which warranted them to display humbleness and seek forgiveness . People were also harsher on themselves than they believed others felt towards them. Overall, the evidence suggests that psychological experiences of guilt and shame are more similar than different.

Is shame actually useful?

OK, so shame and guilt are comparable emotions, and they mostly occur in the presence of others. But does that mean they’re helpful? What do people want to do with these feelings?

To answer this, researchers asked people about instances where they felt guilt, shame, regret, or embarrassment and also asked participants what they wanted to do in response to those emotions. They found that people reported strong motivations not only for social repair (e.g., to make an apology) but also to improve themselves (e.g., “I felt the urge to be a better person”). This motivation for self-improvement was even stronger following shame than for guilt, embarrassment, or regret.

The researchers noted that, in some cases, guilt can be alleviated by simply apologizing but not necessarily changing one’s behavior moving forward. The reason why shame might feel weightier and more difficult to alleviate is because it doesn’t go away with a mere apology—it requires holistic self-development. Shame causes people to want to be better humans. And that’s not easy! Despite the anguish, shame had the strongest potential to promote growth, compared to other emotions. This makes shame a double-edged sword.

The study authors ended their paper with a quote attributed to Blaise Pascal: “The only shame is to have none.” I suggest we all dwell on this idea.

Ferreira, C., Moura-Ramos, M., Matos, M., & Galhardo, A. (2022). A new measure to assess external and internal shame: Development, factor structure and psychometric properties of the external and internal shame scale. Current Psychology, 41 (4), 1892-1901.

Lickel, B., Kushlev, K., Savalei, V., Matta, S., & Schmader, T. (2014). Shame and the motivation to change the self. Emotion, 14 (6), 1049.

Tangney, J. P., Miller, R. S., Flicker, L., & Barlow, D. H. (1996). Are shame, guilt, and embarrassment distinct emotions? Journal of Personality and Social Psychology, 70 (6), 1256.

Dylan Selterman, Ph.D., is an Associate Teaching Professor at Johns Hopkins University in the Department of Psychological and Brain Sciences. He teaches courses and conducts research on personality traits, happiness, relationships, morality/ethics, game theory, political psychology, and more.

• Find a Therapist
• Find a Treatment Center
• Find a Psychiatrist
• Find a Support Group
• Find Online Therapy
• International
• New Zealand
• South Africa
• Switzerland
• Asperger's
• Bipolar Disorder
• Chronic Pain
• Eating Disorders
• Passive Aggression
• Personality
• Goal Setting
• Positive Psychology
• Stopping Smoking
• Low Sexual Desire
• Relationships
• Child Development
• Self Tests NEW
• Therapy Center
• Diagnosis Dictionary
• Types of Therapy

At any moment, someone’s aggravating behavior or our own bad luck can set us off on an emotional spiral that threatens to derail our entire day. Here’s how we can face our triggers with less reactivity so that we can get on with our lives.

• Emotional Intelligence
• Gaslighting
• Affective Forecasting
• Neuroscience