Skip to main content
NCBI home page
Social Cognitive and Affective Neuroscience logoLink to Social Cognitive and Affective Neuroscience
. 2023 Mar 6;18(1):nsad010. doi: 10.1093/scan/nsad010

Voluntary or reluctant? Social influence in charitable giving: an ERP study

Qiang Xu 1,✉,#, Shengnan He 2,#, Zhurong Li 3, Ran Duan 4, Peng Li 5,6
PMCID: PMC10013733  PMID: 36881686

Abstract

Social information has substantial influences on prosocial behavior. In this study, we performed an event-related potential (ERP) experiment to examine the effect of social influence on giving. The participants were allowed to form an initial decision on how much money to donate to a charity provided the program’s average donation amount and to make a second donation decision. Social influence varied in different directions (upward, downward and equal) by altering the relative donation amount between the average donation amount and the participants’ first donation amount. The behavioral results showed that participants increased their donation amount in the upward condition and decreased it in the downward condition. The ERP results revealed that upward social information evoked larger feedback-related negativity (FRN) amplitudes and smaller P3 amplitudes than in the downward and equal conditions. Furthermore, the pressure ratings, rather than the happiness ratings, were associated with the FRN patterns across the three conditions. We argue that people in social situations are more likely to increase their donations owing to pressure than voluntary altruism. Our study provides the first ERP evidence that different directions of social information evoke different neural responses in time course processing.

Keywords: charitable giving, social influence, FRN, P3, social pressure

Introduction

Charitable giving has become a popular topic and has attracted increasing academic interest from various disciplines, including behavioral economics, social psychology and cognitive neuroscience. Traditional economic theory holds that humans have an inherent tendency to pursue the maximization of self-interest; however, charitable giving has revealed that people are frequently prepared to forego self-interest to transfer their material resources to those in need. Why do people voluntarily pay a cost to support and help others? One of the most compelling motivations is that people enjoy giving. For example, one may feel contented as long as public goods increase or get a ‘warm glow’ by helping others (Becker, 1974; Andreoni, 1990; Harbaugh et al., 2007; Kuss et al., 2013). In addition to altruistic motivations, charitable giving is also frequently driven by social pressure, such as the need to conform to social norms or the passive acceptance of a request for help (Andreoni and Rao, 2011; Meer, 2011; DellaVigna et al., 2012). However, these two forms of motivation lead to entirely different behaviors. Altruistic givers experience good feelings from helping and do not shy away from further opportunities to help, whereas givers driven by social pressure are more willing to minimize their altruistic conduct when there is no pressure (Cain et al., 2014).

Researchers have found that social information1 has a considerable effect on charitable giving; that is, people adjust their giving behavior in the direction of conformity when they learn of others’ actions (Shang and Croson, 2009). Previous research has shown that upward social information (e.g. informing donors of the donation amount provided by a more generous donor) effectively increases contributions (Shang et al., 2007). van Teunenbroek and Bekkers (2020) discovered that when individuals were shown an average amount that outperformed the actual average amount in a crowdfunding platform, they donated 17% more than those who had shown the actual information. However, social information does not always boost donation behavior. For instance, some studies have demonstrated that downward social information diminishes donation amounts. Croson and Shang (2008) found that participants lowered their contribution if other donors donated less than they had previously donated. Nook et al. (2016) demonstrated that although others’ generosity increased people’s donations, participants who witnessed others’ stingy behavior donated small amounts. However, not all studies have found the effect that upward social information promotes donating and downward social information reduces giving. Frey and Meier (2004) and Sanders (2017) demonstrated that although upward social comparison increased prosocial intentions, downward social comparison did not reduce prosocial intentions. In addition, some studies indicated that social information does not affect people’s giving (Kubo et al., 2018).

There are two possible explanations for these inconsistent results. First, in many previous studies, participants were instructed to make decisions after viewing social information (Frey and Meier, 2004; Shang et al., 2007; Agerström et al., 2016; Nook et al., 2016; Sanders, 2017). In these studies, the baseline levels of the experimental participants’ willingness to donate before experimental manipulation were unknown. For instance, some participants may have been willing to donate more money than others, even in the upward social information condition, whereas others may have planned to donate less money than in the downward social information. In other words, the effects of social influence may interact with participants’ baseline willingness to donate, thus leading to discrepant results across different studies. Second, the motivation for donating varies across individuals; that is, individuals motivated by voluntary altruism may behave differently from those motivated by social pressure in similar situations. As previous studies on social influence have usually investigated external behavior, people’s motivations underlying giving behavior are unclear. As a result, we have limited knowledge of the psychological mechanisms underlying the effects of social influence.

The purpose of the present study was to investigate the patterns of social influence on giving behavior and their underlying mechanisms. We adopted the experimental paradigm used by Chierchia et al. (2020), which allows participants to choose their donation amount to a charity before viewing the social information. In addition, the participants are allowed to modify their initial donation amount and decide their final donation amount after viewing the social information (upward, downward or equal). Examining the differences between the first and second contribution decisions can provide us with an intuitive understanding of the influence of social information on giving behavior. We further used event-related potential (ERP) technology to monitor individuals’ brain activity while viewing social information. Although people with different incentives respond differently to social information, they may exhibit the same behavioral modifications. As a result, previous laboratory behavior experiments and field experiments have been unable to discern the changes in participants’ reactions. ERP measurements have a high time resolution and can reveal an individual’s time-dependent processing of social information. Thus, by comparing early and late ERP components, it is possible to distinguish potential differences in social information evaluation.

The ERP components most closely associated with the outcome evaluation were the feedback-related negativity (FRN) and P3. The FRN is a negative frontal-central deflection in the waveform, occurring 250–350 ms after the feedback (Miltner et al., 1997). Previous studies have demonstrated that FRN is evoked when obtained feedback differs from the expected feedback, regardless of whether it is positive or negative (Talmi et al., 2013; Sambrook and Goslin, 2014; Heydari and Holroyd, 2016). Holroyd and Coles proposed that the difference between negative and positive feedback stimuli is mainly driven by the latent component, namely reward positivity (RewP) (Cockburn and Holroyd, 2018). The FRN/RewP has been linked with the reward prediction error (RPE)—that is, the deviation between the expected and actual values—in the reinforcement learning literature (Umemoto and Holroyd, 2016). The term RPE has also been frequently applied in broader situations, even when no learning occurs, such as in guessing (Yacubian et al., 2006) or charity donation (Kuss et al., 2013). Inspired by these studies, we assumed that external outcomes (e.g. upward/downward social information) might also elicit an RPE signal when it is mismatched with the expected outcome. Thus, observing the trends in RewP2 could reveal this effect. Another component associated with outcome evaluation is P3, which is a positive wave that occurs 300–600 ms after the results or external feedback are presented (Yeung and Sanfey, 2004). Numerous studies have shown that positive findings result in larger P300 amplitudes than negative results (Bellebaum et al., 2010; Polezzi et al., 2010). P3 may also show a higher level of motivational and emotional appraisal of the stimulus (Yeung and Sanfey, 2004; Li et al., 2010).

In this study, we examined the effect of social information (upward, downward and equal) on charitable donation decisions and elucidated the rapid neural reactions associated with this effect. Our methods allow for a simple test of voluntary altruism versus social pressure in a social context. If people donate mainly voluntarily, upward social information would act as positive feedback by providing a signal that the charity is of good quality or that the level of average altruism is high (Romano and Yildirim, 2001; Potters et al., 2005; Martin and Randal, 2008). This would enhance the perceived trustworthiness and efficacy of that charity, and participants would be more comfortable with their donation decisions when they receive such information (Croson et al., 2009; Reyniers and Bhalla, 2013). In contrast, downward social information may be negative feedback, as it conveys the opposite message. In this case, downward social information would evoke a larger FRN and smaller P3. Correspondingly, participants should report higher happiness ratings for upward social information than for social information in other directions. Conversely, if social pressure is the primary motivator of giving in the social context, upward social information would provide negative feedback because it places undue pressure on people to donate more money (DellaVigna et al., 2012; Name-Correa and Yildirim, 2016). Participants would feel a higher level of pressure in this condition than in others. In contrast, downward social information would not produce this pressure and may allow people to properly lower their own donations. In this condition, downward social information would result in a lower FRN and a greater P3 than upward social information.

Methods

Participants

We recruited 28 undergraduate and graduate students from Shenzhen University for this study (14 women, age range: 18–25 years). All the participants were right-handed and had normal or corrected vision, and none reported a history of mental illness or substance abuse. The study was approved by the Medical Ethical Committee of Shenzhen University. The participants provided informed consent before the experiment. The final reward was an attendance fee (50–60 Yuan) plus the money remaining (from an initial amount of 30 Yuan) after the participants had decided to donate to the charity. Four participants were excluded from further analyses because of extreme donations in over two-thirds of the trials, and one participant was removed from the final data analysis owing to excessive artifacts. Therefore, the data obtained from 23 participants were used for further analysis.

Stimulus materials

A total of 360 charities covering disease relief, poverty alleviation and disaster relief, education assistance, nature conservation, cultural relic protection, equal employment and psychological care were selected from the Tencent Charity in the WeChat applet. These charities were chosen to balance their genres. Another 30 participants independent of the participants in the formal experiment (15 women; age range: 18–25 years) were also enrolled to donate money (0–30 Yuan) to these 360 charities. The participants were informed that one charity would be selected randomly at the end of the experiment, that they would truly donate to it and that the remaining money would be credited to their account. The average donation amount was calculated for each of the 360 charity projects, and 180 charity projects ranked in the middle were chosen as stimulus materials for the formal electroencephalogram (EEG) experiment.

Experimental procedures

The experiment was conducted in a room with electromagnetic shielding, and the participants were provided 30 Yuan to donate to each charity. After the experiment, one randomly selected charity would receive a real donation, and the remaining money that had not been donated (from 30 Yuan) would be deposited in the participants’ personal accounts.

The experimental protocol of a single trial is depicted in Figure 1A. Each trial began with the presentation of a fixation for 1000 ms, followed by a description of the charity’s name and the recipient’s predicament. After the participants had read the message, they could press the space bar. Following this, the participants made their first donation by moving a sliding bar to determine the donation amount. To avoid a biased anchor, the initial position of the cursor was always at 15 Yuan. Every time a participant pressed the ‘F’ or ‘J’ key, 1 Yuan would be subtracted or added to the donation amount. The ‘G’ or ‘H’ key corresponds to subtracting or adding 5 Yuan to the donation amount. After this adjustment, the participants had to press the space bar to confirm their final donation amount. After an interval of 800–1200 ms, the average donation amount of the charity was presented for 2000 ms. Next, a blank screen was presented for 800–1200 ms. Afterward, the participants were required to make a second donation using the same bar as the first donation. The experimental task consisted of 180 trials, including 60 trials for each of the following conditions: upward, downward and equal. The average donation amount was set specifically on the basis of the first donation by an adaptive algorithm, and the specific parameters are listed in Table 1. Some trials with extreme donations were excluded to ensure homogeneity in the same experimental conditions. All the excluded cases are listed in Table 1 (including these trials would not change the main results and the analysis with all the trials is presented in Supplementary Material). Participants could make donation decisions without time limitations and were allowed a short rest period after every 60 trials.

Fig. 1.

Fig. 1.

Open in a new tab

Experimental paradigm. (A) Experiment trial of charitable donations. (B) Control trial.

Table 1.

The parameter settings of the adaptive algorithm and exclusion

First donation (FD) Social information Average donation Excluded
0, 1 Upward Random (FD +5, 29) √
Downward 1 √
Equal 1 √
2–5 Upward Random (FD +5, 29)
Downward Random (1, FD −1) √
Equal FD
6–24 Upward Random (FD +5, 29)
Downward Random (1, FD −5)
Equal Random (FD −2, FD +2)
25–28 Upward Random (FD +1, 29) √
Downward Random (1, FD −5)
Equal FD
29, 30 Upward 29 √
Downward Random (1, FD −5) √
Equal 29 √
Open in a new tab

We designed a control task in order to verify that the differences in EEG between experimental conditions were mainly derived from the processing of the social information underlying the stimulus, rather than the physical attributes of the stimulus or the simple process of numerical comparison. As shown in Figure 1B, each trial began with a fixation (500 ms), followed by the number 15 (800 ms). After an interval of 800–1200 ms, a second number (randomly selected from 0 to 30) was displayed for 1200 ms. The participants were instructed to judge whether the second number was higher or lower than the first number after the second interval of 800–1200 ms. The participants could press the ‘F’ (higher) or ‘J’ (lower) key to make their choice. The control task consisted of 60 trials, including 30 trials with higher numbers and 30 with lower numbers.

Before the formal experiment, the participants were required to practice several trials until they were familiar with the experimental process. After the tasks had been completed, the participants were instructed to rate the pressure and pleasantness of each experimental condition on a 9-point Likert-type scale (pressure level: 1 = very stress-free, 9 = very stressful; pleasantness: 1 = very unpleasant, 9 = very pleasant).

EEG recordings and analysis

Continuous EEG recordings were performed using a 64-electrode head cap (Brain Products GmbH, Munich, Germany) according to the standard international 10–20 recording system (sampling rate: 1000 Hz, band-pass: 0.01–100 Hz). The ground electrode was placed on the medial frontal line between the Fz and FPz. The FCz was selected as an online reference and was offline re-referenced to an average reference. An electrode placed below the right eye was used to record the vertical electrooculogram (VEOG). The impedance of each electrode was maintained <10 kΩ.

The EEG data were preprocessed with the Analyzer 2.0 software (Brain Products GmbH) using a 0.1–30 Hz band-pass filter for EEG data filtering and a corrected baseline of −200 ms. The time window was set at −200–1000 ms. Independent component analysis was performed to eliminate VEOGs, and epochs with amplitudes exceeding ±80 μV were excluded as artifacts. Finally, the EEG epochs were averaged for upward, downward and equal conditions separately. After rejecting the artifacts and removing the trials involving the excluded conditions, the average number of trials for the upward condition was 49 ± 7, those for the downward condition was 47 ± 10 and those for the equal condition was 52 ± 7. In previous studies, FRN effects were found to be most pronounced at the frontal and central electrode sites (Holroyd and Coles, 2002; Heldmann et al., 2008). However, the P3 component reaches its maximum amplitude in the parietal-medial region (Carlson et al., 2009; Cui et al., 2013; Tamburin et al., 2014). The topography of the FRN and P3 in our experiment is in line with our expectations. Therefore, we selected FCz electrodes for FRN analysis and Pz electrodes for P3 analysis.

Statistical analysis

Behavioral data were statistically analyzed using the IBM SPSS Statistics software, version 22 (IBM, New York, NY, USA). Descriptive data are presented as mean ± SE. Repeated-measures analysis of variance (ANOVA) was performed on the donation amount, the subjective ratings of emotion, FRN amplitudes and P3 amplitudes to examine the social influence on the donation amount. Multiple comparisons were corrected using Bonferroni method.

Results

Behavioral results

A two-factor repeated-measures ANOVA revealed a significant interaction between social information and sequence (F (2, 44) = 46.64, P < 0.001, η2 = 0.68). In the upward social information condition, the donation amount of the second donation (12.97 ± 0.75) was significantly higher than that of the first donation (10.17 ± 0.54; P < 0.001). Contrarily, in the downward social information condition, the donation amount of the second donation (9.48 ± 0.63) was significantly lower than that of the first donation (12.04 ± 0.63; P < 0.001). In the equal social information condition, there was no significant difference between the donation amounts of the first (10.96 ± 0.68) and the second (10.82 ± 0.67) (Figure 2A). There was no significant main effect of donation sequence (F (1, 22) = 0.03, P = 0.864) and social information (F (2, 44) = 3.25, P = 0.063). Additionally, we compared the differences in change rates, which were calculated as (second donation − first donation)/(average donation − first donation), between the upward and downward social information using a paired t-test. The results showed that the change rate in the downward condition (0.25 ± 0.04) was larger than that in the upward condition (0.16 ± 0.03), t = 2.32, P = 0.03, Cohen’s d = 0.57 (Figure 2B).

Fig. 2.

Fig. 2.

Open in a new tab

(A) FD and second donation amount in the upward, downward and equal conditions. (B) The donation change rates of the participants in the upward and downward conditions. (C) Ratings of the pressure felt by the participants during the donation task. (D) Ratings of pleasantness felt by the participants. The bars represent SEM; **P < 0.01, ***P < 0.001.

Social information had a significant effect on pressure ratings (F (2, 44) = 32.46, P < 0.001, η2 = 0.60; Figure 2C). Upward social information induced a higher (4.39 ± 0.37) level of pressure than downward social information (2.26 ± 0.34, P < 0.001) and equal information (1.52 ± 0.18, P < 0.001). Nevertheless, there were no significant differences in pressure levels between the downward and equal conditions (P = 0.181). Social information also had a significant effect on pleasantness ratings (F (2, 44) = 24.24, P < 0.001, η2 = 0.52; Figure 2D). The participants were happiest in the equal conditions (7.26 ± 0.23), and there was no significant difference in pleasantness levels between the upward (4.35 ± 0.31) and downward conditions (4.74 ± 0.35; P = 1.00).

ERP results

FRN

Figure 3 shows the grand average ERP waveforms at the FCz electrode site. For the FRN, we analyzed the average amplitudes during 300–350 ms at the FCz site. The effect of social information was significant (F (2, 44) = 5.39, P = 0.008, η2 = 0.20). The FRN levels induced by upward social information (−0.89 ± 0.67) were more negative than those induced by the downward (0.26 ± 0.65, P = 0.034) and equal (0.17 ± 0.72, P = 0.038) social information. However, there was no significant difference between the latter two conditions (P = 1.00).

Fig. 3.

Fig. 3.

Open in a new tab

The grand mean ERP waveforms and topographic map at 290–340 ms for the three experimental conditions (solid lines) and two control conditions (dotted lines).

To further confirm the FRN effect, we calculated the amplitude difference between the upward/downward conditions and the equal condition. The difference waves and topography are presented in Figure 4. We established that the difference waves between the equal and upward conditions (1.27 ± 0.40) in the 300–350 ms window were significantly >0, t = 3.20, P = 0.004, whereas the difference waves between the equal and upward conditions in the 300–350 ms window were not significantly different from 0, t = 0.31, P = 0.762. The location of the largest difference was located in the prefrontal regions.

Fig. 4.

Fig. 4.

Open in a new tab

The differences in the ERP waveforms and topographic map at the 300–350 ms window for the equal–upward condition and the equal–downward condition.

P3

Figure 5 shows the grand average ERP waveforms at the Pz electrode site. For the P3, we analyzed the average amplitudes during the 400–550 ms window at the Pz site. The effect of social information was significant (F (2, 44) = 5.50, P = 0.007, η2 = 0.20). The P3 values evoked by upward social information (4.51 ± 0.47) were smaller than those evoked by downward (5.45 ± 0.56, P = 0.048) and equal (5.36 ± 0.40, P = 0.019) social information. The amplitudes of P3 under the downward and equal social information conditions were not significantly different (P = 1.00).

Fig. 5.

Fig. 5.

Open in a new tab

The grand mean ERP waveforms and topographic map at the 400–550 ms window for the three experimental conditions (solid lines) and two control conditions (dotted lines).

Control experiment

To verify that the ERP differences were mainly derived from cognitive activity rather than the physical attributes of the stimuli, we analyzed the brain electrical activity of the participants in the control group, who had been instructed to compare larger and smaller numbers. For this analysis, we used the average amplitudes during the 300–350 ms window at the FCz site and during the 400–550 ms window at the Pz site. A paired t-test revealed that the differences between the FRN (t (22) = 0.25, P = 0.81) and P3 (t (22) = −0.19, P = 0.85) were not significant.

Discussion

Our primary goal was to investigate the effect of three types of social information (upward, downward and equal) on people’s donation behavior. To our knowledge, this study is the first to investigate the mechanism by which social information affects ERP components. The behavioral results showed that social information had a two-way effect on people’s donation amounts: upward social information encouraged people to donate more, whereas downward social information prompted people to reduce their donation amount. This was consistent with the results of previous studies (Shang and Croson, 2009). Diverse social information also results in divergent emotional experiences. Specifically, upward social information exerts more pressure than downward and equal social information, whereas equal social information results in a higher level of pleasantness in comparison to the other two types of social information. At the neural level, our analysis indicates that upward social information evoked a higher FRN and lower P300 than downward or equal social information. Our results shed light on the temporal course of social influence on charitable donation.

Several studies have focused specifically on the effect of social information on people’s giving behavior; however, these studies have not reached a clear consensus regarding this effect. Some studies have found no evidence of the promotion of prosocial behavior by social information (Kubo et al., 2018), whereas others have found no evidence of the detrimental effects of downward social information (Frey and Meier, 2004; Sanders, 2017). Moreover, some studies have found evidence for both effects (Shang et al., 2007; Nook et al., 2016). Our results support the notion that people adjust their prosocial behavior according to the direction of social information. As abovementioned, various alternative explanations—based on various motivations—have been proposed for this effect. These motivations can fall into two basic categories: voluntary giving and passive compliance with pressure (see Cain et al., 2014 for a review). For people who donate voluntarily, the social influence effect may manifest primarily as a result of a lack of information (Reyniers and Bhalla, 2013). These people are willing to donate but are uncertain of the appropriate amount; therefore, others’ donation information becomes a critical reference. Such people should be pleased when they receive upward social information (signaling the charity’s good quality) and thus raise their donation amount (Romano and Yildirim, 2001; Vesterlund, 2003; Potters et al., 2005). This same logic applies to downward social information; people may believe that the downward social information indicates the untrustworthiness or inefficiency of the charitable project and hence lower their donation amount. However, if people donate to alleviate social pressure, their emotional experiences are quite different. Upward social information creates greater social pressure, which suggests that one is contributing less than the majority of individuals. In this case, people would donate more money under this social pressure (DellaVigna et al., 2012; Name-Correa and Yildirim, 2016). When other people donate an equal or lower amount of money, this type of social pressure is avoided. Regarding the magnitude of the change rate, we established that the change rate in the upward social information condition was significantly smaller than that in the downward social information condition. This is consistent with the results of previous studies (Croson and Shang, 2008). Meanwhile, it is evident that the participants may have been pressured by social norms to change their donation amount. In the upward social information condition, the participants were reluctant to increase their donations at their own expense, but upward social information puts pressure on them to give more money. In the downward social information condition, people can reduce their donation amount to a reasonable degree that complies with social norms while simultaneously serving their personal interests, and consequently, the rate at which people decrease donations in the downward condition becomes greater than the rate at which they increase donations in the upward condition. The fact that upward social information increased pressure in this study implies that social pressure, rather than voluntary altruism, may have been the primary motivation for charitable giving in our experimental environment.

This inference is further supported by evidence at the neural level. Upward social information elicited a larger FRN and a smaller P3 in this study. Previous ERP studies have suggested that FRN and P3 represent the various stages of processing of outcome evaluation (Hu et al., 2017). FRN is stimulated for a quick and rough evaluation at an early stage of the outcome evaluation process (Bellebaum et al., 2010) and could indicate whether the received outcome is good or bad (Hajcak et al., 2006; Gu et al., 2010; Von Borries et al., 2013). Previous studies have also found that losses induce greater FRN than gains (Gehring and Willoughby, 2002; Holroyd et al., 2004; Sidarus et al., 2017). The FRN difference waves between the upward/downward condition and the equal condition also revealed a RewP component at 300–350 ms; this may imply that equal and downward social information is positive feedback for individuals and can provide them with more reward than upward social information. The P3 component represents a more refined and precise appraisal of the stimulus at a later stage of outcome evaluation and is higher when positive feedback is provided than when negative feedback is provided (Hajcak et al., 2005; Gu et al., 2010). Thus, in the early or late stages of outcome evaluation, upward social information is processed as negative feedback, which may be stressful. This is also in line with prior research, which indicated that when people’s decisions or ideas differ from others, FRN increases and P3 decreases (Chen et al., 2012; Yu and Sun, 2013). Our findings—that is, no ERP differences in downward and equal social information conditions—also suggest an intriguing notion: although finding yourself in agreement with the majority is satisfying, being different from others does not always cause stress. Notably, no differences were found in these ERP components in our control experiments, implying that these differences were not generated by the simple processing of numerical comparisons but by the social information behind the number.

These findings have important implications for charitable fundraising. Our results revealed that social pressure was the main motivation for giving when people were given knowledge regarding other people’s giving. Although upward social information can enhance giving behavior effectively, past research has revealed that stress-motivated donors attempt to avoid the same stressful situation, which limits long-term giving and leads to loss of funds (Andreoni and Rao, 2011; DellaVigna et al., 2012; Reyniers and Bhalla, 2013). Charitable organizations can adopt this strategy for one-time or short-term fundraising to raise more money. However, presenting social information that can promote motivation for voluntary giving (such as quality certificates or a good reputation of the charity organization) may be more beneficial for long-term charity programs (Meijer, 2009; Adena et al., 2019). In addition, this is the first study to combine behavioral measures and EEG recordings to investigate people’s responses to different social information. This technique provides more objective evidence to understand the psychological and neural mechanisms underlying people’s social behaviors. However, our findings support the notion that the ERP technique alone has a considerable disadvantage in determining the complicated motives that conceal charitable donation behavior. People’s social behaviors are complicated, and combining several methods allows us to gather more knowledge to better understand them.

Our study has some limitations. For example, because there were individual variations in donation behavior, we could not ensure that all participants were influenced in the same way by social information. However, as the majority of individuals donated between 5 and 25 Yuan, our experimental manipulation should be effective for the majority of individuals. Additionally, to create a highly plausible experimental context, the average donation amount was a randomly selected number in an interval based on the participants’ first donation amount (FD). Therefore, the differences between the average donation amounts and the FDs in the same experimental conditions varied. Although we did not establish significant correlations (Ps > 0.05) between these differences and ERP activity, these differences may add some confounding aspects to the ERP results. Finally, it is important to note that our results were observed in a college student sample, who does not donate regularly and prefer to keep most of the money for themselves in the first donation. Therefore, our findings may not be generalizable to dedicated philanthropists or egomaniacs. These individuals with extreme traits have strong intrinsic altruistic or self-interested motivations to give or avoid giving, and therefore, they are less likely to be affected by social pressure. However, we conducted some exploratory tests and found that there was no significant correlation between participants’ change rate and their FD and no significant difference in change rate between the generous and stingy groups (Supplementary Material). These findings perhaps suggest that the social influence effect observed in our study was not driven by the individual generosity of our participants, except for four extreme ones.

Conclusion

In conclusion, we studied the impact of social information on charitable behavior. The experiment demonstrated that people’s giving behavior was altered in the direction of social information: upward social information increased their donation, whereas downward social information decreased their donation amount. Compared with downward and equal social information, upward social information caused greater pressure and evoked increased FRN and decreased P3. These results suggested that social pressure was the main psychological incentive for behavioral changes in response to social information. These findings contribute to our understanding of the social influence on giving behavior.

Supplementary Material

nsad010_Supp
Click here for additional data file. (601.9KB, zip)

Acknowledgements

This study was supported by the National Natural Science Foundation of China (31671158).

Footnotes

1

Social information is commonly used across the related literature to refer to information regarding another donor’s contribution. In this article, social information only refers to information regarding the average amount of donations made by others.

2

There were three feedback conditions in the present study, and it was not evident which conditions could be defined as ‘positive’ or ‘negative’. Therefore, we analyzed the classical FRN effect on the original waveform rather than on the RewP, which is commonly measured as the difference waveform between the positive and negative feedback conditions.

Contributor Information

Qiang Xu, School of Psychology, Shenzhen University, Shenzhen 518060, China.

Shengnan He, Department of Moral Education, The 21st Primary School Xiangzhou Zhuhai, Zhuhai 519000, China.

Zhurong Li, School of Psychology, Shenzhen University, Shenzhen 518060, China.

Ran Duan, School of Psychology, Shenzhen University, Shenzhen 518060, China.

Peng Li, School of Psychology, Shenzhen University, Shenzhen 518060, China; Shenzhen Key Laboratory of Affective and Social Cognitive Science, Shenzhen University, Shenzhen 518060, China.

Supplementary data

Supplementary data are available at SCAN online.

Data availability

The data that support the findings of this study are available on request from the corresponding author, in consideration of data protection; a formal data-sharing agreement is needed when the data are requested.

Authors’ contribution

Q.X. contributed to conceptualization, formal analysis, visualization and writing—original draft. S.H. contributed to conceptualization, software, investigation and writing—original draft. Z.L. contributed to software, investigation, writing—original draft. R.D. contributed to formal analysis, visualization and writing—review and editing. P.L. contributed to conceptualization, formal analysis, supervision and writing—review and editing.

Conflict of interest

The authors declared that they had no conflict of interest with respect to their authorship or the publication of this article.

References

  1. Adena M., Alizade J., Bohner F., Harke J., Mesters F. (2019). Quality certification for nonprofits, charitable giving, and donor’s trust: experimental evidence. Journal of Economic Behavior & Organization, 159, 75–100. [Google Scholar]
  2. Agerström J., Carlsson R., Nicklasson L., Guntell L. (2016). Using descriptive social norms to increase charitable giving: the power of local norms. Journal of Economic Psychology, 52, 147–53. [Google Scholar]
  3. Andreoni J. (1990). Impure altruism and donations to public goods: a theory of warm-glow giving. The Economic Journal, 100(401), 464–77. [Google Scholar]
  4. Andreoni J., Rao J.M. (2011). The power of asking: how communication affects selfishness, empathy, and altruism. Journal of Public Economics, 95(7-8), 513–20. [Google Scholar]
  5. Becker G.S. (1974). A theory of social interactions. Journal of Political Economy, 82(6), 1063–93. [Google Scholar]
  6. Bellebaum C., Polezzi D., Daum I. (2010). It is less than you expected: the feedback-related negativity reflects violations of reward magnitude expectations. Neuropsychologia, 48(11), 3343–50. [DOI] [PubMed] [Google Scholar]
  7. Cain D.M., Dana J., Newman G.E. (2014). Giving versus giving in. The Academy of Management Annals, 8(1), 505–33. [Google Scholar]
  8. Carlson S.M., Zayas V., Guthormsen A. (2009). Neural correlates of decision making on a gambling task. Child Development, 80(4), 1076–96. [DOI] [PubMed] [Google Scholar]
  9. Chen J., Wu Y., Tong G., Guan X., Zhou X. (2012). ERP correlates of social conformity in a line judgment task. BMC Neuroscience, 13(1), 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chierchia G., Piera Pi-Sunyer B., Blakemore S.J. (2020). Prosocial influence and opportunistic conformity in adolescents and young adults. Psychological Science, 31(12), 1585–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cockburn J., Holroyd C.B. (2018). Feedback information and the reward positivity. International Journal of Psychophysiology, 132, 243–51. [DOI] [PubMed] [Google Scholar]
  12. Croson R., Handy F., Shang J. (2009). Keeping up with the Joneses: the relationship of perceived descriptive social norms, social information, and charitable giving. Nonprofit Management & Leadership, 19(4), 467–89. [Google Scholar]
  13. Croson R., Shang J.Y. (2008). The impact of downward social information on contribution decisions. Experimental Economics, 11(3), 221–33. [Google Scholar]
  14. Cui J.F., Chen Y.H., Wang Y., Shum D.H., and Chan R.C. (2013). Neural correlates of uncertain decision making: ERP evidence from the Iowa Gambling Task. Frontiers in Human Neuroscience, 7, 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. DellaVigna S., List J.A., Malmendier U. (2012). Testing for altruism and social pressure in charitable giving. The Quarterly Journal of Economics, 127(1), 1–56. [DOI] [PubMed] [Google Scholar]
  16. Frey B.S., Meier S. (2004). Social comparisons and prosocial behavior: testing “conditional cooperation” in a field experiment. American Economic Review, 94(5), 1717–22. [Google Scholar]
  17. Gehring W.J., Willoughby A.R. (2002). The medial frontal cortex and the rapid processing of monetary gains and losses. Science, 295(5563), 2279–82. [DOI] [PubMed] [Google Scholar]
  18. Gu R., Ge Y., Jiang Y., Luo Y.J. (2010). Anxiety and outcome evaluation: the good, the bad and the ambiguous. Biological Psychology, 85(2), 200–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hajcak G., Moser J.S., Yeung N., Simons R.F. (2005). On the ERN and the significance of errors. Psychophysiology, 42(2), 151–60. [DOI] [PubMed] [Google Scholar]
  20. Hajcak G., Moser J.S., Holroyd C.B., Simons R.F. (2006). The feedback-related negativity reflects the binary evaluation of good versus bad outcomes. Biological Psychology, 71(2), 148–54. [DOI] [PubMed] [Google Scholar]
  21. Harbaugh W.T., Mayr U., Burghart D.R. (2007). Neural responses to taxation and voluntary giving reveal motives for charitable donations. Science, 316(5831), 1622–5. [DOI] [PubMed] [Google Scholar]
  22. Heldmann M., Rüsseler J., Münte T.F. (2008). Internal and external information in error processing. BMC Neuroscience, 9(1), 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Heydari S., Holroyd C.B. (2016). Reward positivity: reward prediction error or salience prediction error? Psychophysiology, 53(8), 1185–92. [DOI] [PubMed] [Google Scholar]
  24. Holroyd C.B., Larsen J.T., Cohen J.D. (2004). Context dependence of the event‐related brain potential associated with reward and punishment. Psychophysiology, 41(2), 245–53. [DOI] [PubMed] [Google Scholar]
  25. Holroyd C.B., and Coles M.G. (2002). The neural basis of human error processing: reinforcement learning, dopamine, and the error-related negativity. Psychological Review, 109(4), 679–709. [DOI] [PubMed] [Google Scholar]
  26. Hu X., Xu Z., Mai X. (2017). Social value orientation modulates the processing of outcome evaluation involving others. Social Cognitive and Affective Neuroscience, 12(11), 1730–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kubo T., Shoji Y., Tsuge T., Kuriyama K. (2018). Voluntary contributions to hiking trail maintenance: evidence from a field experiment in a national park, Japan. Ecological Economics, 144, 124–8. [Google Scholar]
  28. Kuss K., Falk A., Trautner P., Elger C.E., Weber B., Fliessbach K. (2013). A reward prediction error for charitable donations reveals outcome orientation of donators. Social Cognitive and Affective Neuroscience, 8(2), 216–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Li P., Jia S., Feng T., Liu Q., Suo T., Li H. (2010). The influence of the diffusion of responsibility effect on outcome evaluations: electrophysiological evidence from an ERP study. NeuroImage, 52(4), 1727–33. [DOI] [PubMed] [Google Scholar]
  30. Martin R., Randal J. (2008). How is donation behaviour affected by the donations of others? Journal of Economic Behavior & Organization, 67(1), 228–38. [Google Scholar]
  31. Meer J. (2011). Brother, can you spare a dime? Peer pressure in charitable solicitation. Journal of Public Economics, 95(7-8), 926–41. [Google Scholar]
  32. Meijer M.M. (2009). The effects of charity reputation on charitable giving. Corporate Reputation Review, 12(1), 33–42. [Google Scholar]
  33. Miltner W.H., Braun C.H., Coles M.G. (1997). Event-related brain potentials following incorrect feedback in a time-estimation task: evidence for a “generic” neural system for error detection. Journal of Cognitive Neuroscience, 9(6), 788–98. [DOI] [PubMed] [Google Scholar]
  34. Name-Correa A.J., Yildirim H. (2016). “Giving” in to social pressure. Games and Economic Behavior, 99, 99–116. [Google Scholar]
  35. Nook E.C., Ong D.C., Morelli S.A., Mitchell J.P., Zaki J. (2016). Prosocial conformity: prosocial norms generalize across behavior and empathy. Personality & Social Psychology Bulletin, 42(8), 1045–62. [DOI] [PubMed] [Google Scholar]
  36. Polezzi D., Sartori G., Rumiati R., Vidotto G., Daum I. (2010). Brain correlates of risky decision-making. NeuroImage, 49(2), 1886–94. [DOI] [PubMed] [Google Scholar]
  37. Potters J., Sefton M., Vesterlund L. (2005). After you—endogenous sequencing in voluntary contribution games. Journal of Public Economics, 89(8), 1399–419. [Google Scholar]
  38. Reyniers D., Bhalla R. (2013). Reluctant altruism and peer pressure in charitable giving. Judgment and Decision Making, 8(1), 7–15. [Google Scholar]
  39. Romano R., Yildirim H. (2001). Why charities announce donations: a positive perspective. Journal of Public Economics, 81(3), 423–47. [Google Scholar]
  40. Sambrook T.D., Goslin J. (2014). Mediofrontal event-related potentials in response to positive, negative and unsigned prediction errors. Neuropsychologia, 61, 1–10. [DOI] [PubMed] [Google Scholar]
  41. Sanders M. (2017). Social influences on charitable giving in the workplace. Journal of Behavioral and Experimental Economics, 66, 129–36. [Google Scholar]
  42. Shang J., Croson R., and Reed A. (2007). “I” give, but “we” give more: the impact of identity and the mere social information effect on donation behavior. ACR North American Advances, 34, 421–8. [Google Scholar]
  43. Shang J., Croson R. (2009). A field experiment in charitable contribution: the impact of social information on the voluntary provision of public goods. The Economic Journal, 119(540), 1422–39. [Google Scholar]
  44. Sidarus N., Vuorre M., Haggard P. (2017). How action selection influences the sense of agency: an ERP study. NeuroImage, 150, 1–13. [DOI] [PubMed] [Google Scholar]
  45. Talmi D., Atkinson R., El-Deredy W. (2013). The feedback-related negativity signals salience prediction errors, not reward prediction errors. Journal of Neuroscience, 33(19), 8264–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Tamburin S., Maier A., Schiff S., et al. (2014). Cognition and emotional decision-making in chronic low back pain: an ERPs study during Iowa gambling task. Frontiers in Psychology, 5, 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Umemoto A., Holroyd C.B. (2016). Exploring individual differences in task switching: persistence and other personality traits related to anterior cingulate cortex function. Progress in Brain Research, 229, 189–212. [DOI] [PubMed] [Google Scholar]
  48. van Teunenbroek C., and Bekkers R. (2020). Follow the crowd: social information and crowdfunding donations in a large field experiment. Journal of Behavioral Public Administration, 3, 1–17. [Google Scholar]
  49. Vesterlund L. (2003). The informational value of sequential fundraising. Journal of Public Economics, 87(3-4), 627–57. [Google Scholar]
  50. Von Borries A.K.L., Verkes R.J., Bulten B.H., Cools R., De Bruijn E.R.A. (2013). Feedback-related negativity codes outcome valence, but not outcome expectancy, during reversal learning. Cognitive, Affective & Behavioral Neuroscience, 13(4), 737–46. [DOI] [PubMed] [Google Scholar]
  51. Yacubian J., Gläscher J., Schroeder K., Sommer T., Braus D.F., Büchel C. (2006). Dissociable systems for gain-and loss-related value predictions and errors of prediction in the human brain. Journal of Neuroscience, 26(37), 9530–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Yeung N., Sanfey A.G. (2004). Independent coding of reward magnitude and valence in the human brain. Journal of Neuroscience, 24(28), 6258–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Yu R., and Sun S. (2013). To conform or not to conform: spontaneous conformity diminishes the sensitivity to monetary outcomes. PLoS One, 8(5), 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

nsad010_Supp
Click here for additional data file. (601.9KB, zip)

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author, in consideration of data protection; a formal data-sharing agreement is needed when the data are requested.

Articles from Social Cognitive and Affective Neuroscience are provided here courtesy of Oxford University Press

RESOURCES