The shorter a man is, the more he defends fairness: relative height disadvantage promoting third-party punishment—evidence from inter-brain synchronization

Abstract

Third-party punishment occurs in interpersonal interactions to sustain social norms, and is strongly influenced by the characteristics of the interacting individuals. During social interactions, height is the striking physical appearance features first observed, height disadvantage may critically influence men’s behavior and mental health. Herein, we explored the influence of height disadvantage on third-party punishment through time-frequency analysis and electroencephalography hyperscanning. Two participants were randomly designated as the recipient and third party after height comparison and instructed to complete third-party punishment task. Compared with when the third party’s height is higher than the recipient’s height, when the third party’s height is lower, the punishment rate and transfer amount were significantly higher. Only for highly unfair offers, the theta power was significantly greater when the third party’s height was lower. The inter-brain synchronization between the recipient and the third party was significantly stronger when the third party’s height was lower. Compared with the fair and medium unfair offers, the inter-brain synchronization was strongest for highly unfair offers. Our findings indicate that the height disadvantage-induced anger and reputation concern promote third-party punishment and inter-brain synchronization. This study enriches research perspective and expands the application of the theory of Napoleon complex.

Introduction

Third-party punishment (TPP) refers to situations where a third party not directly affected by a violation willingly incurs a cost to penalize the violator (Fehr and Gächter 2002). Compared with second-party punishment imposed by the direct victim (Fabbri and Carbonara 2017), TPP is more credible and could minimize violations of social norms (Fehr and Fischbacher 2004; Sun et al. 2015). Given that TPP is crucially involved in societal evolution (Lewisch et al. 2011), it is imperative to thoroughly analyze its underlying motivations.

We live in a society where physical appearance is important not just because it influences how others respond to us, but because it also affects how we perceive ourselves (Judge and Cable 2004). Physical characteristics, such as facial appearance, height, skin color, and weight, hugely influence social interactions (Hersch 2011; Little and Roberts 2012). For example, previous research revealed that facial attractiveness influences TPP (Putz et al. 2016). Furthermore, height, as another remarkable physical appearance first observed by others in social interactions, is associated with male attractiveness (Little and Roberts 2012). Height affects peoples’ perception of an individual’s abilities, power, and social status (Yang et al. 2017). Compared with taller people, shorter individuals are generally perceived as having less persuasive power (Young and French 1996) and poorer job performance (Judge and Cable 2004). Height influences not only an individual’s social perception of others but also their cognitive abilities (Case and Paxson 2008; Kanazawa and Reyniers 2009), personality (Melamed 1992), and mental health (Yang et al. 2017). The Napoleon complex is a key manifestation of the influence of height on individual psychology. According to the evolutionary theory of the Napoleon complex, men’s lower height is a disadvantage that makes them feel inferior (Knapen et al. 2018). Under the current cultural ecosystem, due to social pressure and stereotypes, men are generally dissatisfied with the difference between their actual and ideal height (Chen et al. 2006), leading to psychological difficulties. As Prieto and Robbins (1975) suggested, “Since it can be demonstrated that this culture positively values a tall stature, particularly for males, the consequences of a short stature deserve attention from behavioral scientists.” Examining the influence of short height on individuals could help improve their mental well-being (Yang et al. 2017).

Many experts believe that height gives men an economic and social advantage (Stulp and Barrett 2016) as tall people generally have an advantage over short individuals in terms of status, prestige, and leadership (Blaker et al. 2013). Furthermore, tall stature is associated with higher income (Judge and Cable 2004), authority status in the workplace (Gawley et al. 2009), and suitability for military ranks (Mazur et al. 1984). Taller men could also have an advantage in professional and social competitions (Nettle 2002). In addition, height is positively associated with various socioeconomic well-being indicators such as wealth, income, education, happiness, and success (Undurraga et al. 2012). Male tallness is also associated with increased reproductive success, as taller males easily find mates (Nettle 2002). However, the evolutionary theory of the Napoleon complex posits that short men may have developed a flexible status psychology that enables them to exercise behavioral flexibility, compensating for their height disadvantage (Knapen et al. 2018; Wu et al. 2021). For example, shorter men may take more risks when competing against their taller counterparts (Wu et al. 2021). In addition, shorter football referees may compensate for their lack of dominance during a game by issuing more yellow and red cards (McCarrick et al. 2020). Knapen et al. (2018) also discovered that shorter males were more likely to show resource-securing and indirect aggressive behaviors when confronted with a taller same-sex rival. Furthermore, in a romantic relationship, short men are more likely to be jealous of their partners, a phenomenon interpreted as compensating for their height disadvantage (Brewer and Riley 2009). Overall, research has demonstrated that shorter men make up for their height disadvantage by acquiring more resources, increasing indirect aggression, and becoming more sensitive to competitors (Wu et al. 2021). However, can they also compensate for their height disadvantage through other means? TPP is regarded by bystanders as a signal of trustworthiness and prosociality, punishers may receive benefits, such as being preferred as partners, rewards from others (Barclay 2006), and getting more assistance from others in future interpersonal interactions (Nelissen 2008; Horita 2010; Santos et al. 2011). Given that it evolves and ultimately bestows evolutionary benefit to individuals, TPP is considered adaptive (Roos et al. 2014). In this regard, will a third party engage in TPP based on this evolutionary advantage to compensate for their height disadvantage?

Previous research suggested that a personal disadvantage is the leading cause of prosocial behavior (Han et al. 2009). Compared with people in non-disaster areas (advantage), people in disaster areas (disadvantage) are more likely to exhibit cooperative behavior (Rao et al. 2011). Furthermore, studies established that children on other people’s turf (disadvantage) show a more cooperative behavior than children on their own turf (advantage) (Han et al. 2009). However, the influence of a third party’s height disadvantage on TPP, an essential prosocial behavior for maintaining social fairness, remains unclear and should be explored further. As a result, this study will establish whether a third party’s height disadvantage can promote TPP to expand the explanation range of the evolutionary theory of the Napoleon complex and elucidate the specific influence of personal disadvantages on prosocial behavior. It is also noteworthy that the majority of existing research has investigated the factors influencing TPP from the perspective of third parties and violators. However, it was discovered that the recipient’s status information, rather than the violators’ status information, strongly influenced TPP (Cui et al. 2019). This study will delve into a more detailed analysis of how height disadvantage influences TPP from the recipient’s perspective to enrich the research perspective of TPP and comprehensively analyze its motivations. As earlier stated, TPP is a prosocial behavior that maintains fairness. Therefore, exploring how the height disadvantage propagates TPP will provide valuable empirical insights for maintaining social fairness and norms and strengthening effective social governance, as well as more references for promoting prosocial conduct.

Behavioral manifestation of the influence of personal disadvantage on prosocial behavior has gained more attention among researchers. However, the underlying cognitive neural mechanism remains unclear. Moreover, as mentioned earlier, the effect of height disadvantage on male psychology is quite complex, necessitating a thorough investigation of its neural mechanism to understand its influence on TPP. The theta band is related to the cognitive control required to perform TPP (Rodrigues et al. 2015). Furthermore, an increase in prosocial decisions correlates with frontal theta activity (Lavín et al. 2023), and mid-frontal theta activity predicts upcoming fair offers, potentially indicating altruistic motivation or empathy on a physiological level (Rodrigues et al. 2015). Theta band activities are sensitive to the observation factors, in the observed condition, theta power was enhanced relative to finishing the gambling task alone (Huang and Yu 2018). Theta band is also related to anger (Krause et al. 2000). The relative deprivation theory holds that individuals will have negative emotions such as anger when they think they are relatively at a disadvantage (Mishra and Carleton 2015). The activation of delta and alpha bands is also involved in TPP. Delta band activation is linked to the potential influence of the group environment on the emotional factors involved in TPP (Gao et al. 2023). While low-frequency neural oscillations (such as the delta band) may be associated with emotional and motivational processes, the alpha band may be more related to cognitive processes (Zhan et al. 2020) and could aid in punishment decision-making. The alpha band is critical in the top-down processing of physical appearance (Kang et al. 2015). It also integrates and processes social information, ultimately contributing to making effective judgments on social information (Perry et al. 2010). Given that the theta band is closely associated with personal disadvantage-induced anger, it helps reveal the influence of height disadvantage on the internal neural mechanism of TPP. This study attempts to further explore whether the height characteristics of observers and actors affect the theta band and whether the height disadvantage-induced anger will lead to theta band activation.

TPP is a complex behavior that occurs in social interactions that require at least 2 brains for mutual communication (Jahng et al. 2017). In this regard, the precise mechanism underlying how our brains make decisions regarding punishment during real social interactions remains unclear. Due to the intricate nature of social interactions and advancements in neuroscience technologies, researchers explored the inter-brain synchronization (IBS) between recipients and third parties in TPP using hyperscanning technology (Astolfi et al. 2015; Ciaramidaro et al. 2018). Hyperscanning refers to the simultaneous recording of the neural activity of more than 1 person during social interactions (Montague et al. 2002). IBS, which has garnered significant attention in hyperscanning research, refers to the functional connection changes of neural activities in the brain of 2 or more people during a social interaction (Czeszumski et al. 2020). The electroencephalography (EEG)-based hyperscanning technology offers higher temporal resolution, enabling the precise recording of the neural mechanisms underlying real-time social interactions. Astolfi et al. (2015) investigated the IBS between recipients and third parties in TPP, elucidating the impact of diverse interaction scenarios and the unfairness of distribution schemes on IBS. Another study evaluated IBS changes during the process of establishing interpersonal fairness, revealing a higher IBS in the temporoparietal junction (TPJ), which indicates that punishment could enhance interpersonal fairness via IBS promotion (Zhang et al. 2019). To better realize the goals of this study, we further explored how the relative height characteristics of both recipients and third parties affect IBS from the perspective of interactive objects’ height characteristics. Specifically, we employed the EEG-based hyperscanning technology as a reliable tool to comprehensively explore how the relative height difference of both receivers and third parties influences TPP. According to research, IBS originates from similar cognitive and emotional processes (empathy, mentalizing, and so on.) (Tang et al. 2016; Peng et al. 2021). The mentalizing-related regions of the social brain network, such as the TPJ, are an important neural index for measuring interpersonal interaction. The right TPJ (rTPJ) can evaluate violators’ behavior, infer others’ behavioral intentions, and integrate the resulting information, making it crucial in implementing TPP decisions (Young et al. 2011). Moreover, the rTPJ is involved in the trade-off between the altruistic tendency and the cost of altruistic behavior (Morishima et al. 2012), as well as the generation of reputation motivation (Obeso et al. 2018). Furthermore, empathy is closely associated with TPP (Hoffman 2001), and Astolfi et al. (2015) established a link between empathy and the inter-brain connection between the recipient and the third party. Previous research has shown that empathy promotes prosocial behavior by enhancing IBS (Peng et al. 2021). Therefore, this study will further explore the involvement of empathy in TPP promotion.

There is limited research on IBS in men with relative height differences during TPP. Given the unique behavioral characteristics of individuals’ decision-making behavior in social situations, this study will explore the influence of height disadvantage on TPP from the perspective of interpersonal interactions, thereby enriching and improving the research perspective on TPP. Furthermore, this study will comprehensively analyze the causes of this behavior and enhance the understanding and application of the Napoleon theory. In addition, this study will explore the potential IBS mechanism of TPP using EEG-based hyperscanning technology to make up for not examining the neural mechanism of TPP from an individual’s perspective. Because shorter height is a disadvantage for men, and previous studies have revealed that personal disadvantage can promote prosocial behavior, this study assumes that when the third party’s height is lower than the recipient’s, the third party will take TPP more to make up for the height disadvantage. Furthermore, individuals will be angry when disadvantaged, and theta band activation is related to anger; thus, we also assumed that more theta band power would be activated when the third party’s height is lower than the recipient’s. In addition, we assumed that when the third party’s height is lower than the recipient’s, the stronger reputation concern makes the third party to more actively infer the psychological intentions of others, and the IBS of the brain regions associated with the “theory of mind” intensifies.

Materials and methods

Participants

We calculated the optimal sample size using G*Power 3.1.9.7 for a medium effect size repeated measures design (groups = 2, f = 0.25, α = 0.05, 1 − β = 0.80) with the following parameters: F-test, with-between interaction, number of groups = 2, and number of measurements = 6. The resulting optimal sample size (N) was 24. The selection of the above parameter settings was based on prior research (Yang et al. 2023) and conformed to the standards outlined by Cohen (1992). Since various factors, such as equipment or participants could contribute to poor data quality when collecting EEG data, we recruited more subjects in each group. Thirty-four male dyads (M_age ± SD = 22.82 ± 3.01) were enrolled and randomly designated as recipients and third parties in the TPP task. They were divided into 2 groups: Third party’s height is lower than the recipient’s (N = 17 male dyads), and third party’s height is higher than the recipient’s (N = 17 male dyads). The participants received a basic payment of 15 Chinese yuan (~$2.20) and were informed of additional monetary rewards that would be calculated at the rate of 10% per the number of tokens presented on the feedback page in any trial. Ultimately, all the participants received an extra 5 yuan on top of the basic payment. This study was ethically approved by the local ethics committee. All participants provided written informed consent before participating in the experiment.

Design

A 2(Relative Height Difference: third party’s height is lower than recipient’s, third party’s height is higher than recipient’s) × 3(Offers: fair, medium unfair, and highly unfair) repeated-measures ANOVAs, with Offers as within-subject variable and Relative Height Difference as between-subject variable.

Tasks

This study adopted a TPP task based on the dictator game framework (Liu et al. 2018). In the game, the dictator (A) and the recipient (B) jointly got 100 tokens. While the dictator had the right to distribute these 100 tokens, the recipient could only passively accept and cannot refuse the tokens. There were 3 offers: 90:10, 70:30, or 50:50, representing highly unfair, medium unfair, and fair, respectively. Participants were designated as the recipient and third party and were then asked to complete the TPP task on the computer. After observing the dictator’s choice, the participant (as the third party) had 2 options: (i) not to intervene; or (ii) punish the dictator. In each trial, the third party had only 50 intervention tokens (Leliveld et al. 2012; Hu et al. 2015). The third-party participants needed to select the specific number of tokens they would spend if they chose to punish. For every 1 token the third party spent, their payoff decreased by 3 tokens, and if they chose not to intervene, they would retain all 50 tokens. As outlined in a previous study (Gao et al. 2023), the third-party participants spent a specific number of tokens among 6 choices: 5, 10, 15, 20, 25, and 30. All instructions were clearly explained to participants, emphasizing that if the third party chose to pay more tokens and exceeded the number of tokens owned by the proposer after the 1:3 conversion (Leliveld et al. 2012; Hu et al. 2015), the dictator’s token count will not become negative but 0 (Hu et al. 2015; Li et al. 2023).

Procedure

In each session, the researcher led 2 male participants who did not know each other into the lab. To create the relative height disadvantage, we followed a similar strategy as outlined in a previous study (Knapen et al. 2018; Wu et al. 2021). The 2 participants were asked to stand on opposite sides while facing each. The researcher then measured their actual height using a wall-mounted stadiometer (without shoes) and read out the results in front of the 2 participants. Following that, the researcher asked the participants to report their actual height value and answer whether they were taller or shorter than the other participant in the lab. Subsequently, the participants sat side by side in front of the computer 1 m apart in a brightly lit and electromagnetically shielded room to complete the task. They were randomly assigned as the recipient and the third party, and then the examiner emphasized their relative height differences again. The participants were instructed to avoid body movements such as head movements during the experiment.

Figure 1 shows the time course of a trial. Each trial began with the presentation of a fixation cross for 1,000 ms. Afterward, a picture was presented indicating the offer amount (for 3,000 ms). The third-party participants were then invited to make the “accept” or “reject” decision by pressing the “F” or “J” keys on the keyboard without a time limit. The participants were required to press the corresponding number keys on the next page to choose the transfer amount if they selected “punish.” The subsequent trial began 2,000 ms after the feedback appeared. Each participant completed 180 trials divided into 6 blocks of 30 trials each, equally and randomly providing the 3 conditions (fair, medium unfair, and highly unfair).

After completing the experiment, the third-party participants were invited to complete the Chinese version (Zhang et al. 2010) of the Interactional Reactivity Index (IRI-C) questionnaire (Davis 1983). The questionnaire contains 22 items divided across 4 subscales measuring different aspects of empathy: Perspective Taking (PT), Fantasy (FS), Empathic Concern (EC), and Personal Distress (PD). The tool had an acceptable reliability with a Cronbach’s α of 0.73. In filling out the questionnaire, participants rated the statements on a 5-point Likert scale with scores ranging from 1 (almost never or never true) to 5 (almost always or always true).

E‌EG data recordings

We simultaneously and continuously recorded EEG signals of each dyad using two 64-channel EEG systems (NeuroScan Inc, United States of America) with 2 electrode caps containing Ag/AgCl electrodes mounted per the extended international 10–20 system. A band-pass filter of AC 0.05 to 100 Hz (sampling rate 1,000 Hz) was used to amplify EEG signals. Two more electrodes were placed below each mastoid (M1 and M2). An electrode between CZ and FCZ was used as a reference. The midpoint of Fpz and Fz (GND) was set as the ground point. The electrode impedance was maintained below 10 kΩ throughout the experiment. For good recording quality, participants were instructed to avoid eye-blinking and eye movements during stimulus presentation.

E‌EG data preprocessing

The collected EEG data were preprocessed using the EEGLAB toolbox version 14.1 (Delorme and Makeig 2004) in MATLAB R2020b. The EEG data were down-sampled at 500 Hz. The noise was removed using high-pass and low-pass filters of 0.1 and 40 Hz, respectively. Following that, a 50 Hz notch filter was applied. The data were then segmented from −1,000 to 2,000 ms relative to the offer onset, and a mean of the data from −1,000 to 0 ms was used for baseline correction. Channels were replaced using a spherical interpolation algorithm when >20% of the trials were bad for a specific channel. Each participant’s data were subjected to Independent Component Analysis (ICA, Delorme et al. 2007), and independent components associated with artifacts were identified through visual inspection. The artifact-marked epochs were eliminated from all subsequent analyses.

Time-frequency data analysis

Given that commonly used time-frequency methods yield highly comparable results and are considered fundamentally equivalent (Le Van et al. 2001; Bruns 2004), the Short-Time Fourier Transform (STFT) method is more intuitive and intelligible (Zhang et al. 2020). Furthermore, the Hann window is the preferred window function as it tapers the data to zero on both sides, reducing spectral leakage and ensuring a more accurate estimation of frequency content within each time segment (Cohen 2014; Zhang et al. 2020). Herein, time-frequency distributions were estimated using STFT with a Hanning window fixed at 200 ms. At each time-frequency point (t, f), each trial yielded a complex time-frequency estimate F(t, f) extending from −1,000 to 2,000 ms and from 1 to 45 HZ in the time and frequency domains, respectively. The resulting spectrogram, P(t, f) = |F(t, f)|², presents the signal magnitude as a joint function of time and frequency at each time-frequency point. Figure 2 depicts each condition’s grand-averaged power.

The overall time-frequency analysis idea is to use a data-driven method to determine the time of interest, then determine the electrode points of interest in the time of interest, and finally carry out the final repeated measurement variance analysis with the theta power values of several electrodes of interest as dependent variables in the time of interest.

First, we used a data-driven method to determine the time of interest in the theta band (4–7 Hz); that is, we performed 2 (Relative Height Difference: third party’s height is lower than recipient’s, or third party’s height is higher than recipient’s) × 3 (Offers: fair, medium unfair, and highly unfair) repeated-measures ANOVAs using the theta band power values at each time point of each electrode as the dependent variable. After FDR correction, we found that the main effect of the Relative Height Difference and Offers, as well as the interaction effect between the Relative Height Difference and Offers, was significant at specific time points of each electrode (Figs 3–5). Take theta band power value at a certain electrodes (i.e. a certain vertical coordinate value in Fig. 3) and a certain time point (i.e. a certain horizontal coordinate value in Fig. 3) as the dependent variable to performed repeated-measures ANOVAs, the resulting P values of the main effect of Offer are shown in Fig. 3, the resulting P values of the main effect of interaction effect are shown in Fig. 4, and the resulting P values of the main effect of Relative Height Difference are shown in Fig. 5. The white regions in the figure indicate P > 0.05, implying that the main effect of the Offer is not statistically significant. Other colors represent values ranging from 0 to 0.05 (per the color bar), indicating a statistically significant main effect of the Offer at specific electrodes and time points. The * mark represents P values that remain significant after FDR correction. We determined ROI 1 (theta band, 500–600 ms), ROI 2 (theta band, 0–500 ms), and ROI 3 (theta band, 1,600–1,800 ms) based on the significant time ranges shown in Figs 3–5.

Second, we performed 2 (Relative Height Difference: third party’s height is lower than recipient’s, or third party’s height is higher than recipient’s) × 3 (Offers: fair, medium unfair, and highly unfair) repeated-measures ANOVAs using power in the ROI as the dependent variable to determine the electrodes of interest in each time-frequency ROI at each electrode.

In theta band (4–7 Hz, 500–600 ms), the data-driven results show that the main effect of Offers was significant at the FP1, FP2, AF3, AF4, F3, F1, FZ, FC3, FC2, and CPZ electrodes (see Table 1), the main effect of Relative Height Difference was significant at the F4, F6, FC6, and FT8 electrodes (see Table 2). In theta band (4–7 Hz, 0–500 ms), the data-driven results show that the interaction effect between the Relative Height Difference and Offers was significant at the electrodes O2, F(2, 132) = 3.29, P = 0.040, η_p² = 0.05. In theta band (4–7 Hz, 1,600–1,800 ms), the main effect of the Offer was significant at the electrodes O2, F(2, 132) = 4.67, P = 0.011, η_p² = 0.07 and at O10, F(2, 132) = 3.60, P = 0.030, η_p² = 0.05. Many studies have found that the theta band oscillation in the mid-frontal is closely related to TPP (Rodrigues et al. 2020; Gao et al. 2023). Therefore, we selected the above electrodes located in the frontal lobe from the data-driven results. Furthermore, to more concisely carry out multiple comparisons between different levels of independent variables and analyze the simple effect analysis of interaction (instead of calculating many times one by one at so many electrode points), we finally calculate the average theta power (500–6,000 ms) at FP2, AF3, AF4, F3, F1, FZ, FC3, FC2, F4, F6, and FC6, take it as the dependent variables and perform repeated-measures ANOVA again.

IBS analysis

The IBS between 2 participants was evaluated through the phase synchronization of EEG signals (Hu et al. 2017). The Phase-Locking Value (PLV) is a widely used synchronization index. The PLV ranges from 0 to 1, with 0 representing the absence of synchronization and 1 indicating perfect phase-locked oscillations between 2 signals across trials at a specific time-frequency point. The instantaneous phase time series of the EEG signal was obtained by performing a windowed Fourier transform on signals and the power spectra separated by frequency bands. The PLV at a given time (t) and frequency (f) was calculated as the absolute value of the sum of the phase differences between 2 electrodes (j, k) in a dyad across N epochs.

The PLV was calculated for all possible electrode pairs between the 2 brains at specific time and frequency ranges of interest: theta band (0–500 ms), theta band (500–600 ms), and theta band (1,600–1,800 ms). Using a data-driven method, we performed 2 (Relative Height Difference: third party’s height is lower than recipient’s, or third party’s height is higher than recipient’s) × 3 (Offers: fair, medium unfair, and highly unfair) repeated-measures ANOVAs using the PLV calculated for all possible electrode pairs between the 2 brains in each time-frequency ROI as the dependent variable. Data-driven results showed that after FDR correction using the Benjamini–Hochberg procedure, the main and interaction effects were, respectively, significant at specific electrode pairs. The repeated measurement variance analysis was reconducted using the average PLV values of the examined electrode pairs as the dependent variable.

Results

Behavioral results

Punishment rate

The results also revealed a significant main effect of the Offers, F(2, 64) = 352.02, P < 0.001, η_p² = 0.92. Participants were more likely to punish when the fairness of the offer decreased (fair vs. medium unfair, P < 0.001; medium unfair vs. highly unfair, P = 0.003; fair vs. highly unfair: P < 0.001, LSD correction, see Table 3). The main effect of the Relative Height Difference was also significant, F(1, 32) = 12.02, P = 0.002, η_p² = 0.27, when the height of the third party is lower than that of the recipient, the punishment rate will be significantly greater (see Table 3). Moreover, a significant interaction effect was observed between Relative Height Difference and Offers, F(2, 64) = 8.22, P = 0.001, η_p² = 0.20 (see Table 4). Further simple effect test showed that, for medium and highly unfair offers, the punishment rate is significantly greater when the height of the third party is lower than that of the recipient; for fair offers, there was no significant difference between Relative Height Differences (see Table 4, Fig. 6).

Transfer amount

For the transfer amount, a repeated-measures ANOVA revealed significant main effects of both Relative Height Difference (F(1, 32) = 7.59, P = 0.010, η_p² = 0.19) and Offer (F(2, 64) = 313.46, P < 0.001, η_p² = 0.91). The transfer amount was biggest when the offer was highly unfair (see Table 3). When the height of the third party is lower than that of the recipient, the transfer amount is significantly greater than that when the height of the third party is higher than that of the recipient (see Table 3). Besides, the interaction between Relative Height Difference and Offers was also significant (F(2, 64) = 4.86, P = 0.011, η_p² = 0.13). Further simple effect test showed that, for medium unfair and highly unfair offers, when the height of the third party is lower than that of the recipient, the transfer amount is significantly greater than that when the height of the third party is higher than that of the recipient. While such an effect was not found in fair offers (see Table 4, Fig. 7).

Time-frequency analysis result

In this time-frequency of interest (500–600 ms), the brain topological map of each condition is shown in Fig. 8. The main effect of Offers was not significant (F(2, 132) = 2.43, P = 0.101), and the main effect of Relative Height Difference was not significant (F(1, 66) = 2.96, P = 0.09). The interaction between Offers and Relative Height Difference was significant, F(2, 132) = 8.18, P = 0.001, η_p² = 0.11. Further simple effect test showed that, for highly unfair offers, when the height of the third party is lower than that of the recipient, the theta power is significantly greater (M ± SD = 1.57 ± 0.38 dB) than that when the height of the third party is higher than that of the recipient (M ± SD = −0.13 ± 0.38 dB, P = 0.002). Such an effect was not found in fair or medium unfair offers (see Fig. 9).

The IBS of electrodes result

In theta band 0–500 ms, data-driven results show that after FDR correction, the Offers’ main effect was significant at (P6-P2, P8-P2, PO4-P2, P8-POZ, P4-PO4, P6-PO4, O2-PO4, CP6-O2, P2-O2, P4-O2, P6-O2, P8-O2) (see Table 5 and Fig. 10). However, when the PLV value of any electrode pair is taken as the dependent variable, the main effect of the group and the interaction between offer and the Relative Height Difference were not significant (after FDR correction). Taking the average PLV values of these electrode pairs as the dependent variable, the repeated measurement variance analysis was performed again, and the results show that the main effect of the Offers was significant, F(2, 64) = 19.45, P < 0.001, η_p² = 0.38. Further multiple comparison results show that IBS in response to a highly unfair offer (M ± SD = 0.16 ± 0.004) was greater than in response to a medium unfair offer (M ± SD = 0.13 ± 0.003, P < 0.001, LSD correction), and was also greater than in response to fair offer (M ± SD = 0.14 ± 0.005, P = 0.014, LSD correction). The main effect of Relative Height Difference (F(1, 32) = 0.03, P = 0.863) and the interaction between Offers and Relative Height Difference (F(2, 64) = 0.92, P = 0.402) were not significant.

In theta band 500–600 ms, when the PLV value of any electrode pair is taken as the dependent variable, the main effect of the Relative Height Difference, the main effect of the Offers, and the interaction between the Offers and the Relative Height Difference were insignificant (after FDR correction).

In theta band 1,600–1,800 ms, data-driven results show that after FDR correction, the Relative Height Difference’s main effect was significant at (AF3-C4, F5-C4, AF3-CP4, F5-CP4) (see Table 6 and Fig. 11). However, when the PLV value of any electrode pair is taken as the dependent variable, the main effect of Offers and the interaction between Offers and Relative Height Difference were not significant (The P values were corrected with the FDR method with the Benjamini–Hochberg procedure). Taking the average PLV values of these electrode pairs as the dependent variable, the repeated measurement variance analysis was performed again, and the results show that the main effect of the Relative Height Difference was significant, F(1, 32) = 31.73, P < 0.001, η_p² = 0.50. IBS under the condition that the height of the third party is lower than that of the recipient (M ± SD = 0.13 ± 0.003) was greater than under the condition that the height of the third party is lower than that of the recipient (M ± SD = 0.11 ± 0.003, P < 0.001). The main effect of Offers (F(2, 64) = 3.03, P = 0.059) and the interaction between Offers and Relative Height Difference (F(2, 64) = 0.42, P = 0.650) were not significant.

Results of correlation analysis

The bivariate Pearson correlation analyses between the IRI score of the third party and the average punishment rate, the correlation between the IRI score of the third party and transfer amount of offers were conducted. The results show that the IRI score of the third party is positively correlated with the average transfer amount of the offers (Pearson’s r = 0.36, P = 0.038, see Fig. 12) and with the average punishment rate of the offers (Pearson’s r = 0.35, P = 0.041, see Fig. 13).

For fair offers, medium unfair offers, highly unfair offers and the average of 3 offers, the bivariate Pearson correlation analyses between band power (500–600 ms) and transfer amount, the correlation between theta band power (500–600 ms), and punishment rate were conducted, respectively. The results show that the average punishment rate of 3 offers is positively correlated with the average theta band power (500–600 ms) of 3 offers (Pearson’s r = 0.55, P = 0.001, see Fig. 14). What’s more, similar correlation analyses were conducted between the behavior data and PLV values, but there was no significant result.

Mediation effect analysis result

Before the mediation effect analysis, the bivariate Pearson correlation analyses were conducted. The results were illustrated in Table 7.

Mediation effect analysis was conducted via SPSS PROCESS macro (Model 4) to test the hypothesized models. We set the third party’s IRI score as the independent variable, the average punishment rate of 3 offers as the dependent variable, and average PLV in theta band (1,600–1,800 ms) as the mediating variable. The statistical significance of the hypothesized mediation effect was determined by the 95% confidence interval (CI) with 5000 bootstrapping resamples. The indirect effect of IBS as a mediating variable is 0.002(95% Cl = [0.0003, 0.0047]), and the complete mediating effect of IBS between empathy and punishment rate is established. The results above were depicted in Fig. 15.

Discussion

This study draws on the evolutionary theory of the Napoleon complex and adopts the dictator game-based TPP task to further explore the specific influence of personal disadvantage on prosocial behavior. Specifically, we employed the time-frequency analysis technology and EEG-based hyperscanning to explore whether the third party’s height disadvantage can promote TPP and its cognitive neural mechanism. The punishment rate and transfer amount were significantly higher when the third party was shorter than the recipient compared with when the third party was taller. Furthermore, the theta power was significantly greater, but only for the highly unfair offer when a third party was disadvantaged. In addition, the IBS between the recipient and the third-party punisher was significantly stronger in the left frontal and the right central parietal lobes when the third party was shorter than the recipient. The strongest synchronization was observed in the right parietal lobe when comparing the IBS under 3 unfair offers. Moreover, the third-party’s IRI-C scores exhibited significant positive correlations with the transfer amount and punishment rate. A significant correlation was also found between the punishment rate and the theta band power. Mediation analysis further revealed that IBS mediated the effects of the punisher’s empathy ability on TPP.

The behavior analysis results indicated a significant rise in punishment rate and transfer amount when the third party’s height was lower than the recipient’s. The TPP concept arises from the psychology of deterrence aimed at safeguarding personal interests and is considered a form of protecting against future encroachments on one’s self-interests due to other people’s actions that violate social fairness (Hauert et al. 2007; Krasnow et al. 2016). Building upon prior research (Knapen et al. 2018), we induced a relative height disadvantage by instructing participants to compare their heights face-to-face. In this regard, after performing the TPP with their taller counterparts and realizing a height disadvantage, short individuals, as a third party, may be more inclined to perceive this situation as an opportunity to show their ability to maintain social fairness and protect their future interests from infringement (Delton and Krasnow 2017). Individuals’ willingness to pay a price for fairness arises from the anticipated benefits it yields in future interactions (Santos et al. 2011). For example, third-party punishers could be the preferred cooperative partners or may be rewarded by bystanders (Barclay 2006; Nelissen 2008; Horita 2010). When considering whether or not to punish under the receiver’s observation, third parties with a shorter stature are more likely to consider the compensatory effect of future benefits on their height disadvantage. Moreover, this finding is consistent with the promotional effect of a disadvantage on cooperative behavior (Barclay 2006; Nelissen 2008; Horita 2010), which further supports the notion that personal inferiority could promote prosocial behavior (Rao et al. 2011; Miao et al. 2021). Furthermore, consistent with previous research, the punishment rate and transfer amount increased with unfairness, indicating that the offer’s unfairness significantly affected TPP (Sun et al. 2015; Cui et al. 2019).

In this study, we observed that in situations with a highly unfair offer, where the third party’s height is lower than that of the recipient, there is a significant increase in theta band power activation in the mid-frontal region. Relative deprivation is characterized by an individual’s subjective perception of being in a disadvantaged position compared with others, leading to negative emotions like anger and dissatisfaction (Smith et al. 2012). Anger elicits increased the energy in the theta band (Krause et al. 2000), and is associated with prefrontal brain activity (Harmon-Jones et al. 2004). This can explain why third parties with height disadvantage engage in more TPP, and height disadvantage affects third party punishment by activating more theta energy. Theta band oscillation participates in self-reflective thinking during social interaction as evidenced by its role in monitoring individuals’ responses to their own actions (Pablo et al. 2018). Punishing violators may establish a robust reputation, resulting in long-term advantages that outweigh any personal, temporal, or economic expenses (Gintis et al. 2001; Leimar and Hammerstein 2001; Gao et al. 2023). The prevailing knowledge underscores the crucial role of reputation plays in social interactions (Sigmund et al. 2001; Brandt and Sigmund 2006). Establishing a positive reputation requires careful management of one's behavior, aligning with societal expectations and norms. The results suggest that when the third party’s height is lower than that of the recipient, there is an amplified concern about how the recipient evaluates them. This increased sensitivity makes them more susceptible to being influenced by the recipient. Interestingly, in this study, we found that when the height of the third party is lower than the height of the recipient, for highly unfair offers, the theta band power in the mid-frontal is significantly activated. In addition, it is worth noting that activation of the theta band in the mid-frontal region reflects the essential cognitive control over punishment (Cavanagh and Frank 2014; Rodrigues et al. 2020). Consequently, in the context of highly unfair offers, the third party with lower height exerts more cognitive control thereby balancing between safeguarding their economic interests and avoiding being swayed by anger and reputation-driven motivations.

In our investigation, we observed that when the third party’s height was lower than that of the recipient, the intensity of IBS within electrode pairs situated in the prefrontal-right central parietal lobe (C4, CP4) was significantly greater compared with the level observed when the third party was taller than the recipient. It’s noteworthy that CP4 is positioned in the rTPJ (Zinchenko et al. 2021; Oku et al. 2022). Studies have indicated that TPJ can engage in the psychological process of theory of mind and mentalizing (Young et al. 2010; Feng et al. 2021). During social interaction, increased activity in the right TPJ was observed in TPP. rTPJ can judge others’ perceptions of their behavior (Saxe and Kanwisher 2003) and infer others’ intentions, making it essential for reputation concerns. Researchers believe that reputation plays a critical role in TPP (Raihani and Bshary 2015) and serves as an avenue for making prosocial behavior conform to individual interests (Zhan et al. 2022), and it is also a powerful incentive for prosocial behavior (Li et al. 2021). The quest for reputation benefits contributes to the augmentation of individual mentalizing (Tennie et al. 2010), and mentalizing stands as a significant factor in IBS development (Mayseless et al. 2019; Yoshioka et al. 2021). Consequently, the high attention individuals allocate to reputation due to a relative height disadvantage will intensify the IBS between the recipient and the third party. TPP can provide lasting advantages on the third party by enhancing its reputation, yielding an adaptive edge (Zhan et al. 2022). Research on the Napoleon complex suggests that individuals facing a disadvantage may resort to alternative actions, such as participating in risky behavior, to compensate for their height disadvantage (Wu et al. 2021). Our study suggests that the reputation benefits from TPP effectively offset their height disadvantage.

Our results showed that for highly unfair offers, IBS in electrode pairs is highest between the right parietal and the parietal occipital lobe. Existing EEG hyperscanning studies have demonstrated the specific connectivity patterns in relation to different levels of fairness in TPP, a stronger integration has been reported in the highly unfair condition (Ciaramidaro et al. 2018). Researchers have found that the different levels of empathy induced by highly unfair offers result in denser inter-subjects connectivity (Astolfi et al. 2015). The comparable cognitive and emotional processes caused by empathy will promote IBS (Peng et al. 2021). Empathy refers to the capacity to perceive and sense the emotional state of others (Decety et al. 2016). Lee and Hsieh (2014) observed an increase in IBS levels among individuals when exposed to positive or negative movie clips, as opposed to neutral videos. Emotional stimulation seems to enhance neural coupling between individuals, fostering mutual understanding and communication (Hasson et al. 2012). Therefore, in the context of highly unfair offer, the recipient and third party will have the same anger experience, yielding the highest level of IBS between electrode pairs in the right parietal lobe (Findlay et al. 2006).

Individuals with high empathy tend to share and understand the recipient’s feelings. Thus, the stronger the third party’s empathy, the stronger is the motivation to punish violators. These results further demonstrate a significant positive correlation between the third party’s empathy ability and punitive behavior, aligning with previous research findings. Results of the mediation analysis indicate that IBS between the prefrontal lobe and the right central parietal lobe only acts as a mediator for predicting punishment behavior by the third party’s empathy. Notably, IBS in interpersonal interaction promoting prosocial behavior is increasingly being studied (Feng et al. 2020). Previous studies have shown that an individual’s empathy can affect his prosocial behavior (Stocks et al. 2009). Hu et al. (2017) found that interpersonal interaction enhances IBS and promotes prosocial behavior. Hu et al. (2017) reported that interpersonal interaction enhances IBS and promotes prosocial behavior. Our study further reveals that empathy promotes TPP by enhancing IBS between individuals.

Although this study has some limitations, our results provide important ideas for future research. First, we only recruited male participants, as existing research indicates that the Napoleon complex predominantly affects males (Wu et al. 2021). However, this choice restricts the applicability of the current findings to females. Future research should investigate IBS in females within the context of TPP to enhance the comprehensiveness of the outcomes. Second, although PLV is widely calculated in the prevailing literature, PLV is susceptible to the volume conduction effect and has been mentioned by many researchers. We also observed that source localization technology is a novel method that can reduce the influence of volumetric conduction. Given that each method has its advantages and disadvantages, we will consider alternative methods, such as phase lag index, weighted phase lag index (wPLI), and wavelet transform coherence, and combine complementary techniques, such as the incorporation of nonlinear methods, to improve the robustness of IBS computations. Finally, due to the restricted spatial resolution offered by EEG hyperscanning, pinpointing exact sources and evaluating brain mechanisms poses a challenge. Integrating EEG and fNIRS devices for hyperscanning can expand our understanding of brain function in social interactions and cognitive tasks.

Conclusion

There has been extensive research on the neural mechanism of TPP. In the present study, we explored the influence of male height disadvantage on TPP and IBS using EEG hyperscanning. The results demonstrate that the height disadvantage of the male third party will enhance TPP. When a third party’s height is lower compared with that of the recipient, more theta power is activated, suggesting that the relative height disadvantage contributes to individual anger. Moreover, the reputation concern caused by the height disadvantage promotes IBS between the recipient and the third party; the anger caused by the unfair offers will also potentiate IBS. The IBS can mediate empathy-promoting the TPP. Based on the evolutionary theory of the Napoleon complex, our study highlights the promotion of male height disadvantage to TPP. Not only does it support the view that “disadvantage promotes prosocial behavior” (Miao et al. 2021) but it also provides additional confirmation of the evolutionary theory of the Napoleon complex in the realm of prosocial behavior. Furthermore, it delves into the neural mechanism of TPP.

Acknowledgments

We would like to thank the participants who took part in the study, and thanks to anonymous reviewers for their constructive suggestions on the article.

Authors’ contributions

Yujia Meng (Conceptualization, Data curation, Writing—original draft), Shuyu Jia (Data curation, Visualization, Writing—review & editing), Jingyue Liu (Data curation, Methodology, Writing—review & editing), Chenyu Zhang (Methodology, Visualization, Writing—review & editing), He Wang (Data curation, Methodology, Software, Supervision), and Yingjie Liu (Conceptualization, Resources, Supervision, Writing—original draft, Writing—review & editing)

Funding

This research was supported by The Hebei Province education science planning-General funded project. “The moderation of altruistic punishment by social environmental factors and its educational implications” (2203198); The key research project of North China University of Science and Technology in 2023. “The research on social psychological mechanism of interpersonal interaction from altruistic perspective” (ZD-RW-202319); The National education science planning-Youth project of The Ministry of Education. “The influence of social moral factors on pain empathy and its educational implications” (EBA210396).

Conflict of interest statement: None declared.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

References