Happy Faces And Friendly Choices: Analyzing Reactivity During Social Interactions

Pedagogic literature usually emphasizes the importance of being proactive rather than reactive. It is suggested that people who are reactive are controlled by circumstances or the environment, whereas proactive people choose how they are going to respond to things (Page and Page, 2003).

However, some types of reactive behavior make up the core of reciprocity and are indispensable for successful social interactions and cooperation (Mahmoodi et al., 2018; Shaw et al., 2018). Amazingly, apart from aggression, aspects of reactive social behavior during social interactions is virtually unstudied.

For our study, we used a virtual social interaction model to manipulate the emotional display of an interaction partner and to register the actor’s responses to these manipulations. We presented participants with pictures of angry, neutral, and happy facial expressions, and they were then allowed to choose one out of three behavioral responses: “attack,” “avoid,” or “make friends.” Participants’ EEG was recorded throughout the experiment, and afterward, they filled out a personality questionnaire.

The linear mixed-effects analysis of behavioral data showed that the likelihood of participants showing less friendly and more aggressive responses increased with the increase of stimuli aversiveness. In other words, participants’ behavior in this model was partly reactive. To reveal how the association between stimuli valence and response type is mediated by brain activity, EEG data were projected into the source space, and multilevel mediation analysis was performed using emotional stimulus category as the predictor, behavioral response as the outcome, and EEG source-level oscillatory power as the mediator. Event-related theta activity in the right temporoparietal junction (rTPJ) significantly mediated the association between the predictor and the outcome.

It has been repeatedly shown that event-related theta synchronization correlates with emotion processing (Aftanas et al., 2003; Bekkedal et al., 2011; Güntekin and Başar, 2014; Knyazev, 2007). This task, both at the perceptual and at the response stages, was clearly associated with emotion. The rTPJ is involved in taking the perspective of others and inferring their mental states (Saxe, 2006; Van Overwalle and Baetens, 2009). It is considered a key node of the social brain because it plays a critical role in the theory of mind, empathy, and social interactions (Decety and Lamm, 2007; Zaitchik et al., 2010; Tang et al., 2016).

Our data show that, even in a simplified task like this one, reactive social behaviors involve the activation of higher-level cortical social circuits, and we need the ability and willingness to pay attention and correctly interpret nonverbal signs from the interaction partner, as well as to tune our own behavior accordingly. This latter notion is confirmed by the fact that the strength of mediation was positively associated with the personality trait of agreeableness, which explained up to 25% of its variability, suggesting that the brain mechanism underlying reactive social behavior is more active in agreeable individuals.

On the face of it, this association between agreeableness and reactive social behavior may seem counter-intuitive. One would expect it to be associated with friendly behavior independently of a person’s opponent’s display. However, agreeableness is specifically associated with cooperativeness (Denissen and Penke, 2008; Pothos et al., 2011), and reactive behavior in the social interaction model could actually be seen as a kind of cooperation. Overall, individuals automatically use appearance-based cues to predict the trustworthiness of their interaction partner and to condition their behavior on these judgments (Chang et al., 2010; DeSteno et al., 2012; Posten et al., 2014). Being naturally predisposed to cooperation, agreeable individuals seem to be more apt to do just that.

It is suggested that the tendency to pay attention to the mental states of others is a central aspect of agreeableness (Nettle and Liddle, 2008). Converging evidence indicates that highly agreeable people tend to make emotional attribution decisions more quickly and exhibit greater rTPJ activity during emotion attribution decisions compared to less agreeable people (Haas et al., 2015). On the other hand, low agreeableness associated with borderline personality traits is characterized by hypoactivity in the rTPJ during an emotional perspective-taking task (Haas and Miller, 2015).

Our analysis of the direction of mediation effects showed that in agreeable participants, the mediation of a reactive response is mostly driven by a tendency to respond to happy faces with friendly choices. Overall, our data show that reactivity during social interactions should not be dismissed as a kind of inferior behavior driven by circumstance or the environment but should be considered as an important part of reciprocity, which is accompanied by activation of higher-level centers of our social brain and is more pronounced in agreeable individuals who are naturally predisposed to be cooperative.