Social synchronization of brain activity increases during eye-contact

ARTICLE https://doi.org/10.1038/s42003-022-03352-6 OPEN Social synchronization of brain activity increases during eye-contact 1234567890():,; Caroline Di Bernardi Luft 1 ✉, Ioanna Zioga1,2, Anastasios Giannopoulos 3, Gabriele Di Bona Nicola Binetti1, Andrea Civilini4, Vito Latora 4,5,6,7 & Isabelle Mareschal1 4, Humans make eye-contact to extract information about other people’s mental states, recruiting dedicated brain networks that process information about the self and others. Recent studies show that eye-contact increases the synchronization between two brains but do not consider its effects on activity within single brains. Here we investigate how eyecontact affects the frequency and direction of the synchronization within and between two brains and the corresponding network characteristics. We also evaluate the functional relevance of eye-contact networks by comparing inter- and intra-brain networks of friends vs. strangers and the direction of synchronization between leaders and followers. We show that eye-contact increases higher inter- and intra-brain synchronization in the gamma frequency band. Network analysis reveals that some brain areas serve as hubs linking within- and between-brain networks. During eye-contact, friends show higher inter-brain synchronization than strangers. Dyads with clear leader/follower roles demonstrate higher synchronization from leader to follower in the alpha frequency band. Importantly, eye-contact affects synchronization between brains more than within brains, demonstrating that eye-contact is an inherently social signal. Future work should elucidate the causal mechanisms behind eyecontact induced synchronization. 1 School of Biological and Behavioural Sciences, Queen Mary, University of London, London E1 4NS, United Kingdom. 2 Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, The Netherlands. 3 School of Electrical and Computer Engineering, National Technical University of Athens (NTUA), Athens, Greece. 4 School of Mathematical Sciences, Queen Mary University of London, London E1 4NS, United Kingdom. 5 Dipartimento di Fisica ed Astronomia, Università di Catania and INFN, I-95123 Catania, Italy. 6 The Alan Turing Institute, The British Library, London NW1 2DB, United Kingdom. 7 Complexity Science Hub, Josefstäadter Strasse 39, A 1080 Vienna, Austria. ✉email: COMMUNICATIONS BIOLOGY | (2022)5:412 | https://doi.org/10.1038/s42003-022-03352-6 | www.nature.com/commsbio 1 ARTICLE H COMMUNICATIONS BIOLOGY | https://doi.org/10.1038/s42003-022-03352-6 uman and non-human primates’ gaze is drawn to others’ eyes1,2. While non-human primates have a pigmented sclera, human’s sclera are white3. This morphological difference allows humans to extract a wealth of information from our conspecific’s eyes, which may shape our social interactions. For instance, humans can detect eye contact from a longer distance than nonhuman primates4 and use this information to infer other people’s mental states and intentions (for a review see5). The brain regions involved in eye-contact overlap with structures in the social brain network6, including the ventral and medial prefrontal cortex, superior temporal gyrus, fusiform gyrus, cingulate gyrus and amygdala (for a review see7), suggesting that mutual eye contact is key for inferring others’ emotions and intentions. The perception of direct eye contact in humans is consistently found to involve the superior temporal sulcus (STS)8–10, a region, which is a key part of the mentalising network that is involved in tasks that require making inferences about the mental states of others11. Research has made remarkable progress towards understanding how eye contact is processed in a single (perceiver’s) brain, but eye contact is an interactive process between two people. More recently, we have begun to extend this understanding to multiple brains—for example, the synchronization of activity between two brains has been found to increase during eye contact12–14. However, we still do not know how both intra- and inter-brain activity is integrated, nor the functional role of this synchronised activity. To address this, it is important to examine the activity of two brains simultaneously, through a process known as Hyperscanning15–19. A classical Hyperscanning EEG study demonstrated that the brains of two people interacting in an imitation paradigm synchronize in a few frequencies, including alpha mu rhythms, beta, and gamma. Hyperscanning studies have shown that higher synchronization between brains (e.g., interbrain activity) is associated with more effective social interactions20–27. For example, higher phase synchronization has been observed between the brains of parents and infants during direct eye contact22. During direct gaze, they also observed that the adult exerted a stronger influence on infant’s neural activity, evidencing that eye contact might lead to stronger modulation or affect the direction of the synchronization. Directed inter-brain synchronization has been observed in leader-follower scenarios22,28–30, a phenomenon also demonstrated in non-human animals31. Another study29 demonstrated that leaders presented stronger motor-related oscillatory patterns compared to followers when interacting in a finger-tapping task. A computational modelling study32 explained this effect by demonstrating that successful behavioural interaction requires an increase in between-unit coupling (e.g., inter-brain) and a decrease in within-unit (e.g., intra-brain) coupling. For instance, they observed that leader-follower interactions require the follower to have low within-unit coupling whereas the relationship between two leaders tends to result in low between unit coupling. Taken together, these studies suggest that individual brains’ responses might affect the dynamics of interactions, and viceversa. These findings highlight the need to understand how interactions work in the dual brain system, combining both interand intra- brain connectivity. Since eye contact is a key factor in initiating and coordinating human interactions, it is important to determine if eye contact alone (a) plays a role in establishing leader-follower dynamics, and (b) results in directed synchronisation between brains, for instance, from leader to follower. Graph theory can be used to quantify the properties of entire networks with measures that estimate how information flows through their nodes (i.e. brain areas) via their edges (i.e. connections)33. A few studies have exploited graph theory to understand the global and local characteristics of the so-called 2 hyperbrain networks which include both intra- and inter-brain connections34–36. For example, a study36 observed that the brain networks of an uncooperative dyad (two defectors in a prisoners’ dilemma game) contained less interbrain links and were more modular (i.e. stronger connectivity within brains than between brains). Therefore, the current st (...truncated)