Brain activity and prosocial behavior in a simulated life-threatening situation

INTRODUCTION Human societies show the highest levels of complexity and social relationships. During evolution, humans have developed neuronal circuits dedicated to mental abilities that are fundamental to tie social bonds, such as to engage altruistic behaviors that go beyond relatedness and genetic similarities (Boyd 2006). Despite the importance of this social phenomenon, many questions are greatly unsolved (Lieberman 2012, Singer 2012). Here, we aimed at extending the knowledge about the neurophysiology of prosocial decision making, by combining Virtual Reality (VR) with Independent Component Analysis (ICA) of fMRI data. This approach was tailored to avoid two main shortcomings in social neuroscience: on one hand, we provided a contextually rich environment, closer to real-world conditions than classical experimental paradigms (Bohil et al. 2011); on the other hand, we decoded brain activity during a flowing experience, when no a priori models of signal changes were available (Spiers and Maguire 2007, Beckmann 2012). METHODS The behavior of 43 young adults (age 21-30 years; 30 females) during a life-threatening situation was evaluated in a virtual environment, in which the participants had to evacuate a building on fire, in the attempt to save their lives. Toward the end, without being previously informed, participants encountered a trapped virtual human that asked for help. Thus, participants were faced with the dilemma of helping it or not, at the possible cost of their own life (Figure 1). Functional data acquired during the task were preprocessed with SPM8 and the GIFT toolbox (Calhoun et al. 2001) was used to decompose the datasets into sets of independent components (ICs). Statistical tests were carried on IC maps to identify commonalities and differences at a group level. Indeed, participants could be classified in 3 groups considering their behaviors: those who successfully helped the virtual human (SH Group, n=16), who ignored it (NoH group, n=19) or who started helping it, without succeeding (UnSH group, n=8). RESULTS Group ICA revealed several functional brain networks, similar to those previously reported in the literature during both resting state and active tasks (Smith et al. 2009, Bressler and Menon 2010, Arbabshirani et al. 2012). These networks are related both to the processing of sensory features of stimuli and to high-order cognitive functions. Interestingly, two networks showed significant differences between the participants who succeeded in acting prosocially and those who did not (Figure 2). In particular, weaker functional connections were observed between anterior insula (AI) and anterior mid cingulate cortex (aMCC) in the group of participants who acted prosocially, whereas the same group showed greater functional connectivity in a cortical network including the medial prefrontal cortex (mPFC). CONCLUSIONS By combining VR with ICA on fMRI data, we detected patterns of functional connected areas associated with the flowing experience in a highly stressful situation which required engaging in prosocial decision-making. Importantly, we report that prosocial behavior varies between participants and that this variability is predicted by differential connectivity in dedicated functional brain networks. Our findings suggest that the higher degree of functional connectivity among AI and aMCC in the two groups that did not help the virtual human reflects the higher level of personal distress in these participants, consequently resulting in the decision not to risk personal damage and therefore act selfishly. Conversely, the significant result in the mPFC supports the hypothesis that this area is involved in altruistic decision (Waytz et al., 2012). Thus, the interplay between these two networks is likely to determine the behavioral response of participants: the activity of mPFC prompts to helping behaviors, conversely, the AI and aMCC seem to be responsible for the evaluation of risk during the entire task and the prevailing self-oriented choice. REFERENCES Arbabshirani, M.R., Havlicek, M., Kiehl, K.A., Pearlson, G.D., Calhoun, V.D., (2012), 'Functional network connectivity during rest and task conditions: A comparative study', Hum Brain Mapp, vol. no. pp. Beckmann, C.F., (2012), 'Modelling with independent components', NeuroImage, vol. 62, no. 2, pp. 891-901. Bohil, C.J., Alicea, B., Biocca, F.A., (2011), 'Virtual reality in neuroscience research and therapy', Nat Rev Neurosci, vol. 12, no. 12, pp. 752-762. Boyd, R., (2006), 'Evolution. The puzzle of human sociality', Science, vol. 314, no. 5805, pp. 1555-1556. Bressler, S.L., Menon, V., (2010), 'Large-scale brain networks in cognition: emerging methods and principles', Trends Cogn Sci, vol. 14, no. 6, pp. 277-290. Calhoun, V.D., Adali, T., Pearlson, G.D., Pekar, J.J., (2001), 'A method for making group inferences from functional MRI data using independent component analysis', Hum Brain Mapp, vol. 14, no. 3, pp. 140-151. Lieberman, M.D., (2012), 'A geographical history of social cognitive neuroscience', Neuroimage, vol. 61, no. 2, pp. 432-436. Singer, T., (2012), 'The past, present and future of social neuroscience: a European perspective', NeuroImage, vol. 61, no. 2, pp. 437-449. Smith, S.M., Fox, P.T., Miller, K.L., Glahn, D.C., Fox, P.M., Mackay, C.E., Filippini, N., Watkins, K.E., Toro, R., Laird, A.R., Beckmann, C.F., (2009), 'Correspondence of the brain's functional architecture during activation and rest', Proc Natl Acad Sci U S A, vol. 106, no. 31, pp. 13040-13045. Spiers, H.J., Maguire, E.A., (2007), 'Decoding human brain activity during real-world experiences', Trends Cogn Sci, vol. 11, no. 8, pp. 356-365.