Periodic Reporting for period 4 - BRAINCODES (Brain networks controlling social decisions) Reporting period: 2022-03-01 to 2023-02-28 Summary of the context and overall objectives of the project Humans do not act in a social vacuum but our decisions always affect the people around us. Optimal control of behavior therefore requires social decision making, the ability to account for other people’s goals, beliefs, and emotions when choosing our own actions. This ability underlies many behaviors thought to be specifically human, such as trusting, sharing, strategic coordination, or the ability to follow social norms. These behaviors are essential for the welfare of individuals and societies and result in major personal and financial burden when pathologically disrupted in the context of brain disorders or criminal behavior. Understanding the origins, determinants, and pathologies of human social decision making is therefore vital for peaceful and productive societies and is a major aim of several scientific disciplines, including psychology, economics, the social sciences, biology, and medicine.BRAINCODES addresses this overall aim by elucidating the brain mechanisms that causally control our social decisions in the context of strategic, prosocial, and norm-compliant behavior. These human behaviors are thought to have evolved due to the advantage of cooperation within groups and are thought to have shaped a set of dedicated brain processes that allow us to take the perspective of others, trust and share resources even with genetically unrelated strangers, and follow abstract social norms such as fairness and honesty. The set of brain processes dedicated to controlling these social behaviors are often collectively referred to as the “social brain network”. However, it has been very challenging to precisely characterize the specific mechanisms by which our brains allow us to take strategic, prosocial, and norm-compliant social decisions. One reason for this may be that humans are the only species to show these behaviors to the full extent, making it very difficult to study comparable brain mechanisms in animal models. In humans, these brain mechanisms have mainly been studied with functional magnetic resonance imaging (fMRI), which has indeed identified a set of brain areas in which neural activity consistently correlates with aspects of strategic, prosocial, and norm-compliant behavior. However, despite these encouraging results, the specific functions of the different regions in the social brain network remain poorly understood. It is particularly unclear (1) what neural computations are carried out in these areas, (2) how these areas interact as a network to control social behavior, and (3) which aspects of neural processing in the network indeed steer behavior in a causal sense and (4) lead to pathological behavior if disrupted in brain disorders.BRAINCODES will address four open questions by employing a unique multi-method approach that integrates computational modeling of social decisions (to identify neural computations, WP1) with new combinations of multimodal neuroimaging and brain stimulation methods. Using EEG and fMRI, the project first identifies spatiotemporal patterns of functional interactions between brain areas that correlate with the computationally-defined social decision processes (to characterize network interactions, WP2). Combined brain stimulation-fMRI studies will then attempt to affect – and in fact enhance – these social decision-making processes by modulating the identified brain network patterns with novel, targeted brain stimulation protocols and measuring the resulting effects on behavior and brain activity (to identify causally necessary brain processes, WP3). Finally, the project will examine whether the identified brain network mechanisms are indeed related to pathologically altered social decisions in the elderly and in individuals with psychiatric illnesses characterized by maladaptive social behavior (WP4). Thus, BRAINCODES aims to generate a causal understanding of the brain network mechanisms that allow humans to control their social decisions, thereby elucidating a biological basis for individual differences in social behavior and paving the way for new perspectives on how disordered social behavior may be identified and hopefully remedied. Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far In the first 30 months, we have mainly focused on WP1-2 of the project, namely, identifying the decision processes underlying behavior by means of computational modelling and characterizing the underlying brain network processes by means of functional neuroimaging. This work has already led to publications, and we have also collected more datasets that are currently being analyzed or written up. Moreover, we have started WP3-4 based on the initial results, by developing and setting up data collection for experimental studies with brain stimulation methods and in elderly participants and in psychiatric and neurological patients. This work has also already led to a publication, and more studies are being set up and have entered the data collection phase.To lay the groundwork for all aims of the project, we have outlined our theoretical and methodical approach in three literature reviews that have been published. In the neuroimaging studies conducted to address WP1-2, we have conducted studies with functional magnetic resonance imaging (fMRI) and/or electroencephalography (EEG) to identify neural computations and the corresponding brain processes underlying various types of social behaviors: (1) decisions to take responsibility for others as relevant for leadership, (2) decisions whether to trust others, (3) altruistic choice, (4) moral decisions, (5) compliance with social norms, and (6) strategic decision making accounting for other people’s beliefs. Together, these studies highlight various types of neural computations and functional interactions in the social brain network that underlie distinct aspects of social behavior. Moreover, they highlight some key areas that appear to implement these computations and interact with one another (among them the temporo-parietal junction, TPJ, and the lateral and medial prefrontal cortex, lPFC and mPFC). Finally, we have established a computational model that will allow us to investigate, for all these types of behavior, how the information about social context can be integrated with non-social information to control action.Based on the results of the first set of studies, we have begun to test the causal relevance of these identified neural processes for behavior by means of brain stimulation studies (WP3). One key finding that has already emerged from these studies is that the TPJ is indeed causally relevant for moral decisions, by arbitrating the conflict between self-interested and moral motives. Moreover, we have also been able to show with combined brain stimulation and neuroimaging that the lPFC is causally relevant for compliance with social norms because it is able to increase the sensitivity of emotion-related brain areas that process our fear of being punished for breaking the norm. We are now setting up other studies to establish conceptually related causal links underlying trusting and strategic behavior. Finally, to establish the clinical relevance of these processes (WP4), we are testing various groups of people with impaired brain function due to aging, stroke, borderline personality disorder, and autism spectrum disorder. Initial results show that healthy aging leads to a decrease in the use of the neural computations underlying the ability to take strategic decisions that account for other people’s beliefs, but to a concurrent increase in altruism. We are currently investigating the degree to which these abilities (and the corresponding brain processes) are altered in the other clinical groups. Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far) The work conducted in these first 30 months has led to the establishment and validation of a comprehensive set of sensitive measures that index specific social decision processes at the level of neural computations and underlying brain processes. We have already published some of these results and approaches (concerning leadership and trusting decisions) and we are in the process of finalizing and publishing several other studies on strategic, altruistic, moral, and norm-related decision processes. Moreover, we have shown for some of these processes (moral and strategic choice processes) that they are indeed causally relevant for behavior and altered in healthy aging. We will proceed to test this also for the other domains of behavior and in various clinical groups. The measures established by the work conducted in this project go beyond the state of the art because they do more than simply indicate a correlation between brain activity and behavior: They index precisely defined neural computations that are demonstrated to be causally relevant for successful control of behavior in the healthy brain. These measures will therefore be useful for many academic disciplines: They can be used by neuroscientists to study how social and non-social information is combined during decision making, by psychologists to understand individual differences in and the development of social decision making throughout the lifespan, by social scientists to compare predictions of different theories about human social behavior, by neurobiologists to identify precursors of social brain networks in animals, by forensic scientists to study the origins of pathologically dishonest or rule-violating behavior, and by physicians to diagnose and monitor maladaptive social decision making in psychiatric disorders. MRI Scanner
