Laboratory for Social and Neural Systems Research

The phenomenon of empathy entails the ability to share the affective experiences of others. In recent years social neuroscience made considerable progress in revealing the mechanisms that enable a person to feel what another is feeling. The present review provides an in-depth and critical discussion of these findings. Consistent evidence shows that sharing the emotions of others is associated with activation in neural structures that are also active during the first-hand experience of that emotion. Part of the neural activation shared between self- and other-related experiences seems to be rather automatically activated. However, recent studies also show that empathy is a highly flexible phenomenon, and that vicarious responses are malleable with respect to a number of factors--such as contextual appraisal, the interpersonal relationship between empathizer and other, or the perspective adopted during observation of the other. Future investigations are needed to provide more detailed insights into these factors and their neural underpinnings. Questions such as whether individual differences in empathy can be explained by stable personality traits, whether we can train ourselves to be more empathic, and how empathy relates to prosocial behavior are of utmost relevance for both science and society.

Over the last 2 decades, a large number of neurophysiological and neuroimaging studies of patients with schizophrenia have furnished in vivo evidence for dysconnectivity, ie, abnormal functional integration of brain processes. While the evidence for dysconnectivity in schizophrenia is strong, its etiology, pathophysiological mechanisms, and significance for clinical symptoms are unclear. First, dysconnectivity could result from aberrant wiring of connections during development, from aberrant synaptic plasticity, or from both. Second, it is not clear how schizophrenic symptoms can be understood mechanistically as a consequence of dysconnectivity. Third, if dysconnectivity is the primary pathophysiology, and not just an epiphenomenon, then it should provide a mechanistic explanation for known empirical facts about schizophrenia. This article addresses these 3 issues in the framework of the dysconnection hypothesis. This theory postulates that the core pathology in schizophrenia resides in aberrant N-methyl-D-aspartate receptor (NMDAR)-mediated synaptic plasticity due to abnormal regulation of NMDARs by neuromodulatory transmitters like dopamine, serotonin, or acetylcholine. We argue that this neurobiological mechanism can explain failures of self-monitoring, leading to a mechanistic explanation for first-rank symptoms as pathognomonic features of schizophrenia, and may provide a basis for future diagnostic classifications with physiologically defined patient subgroups. Finally, we test the explanatory power of our theory against a list of empirical facts about schizophrenia.

Dynamic causal modeling (DCM) is a generic Bayesian framework for inferring hidden neuronal states from measurements of brain activity. It provides posterior estimates of neurobiologically interpretable quantities such as the effective strength of synaptic connections among neuronal populations and their context-dependent modulation. DCM is increasingly used in the analysis of a wide range of neuroimaging and electrophysiological data. Given the relative complexity of DCM, compared to conventional analysis techniques, a good knowledge of its theoretical foundations is needed to avoid pitfalls in its application and interpretation of results. By providing good practice recommendations for DCM, in the form of ten simple rules, we hope that this article serves as a helpful tutorial for the growing community of DCM users.

We offer a formal treatment of choice behavior based on the premise that agents minimize the expected free energy of future outcomes. Crucially, the negative free energy or quality of a policy can be decomposed into extrinsic and epistemic (or intrinsic) value. Minimizing expected free energy is therefore equivalent to maximizing extrinsic value or expected utility (defined in terms of prior preferences or goals), while maximizing information gain or intrinsic value (or reducing uncertainty about the causes of valuable outcomes). The resulting scheme resolves the exploration-exploitation dilemma: Epistemic value is maximized until there is no further information gain, after which exploitation is assured through maximization of extrinsic value. This is formally consistent with the Infomax principle, generalizing formulations of active vision based upon salience (Bayesian surprise) and optimal decisions based on expected utility and risk-sensitive (Kullback-Leibler) control. Furthermore, as with previous active inference formulations of discrete (Markovian) problems, ad hoc softmax parameters become the expected (Bayes-optimal) precision of beliefs about, or confidence in, policies. This article focuses on the basic theory, illustrating the ideas with simulations. A key aspect of these simulations is the similarity between precision updates and dopaminergic discharges observed in conditioning paradigms.

Mathematical models of scientific data can be formally compared using Bayesian model evidence. Previous applications in the biological sciences have mainly focussed on model selection in which one first selects the model with the highest evidence and then makes inferences based on the parameters of that model. This "best model" approach is very useful but can become brittle if there are a large number of models to compare, and if different subjects use different models. To overcome this shortcoming we propose the combination of two further approaches: (i) family level inference and (ii) Bayesian model averaging within families. Family level inference removes uncertainty about aspects of model structure other than the characteristic of interest. For example: What are the inputs to the system? Is processing serial or parallel? Is it linear or nonlinear? Is it mediated by a single, crucial connection? We apply Bayesian model averaging within families to provide inferences about parameters that are independent of further assumptions about model structure. We illustrate the methods using Dynamic Causal Models of brain imaging data.

The second iteration of the Autism Brain Imaging Data Exchange (ABIDE II) aims to enhance the scope of brain connectomics research in Autism Spectrum Disorder (ASD). Consistent with the initial ABIDE effort (ABIDE I), that released 1112 datasets in 2012, this new multisite open-data resource is an aggregate of resting state functional magnetic resonance imaging (MRI) and corresponding structural MRI and phenotypic datasets. ABIDE II includes datasets from an additional 487 individuals with ASD and 557 controls previously collected across 16 international institutions. The combination of ABIDE I and ABIDE II provides investigators with 2156 unique cross-sectional datasets allowing selection of samples for discovery and/or replication. This sample size can also facilitate the identification of neurobiological subgroups, as well as preliminary examinations of sex differences in ASD. Additionally, ABIDE II includes a range of psychiatric variables to inform our understanding of the neural correlates of co-occurring psychopathology; 284 diffusion imaging datasets are also included. It is anticipated that these enhancements will contribute to unraveling key sources of ASD heterogeneity.

Computational learning models are critical for understanding mechanisms of adaptive behavior. However, the two major current frameworks, reinforcement learning (RL) and Bayesian learning, both have certain limitations. For example, many Bayesian models are agnostic of inter-individual variability and involve complicated integrals, making online learning difficult. Here, we introduce a generic hierarchical Bayesian framework for individual learning under multiple forms of uncertainty (e.g., environmental volatility and perceptual uncertainty). The model assumes Gaussian random walks of states at all but the first level, with the step size determined by the next highest level. The coupling between levels is controlled by parameters that shape the influence of uncertainty on learning in a subject-specific fashion. Using variational Bayes under a mean-field approximation and a novel approximation to the posterior energy function, we derive trial-by-trial update equations which (i) are analytical and extremely efficient, enabling real-time learning, (ii) have a natural interpretation in terms of RL, and (iii) contain parameters representing processes which play a key role in current theories of learning, e.g., precision-weighting of prediction error. These parameters allow for the expression of individual differences in learning and may relate to specific neuromodulatory mechanisms in the brain. Our model is very general: it can deal with both discrete and continuous states and equally accounts for deterministic and probabilistic relations between environmental events and perceptual states (i.e., situations with and without perceptual uncertainty). These properties are illustrated by simulations and analyses of empirical time series. Overall, our framework provides a novel foundation for understanding normal and pathological learning that contextualizes RL within a generic Bayesian scheme and thus connects it to principles of optimality from probability theory.

Although empathy is crucial for successful social interactions, excessive sharing of others' negative emotions may be maladaptive and constitute a source of burnout. To investigate functional neural plasticity underlying the augmentation of empathy and to test the counteracting potential of compassion, one group of participants was first trained in empathic resonance and subsequently in compassion. In response to videos depicting human suffering, empathy training, but not memory training (control group), increased negative affect and brain activations in anterior insula and anterior midcingulate cortex-brain regions previously associated with empathy for pain. In contrast, subsequent compassion training could reverse the increase in negative effect and, in contrast, augment self-reports of positive affect. In addition, compassion training increased activations in a non-overlapping brain network spanning ventral striatum, pregenual anterior cingulate cortex and medial orbitofrontal cortex. We conclude that training compassion may reflect a new coping strategy to overcome empathic distress and strengthen resilience.

Difficulties in social cognition are well recognized in individuals with autism spectrum conditions (henceforth 'autism'). Here we focus on one crucial aspect of social cognition: the ability to empathize with the feelings of another. In contrast to theory of mind, a capacity that has often been observed to be impaired in individuals with autism, much less is known about the capacity of individuals with autism for affect sharing. Based on previous data suggesting that empathy deficits in autism are a function of interoceptive deficits related to alexithymia, we aimed to investigate empathic brain responses in autistic and control participants with high and low degrees of alexithymia. Using functional magnetic resonance imaging, we measured empathic brain responses with an 'empathy for pain' paradigm assessing empathic brain responses in a real-life social setting that does not rely on attention to, or recognition of, facial affect cues. Confirming previous findings, empathic brain responses to the suffering of others were associated with increased activation in left anterior insula and the strength of this signal was predictive of the degree of alexithymia in both autistic and control groups but did not vary as a function of group. Importantly, there was no difference in the degree of empathy between autistic and control groups after accounting for alexithymia. These findings suggest that empathy deficits observed in autism may be due to the large comorbidity between alexithymic traits and autism, rather than representing a necessary feature of the social impairments in autism.

The development of social emotions such as compassion is crucial for successful social interactions as well as for the maintenance of mental and physical health, especially when confronted with distressing life events. Yet, the neural mechanisms supporting the training of these emotions are poorly understood. To study affective plasticity in healthy adults, we measured functional neural and subjective responses to witnessing the distress of others in a newly developed task (Socio-affective Video Task). Participants' initial empathic responses to the task were accompanied by negative affect and activations in the anterior insula and anterior medial cingulate cortex--a core neural network underlying empathy for pain. Whereas participants reacted with negative affect before training, compassion training increased positive affective experiences, even in response to witnessing others in distress. On the neural level, we observed that, compared with a memory control group, compassion training elicited activity in a neural network including the medial orbitofrontal cortex, putamen, pallidum, and ventral tegmental area--brain regions previously associated with positive affect and affiliation. Taken together, these findings suggest that the deliberate cultivation of compassion offers a new coping strategy that fosters positive affect even when confronted with the distress of others.

Humans tend to use the self as a reference point to perceive the world and gain information about other people's mental states. However, applying such a self-referential projection mechanism in situations where it is inappropriate can result in egocentrically biased judgments. To assess egocentricity bias in the emotional domain (EEB), we developed a novel visuo-tactile paradigm assessing the degree to which empathic judgments are biased by one's own emotions if they are incongruent to those of the person we empathize with. A first behavioral experiment confirmed the existence of such EEB, and two independent fMRI experiments revealed that overcoming biased empathic judgments is associated with increased activation in the right supramarginal gyrus (rSMG), in a location distinct from activations in right temporoparietal junction reported in previous social cognition studies. Using temporary disruption of rSMG with repetitive transcranial magnetic stimulation resulted in a substantial increase of EEB, and so did reducing visuo-tactile stimulation time as shown in an additional behavioral experiment. Our findings provide converging evidence from multiple methods and experiments that rSMG is crucial for overcoming emotional egocentricity. Effective connectivity analyses suggest that this may be achieved by early perceptual regulation processes disambiguating proprioceptive first-person information (touch) from exteroceptive third-person information (vision) during incongruency between self- and other-related affective states. Our study extends previous models of social cognition. It shows that although shared neural networks may underlie emotional understanding in some situations, an additional mechanism subserved by rSMG is needed to avoid biased social judgments in other situations.

In its full sense, perception rests on an agent's model of how its sensory input comes about and the inferences it draws based on this model. These inferences are necessarily uncertain. Here, we illustrate how the Hierarchical Gaussian Filter (HGF) offers a principled and generic way to deal with the several forms that uncertainty in perception takes. The HGF is a recent derivation of one-step update equations from Bayesian principles that rests on a hierarchical generative model of the environment and its (in)stability. It is computationally highly efficient, allows for online estimates of hidden states, and has found numerous applications to experimental data from human subjects. In this paper, we generalize previous descriptions of the HGF and its account of perceptual uncertainty. First, we explicitly formulate the extension of the HGF's hierarchy to any number of levels; second, we discuss how various forms of uncertainty are accommodated by the minimization of variational free energy as encoded in the update equations; third, we combine the HGF with decision models and demonstrate the inversion of this combination; finally, we report a simulation study that compared four optimization methods for inverting the HGF/decision model combination at different noise levels. These four methods (Nelder-Mead simplex algorithm, Gaussian process-based global optimization, variational Bayes and Markov chain Monte Carlo sampling) all performed well even under considerable noise, with variational Bayes offering the best combination of efficiency and informativeness of inference. Our results demonstrate that the HGF provides a principled, flexible, and efficient-but at the same time intuitive-framework for the resolution of perceptual uncertainty in behaving agents.

Compassion has been suggested to be a strong motivator for prosocial behavior. While research has demonstrated that compassion training has positive effects on mood and health, we do not know whether it also leads to increases in prosocial behavior. We addressed this question in two experiments. In Experiment 1, we introduce a new prosocial game, the Zurich Prosocial Game (ZPG), which allows for repeated, ecologically valid assessment of prosocial behavior and is sensitive to the influence of reciprocity, helping cost, and distress cues on helping behavior. Experiment 2 shows that helping behavior in the ZPG increased in participants who had received short-term compassion training, but not in participants who had received short-term memory training. Interindividual differences in practice duration were specifically related to changes in the amount of helping under no-reciprocity conditions. Our results provide first evidence for the positive impact of short-term compassion training on prosocial behavior towards strangers in a training-unrelated task.

BACKGROUND: Obesity is caused by an imbalance between energy intake, i.e. eating and energy expenditure (EE). Severe obesity is more prevalent in women than men worldwide, and obesity pathophysiology and the resultant obesity-related disease risks differ in women and men. The underlying mechanisms are largely unknown. Pre-clinical and clinical research indicate that ovarian hormones may play a major role. OBJECTIVE AND RATIONALE: We systematically reviewed the clinical and pre-clinical literature on the effects of ovarian hormones on the physiology of adipose tissue (AT) and the regulation of AT mass by energy intake and EE. SEARCH METHODS: Articles in English indexed in PubMed through January 2016 were searched using keywords related to: (i) reproductive hormones, (ii) weight regulation and (iii) central nervous system. We sought to identify emerging research foci with clinical translational potential rather than to provide a comprehensive review. OUTCOMES: We find that estrogens play a leading role in the causes and consequences of female obesity. With respect to adiposity, estrogens synergize with AT genes to increase gluteofemoral subcutaneous AT mass and decrease central AT mass in reproductive-age women, which leads to protective cardiometabolic effects. Loss of estrogens after menopause, independent of aging, increases total AT mass and decreases lean body mass, so that there is little net effect on body weight. Menopause also partially reverses women's protective AT distribution. These effects can be counteracted by estrogen treatment. With respect to eating, increasing estrogen levels progressively decrease eating during the follicular and peri-ovulatory phases of the menstrual cycle. Progestin levels are associated with eating during the luteal phase, but there does not appear to be a causal relationship. Progestins may increase binge eating and eating stimulated by negative emotional states during the luteal phase. Pre-clinical research indicates that one mechanism for the pre-ovulatory decrease in eating is a central action of estrogens to increase the satiating potency of the gastrointestinal hormone cholecystokinin. Another mechanism involves a decrease in the preference for sweet foods during the follicular phase. Genetic defects in brain α-melanocycte-stimulating hormone-melanocortin receptor (melanocortin 4 receptor, MC4R) signaling lead to a syndrome of overeating and obesity that is particularly pronounced in women and in female animals. The syndrome appears around puberty in mice with genetic deletions of MC4R, suggesting a role of ovarian hormones. Emerging functional brain-imaging data indicates that fluctuations in ovarian hormones affect eating by influencing striatal dopaminergic processing of flavor hedonics and lateral prefrontal cortex processing of cognitive inhibitory controls of eating. There is a dearth of research on the neuroendocrine control of eating after menopause. There is also comparatively little research on the effects of ovarian hormones on EE, although changes in ovarian hormone levels during the menstrual cycle do affect resting EE. WIDER IMPLICATIONS: The markedly greater obesity burden in women makes understanding the diverse effects of ovarian hormones on eating, EE and body adiposity urgent research challenges. A variety of research modalities can be used to investigate these effects in women, and most of the mechanisms reviewed are accessible in animal models. Therefore, human and translational research on the roles of ovarian hormones in women's obesity and its causes should be intensified to gain further mechanistic insights that may ultimately be translated into novel anti-obesity therapies and thereby improve women's health.

This is the final paper in a Comments and Controversies series dedicated to "The identification of interacting networks in the brain using fMRI: Model selection, causality and deconvolution". We argue that discovering effective connectivity depends critically on state-space models with biophysically informed observation and state equations. These models have to be endowed with priors on unknown parameters and afford checks for model Identifiability. We consider the similarities and differences among Dynamic Causal Modeling, Granger Causal Modeling and other approaches. We establish links between past and current statistical causal modeling, in terms of Bayesian dependency graphs and Wiener-Akaike-Granger-Schweder influence measures. We show that some of the challenges faced in this field have promising solutions and speculate on future developments.

Decision-making can be broken down into several component processes: assigning values to stimuli under consideration, selecting an option by comparing those values, and initiating motor responses to obtain the reward. Although much is known about the neural encoding of stimulus values and motor commands, little is known about the mechanisms through which stimulus values are compared, and the resulting decision is transmitted to motor systems. We investigated this process using human fMRI in a task where choices were indicated using the left or right hand. We found evidence consistent with the hypothesis that value signals are computed in the ventral medial prefrontal cortex, they are passed to regions of dorsomedial prefrontal cortex and intraparietal sulcus, implementing a comparison process, and the output of the comparator regions modulates activity in motor cortex to implement the choice. These results describe the network through which stimulus values are transformed into actions during a simple choice task.

What determines reciprocity in employment relations? We conducted a controlled field experiment to measure the extent to which monetary and nonmonetary gifts affect workers' performance. We find that nonmonetary gifts have a much stronger impact than monetary gifts of equivalent value. We also observe that when workers are offered the choice, they prefer receiving money, but reciprocate as if they received a nonmonetary gift. This result is consistent with the common saying, “it's the thought that counts.” We underline this point by showing that monetary gifts can effectively trigger reciprocity if the employer invests more time and effort into the gift's presentation.

Do people intuitively favour certain actions over others? In some dual-process research, reaction-time (RT) data have been used to infer that certain choices are intuitive. However, the use of behavioural or biological measures to infer mental function, popularly known as 'reverse inference', is problematic because it does not take into account other sources of variability in the data, such as discriminability of the choice options. Here we use two example data sets obtained from value-based choice experiments to demonstrate that, after controlling for discriminability (that is, strength-of-preference), there is no evidence that one type of choice is systematically faster than the other. Moreover, using specific variations of a prominent value-based choice experiment, we are able to predictably replicate, eliminate or reverse previously reported correlations between RT and selfishness. Thus, our findings shed crucial light on the use of RT in inferring mental processes and strongly caution against using RT differences as evidence favouring dual-process accounts.

Neural communication is disrupted in autism by unknown mechanisms. Here, we examined whether in autism there are changes in axons, which are the conduit for neural communication. We investigated single axons and their ultrastructure in the white matter of postmortem human brain tissue below the anterior cingulate cortex (ACC), orbitofrontal cortex (OFC), and lateral prefrontal cortex (LPFC), which are associated with attention, social interactions, and emotions, and have been consistently implicated in the pathology of autism. Area-specific changes below ACC (area 32) included a decrease in the largest axons that communicate over long distances. In addition, below ACC there was overexpression of the growth-associated protein 43 kDa accompanied by excessive number of thin axons that link neighboring areas. In OFC (area 11), axons had decreased myelin thickness. Axon features below LPFC (area 46) appeared to be unaffected, but the altered white matter composition below ACC and OFC changed the relationships among all prefrontal areas examined, and could indirectly affect LPFC function. These findings provide a mechanism for disconnection of long-distance pathways, excessive connections between neighboring areas, and inefficiency in pathways for emotions, and may help explain why individuals with autism do not adequately shift attention, engage in repetitive behavior, and avoid social interactions. These changes below specific prefrontal areas appear to be linked through a cascade of developmental events affecting axon growth and guidance, and suggest targeting the associated signaling pathways for therapeutic interventions in autism.

Successful decision making in a social setting depends on our ability to understand the intentions, emotions and beliefs of others. The mirror system allows us to understand other people's motor actions and action intentions. 'Empathy' allows us to understand and share emotions and sensations with others. 'Theory of mind' allows us to understand more abstract concepts such as beliefs or wishes in others. In all these cases, evidence has accumulated that we use the specific neural networks engaged in processing mental states in ourselves to understand the same mental states in others. However, the magnitude of the brain activity in these shared networks is modulated by contextual appraisal of the situation or the other person. An important feature of decision making in a social setting concerns the interaction of reason and emotion. We consider four domains where such interactions occur: our sense of fairness, altruistic punishment, trust and framing effects. In these cases, social motivations and emotions compete with each other, while higher-level control processes modulate the interactions of these low-level biases.