Recent research in social neuroscience has postulated that Theory of Mind (ToM) regions play a role in processing social prediction error (PE: the difference between what was expected and what was observed). Here, we tested whether PE signal depends on the type of prior information people use to make predictions - an agent's prior mental states (e.g., beliefs, desires, preferences) or an agent's prior behavior - as well as the type of information that confirms or violates such predictions. That is, does prior information about mental states (versus behavior) afford stronger predictions about an agent's subsequent mental states or behaviors? Additionally, when information about an agent's prior mental states or behavior is available, is PE signal strongest when information about an agent's subsequent mental state (vs behavior) is revealed? In line with prior research, results suggest that DMPFC, LTPJ, and RTPJ are recruited more for unexpected than expected outcomes. However, PE signal does not seem to discriminate on the basis of prior or outcome information type. These findings suggest that ToM regions may flexibly incorporate any available information to make predictions about, monitor, and perhaps explain, inconsistencies in social agents.
We examined biopsychosocial stress of acute social pain in relation to chronic loneliness. Hypotheses: 1) Cyberball exclusion (vs. inclusion) would be associated with lower cortisol reactivity to a speech task, and 2) loneliness would moderate the relationship between social exclusion and cortisol reactivity to a speech task, such that higher loneliness would be linked to lower cortisol. Participants (n = 31, women, aged 18-25, 51.6% non-Hispanic white) were randomized to be excluded or included in a game of Cyberball, then completed a speech task. Salivary cortisol was measured at baseline, pre-speech, post-speech, and 15 minutes post-speech. Cortisol reactivity was calculated using area under the curve-increase (AUCi). ANOVA revealed a non-significant, meaningful effect of Cyberball exclusion on cortisol AUCi (p=.103, ηp2=.10), accounting for contraceptive use. Moderation analysis revealed among women with high loneliness, women in the exclusion condition had significantly lower cortisol reactivity than women in the inclusion condition (p=.001). For women with low and medium loneliness, there were no significant differences by Cyberball condition. In sum, lonely young women who are excluded may have hypocortisolemic responses to social stress. Results are consistent with literature suggesting that chronic stress is linked to lower cortisol responses, which is linked to negative physical health outcomes.
Self-esteem is an important psychological resource with adaptive values, and numerous investigations have revealed that self-esteem is influenced by perceived social support. However, the potential neural basis linking perceived social support with self-esteem remains unclear. Therefore, we used voxel-based morphometry to explore whether the hippocampus and amygdala function as the neuroanatomical basis linking perceived social support with self-esteem in a cohort of 243 young healthy adults (128 women; mean age 22.64 years, standard deviation 1.01 years). The Social Provisions Scale and Rosenberg Self Esteem Scale were used for the survey. Magnetic resonance imaging was used to measure the gray matter volume of the hippocampus and amygdala. Correlation analysis revealed that those who perceived more social support had higher self-esteem. Notably, mediation analysis showed that hippocampal gray matter volume linked perceived social support with self-esteem. Our study suggests that the hippocampus plays a primary, but not exclusive, role in linking perceived social support with self-esteem, which provides a novel explanation for how perceived social support affects self-esteem from the perspective of cognitive neuroscience.
Children with autism often have difficulty with Theory of Mind (ToM), the ability to infer mental states, and pragmatic skills, the contextual use of language. Neuroimaging research suggests ToM and pragmatic skills overlap, as the ability to understand another's mental state is a prerequisite to interpersonal communication. To our knowledge, no study in the last decade has examined this overlap further. To assess the emerging consensus across neuroimaging studies of ToM and pragmatic skills in autism, we used coordinate-based activation likelihood estimation (ALE) analysis of 35 functional magnetic resonance imaging (MRI) studies (13 pragmatic skills, 22 ToM), resulting in a meta-analysis of 1,295 participants (647 autistic, 648 non-autistic) aged 7 to 49 years. Group difference analysis revealed decreased left inferior frontal gyrus (LIFG) activation in autistic participants during pragmatic skills tasks. For ToM tasks, we found reduced anterior cingulate cortex (ACC), medial prefrontal cortex (MPFC), and temporoparietal junction (TPJ) activation in autistic participants. Collectively, both ToM and pragmatic tasks showed activation in IFG and superior temporal gyrus (STG) and a reduction in left hemispheric activation in autistic participants. Overall, the findings underscore the cognitive and neural processing similarities between ToM and pragmatic skills, and their underlying neurobiological differences in autism.
Socio-emotional interactions are integral for regulating emotions and buffering psychological distress. Social neuroscience perspectives on aging suggest that empathetic interpersonal interactions are supported by the activation of brain regions involved in regulating negative affect. The current study tested whether resting state functional connectivity of a network of brain regions activated during cognitive emotion regulation, i.e., emotion regulation network (ERN), statistically mediates the frequency of social contact with friends or family on psychological distress. Here, a 10-min resting-state functional MRI scan was collected along with self-reported anxiety/depressive, somatic, and thought problems and social networking from 90 community-dwelling older adults (aged 65-85 years). The frequency of social interactions with family, but not friends and neighbors, was associated with lower psychological distress. The magnitude of this effect was reduced by 33.34% to non-significant upon adding resting state ERN connectivity as a mediator. Follow-up whole-brain graph network analyses revealed that efficiency and centrality of the left inferior frontal gyrus and the right middle temporal gyrus relate to greater family interactions and lower distress. These hubs may help to buffer psychological problems in older adults through interactions involving empathetic and cognitive emotion regulation with close family.
Numerous studies show that social cues are processed preferentially by the human visual system and that perception of communicative intentions, particularly those self-directed, attracts and biases attention. However, it is still unclear when in the temporal hierarchy of visual processing communicative cues exert impact on perception and whether their effects are automatic or volitional. Therefore, in the present study, we used electroencephalography (EEG) to investigate the pattern of neural activity associated with processing communicative and individual gestures. Participants (N = 24) were shown animations depicting either biological (BM) or scrambled motion (SM) and were asked to categorize them accordingly. Additionally, BM depicted either communicative or individual actions. The results showed that while early components (N1, N2) are sensitive to differences between BM and SM, the differentiation of neural activity related to the type of action performed by point-light agent (individuals vs. communicative) is observed only for late components such as posterior late positive potential (>500 ms). The findings of the current study show that even in the absence of any top-down effects, social intentions produce long-lasting attentional effects at the later stages of stimuli processing.

