Early Neural Development of Social Interaction Perception in the Superior Temporal Sulcus

The ability to perceive and understand social interactions is one of the most fundamental capacities of human cognition, and it emerges during the earliest stages of development. This ability encompasses not only the basic recognition of social agents and their actions but also the interpretation of intentional behaviors, emotional exchanges, and strategic interactions between individuals. From the first moments of life, infants demonstrate a remarkable sensitivity to social stimuli, which lays the groundwork for increasingly advanced social cognitive abilities that will support their navigation of the social world throughout the lifespan.

Studies have revealed that preverbal infants possess remarkable social perceptual abilities that extend far beyond simple social awareness. Just as other social animals must identify potential allies and avoid threats among their conspecifics, human infants must develop the ability to assess the actions and intentions of others, distinguishing potential friends from foes and identifying appropriate social partners. Adults evaluate others rapidly and automatically based on both behavior and physical features, and Hamlin et al. (2007) demonstrated that this evaluative capacity emerges surprisingly early. Their research showed that even at 6 and 10 months of age, infants can make complex social judgments, displaying clear preferences for individuals who help others over those who hinder them and favoring neutral individuals over those who cause harm.

Further evidence of infants’ social cognitive abilities comes from research on their understanding of imitation in social interactions. Powell and Spelke (2018) conducted a series of experiments with infants as young as 4–5.5 months old, which revealed that even at this early age, infants attach significant social meaning to imitative behaviors. Their results demonstrated that infants possess specific expectations about social relationships based on observed imitation patterns: they anticipate that individuals who imitate others will subsequently approach and affiliate with their imitation targets. The complexity of processing these imitative interactions was evident in how infants’ expectations varied depending on factors such as the number of characters involved and the number of parties initiating affiliative actions. These findings illustrate that infants use imitation as a lens through which they understand social relationships and suggests that the perception of social interactions serves as a fundamental mechanism for interpreting and predicting social affiliations from the earliest stages of development.

Given these early behavioral capabilities, two critical questions emerge: What specific neural mechanisms support these remarkable social perceptual abilities and when do these neural systems develop? These questions are fundamental for understanding how the human brain specializes in processing social interactions and whether the neural architecture for social perception develops alongside infants’ behavioral abilities.

Recent neuroscientific investigations have identified the superior temporal sulcus (STS) as a critical region for processing social interactions. The STS shows selective responses to interacting agents, as opposed to individuals acting independently, suggesting it has a role in social interaction perception (McMahon and Isik, 2023). This selectivity extends beyond mere recognition that social interactions are occurring: the STS demonstrates sensitivity to the nature of these interactions, such as helping behaviors, indicating it has a fundamental role in understanding and navigating the social world (Isik et al., 2017).

In a recent paper published in The Journal of Neuroscience, Im et al. (2025) investigated the neural development of advanced social perceptual ability by analyzing young children's brain responses to social interactions and other visual features during movie viewing, focusing particularly on STS selectivity. Their study addresses a crucial question: How early do children develop specialized neural responses to social interactions, and how do these responses compare to those observed in adults?

By showing naturalistic, narrative stimuli and using voxel-level activity patterns using encoding methods, the researchers made several key findings. First, STS was responsive to faces in all age groups, with adult-like activation patterns appearing in children aged 5–12 years. More importantly, social interaction features were closely associated with STS responses across all age groups, including children as young as 3–4 years old, with only minor differences between the youngest children and adults and no significant differences between 7-year-olds and adults.

These findings contrast with previous research that suggested a lengthier development of social interaction processing. For instance, Walbrin et al. (2020) found that although both children and adults showed selectivity to social interactions in the right STS, only adults exhibited selectivity in the left STS. Their study indicated that adults demonstrated more focal and stronger selectivity bilaterally than children aged 6–12 years, with younger children (6–8 years) showing less selectivity in the right STS than adults. The discrepancies between these findings might be attributed to the level of complexity and the type of social information presented in the stimuli: Walbrin et al. (2020) used human point-light figures, whereas Im et al. (2025) used animated videos. The early development of face selectivity in the STS also conflicts with previous research. Scherf et al. (2007) suggested that face selectivity does not fully develop until adolescence, whereas Im et al.'s more stratified age group analysis revealed adult-like activation patterns in children as young as 5–12 years, with only minor differences in the youngest group (3–4 years).

Recent research has provided additional insights into the role of the STS in social interaction processing. Masson and Isik (2021) demonstrated that the STS uniquely processes social interactions in response to naturalistic stimuli, independently of other covarying features like faces, voices, and theory of mind (i.e., the ability to attribute mental states to others). While both the STS and medial prefrontal cortex (mPFC) respond to broad social-affective features, only the STS shows robust selectivity to social interactions. Furthermore, Masson et al. (2024) found that activity in the STS supports the memory of social interactions, with neural responses during perception being reinstated during recall, suggesting that the STS contains abstract or conceptual representations of social interactions.

Intriguingly, the timing of adult-like STS selectivity for social interactions (around age 7) aligns with important behavioral developments in children's social cognition. Smith et al. (2013) demonstrated that children begin to distribute payoffs equally around age 7–8, the same age range in which Im et al. (2025) found adult-like selectivity for social interaction perception in the STS. Similarly, House et al. (2013) showed that children in middle childhood start to exhibit varying degrees of selflessness and willingness to help others. Such prosocial behaviors depend on perceiving others’ actions and thus might also depend on maturation of STS selectivity for social interactions. Future work should examine the relationship between STS maturation and prosocial behaviors by combining fMRI with behavioral measures in child populations. For instance, studies could use within-subject designs to compare STS activity during the observation of social interactions with children's real-time decisions in tasks involving sharing, cooperation, or third-party punishment. This approach would allow researchers to test whether greater STS selectivity corresponds to more frequent or nuanced prosocial responses. Importantly, given the complexity of prosocial behavior, it is likely that the STS operates within a broader network that includes regions such as the mPFC, temporoparietal junction, and amygdala, all of which have been implicated in theory of mind, moral reasoning, and emotional processing. Future work should therefore adopt a network-based approach to clarify how these systems jointly support the emergence of prosocial behavior during development.

While these findings represent a significant advance in the understanding of social cognitive development and behaviors, some limitations should be considered. The use of regions of interest (ROI) masks—predefined anatomical areas assumed to be functionally relevant—may lack precision compared with using functional localization methods, which identify brain regions based on each individual’s functional responses to specific tasks. The large ROI masks used in the study may encompass areas beyond the specific brain ROIs, potentially capturing signals unrelated to the phenomenon under investigation. A replication using functional localization methods would provide more precise mapping and avoid potential overestimation of areas captured by ROI masks.

In summary, the work by Im et al. (2025) demonstrates the early development and maturing of specialized neural responses to social interactions in the STS region. Their findings revise previous assumptions about the timeline of social cognitive development by showing that children as young as 3–4 years old display patterns of STS activation similar to that in adults when processing social interactions, with near adult-like responses present by age 7. By using naturalistic stimuli, the study refines our understanding of the developmental trajectory of social interaction processing and suggests possible links between early neural specialization and later behavioral shifts in social cognition. These findings contribute to a growing body of work aiming to explain the neural basis of emerging social behaviors in middle childhood.