Assessing the attentional dynamics of Responding to Joint Attention in infancy via online webcam-based eye-tracking

Responding to Joint Attention (RJA) is a foundational social-cognitive skill that unfolds through a sequence of distinct attentional processes: engagement with a social partner, orienting toward a referent, and sustained attention to a shared point of interest. Traditional assessments of joint attention often rely on aggregate behavioral scores that may not capture this fine-grained temporal structure. The present study developed and tested a remotely administered, webcam-based eye-tracking task, delivered via the Labvanced platform, to characterize the attentional dynamics underlying RJA in infants and toddlers aged 10 to 36 months (N = 29). The task decomposed RJA into three sequential phases and included a gaze-contingent delaymanipulation to probe how infants balance attention between the social partner and the cued object under varying temporal demands, quantified through dwell time and gaze re-engagement (entry rate) toward the face, the cued object, and an uncued object. Infants showed a clear attentional hierarchy, with the greatest engagement directed toward the face during the early phases, followed by a gradual reallocation of attention toward the cued object as the referent became identifiable. This temporal profile is consistent with theoretical accounts describing joint attention as an unfolding sequence, in which infants first anchor attention to the social partner before coordinating it with a shared referent. Age-related differences emerged selectively in later phases of attentionalmaintenance, with older infants showing reduced monitoring of the social partner’s face after the partner’s gaze had shifted toward the referent while attention to the target itself remained stable across age, although the wide age range relative to sample size may limit sensitivity to finer-grained developmental change. Together, these findings support the feasibility of remote, webcam-based eye-tracking for capturing fine-grained dynamics of early social attention, supporting progression to future pilot testing of developmental and clinical applications.