The theory of natural pedagogy stipulates that infants follow gaze because they are sensitive to the communicative intent of others. According to this theory, gaze following should be present if, and only if, accompanied by at least one of a set of specific ostensive cues. The current article demonstrates gaze following in a range of contexts, both with and without expressions of communicative intent in a between-subjects design with a large sample of 6-month-old infants (n = 94). Thus, conceptually replicating prior results from Szufnarowska et al. (2014) and falsifying a central pillar of the natural pedagogy theory. The results suggest that there are opportunities to learn from others' gaze independently of their displayed communicative intent.

The development of gaze following begins in early infancy and its developmental foundation has been under heavy debate. Using a longitudinal design ( N = 118), we demonstrate that attachment quality predicts individual differences in the onset of gaze following, at six months of age, and that maternal postpartum depression predicts later gaze following, at 10 months. In addition, we report longitudinal stability in gaze following from 6 to 10 months. A full path model (using attachment, maternal depression and gaze following at six months) accounted for 21% of variance in gaze following at 10 months. These results suggest an experience-dependent development of gaze following, driven by the infant's own motivation to interact and engage with others (the social-first perspective).

Gaze following (GF), the ability to synchronize visual attention with others, is often considered a foundation of social cognition. In this study, GF was assessed while changing the space between an actor's eyes and the gaze target. This was done to address a potential confound in the gold standard GF performance test, namely the spatial bias of the actors’ eye position that occurs when the actor turns the head to look at a target, offsetting the eye position from a centered position toward the attended target. Our results suggest that both 4.5 (n = 27) and 6 (n = 30)‐month‐old infants can follow an actor's gaze regardless of proximity. This is the first demonstration that early GF is not dependent on proximity cues, and our results strengthen previous findings suggesting that GF develops well before 6 months of age. The study was preregistered, and all data and analysis routines can be downloaded with provided links.

No abstract