Effects of a Short Mobile Intervention on Digital Health Literacy in Adolescents and Teachers: Randomized Controlled Trial.

Sighted adults adeptly use vision to recognize and find their way through the local environment or “scene.” The foundations of this ability develop early in infancy and toddlerhood. For example, 18-mo-olds use the geometry of the visual environment (e.g., the layout of walls) to navigate to a previously learned location (), while 8-mo-olds visually distinguish safe vs. unsafe navigational affordances (e.g., steep vs. shallow visual cliffs) (). By adulthood, the human brain contains at least three regions that respond selectively to visual scenes, and which play a causal role in scene processing and navigation (): the occipital (OPA) (), parahippocampal (PPA) (), and medial place areas (MPA; also known as the retrosplenial complex) (, ). How does cortical scene selectivity develop? One theoretical framework emphasizes the role of passive visual exposure to low-level visual features (e.g., refs. –). On this account, regions that will later develop scene selectivity are initially driven by peripheral visual input and/or low-level visual features that covary with such input, including high spatial frequency content and rectilinear features. Passive exposure to the visual statistics of scenes leads to the frequent coactivation of these features across development, eventually driving the development of higher-level scene selectivity (i.e., responses to scenes over and above a simple combination of this initial set of low-level features). We will refer to this as the “passive exposure” framework. Given that all waking life occurs embedded in a scene, one prediction of the passive exposure framework is that scene responses will increase gradually with age. A second framework suggests that infants are born not only with basic sensorimotor functions but also with endogenous motivation to learn about the spatial environment and how to navigate through it (e.g., refs. –). On this view, development of higher-level scene selectivity depends on active experience using visual scene information in the service of functional goals, as when planning and guiding navigation. For example, dark-reared kittens develop greater visual cliff sensitivity when they gain active navigation experience (compared to yoked passive exposure to the same space in a “kitten carousel”) (). Similarly, human infants’ visual cliff sensitivity is related to their locomotor status and ability (), and prereaching infants given active reaching experience (with “sticky mittens”) show greater attention to goal-directed actions than infants without active experience (). We will refer to this as the “active experience” framework. Infants are born incapable of independent navigation, and only gain their first experience actively planning and guiding their own navigation late in infancy (i.e., by learning to locomote). One prediction of the active experience framework then is that scene-selective responses emerge only at or after locomotor onset, when scene information becomes relevant for navigation planning.