Sharing joint visual attention to an object with another person biases infants to encode qualitatively different object properties compared to a parallel attention situation lacking interpersonal sharedness. This study investigated whether merely observing joint attention amongst others shows the same effect. In Experiment 1 (first-party replication experiment), N = 36 9-month-old German infants were presented with a violation-of-expectation task during which they saw an adult looking either in the direction of the infant (eye contact) or to the side (no eye contact) before and after looking at an object. Following an occlusion phase, infants saw one of three different outcomes: the same object reappeared at the same screen position (no change), the same object reappeared at a novel position (location change), or a novel object appeared at the same position (identity change). We found that infants looked longer at identity change outcomes (vs. no changes) in the “eye contact” condition compared to the “no eye contact” condition. In contrast, infants’ response to location changes was not influenced by the presence of eye contact. In Experiment 2, we found the same result pattern in a matched third-party design, in which another sample of N = 36 9-month-old German infants saw two adults establishing eye contact (or no eye contact) before alternating their gaze between an object and their partner without ever looking at the infant. These findings indicate that infants learn similarly from interacting with others and observing others interact, suggesting that infant cultural learning extends beyond infant-directed interactions.

Empathy is commonly considered a driver of prosociality in child ontogeny, but causal assumptions regarding this effect mostly rely on correlational research designs. Here, 96 urban German children (5–8 years; 48 girls; predominantly White; from mid-to-high socioeconomic backgrounds) participated in an empathy intervention or a control condition before prosocial behaviors (polite lie-telling: rating the drawing as good ; prosocial encouragement: utterances interpreted as cheering up the artist) were assessed in an art-rating task. Contrasting children’s empathy at baseline with their empathy after the intervention indicated promoted empathy compared to the control group. Despite the intervention’s effect on children’s empathy, there were no simultaneous changes in prosocial behaviors. At the same time, children’s empathy at baseline was associated with their prosocial encouragement. These results indicate conceptual associations between children’s empathy and prosociality. However, they do not support strict causal claims regarding this association in middle childhood. Further applications of the novel short-time intervention to address causal effects of empathy on prosociality and other developmental outcomes are discussed.