The impact of learning supervision in dynamic face learning conditions

We recognize familiar faces even in unpredictable situations where appearance and viewing conditions differ. In lab-based studies, such variations are termed "within-person variability" and participants are explicitly asked to learn a set of faces which can be called “explicit learning” design. However, in real life, people might not attempt to study faces of new people they meet, which may affect the face learning process. To gain insights into the face learning process, we conducted two experiments implementing both incidental (closer to real life situations in which faces are learned; experiment 1) and explicit learning (experiment 2) procedures.
To study face learning in ecological conditions, we used a new video database showing dynamic faces in controlled changing conditions. Participants learned four women’s faces, with two displaying a stable appearance and the other two a variable appearance (i.e., appearance condition) and were assigned to one of four learning groups, exposed to 3, 6, 9, or 12 videos per identity. The structure of the learning phase was similar in both experiments, except that in experiment 1, participants were asked to provide three ratings (friendliness, boldness and intelligence) of four different women and did not know their memory for those faces would be tested. In the explicit learning they were simply told to learn the four faces. We found that recognition performance was consistently higher under stable conditions with no interaction observed between learning groups and appearance conditions. That pattern was consistent across both experiments. Furthermore, no overall differences in performance were observed between the two experiments. These findings provide valuable insights into how face learning takes place under dynamic yet carefully controlled conditions with supervised learning.