Martina J Speck¹, Maya J Egerton-Graham¹, Charles Collin¹, Isabelle Boutet¹; ¹University of Ottawa

The presence of person information can make learning new faces more efficient (e.g., Schwartz & Yovel, 2018; Wiese & Schweinberger, 2015). We sought to replicate this finding and extend it to face-name associations. We also examined how visual scanning patterns change when participants learn new faces and face-name associations. Based on past literature, we predicted that there would be fewer fixations to a face as it becomes familiar, and that a higher proportion of these fixations would be to the eyes. We also predicted that person information would enhance recollection performance. Participants (N=26) took part in a learning phase, where each face was associated with a name. For half of the faces, additional person information (a hobby) was provided. Lighting conditions were varied across repeated presentations to avoid image-based recognition. Recognition and naming of the learned faces were tested immediately after the learning phase, and after 1-week and 2-week intervals. Faces were presented in a different viewing angle during recall. For scanning patterns, we focused specifically on the number of fixations on different interest areas (eyes, nose, mouth, face). Preliminary results suggest that the total number of fixations to a face is reduced as it becomes more familiar. In addition, the eyes were generally viewed more often than any other area of interest region, regardless of the number of exposures. However, unlike previous research, we did not find that the proportion of total fixations to the eyes increased with familiarity. The presence of person information did not influence on scanning patterns during learning, nor recognition performance at testing. These findings advance our understanding of face recognition, particularly regarding how eye movements can be used as an index of familiarity, and under which conditions person information aids (or does not aid) in forming face representations and face-name associations.

Acknowledgements: This research is funded by a Natural Sciences and Engineering Research Council of Canada Discovery grant (2022-03998) to IB