| When recognizing faces,people perform significantly better with familiar faces than with unfamiliar ones.Recognition of familiar faces is faster and more accurate,while unfamiliar faces are slower and more prone to errors.A multitude of studies have focused on the changes in face mechanisms underlying the differences in recognition between familiar and unfamiliar faces.The changes in holistic processing levels,processing in different facial regions,and changes in face representations have been identified as potential key factors.However,little research has investigated how unfamiliar faces become familiar during interpersonal contact.Laboratory studies suggest that people can form representations of unfamiliar faces after brief exposure,indicating that familiarity could be achieved in a short period of time.However,an ERP study has found that the SFE signal(Sustained Familiarity Effect),which reflects the familiarity effect and peaks in negative amplitude between 400 and 600 ms in EEG signals,stabilized after 14 months of interpersonal contact,suggesting that familiarity is a gradual process.Given this evidence,further research is required to determine the specific processes of face recognition performance and internal mechanisms that underly these changes over time as familiarity develops.This paper quantified familiarity based on the length of time two classmates spent in contact with each other.Three longitudinal studies were conducted on university students for up to 6 months of contact time(1 month,2 months,3 months,and 7 months).The study aimed to investigate the changes in face recognition performance and related internal facial mechanisms as familiarity increased over time.Study 1 aimed to investigate the changes in face recognition performance and holistic processing levels during the process of familiarization.To achieve this,participants were presented with classmates’ faces from different perspectives(0°,30°,60°,90°)and orientation(upright,inverted).The study also explored how threedimensional representations of faces with independent perspectives were formed.The study used complete face pictures as materials for a signal detection task,where participants were asked to judge whether the face presented belonged to their classmates.Results showed that:(1)face recognition performance gradually improved over time,with performance showing a consistent exponential trend as contact time increased.(2)Improvement in holistic processing levels may fluctuate and was not the main factor contributing to the performance improvement.The inverted effect in familiar face recognition remained constant from 1 to 7 months of exposure,which differed from the changes in recognition performance.(3)The independence of visual angle representation gradually broadened from the front to the side.During the 1-7month contact period,recognition performance for faces at 0° and 30° viewing angles were same and higher than for faces at 60° and 90° viewing angles,with 60° having higher recognition performance than 90°.Study 2 investigated changes in recognition performance of individual facial areas(eyes,nose,mouth)from different perspectives(0°,30°,60°,90°)during the familiarization process with classmates’ faces.The study aimed to explore differences and trends in feature processing levels for different regions,as well as possible effects of viewing angles on facial feature processing.Signal detection tasks were employed in which subjects judged whether the presented face area picture belonged to their classmate.Results showed:(1)improved processing levels of facial features accumulate over time during the familiarization process;as contact time increased,subjects’ performance in distinguishing individual face regions gradually improved.(2)The eye area has a characteristic processing advantage that increases with increased contact time.Recognition performance of the eye area alone was better than that of the nose and mouth area alone.Additionally,eye-nose(eye-nose)and eye-mouth(eye-mouth)differences were significantly greater at 7 months of exposure than at other exposure times.(3)The recognition advantage of the eye area is partly due to the presence of more diagnostic information.Eye-nose and eye-mouth differences of 90°were significantly smaller than other three angles.Study 3 examined changes in recognition performance during the familiarization process by blocking different areas(eyes,nose,mouth)of classmates’ faces at various perspectives(0°,30°,60°,90°).The study aimed to investigate the role and variation of different face regions in face representation.Signal detection tasks were used,with area-occluded face pictures as materials.Subjects judged whether presented areaoccluded face pictures belonged to their classmates.Results showed:(1)The processing dominance of the eye region stems in part from its characteristic importance in face representations.Subjects performed significantly worse on eyes-occluded faces than on nose-or mouth-occluded faces.(2)Face representation is refined holistically.Participants’ performance in recognizing faces occluded in different areas exhibited a similar trend with familiarity.(3)The relative role of a face region in representation is affected by its ratio in the representation.Eye-nose(nose-occluded – eye-occluded)and eye-mouth(mouth-occluded – eye-occluded)differences were significantly larger at a 0° viewing angle compared to the other three viewing angles.Overall,this study suggests that face familiarity is a complex process that involves the gradual refinement of eye-centered facial representations.College students’ face recognition performance of classmates increased gradually from one month to seven months of enrollment,following an exponential function.Although holisstic face processing improved within one month,representations that are independent of face perspective formed gradually from frontal to side views.Familiar face representations are eye-specific,and eye feature processing improves with familiarity,with the eye area playing a more prominent role in facial representations. |