The Effect Of Facial Familiarity On The Other-Race Categorization Advantage

Race is an important dimension of face processing.Research has shown that there are differences in cross-race face processing.The other-race categorization advantage is one of its typical manifestations.The categorization of faces by race is faster for other-race than own-race faces,a phenomenon labeled “the other-race categorization advantage”(ORCA).The ORCA has been well established in unfamiliar faces,but few studies have examined whether it still exists in familiar faces.The facial familiarity can be understood from two aspects:(1)semantic familiarity(Nessler,Mecklinger,& Penney,2005): familiarity exists not only at the visual level for faces,but also with associated individual knowledge or emotional experiences.For example,we are not only familiar with the faces of famous people,but also know information about their careers and achievements,and may also have the emotion of liking or disliking them,etc.;(2)visual familiarity: familiarity at the visual level only.For example,you may often meet a stranger and although you are familiar with his or her face,you have never talked or gotten to know each other,and therefore have no relevant knowledge or emotion.The study used different ways of manipulating facial familiarity,aiming to investigate whether there are differences in the effects of two different type of face familiarities on ORCA.In particular,whether the cognitive mechanisms by which semantic familiarity affects ORCA are different and how they manifest themselves specifically at the neural level.Experiments 1a and 1b used celebrity faces as experimental stimuli.A 2(race: other-race,ownrace)× 2(familiarity: unfamiliar,familiar)within-subjects design was used in Experiment 1a to investigate the effect semantic familiarity on ORCA.The results showed that:(1)other-race stimuli were categorized faster than own-race stimuli in the unfamiliar face condition,which means the ORCA was found;(2)the difference between own and other faces in reaction time was not significant in familiar face condition,which means the ORCA was not found.Experiment 1b yielded results consistent with Experiment 1a after controlling possible systematic differences in face stimuli.Experiment 2 used a perceptual learning procedure to familiarize subjects with the face stimuli beforehand.A 2(race: other-race,own-race)× 2(familiarity: unfamiliar,familiar)within-subjects design was used to investgete the effect of visual familiarity on ORCA.The results showed that:(1)the ORCA was found for both unlearned and learned faces;(2)there was no interaction effect was found in Experiment 2.The results of Experiment 2 indicated that familiarity at the visual level alone was not sufficient to affect the ORCA,suggesting that there may be differences in face familiarity across types of faces.Therefore,Experiment 3 used the same research design as Experiment 1a and continued to investigate the effect of semantic familiarity on ORCA.Experiment 3 used celebrity faces as experimental stimuli and incorporated ERP techniques.A 2(race: other-race,own-race)× 2(familiarity: unfamiliar,familiar)within-subjects design was used to further explore the specific manifestations at the neural level.The results showed that:(1)The behavioral results were consistent with Experiment 1a and 1b;(2)both familiar and unfamiliar faces induced greater P2 amplitudes for other-race faces,but the difference between other and own race faces on P2 amplitudes was smaller in familiar face condition.In conclusion,the study shows that:(1)there were differences in the effects of semantic familiarity and visual familiarity on ORCA,with the former causing ORCA to disappear and the latter not affecting ORCA;(2)the effect of semantic familiarity on ORCA may be achieved by modulating early attentional processing of face race information.