| Objective: to study the visual working memory(WM)mechanism of facial expressive identity recognition.In addition,the effects of sad expression at encoding on face identity recognition in depressed individuals was investigated by event-related potential technique.Methods:(1)Our first study consisted of three behavioral experiments.The modified delayed discrimination task was used to investigate the effects of the appearance of sad and happy expressions at different stages(Using happy facial expression as baseline level(see Jackson,Wu,Linden,& Raymond,2009)and compare sad expression with them)of the task on the WM of facial identity.In experiment 1,sad and happy expressions were presented at the encoding stage,and the WM performance of 25 subjects(13 women)was examined.In experiment 2,sad and happy expressions were presented at the retrieval stage,and the WM performance of 23 subjects(16 women)was examined.To investigate whether emotional information was discarded at the end of encoding or retention of the WM,in experiment 3,sad and happy expressions were still presented at the encoding stage,but an emotional word categorization task was inserted at retention interval.The effects of emotional consistency between words and faces on the WM performance of 34 subjects(21 women)was investigated.(2)The second study also adopted the modified delayed discrimination task.Same as experiment 1 of the first study,sad and happy expressions were presented at the encoding stage,and the effects of sad expression at encoding on the face identity recognition in depressed individuals were investigated.Event-related potentials were studied in 21 control group(9 women)and 20 depressive subjects(11 women).Results:(1)The first study included three behavioral experiments.The results of experiment 1 showed that the WM of sad face identity was worse than that of happy face,while in experiment 2 we found that there was no difference between sad face identity and happy face identity.In experiment 3,the emotional consistency between words and faces does not support the memory of face identity,that is,the facial emotion presented at the encoding stage cannot be maintained throughout the WM maintenance interval.(2)In the second study,the results of behavioral studies showed that the WM performance of depressive group and control group for sad facial identiy were lower than that of happy face identity.Moreover,the overall hits rate(Hits,match trials for which participants correctly responded ‘‘yes’ ’,that is,match-decision ability)of depressive group was lower than that of control group.The ERP results of the hit response showed that both the depressive group and the control group had larger amplitudes of vertex positive potential(VPP,150-200ms)and P3b(300-450ms)for the sad face identity compared to happy face identity.In addition,the interaction between group and facial emotion on VPP amplitudes were significant,which further indicated that the VPP amplitudes of depressive group were significantly lower than that of control group for sad face identity,but the main effect of VPP amplitudes in two groups was not significant for happy face identity.Meanwhile,the interaction between group and facial emotion on P3 b amplitude was marginally significant,which further indicated that,for depressive group,the main effect of P3 b amplitude for different expressive facial identity was not significant,but for control group,the P3 b amplitude of facial identity under sad expression was significantly larger than that of the happy expression.Conclusion:(1)Sad expression at encoding weakens recognition of facial identity in visual working memory.(2)When the negative expression was presented at encoding in visual working memory,the decreased ability of the depressed individuals to match face identity was related to the insufficient processing of the early encoding. |
PDF Full Text Request