An EEG Study On Negative Cognitive Bias In Depression

Depression,characterized by sustained negative affect and loss of pleasure,is among the most prevalent of all psychiatric disorders.Cognitive theories of depression have predicted that depressed individuals had cognitive biases processing of emotional material.A lot of studies have found that depression is characterized by biased processing of emotional information,which may be related to its emotion dysregulation.However,it is still unknown how the early perceptual processing is modulated by emotional stimuli and the time course of attentional disengagement from emotional facial stimuli in depression.To sort out these questions,we recruited 22 depressed patients and 22 healthy controls to take in a cued target-response task with emotional facial expression as stimuli and three types of cue-target intervals(CTIs).Both behavioral and EEG data were collected from each subject while performing the task.The behavioral results showed that depressed patients had shorter response time for sad faces than for neutral faces,which suggested that cognitive bias of emotional facial expression in depression may had an influence on the subsequent digit target selection task.Moreover,both depressed patients and healthy controls had shorter response time in conditions of CTI = 1000 ms and 1500 ms than in condition of CTI = 350 ms,but had no significant difference in response time between conditions of CTI =1000 ms and CTI =1500 ms.This result implied that both depressed patients and healthy controls completed their attentional disengagement from emotional facial expression between 350 ms to 1000 ms after cue offset.The event-related potential(ERP)results showed that depressed patients had larger P1 amplitudes than healthy controls,which imply that early perceptual anomaly for face processing in depression may occur as early as the P1 stage.There was no significant interaction between the emotion types and groups on P1 amplitudes,which suggested that cognitive biases in depression might not arise yet.Following the P1 stage,N170 amplitudes for sad faces were larger than for other emotion types in depressed patients,whereas N170 amplitudes for happy faces were larger than for other emotion types in healthy controls.These results implied that depressed patients may have a perceptual bias associated with sad emotions,which may be detectable from the N170 time-window.The contingent negative variation,a well-documented ERP marker of cue-induced expectation of the forthcoming target,clearly appeared in the conditions of CTI = 1000 ms and 1500 ms,but was absent in the condition of CTI = 350 ms.These results together implied that both depressed patients and normal controls completed their attentional disengagement from emotional facial expression between 350 ms to 1000 ms after cue offset.