Abstract: Depression is a commonly-occurred mental disorder. Researchers have highlighted the attentional bias of depressive disorders, although results have been mixed. The cue-target task has often been used to explore attentional bias; a particular phenomenon revealed by such studies is the inhibition of return (IOR). However, cue-target task has seldom been used so far in the study of depressed patients. The aim of the present study was to investigate the IOR phenomenon in depressed individuals in cue-target task using emotional faces as cues.
Control participants who had never suffered depression (NC), participants who had experienced at least two depressive episodes in their lives but were currently remitted (RMD), and participants diagnosed with a current major depressive disorder (MDD), were recruited using BDI, SDS, HAMD and CCMD-3 as tools. Seventeen participants in each group completed a cue-target task in a behavioral experiment that comprised three kinds of experimental condition, two cue types and four face types. Each participant also completed a simpler cue-target task in an event-related potential (ERP) experiment. In this task, a target appeared after a cue and the participant responded to its location.
In the behavioral experiment, it was found that when the stimulus onset asynchrony (SOA) was 14 ms, the NC and RMD participants had IOR effects for all faces and MDD participants for angry and sad faces. When the SOA was 250 ms, all three groups all had cue validity for sad faces but the effect was much more marked for the MDD group. When the SOA was 750 ms, the NC participants had an IOR effect for sad faces, the RMD participants had cue validity for angry, happy and sad faces, and the MDD participants had cue validity for sad faces and an IOR effect for angry faces. In the ERP experiment, the NC participants showed a bigger P3 amplitude for happy cue than the other groups, a smaller P1 amplitude for happy faces in the invalid cue condition than for other faces, a smaller P1 amplitude for sad faces in the valid cue condition than for other faces, a bigger P3 amplitude for happy faces in the valid cue condition compared with MDD participants, and a bigger P3 amplitude for sad faces in the invalid cue condition compared with other groups. The RMD participants had larger P3 amplitude for sad cue than for other faces, larger P3 amplitude for happy faces in the valid cue condition compared with MDD participants, and smaller P3 amplitude for sad faces in the invalid cue condition compared with NC participants. The MDD participants had a larger P1 amplitude for sad cue compared with other groups, a larger P3 amplitude for sad cue than for other face cues, a smaller P3 amplitude for sad faces in the invalid cue condition compared with NC participants, and a smaller P3 amplitude for happy faces in the valid cue condition compared with other groups.
It can be concluded that the MDD participants had cue validity and deficient IOR for negative stimuli. The deficient inhibition of negative stimuli renders them unable to eliminate the interference of negative stimuli and causes the maintenance and development of depression. The RMD participants had cue validity and deficient IOR for both positive and negative stimuli, which enables them to perceive positive and negative stimuli sufficiently and to maintain emotional balance.

Key words: depression, attentional bias, inhibition of return, cue-target task