Background Social cognitive deficits are importantly involved in the pathogenesis and the consequences of Major Depressive Disorder (MDD). Deficit in "reading" social interactions has been found in acutely depressed patients, but it is still unknown whether it persists in the remission phase, contributing to psychosocial impairment. In this study, we hypothesize deficit in action co-representation, an automatic and dedicated social process, which reflects the ability to represent one's own and another person's actions, both in the acute and in the remission phase of depression. We also expect a positive and significant correlation between the deficit in co-representation and social functioning. Materials and Methods The sample included 35 patients with MDD, evaluated during the acute phase of depression; 25 patients with remitted MDD, re-evaluated during the euthymic phase (ET) and 38 healthy controls (HC). Participants completed the Joint Simon task (JST), which consists in the distribution of the complementary go-nogo parts of the standard Simon task across two participants. Subsequently, as control condition, they were administered the Standard Simon task (SST), which is a behavioural measure of conflict resolution. Participants are asked to respond to visual stimuli (shapes) by making a rightward response to one stimulus and a leftward response to another. The side of the display on which the shapes appear influences participants’ speed of responding by either matching (i.e., corresponding trials) or not matching (i.e., non-corresponding trials) the side (left or right) associated with the shape. Psychometric tests were used to assess social and global functioning. Results Considering the JST, although we found a significant main effect of the Condition (Within subject effect: F1,95=1.09, p=.03), only HC displayed a Joint Simon effect, with significant difference between corresponding and non-corresponding trial (MDD: F1, 95=4.39 p=.25; ET: F1, 95=.63 p=.43; HC: F1, 95=4.39 p=.04). By contrast, in the SST corresponding stimuli were processed faster as compared to non-corresponding stimuli (Within subject effect: F1, 95=125.81, p<.001) in each diagnostic group (MDD: F= F1,95= 69.32, p<.01; ET: F1,95= 16.48, p<.01; HC: F1,95=58.04, p<.01 ), with significant Simon Standard effect. Considering both tasks patients with MDD, also in phase of remission, reported greater reaction times as compared with HC (Between subject effect: SST: F2,95=14.58, p<.01; JST: F2,95=20.68, p<.01). The 'missed-JSE', only for the ET group, was positive correlated with the social functioning scale (r =, 4; p <, 05) with a tendency towards significance with the global functioning scale (r =, 4; p <, 07) Conclusion Patients with MDD did not perform the social task with a relevant JSE, either in the acute phase of the disease or after symptom remission, presenting a "missed-JSE". The absence of a JSE in patients, as compared with HC, could be interpreted as a deficit of action co-representation, which seems to represent a stable element in MDD related with social dysfunction, regardless of symptoms remission