The purpose of this study was to examine the relative contributions of antipsychotic medication, negative symptoms and executive functions to impairment in social functioning in a sample of outpatients with stable schizophrenia. One-hundred and sixty-eight consecutive outpatients with stable schizophrenia were enrolled in a cross-sectional study. We performed a path analysis using multiple regression technique in order to assess the specific effect of antipsychotic type (first-generation antipsychotics versus second-generation antipsychotics) on social functioning and the possible mediating role of executive functions and negative symptoms. Our findings suggested that (i) second generation antipsychotics (SGAs) use predicted better social functioning (Beta=.24, p=.003) and better executive functions (Beta=.25, p=.003); conversely SGAs use was not associated with lesser negative symptoms (Beta=.00, p=.981); (ii) impaired executive functions and severity of negative symptoms were associated with worse social functioning (Beta=.19, p=.016; Beta=.28, p=.001); (iii) when we inserted in the model Positive and Negative Syndrome Scale - Negative Symptom subscale (PANSS-N) and Wisconsin Card Sorting Test - number of achieved sorting categories (WCST-cat), the former failed to show a mediation effect, while the latter seemed to mediate partially the effect of SGAs on social functioning. Taken together, the present results suggest that it is critical to examine individually executive functions and negative symptoms because they seem to relate to social functioning in different and independent ways and thus might represent separable treatment targets. Furthermore, social functioning appears a complex outcome multiply determined with no single predictor variable explaining a sufficient amount of variance.