The primacy-recency effect in impression formation was explored in the present study. Sixty male subjects indicated the favourability of impression formed about eight hypothetical persons, each described by a set of four adjectives. Sets contained either three positive and one negative or three negative and one positive adjective. The independent variables: A (medium of presentation), B (order of presentation) and C (dependence-proneness) were manipulated in a 2 x 2 x 3 factorial design. When three negative and one positive adjective was used, the main effects of A (P < .05) and C (P < .01) and A X B interaction (P < .01) was found to be significant. When, three positive and one negative adjective was used, the main effects of A (P <.01) and C (P <.01), and A x C (P < .05) and A X B X C (P < .05) interactions were found to be significant. No evidence of significant effect of B was found. The influence of sequential and collective presentation, the mediational importance of order of presentation and dependence proneness and the nature of verbal information used has been discussed.