The main objective of this study was to ascertain if the effectiveness of conversational narrations and non-conversational narrations in multimedia environment will be mediated by learners' field dependence and gender. 53 participants (25 field dependent and 28 field independent subjects) were randomly divided to interact with either one of the two multimedia lessons on C-Programming: conversational narrations or non-conversational narrations. Learning achievements of participants were then assessed in three different measures: drawing, terminology and comprehension. A 2 (narration types) x 2 (field dependence) MANOVA showed a significant interaction effect between field dependence and narrations types on comprehension scores; field independent learners in conversational group significantly outperformed field independent learners in non-conversational group on comprehension scores. Field dependence was also shown to be a factor in learning achievement; field independent learners outperformed field dependent learners on drawing and retention measures. A 2 (narration types) x 2 (gender) MANOVA showed no interaction effect between gender and types of narrations; however, male participants significantly outperformed female participants on drawing and comprehension measures.