Men's emotional expression was assessed using Deaux and Major's (1987) interactive model of gender-related behavior as a theoretical framework. This model explains gender differences in behavior as a function of proximal forces and contextual variables such as, activated gender-related schemata regarding oneself and others, and situational cues. Male college students, categorized as high and low on the Masculine Gender Role Stress (MGRS) scale (Eisler & Skidmore, 1987), interacted consecutively with a male and a female confederate. Confederates portrayed either gender-consistent (GC) or gender inconsistent (GI) cues regarding their desire for emotional expression from subjects. Verbal and nonverbal measures were rated from videotapes of the interactions. Speaking and listening roles were analyzed separately. It was hypothesized that men who appraise violations of the traditional masculine role as stressful (high MGRS) would demonstrate less emotionally expressive behavior than other men (low MGRS). Specifically, under GC cue conditions (i.e., emotional expression more appropriate to female than male confederates), high MGRS men were expected to be more expressive to female expressive to both confederates. Under GI cue conditions (i.e., emotional expression more appropriate to male than female confederates), it was hypothesized that low MGRS men would become more expressive to male than female confederates, while high MGRS men would be equally expressive to male and female confederates. Results indicated that under GC conditions, both high and low MGRS men were more expressive to female than to male confederates. As predicted, under GI conditions, this pattern was attenuated or reversed, confirming that contextual variables impact displays of emotional expression. Furthermore, anxiety expression also varied with cue conditions and MGRS level. Verbal results demonstrated support for the attenuation of expressive differences under GI conditions, but this did not vary as a function of MGRS level. Patterns of expression varied as a function of speaking and listening roles, particularly for low MGRS men. This study lends support to the Deaux and Major's hypothesis that gender-related behaviors are influenced by contextual and proximal factors. Implications for this model, as well as the MGRS construct, are discussed. / Ph. D.