Self-disclosure as a function of reward-cost outcomes.

Fifty-six sailors individually participated in a lengthy interaction session, ostensibly with another sailor (an experimenter confederate). They were subjected to one of four reward/cost interaction histories (Continuous Positive, Later Positive, Continuous Negative, and Later Negative). In addition, Ss expected to be in either a long-term or short-term relationship with their partner (Non-Withdrawal or Withdrawal). Disclosure measures included average time talked, breadth and depth of interpersonal exchange. Results confirmed earlier studies to the effect that (1) more disclosure occurred in nonintimate versus intimate areas, (2) disclosure varied directly as a function of interpersonal reward/cost factors, and (3) the impact of reward/cost factors was primarily in intimate topical areas. An extension of past findings indicated a "contrast effect" that is, shifts in reinforcement which were inconsistent with prior established expectations produced exaggerated behaviors. This happened in the NonWithdrawal condition, and only partially in the Withdrawal condition. These results were seen as consistent with findings from the gain-loss model and earlier studies on social penetration processes.