Communication Research

Complaints about media coverage of presidential elections have raised concerns that the public may support restrictions on election campaign messages. This study, based on a nationwide telephone survey (N = 549), investigated perceptions of media influence and support for campaign message restrictions during the 1996 presidential campaign. The study was theoretically grounded in the third-person effect approach. The results confirmed the perceptual component hypothesis for individuals to perceive greater media influence on people other than themselves. They also confirmed the behavioral component hypothesis for third-person perception to lead to support for restrictions on election messages. The implications of the results are discussed.

Blaming outgroups for social problems is a common discursive practice. However, it is not clear whether this discourse represents ethnocentric bias or realistic group conflict. To compare these explanations, the authors studied 14 weeks of local television news on three stations in Philadelphia. The authors found that persons of color were heavily presented in stories about crime, and within those stories were more likely to be presented as perpetrators of crime than as persons reacting to or suffering from it. An analysis using recent homicide rates in Philadelphia and a rational model of expected victimization indicated that contrary to a realistic-conflict explanation, White actors were overpresented as victims of violence compared to their roles as perpetrators, and persons of color were overpresented as perpetrators of violence against White actors. These patterns were consistent with the hypothesis that the news media engage in a discourse of ethnic blame that is independent of realistic group conflict.

Skill acquisition follows a pattern of rapid improvement early on, followed by decreasing performance increments as the skill continues to be practiced, a pattern captured by the “power law of practice,” (P = BN^α). The purpose of these studies was to extend previous research by investigating the impact of individual differences on the course of adult communicative-skill acquisition. In Study 1, participants learned a sequence for describing geometric arrays and then executed 90 performance trials. Individual differences in working memory capacity, speed of information processing, and psychomotor ability were assessed. It was found that counter to expectations, speed of information processing was negatively correlated with overall learning rate and that psychomotor ability was related to asymptotic performance. In Study 2, employing a more complex communication task, speed of information processing was positively correlated with learning rate, working memory capacity predicted initial performance speed, and psychomotor ability was associated with asymptotic performance.

Most research has employed methods that treat interpersonal arguments as though they occur in a single episode or that ignore their episodic nature. However, limited research suggests that a relational argument may be repeated and can be viewed as serial. The authors' perspective assumes that the perceived resolvability of a serial argument is a better predictor of relational quality than is the frequency with which the argument; the results of a survey of individuals involved in a dating relationship support this assumption. Perceived resolvability is negatively related to (a) arguments arising from violated expectations, (b) counter-complaining and partner-initiated demand-withdrawal cycles, (c) predictability of argumentative episodes, (d) mulling, (e) overall amount of discord in the relationship, and (f) withdrawal from the partner after an episode. The extent of constructive communication that occurred during the initial confrontation is positively related to perceived resolvability.