Personality and Social Psychology Bulletin

Sanna, Lawrence J.; Turley, Kandi Jo; Mark, Melvin M.

doi: 10.1177/0146167296224001pmid: N/A

Research indicates that effortful performances are reduced when participants cannot be evaluated, relative to when they can be evaluated. It was hypothesized that mood would interact with goals to attenuate such reductions in performance. As predicted, when participants' goal was to do as much as they could, those in negative moods put forth more effort and persisted longer than those in positive moods and performed equally well whether or not they could be evaluated. In contrast, as predicted, when participants' goal was to continue until they no longer enjoyed the task, those in positive moods put forth more effort and persisted longer than those in negative moods; no-evaluation and evaluation conditions did not differ. For those in positive moods asked to do as much as they could and those in negative moods asked to continue until they no longer enjoyed the task, no-evaluation participants performed worse than evaluation participants.

Kite, Mary E.; Whitley, Bernard E.

doi: 10.1177/0146167296224002pmid: N/A

Meta-analytic techniques were used to compare men's and women's attitudes toward homosexual persons, homosexual behaviors, and gay people's civil rights. As expected, size of sex differences varied across these categories. Men were more negative than women toward homosexual persons and homosexual behavior, but the sexes viewed gay civil rights similarly. Men's attitudes toward homosexual persons were particularly negative when the person being rated was a gay man or of unspecified sex. Women and men evaluated lesbians similarly. Ratings of homosexual persons and homosexual behavior were least likely to differ by subject sex for samples of nonprofessional adults. In addition, sex role attitude mediated sex differences in attitudes toward homosexuality. Biases in the research literature and areas that deserve further attention, such as the confounding of sample with measurement strategy and the tendency to study gay men or targets of unspecified sex, are discussed.

Spencer, Stacie M.; Norem, Julie K.

doi: 10.1177/0146167296224003pmid: N/A

This research concerns the interaction between performance strategies and imagery conditions. Defensive pessimists (DPs, who set low expectations, feel anxious, and rehearse possible outcomes) and strategic optimists (OPTs, who set high expectations, feel calm, and avoid reflecting) were randomly assigned to three conditions: a coping imagery condition (imagining correcting mistakes), a mastery imagery condition (imagining a flawless performance), or a relaxation condition (relaxation imagery). DPs performed better in the coping imagery condition, which was similar to their typical strategy, than in the relaxation condition, which was hypothesized to interfere with their performance. The opposite was true for OPTs. Both groups performed worst in the mastery imagery condition. Discussion highlights the importance of both reflective process and thought content and the costs and benefits of each strategy.

Anderson, Craig A.; Anderson, Kathryn B.; Deuser, William E.

doi: 10.1177/0146167296224004pmid: N/A

A general framework for studying affective aggression, integrating many insights from previous models (e.g., those of Baron, Berkowitz, Geen, and Zillmann), is presented. New research examining effects of extreme temperatures and photos of guns on arousal, cognition, and affect is reported. Hostile cognition was assessed using an automatic priming task (i.e., Stroop interference). Hostile affect was assessed with the State Hostility Scale. Positive and negative affect, hostile attitudes, perceived comfort, and perceived arousal were also assessed. As expected, hot and cold temperatures increased state hostility and hostile attitudes, and viewing guns did not. As expected, viewing guns primed hostile cognitions and extreme temperatures did not. Theoretical implications of these results and societal implications of the general framework are discussed.

Uleman, James S.; Hon, Alex; Roman, Robert J.; Moskowitz, Gordon B.

doi: 10.1177/0146167296224005pmid: N/A

Three experiments obtained evidence that spontaneous trait inferences (STIs) occur on-line, at encoding. In each, participants read many sentences on a computer screen. After each paragraph, they indicated whether it included a test probe word. Paragraphs that imply but do not contain traits should increase errors or reaction times (RTs) to trait probes. In Experiment 1, trait-implying paragraphs produced more errors than control paragraphs, supporting the hypothesis. In Experiments 2 and 3, with feedback on each trial, longer RTs supported the hypothesis. STIs had the same effects as McKoon and Ratcliff's "predicting inferences. " Unexpectedly, participants gained control over STIs and predicting inferences, so that RT differences (and error differences in Experiment 1) declined over trials. Analyses of reading times in Experiment 3 ruled out several alternative explanations. Results demonstrate that social inferences can occur spontaneously at encoding and suggest that immediate feedback may make control possible.

Newman, Leonard S.

doi: 10.1177/0146167296224006pmid: N/A

The dispositionist bias manifests itself when (a) behavior is over attributed to dispositions and (b) contextual factors are underused when predicting behavior. The psychological processes underlying the former bias have been more thoroughly examined than the latter. Three studies support the hypothesis that the trait implications of past behavior function as heuristics used to predict high levels of cross-situational consistency. Subjects in Experiment 1 used both dispositional and situational information to predict behavior, but the level of consistency predicted was inversely related to how much time was spent integrating the available information. In Experiment 2, attributionally complex subjects were less likely to predict high levels of behavioral consistency and were more sensitive to the context in which behavior was to occur. Imposing a cognitive load in Experiment 3 increased noncomplex subjects' tendency to predict consistency across situations.

Herek, Gregory M.; Capitanio, John P.

doi: 10.1177/0146167296224007pmid: N/A

In a two-wave national telephone survey, a probability sample of English-speaking adults indicated their attitudes toward gay men at Wave 1 (1990-91; n = 538) and toward both gay men and lesbians approximately 1 year later (n = 382 at Wave 2). At Wave 1, heterosexuals reporting interpersonal contact (31.3%) manifested more positive attitudes toward gay men than those without contact. Their attitudes were more favorable to the extent that they reported more relationships, closer relationships, and receiving direct disclosure about another's homosexuality. At Wave 2, these findings were generally replicated for attitudes toward lesbians as well as gay men. Cross-wave analyses suggest a reciprocal relationship between contact and attitudes. Theoretical and policy implications of the results are discussed, with special attention to the role of interpersonal disclosure in reducing stigma based on a concealable status.

Diksterhuis, Ap; van Knippenberg, Ad

doi: 10.1177/0146167296224008pmid: N/A

The assumption was tested that organization in memory of behavioral information and recall depends on the descriptive relatedness of consistent information with inconsistent information. Subjects read stereotype-consistent and stereotype-inconsistent behavioral descriptions implying the same trait dimension (e.g., intelligent and stupid behaviors) or different trait dimensions (e.g., intelligent and aggressive behaviors). It was hypothesized that if stereotype-inconsistent behaviors were associated with consistent behaviors because of shared trait implications, these behaviors would be recalled better than stereotype-consistent behaviors. Conversely, it was expected that if subjects were provided with inconsistent information that, because of differential trait implications, could not be associated with consistent information, inconsistent information would be stored separately, and recall of this information would be worse. These predictions were corroborated in a recall task. Conditional recall probabilities and clustering scores supported the proposed underlying organization of information in memory.