Can an experience that simply affirms a valued aspect of the self eliminate dissonance and its accompanying cognitive changes? Three experiments used the conventional forced-compliance procedure to test this question. In the first experiment, some subjects were allowed to affirm an important, self-relevant value (by completing a self-relevant value scale) immediately after having written unrelated dissonant essays and prior to recording their attitudes on the postmeasure. Other subjects underwent an identical procedure but were selected so that the value affirmed by the scale was not part of their self-concept. The value scale eliminated dissonance-reducing attitude change among subjects for whom it was self-relevant but not among subjects for whom it was not self-relevant. This occurred even though the value scale could not resolve or reduce the objective importance of the dissonance-provoking inconsistency. Study 2 showed that the self-affirmation effect was strong enough to prevent the reinstatement of dissonance. Study 3, testing generalizability, replicated the effect by using a different attitude issue, a different value for affirmation, and a different measure of dissonance reduction. These results imply that a need for psychological consistency is not a part of dissonance motivation and that salient, self-affirming cognitions may help objectify our reactions to self-threatening information.

In three experiments, the subjects' task was to decide whether each of a series of words connoted something good (e.g., fame, comedy, rescue)or bad (stress, detest, malaria), One-half second before the presentation of each such target word, an evaluatively polarized priming word was presented briefly to the nondominant eye and was masked dichoptically by either the rapidly following (Experiment 1) or simultaneous (Experiments 2 and 3) presentation of a random letterfragment pattern to the dominant eye. (The effectiveness ofthe masking procedure was demonstrated by the subjects' inability to discriminate the left vs. right position of a test series of words.) In all experiments, significant masked priming effects were obtained; evaluative decisions to congruent masked prime-target combinations (such as a positive masked prime followed by a positive target) were significantly faster than those to incongruent (e.g., negative prime/positive target) or noncongruent (e.g., neutral prime/positive target) combinations. Also, in two of the three experiments, when subjects were at chance accuracy in discriminating word position, their position judgments were nevertheless significantly influenced by the irrelevant semantic content (LEFT vs. RIGHT) of the masked position-varying words. The series of experiments demonstrated that two very different tasks-speeded judgment of evaluative meaning and nonspeeded judgment of word position-yielded statistically significant and replicable influences of the semantic content of apparently undetectable words. Coupled with previous research by others using the lexical decision task, these findings converge in establishing the reliability of the empirical phenomenon of semantic processing of words that are rendered undetectable by dichoptic pattern masking.

Can alcohol make people more helpful, and if so, how? We hypothesized that alcohol would increase helping when, if the person were sober, the helping response would be under high inhibitory conflict--that is, when it would be affected by strong instigating and inhibiting pressures. Alcohol's damage to inhibitory processing should then allow instigating pressures to have more influence on the response, increasing helping. We expected that under low inhibitory conflict, when either or both of these response pressures would be weak under sobriety, alcohol would have little effect on helping. In two experiments we examined this reasoning. In Study 1, a mild dose of alcohol increased helping among high-conflict subjects pressured to help with a task they did not like, but did not increase helping among low-conflict subjects who either liked the task or were weakly pressured to help. In Study 2, a somewhat stronger dose of alcohol increased helping among all high-conflict subjects pressured to help with an undesirable task, yet again had no effect among low-conflict subjects weakly pressured to help. These studies provided an experimental demonstration of the role of inhibitory response conflict in mediating alcohol's social effects, and show that this process generalizes to prosocial behavior. Additional evidence from both experiments helped to rule out alternative explanations concerning drinking expectancies, alcohol's ability to enhance mood, and its ability to make the task more bearable.

No abstract

No abstract