The present study examines personality correlates of personal collectivism (allocentrism) both in an individualist culture (the United States) and in collectivist cultures (Korea and Japan). It was expected those with higher allocentric tendencies would be more concerned with rewards and punishments from in-group members and thus have a lower need for being unique as compared to those with lower allocentric tendencies. Undergraduates in the three countries were administered a Collectivism Scale, the Affiliative Tendency and Sensitivity to Rejection Scale, and the Need for Uniqueness Scale. The results confirmed that allocentric tendencies are associated with higher affiliative tendency, higher sensitivity to rejection, and lower need for uniqueness in all of the three countries. These results support the concurrent validity of the Collectivism Scale across cultures.

Two studies examined self-appraisals in Japanese and Canadian samples. Study 1 included open-ended self-descriptions; Study 2 incorporated indirect measures of self-enhancing tendencies. In Study 1, the content analysis assessed spontaneous evaluations of self and others, private and relational self-statements, reflected appraisals, temporal and social comparisons, and evaluations of objects and events. Canadian participants typically provided self-enhancing self-descriptions. Japanese participants were generally evenhanded rather than self-critical or self-enhancing, although they were more favorable about relational than private aspects of self. In Study 2, Canadian participants reported that proud events felt closer in time and were easier to recall than similarly distant embarrassing events. Japanese participants reported that embarrassing and proud events felt equally far away and were equally memorable. The two studies provide evidence that Canadians possess stronger self-enhancing motivations than do Japanese and enable a cross-cultural analysis of several social psychological theories of self-appraisal.

This study addressed interpersonal factors affecting group entrapment and also attempted to delineate a conceptual link between collective entrapment and I. L. Janis's (1972Janis's ( , 1982 notion of groupthink. Two experiments were conducted in which 3-person groups were assigned either majority or unanimity rule as an official consensus requirement for their initial decision. It was expected and confirmed that groups whose initial decision processes were guided by unanimity rule were entrapped more often to the chosen course of action than were groups with majority rule. The results also suggested that homogeneity of members' opinions at the outset of interaction and group's rationalization norm were responsible for the observed difference. Discussion is focused on the implications of these findings for administrative decision contexts and their conceptual link to the notion of groupthink.Completing Tennessee-Tombigbee is not a waste of taxpayer dollars. Terminating the project at this late stage of development would, however, represent a serious waste of funds already invested.-Senator Sasser, November 4,1981 (Arkes & Blumer, 1985) This statement seems to illustrate most eloquently the nature of a phenomenon called entrapment. Psychological entrapment refers to a faulty decision-making process "whereby individuals escalate their commitment to a previously chosen, though failing, course of action in order to justify or 'make good on' prior investments" (Brockner & Rubin, 1985, p. 5). Delaying withdrawal from a financially failing venture and loaning additional funds to an objectively unpromising project ("throwing good money after bad") are among typical consequences of such decision-making processes. These decision-making processes are also irrational from the perspective of traditional economic theories (e.g., von Neumann & Morgenstern, 1944). Classical economic theories assert that decision makers evalu-

Procedural influence in consensus formation across multiple subgroups was examined. When a cross-sectional decision must be established by a joint conference of all members, 2 consensus procedures are conceivable. The 2-stage procedure encourages each subgroup to coordinate its members' opinions prior to the joint conference; discussion follows 2 steps, first at the subgroup and then at the entire group level. In contrast, the single-stage procedure calls for the joint conference immediately to minimize the opportunities for opinion exchange at subgroups. It was predicted and confirmed that, compared with the single-stage method, the 2-stage procedure enhanced the power of local majorities at the subgroup level and hampered the tendency for the global majorities to prevail in the final decision. Numerical simulations further explored theoretical implications of distinguishing the local and global majorities for collective decision making.Collective decision making is often guided by a variety of formal procedures imposed by a set of bylaws (e.g., quorum rules, voting methods), or informal procedures that social convention or a chairperson sets somewhat discretionarily (e.g., agenda structure, straw polls). Although these procedures are usually justified as means to facilitate consensus processes, their neutrality for the ensuing consensus is not necessarily guaranteed. As pointed out by social-choice theorists (Arrow, 1963;Black, 1958;Fishburn, 1973), even intuitively appealing rules can in principle yield a collective choice that is sometimes counterintuitive and even unacceptable in terms of democratic values. It is therefore essential to explore unintended but potentially hazardous consequences that commonly used decision procedures may cause. The traditional approach to this issue, conducted mainly by economists and political scientists, has been to focus on relatively large decision-making bodies (electorates, congresses), using analytical methods (cf. Ordeshook, 1986). However, rather recently, social psychologists have started empirical assessment of the issue in the context of interactive, small-group decision making (

The experiment was conducted to investigate the effects of smiling expression on cognitive and emotional processes during the introduction of negative mood and cognition by self-focused attention. The mechanisms underlying such effects were also examined, with reference to Interacting Cognitive Subsystems framework (Teasdale & Barnard, 1993). We induced the self-focused attention for all 33 participants but the timing and type of facial expressions manipulated differed among three conditions: control condition (required to move the facial muscles which are unrelated with smiling), buffer condition (required to "smile" before the selffocused attention), and attenuation condition (required to "smile" after the self-focused attention). The results showed that the negative mood was increased in the control group while it was decreased in two experimental groups. Furthermore, the positive mood was decreased in the control group, and increased in the buffer group. The contents of spontaneous thought during experiment were more positive among two "smile" conditions than control condition. These results suggest the importance of smiling before and during negative self-focused attention.