Personality and Social Psychology Bulletin

The conflict between applied and theoretical social psychology is reviewed. It is argued both that this applied-theoretical distinction is a pernicious one which diverts attention from more basic issues of generality and validity and that the applicability of mainstream social psychology is disappointingly slight. While developments in statistical techniques and computer data processing have increased the opportunities for larger- scale, multivariate research which may reflect more accu rately the complexities of the natural world, most reports in the major journals of the field continue to be of small, laboratory experiments using college students as subjects. The prospects for future change in orientation are discussed.

This paper reports the findings of a series of studies dealing with the effects of videotaped trial materials on jurors' re sponses to trial proceedings. In general, jurors who viewed live and videotaped presentations responded comparably with the single exception of retention of trial-related information. Here, jurors who viewed videotaped testimony retained more in formation from the latter portions of the testimony than jurors who viewed the same testimony live. Moreover, jurors who view ed the testimony on monochromatic tape retained more informa tion than jurors who viewed the same testimony on colored tape. Finally, within the procedural confines of these studies, there was no evidence that deletion of varying amounts of inadmissible testimony significantly affected jurors' verdicts or their perceptions of the credibility of the offending attorney. Several important limitations and problems of the studies are also discussed.

Demographic, educational, and research information was abstracted from vitas of 183 applicants for Assistant Professor social psychology positions in 1974 and 1975. Significant differences were found between male and female applicants on criteria indicating research productivity. Some possible explanations are explored.

The distinction between scale value and weight in integration theory is elucidated by considering parallels in Newtonian Mechanics. It is shown that if the impact or force of a piece of information varies from situation to situation, at least a two parameter model, one of which is not a constant, is necessarily required. This is true even for a change of scale value theory. It is further argued that direct empirical proof for the assumption of scale value constancy is not necessary. The validity of the model rests on goodness of fit as must alternative theoretical formulations.

The amount tipped by 396 groups of restaurant diners was a function of the number of people eating together as well as the size of the bill. One-third of the variability in tipping was explained by the norm that tip should equal 15% of bill. In addition, consistent with a new theory of division of responsibility, variation around this norm was an inverse power function of group size, specifically, 18%/N'²².

Three algebraic models for inequity were shown to be simply testable in terms of their additivity structure. The model from integration theory correctly represented the sequence of comparisons, first between persons, then between input and outcome. Also, it gave a good though not perfect account of the additivity structure.

COMPAR is a program which generates descriptions of fictional other persons based on a subject's description of himself. The subject sits at a computer terminal, where he completes an at titude questionnaire. A computer file controls the number of fictional others that will be generated and the degree of atti tudinal similarity between the subject and each of the others. COMPAR automates an otherwise tedious procedure, and subjects enjoy participating. It can be used for a variety of studies. The program deck and instructions are available from the authors.

This field experiment demonstrated that females who were punished for helping in one situation were less likely to help on a subsequent occasion than were females who were reinforced for having helped. In addition, subjects helped a high-dependent person more often than they did the low-dependent person. While subjects usually helped directly, this was less likely to occur for those subjects who were punished for having helped and then were confronted with a low-dependent person. The results were interpreted from a cost-reward model of helping behavior.

Results supportive of Argyle and Dean's (1965) equilibri um theory prediction, relating eye contact and interpersonal distance, have been questioned on methodological grounds. Specifically, the criticism suggests that eye contact only appears to increase with in creased distance between interacting pairs because the observer dis tance also increases and results in overestimation due to decreased discriminability. An evaluation of the evidence on the measurement artifact itself suggests that it is a concern only in contrived set tings involving programmed directional gaze and not in situations involving visual interaction between naive subjects.

Zajonc' theory that the presence of a coactor is suffi cient to facilitate a dominant response was compared with Cot rell's theory that facilitation does not occur unless coactors compete. During 150 baseline trials 80 males individually predic ted the occurrence of two stimuli on a PDP-12 computer screen. For 150 test trials subjects were divided among alone-noncompetition, alone-competition, dyad-noncompetition, and dyad-competition con ditions. In addition, an alone-incentive group was paid 1c per correct prediction in order to assess the effects of drive on the dominant prediction. The results supported Cottrell: only the alone-incentive and dyad-competition groups showed facilitation of the dominant prediction from baseline to test session.

Food sharing behaviors in the white rat were examined in two studies using a 2 X 2 X 2 factorial design with depriva tion of the free roaming animals (deprived or not food deprived), deprivation of the tethered target animals (deprived or not food deprived), and food availability during testing (available or not available) as the independent variables. In Experiment I a single animal served as the tethered target while in Experiment II two animals served as tethered targets. Food sharing and eating behaviors for deprived animals were much like those of the wild rats described by Barnett (1963). The nondeprived animals neither ate nor retrieved-food. No support was generated for an altruism hypothesis which required that the deprivation state of the tethered animal affect retrieval rates by the free roaming animals. The presence of food decreased contact times for the deprived, free roaming animals but not for the non deprived ones (p < .10 Exp I; p < .05 Exp II). In Experiment II more time was spent with nondeprived (larger) targets than with deprived (smaller) targets (p < .001).

A 2x4 split plot factorial experiment investigated the effects of person vs. stimulus attribution and response category (i.e., accomplishments, opinions, emotions and actions) on requests for consistency, consensus and distinctiveness informa tion. Requests for all three types of information were signifi cantly affected by both independent variables, alone and in interaction. Person attribution led to more requests for con sistency and distinctiveness information than stimulus attribu tion, while stimulus attribution led to more requests for consensus. Under person attribution, actions and accomplishments produced more consistency requests than emotions and opinions, while accomplishments produced greatest concern with distinc tiveness. Under stimulus attribution, accomplishments led to greatest concern with consensus.

College subjects were administered Rotter's Interpersonal Trust Scale and a modified version of Bieri's grid procedure for measur ing cognitive complexity for positive and negative social stimuli. As hypothesized, low trusters had significantly more differentiated cognitions for socially distant persons than high trusters. Contrary to hypotheses, socially close persons yielded significantly lesser complexity for high trusters than low trusters and significantly greater complexity for females than males. It was concluded that low generalized expectancies for interpersonal trust might be more adaptive and less problematical than previously contended.

The amount and content of the information exchanged in an interpersonal interaction was examined. Both the medium of exchange and the content of the messages had significant effects on the subjects' responses but personality measures did not. These results are discussed in terms of recent theoretical developments in the area of interpersonal exchange.