A brief description of recent and current research projects conducted by the Social Psychology Research Interest Group (SPRIG) Lab is given below:


INFORMATION SEARCH AND IMPRESSION FORMATION

Four experiments have been completed on the factors which affect information acquisition during impression formation. All studies found that participants preferred to search trait and behavior attributes significantly more than appearance and demographic attributes when forming an impression of a target person. Although this search pattern was obtained for all gender combinations of participant-target pairs, male participants accessed more appearance information about female targets than did any other participant-target combination. The infrequent use of appearance information was surprising given the importance of appearance in interpersonal relations. Additional findings replicated the preference for trait over appearance information in multicategorical searches (both appearance and trait information available) but found a preference for appearance over trait information in unicategorical searches (only appearance or trait information available). This finding was interpreted in terms of a fundamental preference for more stable trait information and that other information such as appearance, behavior, and demographic data is used primarily to enable a trait inference. It was also determined that information searches were shorter with negative vs positive information (negativity effect) and with extreme vs neutral information (polarity effect). These findings were interpreted in terms of a risk-avoidance impression formation strategy.

PSYCHOLOGY AND LAW

Two experiments have been completed to determine whether defendant status is a harmful liability or a protective shield. Both studies found that a defendant's high status leads to increased sanctions by mock jurors for professionally related offenses (status liability effect) and to decreased sanctions for professionally unrelated offenses (status shield effect). These status effects were obtained for both criminal and civil cases. A third experiment is being planned to determine whether the effects of professional relatedness are a special case of "status relatedness," which refers to whether a defendant uses his or her basis of status to commit a crime.

An experiment tested two aspects of eyewitness identification relating to sex differences and the weapon focus effect. An own-sex identification bias was found similar to the well-documented own-race identification bias, with own-sex identification more accurate than other-sex identification. An alternative explanation for the weapon focus effect based on object salience was not supported. A second experiment examined competing theories of arousal vs. attentional processes and object enhancement vs. reduced perceptual processing as explanations for the weapon focus effect. Results supported the attentional processes and reduced perceptual processing models suggesting that weapon focus may be a special instance of a more general salient object effect.

An experiment with mock jurors simulated important elements of the OJ Simpson murder case. A Black or White high or low status defendant was alleged to have murdered his estranged White wife. Black or White mock jurors read summaries of testimony from prosecution and defense witnesses and rendered mock verdicts and other related judgments. A leniency bias was demonstrated by Black mock jurors when judging the Black defendant regardless of his social status. No other mock juror/defendent race combination displayed this leniency bias. The results are related to the racial composition and defendant's race in the OJ Simpson case.

Two other studies examined the relative influence of defendant motive, ability, and opportunity to commit a crime on jury decisionmaking and the relative influence of eyewitness vs physical criminal evidence on jury decisionmaking. The former confirmed that motive, ability, and opportunity influenced mock jurors' verdicts and sentencing options. In addition, sentencing options were affected by motive only when opportunity was also present. Personal evaluations of the defendant were affected only by evidence of motivation. Results were discussed in terms of implications of these three factors for prosecutors and defense attorneys in criminal proceedings. The latter study confirmed that mock jurors' verdicts and evaluative judgments were influenced to a greater extent by physical evidence than by eyewitness testimony. Strong evidence produced more guilty verdicts than weak evidence. However, combining strong evidence of both types was no more effective than presenting strong evidence of either one. Results were discussed in terms of implications of these factors for prosecutors and defense attorneys in criminal proceedings.

Two additional studies were recently completed to content analyze open-ended linguistic reports of mock eyewitnesses in the weapon focus paradigm. Mock witnesses gave a complete description of everything remembered from a video depicting a classroom intrusion (uncued recall) and everything remembered about the target person (cued recall). Recall reports were obtained immediately, one week after, or one month after viewing the videotape. The witness reports were content analyzed by two independent coders for four social encoding categories, including appearances, behaviors, traits, and demographics Results indicated that witness reports emphasized appearance and behaviors with little mention of traits or demographic characteristics An interaction between type of recall and social encoding category indicated that uncued recall emphasized target person behaviors, whereas cued recall emphasized target person appearances Not surprisingly, more total recall of target person characteristics was obtained under cued than uncued recall and the amount of target person information recalled decreased over the three recall intervals. Most recall loss occurred after one week with little further loss after one month. Although the target person was shown during only 27.3% of the video, 46% of the total information recalled by mock witnesses was about the target person under uncued recall and 93% was about the target person under cued recall. The results have implications for the types of information sought by criminal investigators and the interview conditions under which such information is most likely obtained.

Further research is being planned to examine factors which influence the accuracy of collaborative eyewitnessing, as well as to determine if there are differential position effects in correctly identifying target persons in simultaneous and sequential photospreads.



JURY TRAINING

To address criticisms of the American jury system, a recently complete study compared trained vs untrained individuals on a mock jury decision-making task. Training consisted of completion of a university course on psychology and law focusing on jury decision-making. It was predicted that juror training would enhance juror competence, motivation, and satisfaction with participation in the legal process. Mock jurors were shown a videotape of a rape trial and participated in jury deliberations. The evidence contained in the case was contradictory and did not meet the reasonable doubt standard of proof. Trained jurors voted for acquittal more often than untrained jurors. Additionally, trained jurors were more satisfied, were more confident that their jury reached a correct verdict, and believed more that their jury's decision was based on the evidence presented than did untrained jurors. Content analysis of jury deliberations found that trained jurors were more task oriented and focused on relevant evidence than untrained jurors. The feasibility of implementing a juror training program prior to jury service was discussed.

A second study attempted to determine if the biasing influences of pretrial publicity (PTP) can be mitigated by prior jury training. The effects of PTP from physical and witness evidence on decisions made by trained and untrained mock jurors were compared. It was predicted that physical PTP would influence mock jurors' opinions more than witness PTP, and that juror training would be an effective technique for reducing the prejudicial effects of PTP. Mock jurors viewed a videotaped rape trial and participated in jury deliberations. Training consisted of completion of a university course on psychology and law focusing on jury decision-making. As expected, mock jurors who received PTP based upon physical evidence voted guilty more often than those who received witness PTP and those in a control condition receiving no PTP. Both types of PTP influenced untrained mock jurors' punishment preferences, satisfaction with their jury's verdict, perceptions that their jury's verdict was based upon the evidence vs. personal feelings, and the assessment that all members of the jury had their say. Trained mock jurors' opinions on these measures were unaffected by PTP. Deliberations of trained mock juries were more task oriented and focused on relevant evidence and legal issues than the deliberations of their untrained peers. Limitations of the findings of this mock jury study were discussed.

ATTRIBUTION OF CAUSALITY

An experimental investigation on the structure and dimensions of perceived causality examined how naive perceptions of performance outcomes are structured in temporally linked causal chains that differ in terms of causal locus (internal vs external), variability, and controllability. Participants specified the causes necessary to explain outcomes differing in severity and valence of consequences. Most explanations were three causes in length arranged in a temporal chain. Internal, stable, and controllable causes were used more than external, unstable, and uncontrollable causes. Results were consistent with the fundamental attribution error and the need for predictability and control. It is suggested that complex explanation of necessary causality may be more common than previously assumed, but that dimensional structure remains consistent with established theory.


Another study on attributional chaining investigated how persons link together sequential causes to explain everyday events. Participants were shown a powerpoint presentation consisting of a positive or negative outcome that occurred in one of three domains (achievement, accident, interpersonal) and the proximal cause of that outcome. They were then asked if that completely explained the outcome, or if more explanation was necessary. If so, they were given a medial cause that preceded and led to the proximal cause and again were asked if the outcome had been completely explained. If not they were given another medial cause that preceded and led to the first medial cause. If they wanted more explanation, they were given a distal cause that preceded and led to the second medial cause. Each successive cause alternated the locus dimension (internal vs external). Records were kept of the number of chains used by participants, average length of chains, locus of preferred terminating node, and estimates of outcome forseeability. The results indicated that perceivers chained more often than not. There was less chaining and shorter chains used to explain accidents than achievement and interpersonal outcomes. Negative outcomes led to more and longer chains than positive outcomes for the achievement and interpersonal domains (consistent with the findings in studies of simplex causal attribution that negative outcomes require more explanation than positive outcomes), but this was reversed for accidents where positive outcomes led to more and longer chains than negative outcomes. This was explained by the likelihood that positive accidents are less expected than negative accidents, thus requiring more complex explanations. Thus, the accidental domain seems to be fundamentally different than the achievement domain in terms of eliciting causal attributions. More explanations were terminated at dispositional (internal) nodes than at situational (external) nodes, extending the fundamental attribution error to include complex causal explanations. Mean forseeability was lowest in the accidental domain and was lower for positive than for negative outcomes. This may be why positive accidents produced more and longer chains.

HOMOPHOBIA

A field research technique was used to develop an unobtrusive measure of homophobia. Both men and women were less likely to help a perceived male homosexual than a perceived male heterosexual, contradicting the usual survey findings that men are more homophobic than women. Results are interpreted in terms of both cognitive and affective determinants of homophobia.

CENTRALITY PREFERENCES IN CHOICE BEHAVIOR

Three explanations were explored for the finding that people prefer the middle option rather than the extremes when choosing from an array of similar options. In Study I, 68% chose the middle item from a set of three highlighters and three surveys, whereas 32% chose an item from either end, p < .0001. In Study II, 71% selected the middle chair from a row of three chairs that were either all empty, or else had a backpack occupying either one of the two end chairs, p < .0000. Results support a minimal mental effort principle rather than a preference for symmetry rule. In Study III, 37% of the items recalled from a 3-poster collage were from the center poster, whereas 33% and 30% were from the left and right posters respectively, p < .11. Follow up internal analyses lent additional support to a focus of attention explanation.

Return to home page.