The Effects of Hindsight Bias and Causal Attribution on Human Response to Environmental Events1

This research examined the influence of hindsight bias and causal attribution on perceptions of a technological disaster. After reading a fictitious account of a toxic substance spill near a populated area, subjects were provided with information that disease rates had either increased or had remained stable (hindsight conditions), or were presented with no outcome information (control condition). Subjects were then asked to predict the likelihood of increases in disease rates and to make causal attributions regarding the target company and residents of the disaster community. When compared to subjects provided with either no outcome information or with information that disease rates remained stable (positive-outcome condition), subjects told that disease rates had increased (negative-outcome condition) showed elevated predictions regarding future disease rates, ascribed greater responsibility for the accident to the target company, and reported more anger toward the company and greater sympathy for the residents. Subjects receiving positive outcome information and no outcome information did not significantly differ on these measures. In addition, results from a path analysis supported the efficacy of attribution theory to account for the cognitive, affective, and behavioral consequences resulting from hindsight bias following a negative environmental event.     

Locus of causality and the use of information in the development of causal attributions1

Two studies were conducted to examine whether attributions made about events may be influenced by individual assumptions regarding causation that are age related. In Study 1, 96 subjects at three age levels (four and five years, eight and nine years, and college students) observed a target actor on videotape select an item from an unseen array, and four other actors either agree (high consensus) or disagree (low consensus) with the choice. Subjects were asked to decide why the actor liked the chosen object best—because of something about the actor (person attribution) or because of something about the item (entity attribution). The results showed that perceived locus of causality shifted from entity to person attributions with age. In addition, subjects at all ages were able to utilize the consensus information when they had no opportunity to form their own impressions about the items in the array. In Study 2, 126 subjects at four age levels (five and six years, seven and eight years, nine and ten years, and high school students) chose an item from among an array for themselves and responded to a person (self)/entity attribution question regarding the locus of their own choice. The entity to person shift with age was again found and was supported by additional measures. The results are discussed in terms of children's causal reasoning capacities and social environmental factors affecting developmental change in social judgments.     

Secondhand information and social judgment

This research investigates how secondhand impressions of other people differ from those based on firsthand information. It was hypothesized that secondhand impressions are often more extreme because secondhand accounts of another person's actions frequently fail to convey adequately the role of mitigating circumstances and situational constraints in producing that person's behavior. Experiments 1 and 2 tested this hypothesis by exposing “first generation” subjects to information about a target person, having them rate the target on several trait and attribution scales, and having them describe the target person's actions to a group of “second generation” subjects. As predicted, second generation subjects made more extreme ratings of the target than their first generation counterparts. Content analyses of the accounts transmitted by first generation subjects indicated that they did indeed underemphasize various situational qualifications of the target persons' behavior. Experiment 3 extended these findings by demonstrating that people's impressions of someone they have often heard about from a friend (but never met) are more extreme than their friends' more informed impressions.     

Facilitation of social interaction through deindividuation of the target

It was hypothesized that actors want their perception of a target to be consistent with the type of interaction they expect. It was predicted that subjects expecting to aggress would deindividuate their target through the selective recall of deindividuating information. Conversely, subjects expecting a prosocial interaction should individuate the target. Further, angry subjects should deindividuate the individual who angered them. Male subjects were either angered or not angered by an experimental confederate and then given the opportunity to either shock, reward, or have no interaction with him. Subjects recalled information about the confederate either prior to or after the learning task. Subjects expecting to aggress deindividuated the target, whereas subjects expecting a prosocial interaction individuated him. Angry subjects deindividuated the target, nonangry subjects did not. Since the selective recall of information occurred prior to the interaction, the deindividuation (individuation) was aimed at facilitating future behavior rather than justifying it.     

An Attributional (Causal Dimensional) Analysis of Perceptions of Sexual Harassment1

Female and male subjects read one of three scenarios depicting a social-sexual interaction between a man and a woman. The scenarios varied in severity of the stimulus male's behavior according to whether sexual harassment had occurred. Subjects then rated the offensiveness of his behavior and whether it constituted sexual harassment. Subjects also rated the stimulus male's behavior on Weiner's (1986) three basic dimensions of causality: locus of causality, or the perceived cause of the behavior as internal or external; stability, or the likelihood of its recurrence; and controllability, or the amount of volitional influence he exercised during the interaction. Additionally, subjects reported their anger at the target male and sympathy for the target female. Findings generally indicated that for both women and men, as the scenarios increased in severity, these ratings were elevated. Gender comparisons also showed elevations in these ratings for female but not for male subjects, and only in the scenarios of lesser severity. Regression analyses revealed attribution variables—control, stability, and anger—as reliable predictors of perceived offensiveness and harassment in the most severe scenario, whereas affects were the most reliable predictors in scenarios of lesser severity. The implications of these findings for perception of sexual harassment are discussed.     

因果判断中经验与共变信息的结合及各自作用

研究使用相继呈现信息的方法控制了被试获得信息的顺序,从而将经验和共变信息在因果判断中所起到的作用分离了出来。结果表明：（1）个体综合两种信息进行因果判断的过程既不是简单的相加操作,也不是使用经验信息控制共变信息的进入,而是先判断经验信息,再判断共变信息是否与之一致,当出现不一致的情况时又重新考虑经验信息。（2）在改变先前判断的过程中,经验信息所起的作用更大,其中又以当其证明待判断原因不可信时所产生的改变更大。     

Equity and Self-Interest: Effects on Perceptions of Intent,Liking, and Desire to Reciprocate

In two experiments subjects were given information about how one person in a dyadic interaction had allocated outcomes to self and the other participant. The interaction was depicted as competitive in Experiment 1 and noncompetitive in Experiment 2. Subjects made judgments about the allocator's intentions and the recipient's liking for the allocator and desire to reciprocate outcome distribution. Subjects were instructed to respond toward the recipient in either an objective or empathic manner. Subjects logically utilized disparity in outcome distribution in determining the allocator's intent to benefit him/herself. Overall, equitable distributions led to the highest judgments of recipient liking; as inequity increased, ratings about liking decreased. However, empathic subjects in Experiment 2 showed different patterns of rated recipient liking as a function of the direction of the inequity. Judgments about intent to reciprocate were also affected by the orientation of the observer. The results were discussed with particular reference to differential thresholds for advantageous versus disadvantageous inequity. In addition, mathematical models describing the dependent measures were derived and validated, providing results consistent with past research utilizing ratio models of judgment within the context of equity theory.     

The role of observer attitudes in judgments of blame in cases of wife assault

This study examines the importance of observer characteristics in determining blame in cases of wife assault. Four independent variables (observer's attitudes toward sex roles, observer sex and age, and victim behavior) were assessed for their influence on the blaming judgments of 128 participants. Subjects completed a questionnaire that contained demographic items and six wife assault vignettes that varied in level of victim provocation (low or high). Questions about blame of the husband and wife followed each vignette. An attitudes toward women scale (AWS-B) was then administered. The main hypothesis, that subjects with traditional attitudes would blame the victim more and the perpetrator less for the assault than their egalitarian counterparts, was supported, as was the prediction of an interaction between provocation and AWS-B. The results are discussed in light of the role of observer attitudes in attribution models.     

Person, situation, and person-by-situation interaction components in person perception

The hypothesis that people store situation and person-by-situation interaction information as well as trait information about one another's behavior was tested by administering modified S-R Inventories of Anxiousness and Hostility to 54 male and 59 female undergraduates. Subjects rated the strength of affect that would be evoked by a series of situations in the two same-sex college students whom they knew best, the best-known opposite-sex person, a casually known same-sex person, and the “average college student.” Indices were constructed to measure the extent of (a) perceived average differences among people (traits), (b)perceived average differences over situations, and (c) perceived person-by-situation interaction. Analyses of variance showed that: (a) the ratings of the two best-known same-sex persons and the best-known opposite-sex person contained more interactive information than did the ratings of a casually known person and (b) females perceived more situational variability in the behavior of their own sex than did males. The amount of perceived person-by-situation interaction did not differ as a function of sex of rater or affect to be rated. It was concluded that the naive psychologist, like many professional psychologists, is an interactionist.     

“And Who Is My Neighbor?” II: Quest Religion as a Source of Universal Compassion

Are those with a high‐quest orientation to religion less likely to help a person if that person's behavior violates their values of open‐mindedness and tolerance? If so, is it because they have antipathy toward the person or toward the behavior? To answer these questions, sixty undergraduate women were given the opportunity to help either of two same‐sex peers win a monetary prize. About one peer, they knew nothing; from the other, they had received two self‐disclosing notes. The first note either did or did not reveal that the discloser was intolerant of gays; the second revealed that the discloser wanted the money for an activity that either would or would not promote intolerance of gays. Participants scoring high on measures of quest religion helped the intolerant discloser less than the discloser who was not intolerant when their help would promote intolerance; they did not help the intolerant discloser less when their help would not promote intolerance. These results suggest that a high‐quest orientation is associated with antipathy toward the value‐violating behavior (intolerance), not toward the value‐violating person. The scope of the associated compassion seems relatively broad.     

The Effects of Task Characteristics on Covariation Assessment: The Impact of Accountability and Judgment Frame

The impact of personal accountability and judgment frame on strategies for assessing covariation between two binary variables was examined in two experiments. Using a task designed to discriminate between the use of four different strategies varying in degree of sophistication it was found that subjects accountable to an audience with unknown views displayed use of more complex strategies than subjects who were not accountable. In addition, contrary to a recent attribution model (Cheng & Novick, 1992), subjects were less likely to use the conditional probability rule when the judgment question was framed in terms of causality rather than covariation, and covariation judgments did not always parallel causality judgments.     

Information availability,goodness of outcome,and attributions of causality

H. H. Kelley's (American Psychologist, 1973, 28, 107–128) framework for studying attribution processes is introduced as a means of accounting for characteristic asymmetries in success/failure attributions. It is argued that while success/failure asymmetries should occur in the presence of single-observation information, asymmetries should be eliminated when individuals are allowed to observe the covariation between their own actions and outcomes. Subjects participated in a 15-trial stock market simulation in which type of information (single-observation or covariation) and goodness of outcome (relative success or failure) were manipulated. The obtained results supported the experimental hypotheses. Given single-observation information, subjects were more likely to accept personal responsibility for good than for poor outcomes. However, subjects' attributions were not affected by goodness of outcome when they were provided with covariation information. The implications of these findings are discussed in terms of the current debate between motivational and information-processing explanations of asymmetries in success/failure attributions.     

Accuracy of person perception: do people know what kinds of impressions they convey?

Do people know what kinds of impressions they convey to other people during particular social interactions? In a study designed to answer this question, subjects interacted individually with three partners on each of four different tasks. After each interaction, participants reported their impressions of the other person's likability and competence. They also postdicted the impressions they believed they conveyed to the other person along the same dimensions. Accuracy was computed as recommended by Cronbach (1955) and by Kenny's (1981) Social Relations Model. Subjects could tell to a significant degree how the impressions they conveyed to their partners changed over time (time accuracy) and how they changed over time in different ways with different partners (differential accuracy). They could also tell how their competence was differentially perceived by different partners (dyadic accuracy). However, they were not very accurate at discerning which partners perceived them as most competent or most likable across all interactions (person accuracy). Subjects believed that they conveyed similar impressions of themselves to all of their partners, although actually partners evidenced little agreement with each other in their impressions of a given subject. The implications of these findings for symbolic-interactionist theories of the development of the self and impression-management perspectives on social behavior are described.     

Discounting Attributions and Multiple Determinants

The discounting principle in attribution theory was considered a well-established phenomenon until recently, when both the empirical and theoretical basis for discounting have been questioned. Many instances of strong discounting have used measures that constrain explanations, such as a forced-choice and bipolar measures. Two studies were performed simulating Thibaut and Riecken's study of attributions for target persons' compliance to a request, which used a forced-choice measure. In the present studies, a range of measures were used. Responses of two samples of Oxford students and one sample of Oxford teachers indicate that multiple causation was perceived and subjects did not greatly discount either of two relevant causes, but there was nonetheless a compensatory relation between the causes. The results also indicate that the attributions were affected by the social categories of the target person and the subjects.     

The effects of empathic instructions upon donating behavior: Sex differences in young children

First-grade children were given either empathy instructions (in which they were told to imagine themselves in the target person's place) or neutral instructions (in which they were told to listen to what the target person did). For half the subjects, the target person was of the same sex; for the other half, the target person was of the opposite sex. Subjects were then given a chance to anonymously donate money to the target child, who was presented as being in need. Liking for the target child, subject's perceptions of the target child's emotional state, and subjects' report of their own emotional states were also assessed. Empathic instructions increased donating behavior for males, but not for females. Accuracy of perception of the target person's state was generally high. Self-reports of emotional states congruent with that of the target tended to be associated with increased donating behavior only for males, but these self-reports were greater for females than males, and greater for both sexes with same-sex target persons. Liking also was enhanced by gender similarity. These results are discussed in terms of a proposed relationship among empathy, prosocial behavior, sex, and age.We wish to express our gratitude to the staff and students of Wannamaker and North Fairview Elementary Schools of Topeka, Kansas, for their assistance in carrying out this research, and also thank C. Daniel Batson for his helpful suggestions and comments.     

指向具体人物对象的人际信任:跨文化比较及其认知模型

本文对指向具体人物对象的人际信任作了跨文化实证比较研究,并探讨了人际信任的认知模型。使用人际信任行为量表(ITBS)测量了中国、香港及美国三地大学生的人际信任。根据三组被试对诸如父亲、哥哥、同班同学、医生和陌生人等20个人们生活中经常能够接触的人物信任程度的高低,可将这些具体人物对象归为“亲人”,“熟人”和“陌生人”三个群体。三组被试对“亲人”群体的人际信任没有显著差异,但中国被试对其它两个人物对象群体的信任要高于美国和香港被试,而后两组被试的差别则不显著。另外,利用多元回归法在人际信任与六种认知变量间建立了模型关系,发现其中只有两个变量对模型作出主要贡献,即:(1)对由信任行为而得到得回报的期望值越高;(2)得不到回报而引致损害的严重性评估越低,则一个人作出某一信任行为的可能性就越大。被试的文化背景和社会工业化程度对人际信任认知模型没有影响。