|
Good Feelings Produce Bad Judgment: The Effects of Mood and Target on the Fundamental Attribution Error
Wingyun Mak University of California, Los Angeles
Abstract This experiment explores the effects of mood (positive, neutral) and target (self, other) on the occurrence of the Fundamental Attribution Error (FAE). The actor-observer bias is also investigated. After receiving a positive or neutral mood manipulation, participants read two fictitious scenarios detailing a negative event with oneself or another person as the actor in the scenario. Each actor functioned as a target of evaluation. Levels of dispositional versus situational attributions were assessed, as were attributions of responsibility. Results showed that positive mood increased the occurrence of the FAE when judging others, whereas it reduced the occurrence of the FAE when judging oneself. Implications for this study are discussed in terms of the significance of attributional judgment within social relationships. Good Feelings Produce Bad Judgment: The Effects of Mood and Target on the Fundamental Attribution Error Attribution, defined as the process of explaining the causes of people�s behavior, including one�s own (Bernstein, Clarke-Stewart, Penner, Roy, Wickens, 2000), can often lead to mistaken judgments. Rather than recognizing the power of situations, we often attribute a person�s behavior to dispositional factors and ignore the influence of external forces. In a study in which subjects were assigned to write either a pro-Castro or anti-Castro essay, Jones and Harris (1967) found that judges were more likely to interpret the pro-Castro essays as a direct reflection of the subjects� personal beliefs despite knowing that the subjects had been assigned to write with that point of view. This attribution of behavior to intrinsic personality characteristics is called the Fundamental Attribution Error (FAE) (Ross, 1977), or the correspondence bias (Gilbert & Malone, 1995), and has been observed in many people (Nisbett & Ross, 1980). Furthermore, the actor-observer bias (Jones & Nisbett, 1972) indicates the significance of the target of evaluation during the process of attribution. The FAE occurs more frequently when judging others than when judging oneself. Observers pay more attention to the actor in a situation, rather than the situational factors that may affect his or her behavior. When making judgments about others, the observer makes dispositional attributions based upon simple heuristics or stereotypes that require minimal cognitive effort. When judging the self, however, one seeks to minimize dispositional causation, resulting in a natural tendency to view situational constraints as the cause of one�s own shortcomings. Attribution error is an important issue because it can result in social problems (e.g. different interpretations of the same homicide case, misunderstandings in the workplace, misunderstandings within a marriage). Orvis et al. (1976) asked college-aged couples to list a series of behaviors for which the partners had different explanations. When asked to list their reasons for those behaviors, each individual was more likely to attribute their own behavior to situational factors while they were more likely to attribute their partner�s behavior to dispositional factors. The fallibility of attribution may lead to misunderstanding and more serious chasms in the relationship, potentially resulting in strife or the termination of the relationship. As such, attribution error can have lasting negative effects (e.g. Heradstveit & Bonham, 1996). Mood affects decision-making and, potentially, attribution. Forgas, Bower, and Moylan (1990) found that happy people are more likely to attribute their successes to intrinsic characteristics and their failures to external causes. Although positive affect can increase cognitive flexibility and creativity (Isen et al., 1992), people in a positive mood may also dedicate less attention to information processing and rely more heavily on stereotypes or other generic categories (Isen, 1987). Boderhausen, Kramer, and Susser (1994) included a mood manipulation (e.g. positive and neutral) before asking their participants to rate the guilt of a criminal who was given an ethnic name (e.g. Juan Garcia) or a generic name (e.g. John Garner). They found that participants with an induced happy mood were likely to judge the criminal with an ethnic name more severely than the criminal with the generic name. However, there was no significant difference between the two ratings by the participants in a neutral mood. These results suggest that positive mood may cause participants to rely less on evidence and make judgments based upon stereotypes. Because mood has a significant impact on one�s judgment in social situations, it likely also plays a crucial role in attribution. Few studies have explored mood effects on the Fundamental Attribution Error (Forgas, 1998), and none have addressed the lingering question of how the target of evaluation might affect the FAE. Because positive affect reduces a person�s cognitive effort (Clark & Isen, 1982), Forgas (1998) found an increase in the occurrence of the FAE in happy subjects who were judging the behavior of others. In this case, attribution was highly dependent on the subject�s mood rather than the target of his judgment. In the present study, we explore whether both mood (e.g. good vs. neutral) and target (e.g. self vs. other) affect the occurrence of the FAE. After a short mood manipulation, participants are presented with two fictitious scenarios and then asked to judge the actor in the scenario. Consistent with the actor-observer effect, we predict that there will be a main effect of target such that there will be a low occurrence of the FAE during self-evaluation and a high occurrence of the FAE when judging others. Mood should also produce a main effect such that positive mood should elicit a higher occurrence of the FAE than the neutral mood. An interaction should exist such that the rate of the FAE will be highest when the participant is in a good mood and evaluating someone else. Method Participants Twenty-four students (12 female, 12 male) were randomly selected from the University of California, Los Angeles, to fill out the prepared packets. Participants were not paid. Design and Procedure The experiment was a 2 � 2 mixed design, with mood (positive, neutral) as the between-subjects independent variable and target (Jamie, You) as the within-subjects independent variable. Participants were presented with a packet with these directions on the cover sheet: �Please read the packet and follow the corresponding instructions. Thank you.� The second page contained a list of statements as the initial part of the mood manipulation. Half of the packets contained positive statements, and the other half contained neutral statements. The positive statements were: �1. I can do anything I put my mind to. 2. I am feeling good today. 3. Smiling is good for my health. 4. I will not allow trivial problems to bother me. 5. I am happy with who I am. 6. Each new day brings new possibilities. 7. I am loved.� The neutral statements were: �1. UCLA was established in 1919. 2. The earth rotates around the sun. 3.Refrigerate after opening. 4.There are 26 letters in the alphabet. 5. Yellow and blue make green. 6. Humans breathe oxygen. 7. Books are made of paper.� Students were instructed to read these statements. The second part of the mood manipulation was continued on the third page where participants read the following instructions: �In a paragraph or two, describe the happiest experience you have had. What was the event? What led up to it? How did you feel? What was its significance?� (positive mood manipulation) or �Write about two UCLA landmarks that you would recommend a visitor to see. Include the location, general description, and significance of each.� (neutral mood manipulation). After the mood manipulation, participants were presented with two scenarios, one about a fictitious person with the gender-ambiguous name, �Jamie,� the other about the participant, referred to as, �You.� We created two scenarios where we alternated the subject of the story and the order in which the scenarios were presented in the packet. Scenario 1: �After a late night out and a faulty alarm clock, ____ wake up late and must rush to work. After a fast shower and dressing quickly, ____ proceed to the car only to slip and fall on an oil slick left by another car in the street. ____ clothes are stained and torn, so ___ go back into the house to change. Once dressed again, ___ get into the car and head to work. On the way to work, ___ neglected to stop at a stop sign, which is obscured by a tree. Consequently, ___ get pulled over by a police officer and receive a ticket for neglecting to stop at the sign. Although most certainly late, ___ continue to rush to work, in the hopes that the supervisor hasn�t arrived.� Scenario 2: �___ is a college student who lives with three other roommates in a two-bedroom apartment. When they decided to live together, they agreed to share household chores. ___ understood that they would each handle the various chores on a weekly rotation. However, because of financial constraints, ___ must take a heavy workload of 20 units to graduate early. As result, ___ has less free time. Despite such a hectic schedule, ___ will not miss watching the daily back-to-back reruns of Seinfeld. After a month of failing to complete the household chores, the roommates confronted ___ about the unfinished chores. During this discussion, the roommates were reminded of last quarter when ___ covered most of the chores for them when they were busy. The discussion ended unresolved. After reading each scenario, each participant was asked to answer six questions (e.g. To what extent is the mess Jamie�s responsibility? How lazy is Jamie?) about the previous scenario (see Appendix 1). On a seven-point Likert scale, participants made attribution judgments about the subject of the story (1-not at all, 7-completely), where higher values were associated with dispositional attributions and lower values were associated with situational attributions (except for Question 5 where the values were reversed). After participants returned the packet, they were given a debriefing sheet, outlining the purpose and hypotheses of our study. Results Mood manipulation In response to the mood manipulation tasks, participants, on average, wrote a 3-5-sentence paragraph dealing with either a positive or neutral topic. �Positive� paragraphs included details such as, �Even though it was tough to get used to, overall the trip was so beneficial for me emotionally and just in general with the person I became after the experience,� expanding on how the event affected the participants� feelings and thoughts. �Neutral� paragraphs included details such as, �The Bruin Bear is the school�s mascot which is a symbol that unites or represents UCLA students,� providing information about the significance of the landmarks without linking it to a participant�s feelings or thoughts about it. Attribution tasks A preliminary analysis, using a two-way analysis of variance (ANOVA), showed that there was no significant effect of scenario on the data. We predicted a main effect of mood such that a positive mood should elicit a higher occurrence of the FAE than a neutral mood. Using a two-way ANOVA, we evaluated the effects of mood on attribution collapsed across scenarios. There was a main effect of mood: judgments made during a positive mood produced more dispositional attributions (Ms = 4.08, 3.96 for positive and negative moods, respectively, F (1,22)=3.31, p<. 09). We found a marginally significant effect of mood on the FAE such that positive mood increased its occurrence (see Table 1). We also predicted that there would be a main effect of target such that there would be a lower occurrence of the FAE during self-evaluation and a higher occurrence of the FAE when judging others. There was a main effect of target: Judging Jamie produced more dispositional attribution than judging the self (Ms = 4.25, 3.78 for Jaime and self, respectively, F (1,22)=3.214, p<. 09). Using a repeated measures analysis, we observed a marginally significant effect of target on the FAE such that judging others (e.g. �Jamie�) would increase the occurrence of FAE (see Table 2). Lastly, we predicted an interaction between mood and target such that the rate of the FAE would be highest when the participant was in a positive mood and evaluating Jamie. The results yielded a marginally significant interaction between mood and target on the occurrence of the FAE, F (1,22)=3.871, p<. 07 (Ms = 4.569, 3.583, 3.933, 3.979 for pos-Jamie, pos-self, neut-Jamie, neut-self, respectively; see Figure 1). As predicted, subjects were more likely to attribute negative life events to dispositional rather than situational factors when judging Jamie while in a positive mood. Because of the marginally significant effect of mood on attribution, we conducted post-hoc tests on each question within the dependent variable measures. Using an independent samples t-test, we found that two questions (1 & 3) offered stronger evidence in support of our hypothesis. The results of question #1 showed that when participants were making judgments, they were more likely to assign responsibility to Jaime (i.e. Jaime was responsible for his actions) than assign responsibility to themselves, t (22)=3.358, p<. 003. Results for question #3 showed that people were more likely to see Jamie as lazy and careless, as opposed to themselves, t (22)=2.128, p<. 045. DiscussionIn support of our original hypotheses, the results of the present experiment suggest that positive mood increases the occurrence of the FAE while judging others. However, when judging the self, participants in the positive mood were actually less likely to commit the FAE, although the difference in means between positive and neutral moods in judging oneself was less than the difference in means between positive and neutral moods while judging Jamie. The results of the present experiment reaffirm the conclusions of Forgas (1998), that temporary positive moods increase a person�s tendency to commit the FAE. However, instead of merely exploring the effects of mood on judgment, the present experiment incorporated the factor of target and showed that positive affect does not increase the occurrence of the FAE during self- evaluation. This extension of Forgas� (1998) findings shows how attributional judgment is substantially influenced by the actor-observer bias as well as mood, specifically seen in the higher levels of dispositional judgment for Jamie and the higher levels of situational judgment for the self. One of the major limitations of the study is the small sample size and the lack of diversity within the group. Our sample was entirely made up of college students, and because most of them were likely preoccupied with other more pressing issues, it is plausible that they did not put too much effort into reading and answering the questions within the packet. A larger, more diverse sample size, including middle-aged and older participants from different backgrounds, may have strengthened the results and yielded more conclusive evidence of the effect mood and target of evaluation have on attribution. A monetary incentive may have helped as well. However, the fact that there was marginal significance from such a small sample size indicates that this is a promising area of research. Another weakness of our study may be that the scenarios and the mood manipulation did not generate a realistic reaction from the participant. Because of their fictitious nature, judgments based upon these scenarios may not have been as accurate as they would have been in actual situations. Instead of creating fictitious scenarios, perhaps seeking participants who are experiencing certain difficulties in their social relationships (e.g. with a problematic co-worker or teacher or family member) might increase the validity of the study. Although this might require additional measures to screen out extraneous variables, the real situations will elicit more accurate feelings and reactions from the participants. Similarly, the mood manipulation seemed to lack effectiveness. Participants were not visibly affected by the two-part task to induce either a positive or a neutral mood. We did not have a measure to check if the manipulation was successful or if it lasted throughout the duration of the task. However, we decided to forego the manipulation check in hopes of reducing the transparency of our design. A manipulation check may also have eliminated the mood it was originally set to check. Further research should be done to study why mood has an opposite effect on self-evaluation. Although our results suggest that neutral mood increases dispositional attributions (i.e. Positive mood will reduce the occurrence of the FAE in self-evaluation.), a possible explanation behind this phenomenon is that mood may not be such an important factor when judging oneself. The tendency to attribute one�s shortcomings to situational factors may have a stronger effect on how people evaluate oneself, regardless of mood. People have more at stake when judging the self, thus mood may not sufficiently cause them to neglect maintenance of a favorable perception of the self. Additionally, because both of the fictitious scenarios presented negative situations, we cannot be sure if the decrease in the FAE when judging the self is a result of the actor-observer effect or a self-serving bias. The negative scenarios may have provoked participants to feel defensive, thus resulting in a decrease in the FAE when judging themselves. To eliminate this confound, future studies should include, in addition to negative scenarios, neutral and positive scenarios to see if the type of scenario has an effect on the occurrence of the FAE. Understanding the patterns of people�s judgments can offer insights that will likely improve interpersonal relationships. Because attributions are made everyday, many of which have major repercussions, research on the factors that affect attribution is extremely important. In the workplace, erroneous attribution may cause misunderstandings and lead to termination of employment. In a friendship, an erroneous attribution may harm a meaningful relationship. Divorce may become the solution to blame that was carelessly assigned. This type of research will inform people of their tendencies to make faulty attributions while judging others. Studying the interaction of mood and target during judgment will raise awareness of our biases, despite our belief that we can produce objective judgment. The awareness, alone, may reduce unnecessary grief and misunderstanding in our society.
References Bernstein, D.A., Clarke-Stewart, A., Penner, L.A., Roy, E.J., Wickens, C.D. (2000). Psychology. Boston: Houghton-Mifflin. Boderhausen, G.V., Kramer, G.P., & Susser, K. (1994). Happiness and stereotypic thinking in social judgment. Journal of Personality and Social Psychology, 66), 621-632. Clark, M.S., & Isen, A.M. (1982). Towards understanding the relationship between feeling states and social behavior. In A.H. Hastorf & A.M. Isen (Eds.), Cognitive social psychology (pp. 76-108). New York: Elsevier-North Holland. Forgas, J.P. (1998). On being happy and mistaken: Mood effects on the fundamental attribution error. Journal of Personality and Social Psychology, 75(2), 318-331. Gilbert, D.T., & Malone, P.S. (1995). The correspondence bias. Psychological Bulletin, 117, 21-38. Heradstveit, D., & Bonham, G.M. (1996). Attribution theory and Arab images of the Gulf War. Political Psychology, 17, 271-292. Isen, A. (1987). Positive affect, cognitive processes and social behavior. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol. 20, pp.203-253). San Diego, CA: Academic Press. Isen, A.M., Niedenthal, P.M., & Cantor, N. (1992). An influence of positive affect on social categorization. Motivation and Emotion, 16(1), 65-78. Jones, E.E., & Harris, V.A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology, 3, 1-24. Jones, E.E., & Nisbett, R.E. (1972). The actor and the observer: Divergent perceptions of the causes of behavior. In E.E. Jones et al. (Eds. ), Attribution: Perceiving the Causes of Behavior (pp. 79-94). Morristown, NJ: General Learning. Nisbett, R.E., & Ross, L. (1980). Human Inference: Strategies and Shortcomings of Social Judgment. Englewood Cliffs, NJ: Prentice Hall. Orvis, B.R., Kelly, H.H., & Butler, D. (1976). Attributional conflict in young couples. In J.H. Harvey, W. Ickes, & R.F. Kidd (Eds.), New Directions in Attribution Research (Vol. 1, pp. 353-386). Hillsdale, NJ: Erlbaum.
Ross, L. (1977). The intuitive psychologist and his shortcomings:
Distortions in the attribution process. In L. Berkowitz (Ed.), Advances
in Experimental Social Psychology (Vol. 10, pp. 173-220). San Diego, CA:
Academic Press. Author's Note
This project is successfully completed largely because of the creativity and
hard work of Daniel Covarrubias, Rebecca Pedooem, Karen Yi, and Shirin Nassi. Appendix 1. Questionnaires for the Dependent Measure Scenario #1:
1.To what extent is/are ___ responsible for ___ tardiness? 1 2 3 4 5 6 7 not at all completely
2. How likely will ___ be late again? 1 2 3 4 5 6 7 not at all completely
3. How careless is/are ___? 1 2 3 4 5 6 7 not at all completely
4. How negligent is/are ___? 1 2 3 4 5 6 7 not at all completely
5. How organized is/are ___? 1 2 3 4 5 6 7 not at all completely
6. How lucky is/are ___? 1 2 3 4 5 6 7 not at all completely
Scenario #2: 1.To what extent is the mess ___ responsibility? 1 2 3 4 5 6 7 not at all completely
2. How likely is/are ___ going to continue to ignore the chores 1 2 3 4 5 6 7 not at all completely
3. How lazy is/are ___? 1 2 3 4 5 6 7 not at all completely
4. How messy is/are ___? 1 2 3 4 5 6 7 not at all completely
5. How diligent is/are ___? 1 2 3 4 5 6 7 not at all completely
6.How stressed is/are ___? 1 2 3 4 5 6 7 not at all completely
Table 1.
Table 2.
Figure Captions Figure 1. Mean scores for attributional questions as a function of positive (1) and neutral (2) moods.
Copyright 2003 by the Undergraduate Psychology Journal. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
