INTRODUCTION Most people do not realize how they make decisions or what factors can influence their social judgment. The fundamental attribution error (FAE) occurs when people underestimate the impact of situational forces and attribute a person’s behavior to dispositional factors. The present research looked at possible factors that increase or decrease the likelihood of committing the FAE. The FAE was first documented in an experiment by Jones and Harris (1967). Participants were asked to read an article that was either supportive or was against Fidel Castro, and were told that the author of the article either chose freely to write about the topic or was assigned the topic. When the author freely chose to support Castro, the participants logically inferred that the author supports Castro. However, participants made the same inference when the author was forced to support Castro. Some have theorized that the FAE is a cognitive shortcut (see Forgas, 1998). At least in the short-term, it may be much easier to judge a person based on internal characteristics rather than taking into account all possible factors. To accurately judge a person’s behavior would involve considering the person’s personality as well as the situation surrounding the behavior. This kind of judgment would take more effort to process. Thus if a judge is forced to think about the actor’s behavior, there would be less likelihood of committing the FAE. However, social judgments are often made quickly and on the spur of the moment. So, what possible short-term factors could influence the likelihood of the FAE and how could these factors be manipulated to decrease social misjudgment?Forgas (1998) demonstrated that temporary mood can influence dispositional attributions. To induce a happy or sad mood, subjects were given false feedback on a verbal test. Afterwards, participants read a coerced essay on an unpopular opinion. Participants in a happy mood were more likely to commit the FAE whereas participants in a sad mood were less likely to commit the FAE. This finding was also replicated in a field study at a movie theater. Movie goers who saw either a happy or sad film were later asked to read a coerced essay and then to judge the author. Again, those supposedly in a happy mood were more likely to commit the FAE than those in a sad mood. Forgas explains that happy moods inhibit a deliberate thought process, thereby making social judgments less accurate. The FAE is a heuristic used to simplify a complex situation. When in a happy mood, the person is less motivated to understand the situation and more likely to use superficial processing strategies such as the FAE. The way in which a situation is framed affects the way a person remembers and evaluates the situation. Kuvaas and Selart (2004) gave research subjects a simulated business scenario. Subjects were told that the business scenario was either highly successful in the past (positively framed) or not successful (negatively framed). The positively framed scenario produced lower recall and worse evaluation techniques in participants. The opposite was true when the scenario was negatively framed. This suggests that the way in which a situation is perceived affects how a person reacts and interprets the situation. Likewise, there may be cognitive or mood variables that affect how a person judges another’s dispositional qualities.That is, positive affect can deplete one’s cognitive resources (Oaksford, Morris, Grainger, & Williams, 1996). Participants were asked to get into either a happy or a sad mood with the aid of a video clip. After this, participants had to perform the Tower of London puzzle. In this task different colored blocks on pegs have to be moved from an initial configuration to a goal configuration in as few moves as possible. When in a positive mood, participants made more moves to complete the task than the control group. Oaksford and his colleagues argued that a positive mood impaired the subjects’ planning and memory efficiency. In a similar fashion, it’s possible that mood may adversely affect the way a person makes social decisions (Forgas, 1998). But it is also possible to minimize mood effects. Ambady & Gray (2002) hypothesized that participants in a sad mood would be more accurate at making social judgments when given a cognitive load task. In other words, giving the participant something to think about would erase whatever mood effects there happen to be. To induce mood participants were shown a segment from a sad movie. After this the subjects had to judge the relationship status of couples from 15 second video clips. Before seeing the video clips, some participants were given a cognitive load task (counting backward from 1000 by intervals of 7). These subjects were more accurate at judging the relationship status of the couples than those not given the cognitive load task. Further, the subjects given the cognitive load task performed at the same level as participants in a neutral mood who were not given the cognitive load task. In a way, giving the participant something to think about diminished the manipulated mood effect. The above studies suggest that mood reduces the cognitive processing of information. Some researchers argue that a happy affect impairs social judgment and decision making (Forgas, 1998; Oaksford, Morris, Grainger, & Williams, 1996). It is also possible to eliminate induced mood effects by giving participants a cognitive task (Ambady & Gray, 2002). If Forgas (1998) is right that a positive mood will increase the likelihood of the commission of the FAE, then it may be possible to eliminate this effect by making the subject think about the situational factors. Mak (2003) attempted to do this in an experiment on the effects of mood (i.e. positive or neutral) and target (i.e. self or other). By making the participant think the situation is about the self, there would be less likelihood of making dispositional attributions. To induce mood, subjects read either positive or neutral sentences and then wrote essays about either a happy memory or about a landmark. After this the subjects were presented with two scenarios. One scenario was about a person named “Jaime” ( the other), the other scenario was about the participant (the self). When judging Jaime, participants were more likely to commit the FAE when in a positive mood. However, participants in a positive mood were less likely to make dispositional attributions when judging oneself. The problem with this study is that the target variables are confounded with the actor-observer bias, which occurs when a person attributes situational factors while explaining his own behavior. In Mak’s study we do not know if the participants were making situational attributions because they were forced to think through the scenario, or if we are simply observing the actor-observer bias.The current research attempts to unravel this confound and replicate Mak’s findings. Instead of looking at target, I suggest studying personal relevance. Specifically, what happens to the FAE when the subject is personally involved in the situation? Three hypotheses are proposed: a) people in a positive mood will be more likely to commit the FAE; b) people in a personally relevant situation will be less likely to commit the FAE; and c) when in a positive mood people will be less likely to commit the FAE if in a personally relevant situation. This study will also look at how agreeableness is related to committing the FAE.
MOOD MANIPULATION To manipulate mood in the laboratory, Forgas (1998) gave the participant false feedback on a verbal abilities test. Forgas was successfully able to induce positive, neutral, and negative moods in his subjects using this technique. There are two disadvantages to giving false feedback. For one thing, the participant is being deceived; deception should be avoided when possible in doing research. Second, the false feedback may be manipulating the participant’s self-esteem instead of affect. That is a confound that is easily avoidable. Mak (2003) had participants read sentences that were either positive (e.g. “I am feeling good today”) or neutral (e.g. “the earth rotates around the sun”). After this, the participants were asked to write about a positive mood manipulation (i.e. a happy memory) or a landmark (i.e. neutral mood manipulation). Mak did not conduct a manipulation check, so it is impossible to know if her manipulations worked, but the results of her study were moderately significant. Petty, Schumann, Richman, and Strathman (1993) had subjects write about a “positive life event” to induce mood. Participants reported feeling significantly more positive on a mood index than the participants in the control condition, who had listened to classical music instead of writing about a life event. An advantage to using the above technique is that it can easily be done in a large group setting, as opposed to giving false feedback or showing participants video clips (Forgas, 1998; Ambady & Gray, 2002).
Participants were 53 undergraduate students enrolled a general psychology course at Winona State University. There were 43 females and 10 males. Their ages ranged from 18 to 22, with an average of 19.25 (SD = 0.92). For their participation, extra credit was given.
The materials consisted of a five page packet. On the first page the participants were asked to write a short paragraph. In the positive mood conditions, the paragraph had to be about a happy memory in the participant’s life. In the neutral mood conditions, the participants were asked to describe a landmark in the United States. The next page informed participants that there would be an essay that was written by a senior at Winona State. The participants were informed that the author of the essay, Ms. Thompson, was told to take a specific viewpoint.Ms. Thompson’s essay began on the next page. The essay argued why college tuition should be raised, which was judged by the researcher to be an important issue that university students would feel strongly about. In the personally relevant situation conditions, the essay discussed the tuition raises at Winona State. In the irrelevant situation conditions, the essay discussed tuition raises at a similar nearby university.The next page of the packet was a short questionnaire with a seven point Likert scale, with seven being the highest score. Three questions probed the commission of the FAE: 1) how much does Ms. Thompson agree that tuition should be raised? 2) How likely is Ms. Thompson to support a tuition raise? and 3) Do you think Ms. Thompson is highly active in politics? These questions were designed to measure what the participant thinks the author believes about tuition raises. The third question looks more at how the participant views Ms. Thompson’s behavior (i.e. if she is not active in politics, then perhaps she does not fully grasp the issues around tuition raises). A forth question asked if the participant agreed with the author. The last page of the packet was the mood manipulation check. Three positive adjectives and three negative adjectives were listed, and the participants had to indicate on a seven point Likert scale how much the adjectives described their current mood.
The design is a 2(positive mood vs. neutral mood) x 2(personally relevant situation vs. irrelevant situation). The participants were gathered as a large group in a class room setting. The experimenter handed out the packets and instructed the participants to not skip ahead in the packet. The experimenter also verbally noted that the essay topic was assigned to the author. Participants were given as much time as they needed to fill out the packet. Once they were done, the packets were collected. Participants were debriefed and thanked for their time.
MOOD MANIPULATION CHECK
The mood manipulation was assessed by having participants rate six adjectives on a seven point Likert scale according to how they felt. The ratings of the positive adjectives (cheerful, ecstatic, and joyful) were highly correlated with each other and when combined were highly reliable (Cronbach’s alpha = .888). The negative adjectives (sorrowful, dejected, and somber) were not as well correlated and were not as reliable (Cronbach’s alpha = .617). The negative adjectives were reversed scored and then averaged together with the positive adjectives to create a mood manipulation check score. The mean of the mood manipulation check in the positive mood condition was 4.81 (SD = 1.16). In the neutral mood condition the mean was 4.22 (SD = 1.20). An independent samples t-test revealed no significant difference between the positive and neutral mood conditions, t(1,34) = 1.46, p = .154.
Three items on the dependent measure questionnaire were designed to measure the commission of the FAE: does the author agree tuition should be raised, how likely is the author to support a raise, and how active is the author in politics. These items appeared to have sufficient face validity and were well correlated with each other. The items were combined into a single score measure the FAE (Cronbach’s alpha = .773). A higher score on this measurement indicates that the person was likely committing the FAE.See Table 1 for cell means and standard deviations. An 2 x 2 analysis of variance (ANOVA) revealed no significant interaction effect for the mood and personal relevance conditions, F(1,49) = 1.86, p = .179. The positive conditions had a mean score of 4.89 (SD = 1.39), whereas the neutral mood groups had a mean of 5.29 (SD = .98). These groups were not significantly different, F(1,49) = 1.37, p = .248. The relevant conditions had a mean of 5.08 (SD = .94) and the irrelevant conditions had a mean of 5.12 ( SD = 1.42). Again these groups were not significantly different, F(1,49) = .008, p = .931.
AGREEMENT WITH THE AUTHOR
Participants were asked how much they agreed with the author of the essay. In the positive mood condition, there was a mean agreement score of 4.04 (SD = 1.37). In the neutral mood condition there was a mean agreement of 3.61 (SD = 1.64). A one-way ANOVA revealed no significant difference between these two groups, F(1,49) = 2.31, p = .323. No significant difference was found for the relevant conditions (mean = 3.54, SD = 1.50) or irrelevant conditions (mean = 4.07, SD = 1.52), F(1,49) = 1.65, p = .206. A 2-way ANOVA revealed no significant interaction effect, F(1,49) = .19, p = .67. I also wanted to see if agreement with the author was somehow related to the likelihood of committing the FAE. For example, are participants who disagree with the author more likely to commit the FAE? There was no significant correlation between how much the participants agreed with the author and the commission of the FAE, r = -.06.
DISCUSSION The results of this study do not support the hypothesis that positive affect and personal relevance interact to alter the likelihood of committing the FAE. Lack of main effects for the personal relevance conditions suggests that a person’s perception of the situation may not have a role in the commission of the FAE. Theoretically that makes sense, as when one commits the FAE one is making a judgment about the actor in the situation and not the situation itself. The positive mood groups were less likely to commit the FAE than the neutral mood groups, although this difference was not significant. From the manipulation check, it appears that the manipulation was not strong enough to produce differences in mood. Also the manipulation check itself did not work out very well, as it was apparently too ambiguous. Some of the participants were confused about how to fill out the sheet, consequently much data was not included in the manipulation check. And the negative adjectives, particularly dejected and somber, were not well understood by the participants (e.g., some participants reported feeling both cheerful and somber). Perhaps a better way to manipulate a happy mood would be to use humor, either in the form of a video clip or having the subject read funny jokes and stories. Ambady and Gray (2002) used a video clip of a comedian telling jokes to induce a positive mood in participants. This manipulation was very successful in their study. There was no correlation between agreement with the author and committing the FAE. This indicates that judging a person to hold a belief is not the same as being persuaded to hold the same belief. It can be noted that when in a positive mood participants were more likely to agree with the author that tuition should be raised, but this effect was not significant. Also participants were less likely to agree with the author in the personally relevant situation, again this effect was not significant. This study shows that affect and personal relevance do not predict the commission of the FAE. However, the mood manipulation was not strong enough to elicit happiness in research participants. If this study is replicated, a stronger methodology should be employed.
REFERENCES Ambady, N. & Gray, H. M. (2002). On being sad and mistaken: Mood effects on the accuracy of thin-slice judgments. Journal of Personality and Social Psychology, 83(4), 947-961.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.Jones, E. E., & Harris, V. A. (1967). The attribution of attitudes. Journal of Experimental Social Psychology, 3, 1-24.Kuvaas, B. & Selart, M. (2004). Effects of attribute framing on cognitive processing and evaluation. Organizational Behavior and Human Decision Processes, 95, 198-207.Mak, W. (2003). Good feelings produce bad judgment: The effects of mood and target on the fundamental attribution error. UCLA Undergraduate Psychology Journal, 2(1).Oaksford, M.; Morris, F.; Grainger, B.; & Williams, J. M. G. (1996). Mood, reasoning, and central executive processes. Journal of Experimental Psychology: Learning, Memory, and Cognition, 22(2), 476-492.Petty, R. E.; Schumann, D. W.; Richman, S. A.; & Strathman, A. J. (1993). Positive mood and persuasion: Different roles for affect under high- and low-elaboration conditions. Journal of Personality and Social Psychology, 64(1), 5-20.
Cell Means and Standard Deviations of Mood and Personal Relevance Conditions Mood ________________________Personal Relevance Positive Negative___________________Personally Relevant n = 12 n = 14 x-bar = 5.11 x-bar = 5.05 SD = 1.07 SD = .86 Personally Irrelevant n = 13 n = 14 x-bar = 4.69 x-bar = 5.52 SD = 1.65 SD = 1.07