(1977), the vast majority would pick the sweater farthest to the right. When asked to explain why, most would likely state that the one on the right was of better quality than the other three. You would be unaware that the physical placement of the sweaters had an impact on your choice. Why? Because humans often have little access to and knowledge of our internal processes. Nisbett and Wilson (1977) have provocatively argued that, although we know the result of our thought processes (e.g., what our third-grade teacher’s name was), we often do not know how we arrived at that result.
All of us engage in introspection in our efforts to better understand ourselves.
David Grossman / Science Source.
In the fascinating book Strangers to Ourselves, Timothy Wilson (2002) reviews the vast research literature on this topic that provides convincing evidence regarding the limits of introspection. Wilson shows that not only are we unaware of how are thoughts are produced, we often don’t know who we are, what we feel, or why we do what we do! In another study, participants watched a film either while a very loud power saw was operated just outside of the room or with no distracting noise (Nisbett & Wilson, 1977). Respondents were asked to rate the film on a number of dimensions, and the distraction participants also indicated whether or not the noise affected their evaluations. A majority of the distraction participants reported, erroneously, that the noise had in fact lowered their evaluation of the film. Here, participants believed a stimulus affected them when in fact it did not, and again, introspection failed to uncover the truth.
There is another way in which introspection can fail us: The process of thinking about how we feel can itself change the way we feel (T. D. Wilson et al., 1993). In one study, participants evaluated two artistic and three humorous posters and later had the opportunity to bring one of them home after the experiment. Participants in the reasons condition described why they liked each of the posters, whereas those in the control condition did not. Wilson et al. predicted and found that participants in the reasons condition were more likely to bring home a humorous poster than were control participants, most likely because it was easier to provide a rationale for preferring the humorous poster to the art poster: it was funny. When all participants were asked at the end of the semester how happy they were with their poster choice, those who had earlier listed reasons were less satisfied than the control participants, especially when they had chosen and justified selection of a humorous poster. The amusing effects of the comical poster—although humor was the initial reason for choosing it—seem to have worn off during the semester. Think about it—how many times can you laugh at the same joke?! Wilson et al.’s (1993) study demonstrates that analyzing reasons for preferences can undermine the pleasure produced by those preferences: Introspection can reduce satisfaction with one’s decisions. Similarly, another study found that analyzing reasons for liking one’s romantic partner decreased satisfaction with that partner (T. D. Wilson & Kraft, 1993).
Figure 4.1 Introspecting About Reasons Can Undermine Satisfaction
Source: Adapted from Wilson, T. D., Lisle, D. J., Schooler, J. W., Hodges, S. D., Klaaren, K. J., & LaFleur, S. J. (1993). Introspecting about reasons can reduce post-choice satisfaction. Personality and Social Psychology Bulletin, 19, 331–339.
Clearly, using introspection as a way to understand our thoughts, feelings, and behaviors has its limits (Corallo et al., 2008; E. A. Locke, 2009). Not only might we fail to understand why we act as we do, even thinking about the reasons for liking something can reduce that liking. And here we see that how our understanding of the self is inextricably tied to the limits of our reasoning and the nature of rationality itself. Attempting to achieve rationality can have interesting, if unintended, consequences for the self.
Introspection: Looking internally at the self to examine who one is, how one feels, and so forth
Self-Perception
As we’ve seen, introspection is an imperfect way to gain knowledge about who we are. What other means are at our disposal for gaining self-understanding? Well, how do you get to know other people when you don’t have direct access to their inner processes? One way is to simply ask them, but then of course you are relying on the questionable veracity of their introspection. Another strategy is to observe their behavior to see how they act in a variety of situations and under various conditions and use this as a guide to determine their attitudes and beliefs. Although not a perfect method, merely watching them could give us insights that asking them could not. The social psychologist Daryl Bem suggested that we can use the same method for gaining insight into our own selves: observe our own behavior. According to Bem’s (1967) self-perception theory, you can infer your attitude in this same way that a third party might do so: by watching your own behavior (Olson & Stone, 2005; Yee & Bailenson, 2009). Let me say that again: Bem argues that there are times when we rely on observations of our own behavior to figure out what our attitude, emotions, and personality traits are. This is particularly true when our attitudes are weak or ambiguous (Bem, 1967).
Take for example the results of a study of environmental attitudes by Chaiken and Baldwin (1981). Based on their responses to a survey completed earlier in the semester, participants were classified as either holding well-defined or poorly defined attitudes toward protecting the environment. During the subsequent experimental session, participants were led to focus either on their past pro-ecology behaviors or their past anti-ecology behaviors. Finally, they again responded to several questions in which they indicated the extent to which they consider themselves environmentalists. Chaiken and Baldwin predicted that individuals with weak attitudes would, when completing the final attitude measure, infer their attitude toward the environment from the behaviors that they focused on. In this way people with weak attitudes would “observe” their own behavior to determine their attitude. Consequently, those focused on pro-ecology behaviors should identify themselves as pro-ecology, whereas those focused on anti-ecology behaviors should lean toward the anti-ecology attitudes. In contrast, individuals with strong preexisting attitudes would not need to resort to self-perception to infer their attitudes and therefore would not show any effects of the experimental manipulation.
The findings were as predicted: Self-reported attitudes corresponded to whichever type of behavior participants with weakly defined attitudes concentrated on but not for those with previously well-defined attitudes. In short, consistent with self-perception theory, participants with weak attitudes relied on their own behavior to infer their attitudes (Chaiken & Baldwin, 1981). Two recent extensions of self-perception theory have shown that people’s perception of their own avatars—their virtual selves inserted into computer games or online social media—can affect their self-concept as well as their behavior (Yee & Bailenson, 2009) and that people may use their observation of the behavior of others to learn about themselves (see Figure 4.2).
Other research has demonstrated that people may also infer their motivation from their behavior, which in turn has implications for whether they will engage in subsequent related behavior. Before discussing this research in class, I present my students with the following scenario: Say my daughter loves to read and does so with great frequency on her own (which is true). Now say I decide to reward her for reading over the summer by giving her $4 for every book she reads (which I don’t). By the end of the summer, I am deeply in debt to her and give her the money she earned. When school starts and I stop rewarding her for reading, is she likely to freely read even more, read about the same, or read less than she did before being offered the money? Most students believe that she will read more because she has been rewarded. In all likelihood, however, my daughter’s reading frequency would likely decrease, because she has now associated reading with money, and when the money stops, so too will the reading. As a consequence of my paying her to read,
