Cognitive Busyness and Dispositional Inferences

Drawing Dispositional Inferences about Others

The Effects of Cognitive Busyness. We sought to demonstrate that cognitive busyness disables the perceiver’s ability to use information about aspects of the situation that might have facilitated or impeded a target’s behavior (Gilbert, Pelham, & Krull, 1988). We showed participants clips from a videotape of a female target discussing a variety of experimenter-assigned topics with a stranger, and asked them to estimate the target’s level of trait anxiety. In most of the clips, the target appeared extremely anxious. Participants were told that to protect the privacy of the target person, the videotape would be shown without any sound. However, participants were told that they would be able to tell which topic the target was discussing in a given clip because the topic would appear in subtitles at the bottom of the screen. For participants in the anxious topics condition, most of the subtitles indicated that the target was discussing anxiety-provoking topics (e.g., her sexual fantasies), whereas for participants in the relaxing topics condition, most of the subtitles indicated that the target was discussing mundane topics (e.g., world travel). Under normal conditions, one would expect participants in the anxious topics condition to attribute the target’s apparent anxiety to the nature of the topics she was being asked to discuss, and thus to make less extreme ratings of the target’s trait anxiety than participants in the relaxing topics condition.

However, the circumstances were not normal. Half the participants in each condition were required to perform an extra task while watching the video clips. Specifically, participants in the busy condition were instructed to memorize the discussion topics in their proper sequence while viewing the tape, whereas participants in the non-busy condition were given no such instruction. Memorization requires rehearsal, and rehearsal requires resources. As such, the unique prediction derived from our model was that those participants who were asked to memorize the discussion topics would be the least likely to use them when drawing inferences about the target. After viewing the videotape, participants rated the target’s trait-anxiety on scales that were anchored at the endpoints with such phrases as, “She is probably comfortable (uncomfortable) in social situations” and “She is a calm (nervous) sort of person.” We stressed that participants should indicate what kind of person the target is “in her day-to-day life,” and not just “how she was acting” in the video clips. Finally, we asked all participants to recall the discussion topics.

The authors report the following pattern in Table 9.1, which is not reproduced here: nonbusy participants used the situational constraint information (i.e., their knowledge of the topics) when estimating the target’s trait anxiety, but busy participants did not. Although busy participants were less likely than nonbusy participants to use this information, they were actually more likely to recall it. In short, participants who were asked to memorize the topics that a target was discussing drew more dispositional inferences about that target than did participants who were not asked to memorize this information. This finding is consistent with our suggestion that initial characterizations require fewer resources than do subsequent corrections.