The maintenance and change of stereotypes following memory-based and on-line processing Author Dexter Taylor Publication Year 2001 Type Thesis Abstract This research extends previous work showing that the correspondence between memory and judgment depends on whether a perceiver's processing objective is online or memory-based (Hastie & Park, 1986). On-line processing involves the integration of information into a summary judgment as it is encoded. Memory-based processing involves encoding information without a goal of making a summary judgment; later, after encoding, information is retrieved from memory with the goal of making a summary judgment. Several researchers have found that judgments preceded by memory-based processing have a higher correspondence with the content of recalled information than judgments preceded by on-line processing. It was hypothesized that processing objectives would moderate stereotype change, and that the effect of processing objectives on stereotype change would be mediated by the accessibility of exemplars. Judgments preceded by memory—based processing were predicted to change less than judgments preceded by on-line processing. Smith (1990) and Sherman (1996) have suggested that there is a parallel between the literature on processing objectives and the literature on mental representations. This research also tested the assumption that the differences between memory-based processing and on-line processing parallel those between exemplar-based and abstraction-based models of mental representation. The experiments used a paradigm whereby participants formed a negative impression of a group after reading a set of socially undesirable behaviors performed by members of the group. Either a memory-based or an on-line judgment task preceded participants' first ratings of the group. Later in the experiment, participants' initial negative impressions were challenged by the presentation of a set of socially desirable behaviors performed by other members of the group. Experiment 1 demonstrated that judgments following memory-based processing change significantly less when challenged by inconsistent information relative to judgments following on-line processing. It is argued that judgments following online processing are less likely to activate exemplars than judgments following memory-based processing. Consequently, judgments following on-line processing change comparatively more when challenged by new, inconsistent information. Experiment 2 tested whether the accessibility of exemplars mediated the effect of processing objectives on attitude change. Experiment 2 did not replicate the effect of processing objectives on attitude change found in Experiment 1; therefore, the role of accessibility as a mediator could not be tested. The findings of Experiment 1 have implications for research on social cognition and stereotyping, and could lead to the development of a comprehensive model on the effect of processing objectives on attitude formation, maintenance, and change. Keywords memory based processing, on line processing, summary judgments, stereotype change Academic Department Department of Psychology Degree Doctor of Philosophy Pages 5061 Thesis Type Dissertation University University of Maryland URL External link to reference Full text The following is an excerpt of the intervention methodology. For more information, please see the full text of the article on the publisher's website or through your institution's library. Experiment 1 Method Participants The participants in this study were 61 introductory social psychology students [...]. [...] Procedure Participants [...] were told that the experiment concerned verbal expression and comprehension. [...] Participants were presented with one of four questionnaires. The four questionnaires included either the during-exposure or the after-exposure judgment task manipulation, and either version 1 or version 2 of the first set of behavioral information. In all conditions the first set of behavioral information was presented on a single page [...] with the instructions. Before participants began to complete the questionnaire, the experimenter informed participants that they should complete the questionnaire in order, and that they were not to turn back after completing a page of the questionnaire. After this point, all instructions were presented in the text of the questionnaire. Participants were instructed to make trait evaluations of the target group under two types of judgment tasks, either during exposure to the first set of behavioral information or after exposure to the first set of behavioral information. Participants who were informed of their judgment task after exposure to the first set of behavioral information were asked initially to only proofread the set of behaviors. [...] Participants in this condition were asked to describe the proofreading errors they discovered in the set of sentences presented earlier. [...] Participants in this condition were informed that the set of behaviors that they read previously were behaviors performed by members of a real social group. The group was described generically as "Group A." On this page participants were also asked to make trait ratings of "Group A." The participants were asked to rate the target group on [...] four [...] traits. In the during-exposure judgment task condition, participants were told from the outset that the behaviors they were about to read were performed by members of a real social group [...]. On this same page they were informed that their task was to consider the degree to which they thought the group was pleasant, selfish, considerate, and rude. The first set of behavioral information was listed at the bottom of this page. On the next page, participants were instructed to rate the group on the four target traits. After this point, participants in all conditions engaged in the same tasks [...]. [...] Participants were presented with fourteen sentences on a wide variety of topics. Each sentence had a blank space that implied that a single word was missing. Participants were asked to complete the sentence; [...]. These questions bore no relationship to the behavioral information. [...] Participants were presented with a second set of behaviors on the following page. Participants in both conditions were told that this second set of behaviors were additional behaviors performed by other members of Group A [...]. This second set of behavioral information was evaluatively inconsistent with the first set of behavioral information [...]. After reading the second set of behavioral information, participants completed a second rating of "Group A" on the following page. [...] [...] participants were asked to complete a recognition task. This recognition task measured the recognition of the ten behaviors presented in the first set of behavioral information. Twenty items were listed in the recognition task; ten of the items were the same behaviors that the participant read earlier in the first set of behavioral information, the remaining ten items were distractor items. [...] Experiment 2 Method Participants The participants in this study were [...] introductory psychology students [...]. [...] Stimulus Materials [...] In this experiment [...] only the second version of the first set of behavioral information was used [...]. [...] Behavioral information was presented such that desirable and undesirable behaviors were evenly distributed throughout the presentation [...]. Procedure [...] Participants were told that there was one task that required a hand written response, a sentence completion task. The experimenter told participants that an answer sheet for the sentence completion task and a pencil were provided for them and placed beside the computer monitor. Then participants were seated in front of a computer terminal where they were presented with a computerized questionnaire. [...] All subsequent instructions to the participants were presented in the text of the questionnaire. Participants were asked to form an impression of the target group either during exposure to the first set of behavioral information or after exposure to the first set of behavioral information. [...] In this study [...] the time during which participants were allowed to read each behavioral statement was held at 20 seconds. [...] This was true for both judgment task conditions. [...] Participants in the after-exposure judgment task condition were not required to record their responses to the proofreading request. [...] This written response was eliminated in this experiment in order to hold the duration between the presentation of behaviors and the first trait ratings constant across the judgment task conditions. Immediately after the presentation of the first set of behavioral information, participants in the after-exposure judgment task condition were informed that the set of behaviors were actually behaviors performed by members of a real social group. Participants were then asked to make the same trait ratings presented in Experiment 1. [...] [...] In the during-exposure judgment task condition participants were told from the outset that they would be presented with a set of behavioral information consisting of behaviors performed by a real social group, Group A. [...] In both the after-exposure judgment task condition and the during-exposure judgment task condition the latencies in trait ratings were recorded and used as a dependent measure. [...] This measure served as a test for evidence of memory-based processing [...]. [...] Participants in both judgment task conditions followed the same procedure. [...]. In this study, the sentences were presented individually on the computer screen for 20 seconds. [...] [...] Participants were told that they would be presented with a second set of behavioral information ostensibly described as additional behaviors performed by other members of Group A. Participants were told that after the presentation of the second set of behavioral information they would be asked to make second ratings of the target group based on both the first and the second set of behavioral information. [...] To test for the accessibility and availability of the first set of behavioral information in memory, participants performed a recognition task [...]. Type of Prejudice/Bias Minimal/Artificial Country United States Method Lab Setting College/University Google ScholarBibTeX
