Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
Two studies investigating how cognitive processing affects impressions formed from short and long video exposures. The first study found that participants formed more negative impressions from short clips of an invariant negative video compared to the full video. However, when participants engaged in a thought-listing task prior to assessing impressions, the difference in impressions was eliminated. The second study explored cognitive load during short exposures and found that cognitive load did not make a difference in impressions for clip conditions, but there was a significant interaction between valence and cognitive load in the thought-listing-first condition. Participants under cognitive load formed more positive impressions from negative clips and more negative impressions from positive clips.
What you will learn
Typology: Thesis
1 / 76
This page cannot be seen from the preview
Don't miss anything!
A thesis presented to the faculty of the College of Arts and Science of Ohio University
In partial fulfillment of the requirements for the degree Master of Science
Matthew J. Lindberg June 2007
This thesis entitled AFFECT AND ADJUST: CHANGE IN PROCESSING OF VALENCED STIMULI OVER TIME
by Matthew J. Lindberg
has been approved for the Department of Psychology and the College of Arts and Sciences by
G. Daniel Lassiter Professor of Psychology
Benjamin M. Ogles Dean, College of Arts and Science
Acknowledgments I wish to express my gratitude to my advisor, G. Daniel Lassiter, and members of my committee for their support and assistance during the course of this project. I would also like to thank Katie Golden for all of her enduring support, enthusiasm, and patience with me through this entire process.
List of Figures
Figure 1. Hypothetical Distribution of Life Experiences.............................................. 71 Figure 2. Study 1 Predicted Interaction ........................................................................ 71 Figure 3. Study 2 Predicted four-way interaction......................................................... 72
a series of written personality adjectives, more exposure does not necessarily equate to more extreme impressions. In fact, other evidence from research using direct observation—exposing participants to videos or photographs rather than lists of adjectives—indicates that impressions can become less extreme over longer exposures (Willis & Todorov, 2006; Kunda, Davies, Adams, & Spencer, 2002). In the domain of interpersonal attraction, research on the gain-loss model (Aronson & Linder, 1965) examines the sequential effects of positive and negative feedback. Aronson and Linder (1965) found that participants were more attracted to an evaluator who provided first negative then positive feedback, than one who provided consistently positive feedback. The mediators for this finding that have been explored focused on the change (variant) conditions, and only inconsistent and inconclusive results have been obtained (Mettee & Aronson, 1974). The inability to account for the gain-loss effect by focusing on change conditions suggests that the underlying mechanism may have more to do with what is happening in the invariant conditions. Why might repeated exposure to similarly valenced material lead to less extreme judgments? The current research seeks to answer this question in terms of the two information-processing systems of the cognitive-experiential self-theory (CEST; Epstein, 1994). The cognitive-experiential self-theory states that people navigate their environment by two information-processing systems: an automatic experiential system and a deliberate rational system. The automatic experiential system is affect laden, and responds to information in relation to previous life experiences. The deliberate rational
system is logical and processes information independent of previous experience and emotion. At minimal exposure levels, the rational system will not have time to activate, leaving impressions to be formed using the experiential system, generally leading to more extreme affect-laden impressions. However at longer exposures, when the information is invariant, the rational system will have an opportunity to engage and potentially adjust for the experiential system, leading to more moderate impressions. Two studies explored the hypothesis (referred to as “affect and adjust”) that longer exposures of invariant valenced information lead to more moderate impressions. Study 1 had participants form an impression of a young woman from either three 4-s clips or three 10-s clips taken from an ambivalent (positive and negative) video or an invariant negative video. Additional participants formed their impression after viewing the entire 5-min video (either the ambivalent or invariant negative). The results of Study 1 offer initial support for the hypothesis that less exposure leads to more extreme judgments. Participants that viewed clips taken from an invariant negative video formed more negative impressions than participants that watched the complete video. Study 2 extended the results of Study 1 by examining positively valenced information, as well as exploring the processing differences that may occur during short and long exposures. Participants watched three 4-s clips or an entire 5-min video of either positive or negative material. Two approaches were utilized to explore the role of cognitive processing. The first, focusing on full-length-video conditions, sought to disrupt cognitive processing during a given exposure through a cognitive-load task, whereas the second, focusing on clip conditions, sought to induce cognitive processing
Introduction Everyday we are exposed to people whom we have never met or seen before. Whether a short interaction with a cashier at a convenience store, or walking by someone sitting on a park bench, these moments of exposure are often fleeting. Many studies have demonstrated that such brief and minimal encounters (or thin slices of experience) often provide sufficient information on which to make accurate assessments of others (Ambady, Bernieri, & Richardson, 2000; Ambady & Rosenthal, 1992). Nonetheless, it remains unclear what processing underlies such judgments and whether it is fundamentally different from the kind of processing involved in judgments based on longer exposures. In a seminal study on impression formation, Asch (1946) gave participants a series of adjectives describing a fictitious person and then had them write a short description of the individual. By manipulating the order of adjectives, Asch was able to demonstrate that a simple additive account of the information was not a suitable explanation for the data, which demonstrated a primacy effect. Anderson (1981) formulated the Information Integration Theory (IIT) to account for the way people integrate stimulus information to form an impression. Anderson argued that people form impressions by using a simple algebraic rule, whether additive or averaging, to integrate stimulus information. Despite the initial discounting of an additive model by Asch (1946), research by Anderson and colleagues has found both advantages and disadvantages for additive and averaging models in person perception.
Anderson (1965) conducted the first study to directly pit an additive model against an averaging model. Participants rated likeability of a hypothetical person after receiving two or four personality-trait adjectives used to describe the individual. Contrary to an additive model, participants liked a person described by two highly favorable adjectives followed by two moderately favorable adjectives (HHM+M+) less than someone described by only two highly favorable adjectives (HH). This pattern tends to support an averaging model. However, people described by four highly favorable adjectives (HHHH) were liked more than people described only two (HH), which is in contradiction to an averaging model, which would have predicted that the two people would have been equally liked. This last set-type comparison is a demonstration of the set-size effect. Within the personality-adjective-task paradigm, an increase in the number of adjectives leads to more extreme impression judgments, even when the mean affective value is held constant. Sloan and Ostrom (1974) reviewed over 20 studies and found the set-size effect to be robust and reliable. When it comes to understanding someone and being able to make causal attributions about their behavior, the general assumption is that more information is needed rather than less. Kelley’s theory of causal attribution, the covariation model, assumes that people examine behaviors across time and situations in order to assess their underlying cause (Kelley, 1971). Fundamental to the examination of covariation are the inferential tests of consensus, distinctiveness, and consistency. Important to the current discussion, two of these inferential tests, distinctiveness and consistency, require considerable information. Distinctiveness is the extent to which a response is elicited by a
would give more time to change processing from automatic to more controlled, potentially leading to more correction. If underlying dispositions are well represented in most things that we do, as thin- slice researchers argue (Ambady, Bernieri, & Richardson, 2000), then a lengthy exposure to someone is simply more of the same general information. Following the set-size effect, one would expect that longer exposures would yield more extreme impressions and judgments. However, Ambady and Rosenthal (1992), in a meta-analysis of 38 studies, found that the accuracy of judgments based on clips of 4- and 5-min observations did not differ from those based on 30-s observations. This implies that when forming an impression of someone from a video, rather than a series of personality adjectives, more exposure does not necessarily equate to more extreme impressions. In fact there is evidence, from seemingly disparate lines of research, that suggests the opposite is often the case—less, rather than greater, exposure to valenced stimuli leads to more extreme impressions. Willis and Todorov (2006) were able to demonstrate that participants could sufficiently form an impression after 100-ms exposure. Of interest to the current argument, the results also demonstrated that when exposure increased from 100ms to 500ms, participants’ impressions became less extreme. Similarly, minimal exposure (315 ms) to Black faces typically leads White participants to automatically activate a negative stereotype and experience negative affect (Fazio, Jackson, Dunton, & Williams, 1995). However, Kunda, Davies, Adams, and Spencer (2002) were able to demonstrate that the robust effect of automatic stereotype activation has a time course. After viewing only 15 s of an interview with a Black person,
participants showed automatic activation of the stereotype of Black people. However participants that had viewed 12 min of the interview demonstrated no stereotype activation. Participants had an initial reaction to viewing a Black person that activated the stereotypes associated with Black people, however these initial effects completely disappeared at the longer exposure interval. Combined, these studies suggest the possibility that longer exposure to an actual stimulus person will have an opposite effect on impressions than what has heretofore been found with paradigms that increased the set size of a list of trait adjectives. The attribution theories of Quattrone (1982) and Gilbert, Pelham, and Krull (1988) propose that behaviors are automatically identified and characterized in a dispositional manner, however neither theory specifically addresses the issue of the valence of stimuli. The correction phase of both theories would require subsequent information about the situation to warrant correction. In neither theory would the initial identification or characterization of information be expected to change without the presence of other information. However, the processing systems of the cognitive- experiential self-theory (CEST) are capable of explaining how the same information can be processed differently over time. The application of CEST to person perception suggests that minimal exposure would engender more extreme impressions as a result of the emotional and experiential aspect of automatic processing. Because the rational cognitive process is sluggish and is delayed in activation, the CEST would predict that longer exposures would provide time and opportunity to reconsider and correct for the
Aronson and Linder (1965) suggested two entirely different explanations for the gain-loss effect. One was primarily affective, and the other was primarily cognitive in nature, with both emphasizing the importance of change in the sequence of feedback. The affective explanation argues that the response to positive evaluations following negative evaluations not only creates positive affective reactions due to the positive evaluations, but also serves to eliminate the negative drive states that were engendered by the initial negative evaluations. The cognitive explanation argues that evaluations that change demonstrate more discernment or discrimination. The theoretical focus on the change (gain or loss) conditions is understandable as Aronson and Linder were seeking to demonstrate the “gain-loss” effect. However this leaves out the possibility that the results of the gain-loss effect are due to changes in processing in the invariant conditions. Later, Mettee and Aronson (1974, p.267) formalized the aforementioned model using two postulates. The first postulate, the focus of the first demonstration by Aronson and Linder (1965), states that “affective change, or switch, will augment the potency of the postswitch evaluations.” The second postulate states “affect uniformity engenders redundancy and attenuates the potency of evaluations occurring in the later part of a sequential series.” Whereas Mettee and Aronson (1974) argue that both postulates are essential to the gain-loss model, they acknowledge that the latter postulate is often overlooked. Although a number of mediators have been explored, all have focused on gain and loss conditions. Similar to Willis and Todorov (2006) and Kunda et al. (2002), research on the gain-loss effect could be interpreted as a change in processing over time to invariant information. However, research on the gain-loss effect has sought to discover
the underlying mediators in terms of gain and loss conditions. An understanding of the mediators proposed and their inability to account for gain-loss effects suggests that it is time to further explore the change in processing that could occur when one continues to observe an invariant behavior sequence. Mediators of the gain-loss effect. Aronson and Linder (1965) suggested four potential mediators to account for the change in affect participants experience in reaction to differential evaluations. In a review of the gain-loss literature, Mettee and Aronson (1974) elaborated on the four mediators and the empirical support. The first mediator, focusing on the evaluations that participants receive, suggests an evaluation contrast effect, with the affect induced by the second-half evaluations being strengthened due to their contrast to the initial evaluations. The second mediator focuses on the evaluator, arguing for an evaluator discernment effect to account for the gain-loss effect. The reversal of evaluations in the variant sequences can be interpreted as evidence for the credibility and discriminativeness of the evaluator. The last two mediators, a recipient-anxiety effect and a recipient-competence effect, both focus on the recipient of the evaluation, with the former being more affective in nature and the latter more cognitive. The recipient-anxiety mediator focuses on alterations in experienced anxiety as a result of variant feedback. In the gain condition (- +), the participant first experiences anxiety as a result of the initial negative evaluation, then subsequently experiences a reduction in anxiety due to the positive second evaluation. This relief leads to more favorable ratings of the evaluator than in the invariant positive (+ +) condition in which there was no anxiety inducement or
