Memory for Positive, Negative, and Comparison Ads: Studying Semantic Associations Between Candidates and Issues Using EEG

Political advertisements are a ubiquitous force during political campaigns. The 30-second TV spot is a particularly prominent form of political advertising, securing the bulk of advertising expenditures. Although political TV ads arouse contempt and condemnation from politicians, the media, and citizens, research has shown that these ads offer citizens substantive information about political candidates. Furthermore, survey and experimental research indicates that individuals acquire information from political advertisements. A controversy remains over which types of ads are most likely to instill knowledge among the electorate. This dissertation examines memory differences for Positive, Negative, and Comparison political TV ads. It differs from previous research in a number of respects. First, comparison ads are included as a separate political ad type category, rather than being neglected entirely or subsumed into the larger category of negative ads. Second, extant political advertising research uses various criteria for defining positive and negative ads. Definitions within and across studies frequently conflate intended and actual emotional responses, which may account for inconsistent empirical results. This dissertation proposes that Negative Ads and Comparison Ads should lead to greater Memory relative to Positive Ads, according to the negativity bias (Hypothesis 1 and 2, respectively). However, recognizing that humans are motivated information processors, this dissertation tests for moderating effects between and among variables (in particular, Ad Type, Ad Sponsor, and Partisanship) that may generate divergent emotional responses to the same messages (Research Question 1). Third, some researchers have suggested that emotional arousal (rather than valence) drives information processing and memory. This dissertation examines whether the effect of Ad Type (independent or moderated) on Memory remains when emotional arousal is controlled statistically (Research Question 2). Fourth, the measure of political knowledge used here tests for associations between candidates and issues in participants’ memory. By examining memory linkages between candidates and issues, this measure captures a critical and fundamental aspect of knowledge, and also focuses on information that citizens may find most relevant in evaluating political candidates. Lastly, this dissertation uses encephalography (EEG) recordings to study information processing of ad messages. Neurological activity in the gamma band frequency range (36-44 Hz) is suggested as indexing semantic processing, a basic component of knowledge formation. Gamma band power is proposed as a mediator between ad exposure and memory (Hypothesis 3). Results indicated no main effect of Political Ad Type on Memory for issues presented in political TV ads. Memory for Positive Ads was not statistically different from memory for Negative Ads or Comparison Ads (H1 and H2 not supported). Results indicated a statistically significant interaction between Ad Type and Partisanship (RQ1). Compared to Positive Ads, Negative Ads led to a decrease in memory among Republicans (statistically significant conditional effect), whereas Negative Ads led to an increase in memory among Democrats (marginally significant conditional effect). This interaction effect remained when emotional arousal was controlled in analyses (RQ2). Although no evidence was found to support a mediating role of Gamma (H3 not supported), results showed a statistically significant independent effect of Gamma on Memory.

Name thesis:degree_name

Doctor of Philosophy

Level thesis:degree_level

doctoral

Discipline thesis:degree_discipline

Communication

Grantor dc:publisher

The Ohio State University

Year dc:date

2013

Author dc:creator

• Morey, Alyssa C.

Contributors dc:contributor

• Eveland, William

• Communication
• Cognitive Psychology
• Political Science
• Neurosciences
• Political Ads
• Memory
• EEG
• Gamma
• Cognitive Neuroscience

Statement dc:rights

• unrestricted
• This thesis or dissertation is protected by copyright: all rights reserved. It may not be copied or redistributed beyond the terms of applicable copyright laws.

Language dc:language

English

Repository record dc:identifier

http://rave.ohiolink.edu/etdc/view?acc_num=osu1365611112

OAI identifier oai:identifier

oai:etd.ohiolink.edu:osu1365611112