The upside of evaluative inconsistency : when contradictions foster persuasion and protect the self

Placeholder Show Content


Decades of past research point to the downside of evaluative inconsistency. Generally, this past research suggests that evaluative inconsistency is an unpleasant state that can result in negative affect. Consequently, when people have inconsistent reactions to evaluative targets, they often attempt to reduce their ambivalence in various ways (e.g., paying careful attention to new information that might help them resolve their ambivalence). Similarly, when people detect inconsistency in others' views, past research indicates that they will be less open, or more resistant, to those views. This dissertation takes a different track and explores the possibility that evaluative inconsistency can, in some instances, offer specifiable benefits. In particular, in two separate bodies of work, I examine the conditions under which, and mechanisms through which, evaluative inconsistency can protect the self and persuade others. In one series of studies, I have unveiled a contradiction effect, whereby changing one's stated opinion can, under specifiable conditions, introduce a persuasive advantage. My studies suggest that contradicting oneself can enhance one's persuasiveness relative to offering just a single message with no contradiction or even offering multiple messages that are consistent across time. I observe this effect across several distinct contexts including medical decisions, university policy assessments, and real consumer choices. I propose an attributional account to explain these effects, whereby conflicting messages, or opinion shifts, stimulate attributional reasoning by virtue of their unexpectedness. Extensive literature points to the role of expectancy violations in fostering attributional reasoning. I apply this logic to the current context, postulating that contradictions can promote persuasion as a function of the attributions they elicit. I show that contradictions' effect on persuasion is mediated by perceptions that the contradicting source had considered more information and engaged in more extensive thinking about the message topic. In addition, I also establish several moderators that point to an attributional account. The contradiction effect emerges only when strong arguments support the opinion shift, when that shift comes from a single source, and when trust is high. Under weak arguments, multiple source, or low trust conditions, the effect disappears or even reverses. Thus, only when conditions encourage favorable attributions does the effect emerge. In the second body of work, I demonstrate that there are situations in which people desire to be ambivalent. Specifically, when people want an object or outcome that is uncertain, such as a coveted job, house, or admission to a prestigious school, they will cultivate ambivalence in order to protect their feelings in the event that they fail to get what they want. I find that people are most likely to generate ambivalence when they are least certain that they can obtain their desired target. Depending on the outcome, this cultivated ambivalence can either be useful (when people fail to obtain the desired target) or backfire (when people obtain the desired target). In other words, cultivating both positive and negative thoughts and feelings towards a desirable target can potentially protect one's feelings in the event that it is not obtained. However, if the desirable target is obtained, these cultivated thoughts and feelings can be detrimental, as the person has now endorsed of the drawbacks of their object of desire. Across educational, employment and consumer choice settings, I show that when people fail to obtain their desired target, the more ambivalent people are prior to finding out the outcome, the better they feel about themselves, whereas the opposite pattern emerges when people obtain their desired target. In that case, the more ambivalent people are, the worse they feel about themselves once they find out the outcome. Interestingly, while past research has shown that ambivalence is usually associated with discomfort, I show that it depends on the outcome. That is, ambivalence before the outcome is known is associated with more discomfort and less favorable thoughts only when people obtain their desired target. In order to further establish the causal link between ambivalence and self-protection, I directly manipulate ambivalence and rule out an alternative attitude account of my results by showing that the protective utility of ambivalence is independent of attitude. Finally, I explore whether ambivalence reduces risk aversion. I posit that the protective utility of ambivalence allows people to open up to failure. In particular, I demonstrate that for those who are in the midst of a job search, the more ambivalent they are about their future job opportunity, the more willing they are to take risks (e.g., negotiate various aspects of the position). In the same vein, I find that people are most likely to make a lower bid on a desired house to the extent that they are ambivalent. A voluminous body of research has highlighted the negative aspects of evaluative inconsistency. Together, the two streams of research outlined in this dissertation suggest that evaluative inconsistencies in both the self and others can sometimes be beneficial. Understanding people's responses to inconsistency within self and in others is an important part of predicting and shaping their attitudes toward objects and issues they encounter in their daily lives. In the current dissertation I sought insight into this issue by exploring the effect of contradictory messages on persuasion as well as the effect of ambivalence on the self. The former counterintuitive effect provides unique theoretical insight into the complex relationship between source perceptions and message consistency effects in persuasion and the latter expands our understating of the consequences of being ambivalent.


Type of resource text
Form electronic; electronic resource; remote
Extent 1 online resource.
Publication date 2014
Issuance monographic
Language English


Associated with Reich, Taly
Associated with Stanford University, Graduate School of Business.
Primary advisor Tormala, Zakary L
Primary advisor Wheeler, S. Christian
Thesis advisor Tormala, Zakary L
Thesis advisor Wheeler, S. Christian
Thesis advisor Shiv, Baba, 1960-
Advisor Shiv, Baba, 1960-


Genre Theses

Bibliographic information

Statement of responsibility Taly Reich.
Note Submitted to the Graduate School of Business.
Thesis Thesis (Ph.D.)--Stanford University, 2014.
Location electronic resource

Access conditions

© 2014 by Taly Reich
This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).

Also listed in

Loading usage metrics...