Charles S. Taber

Elements of Reason: Three Steps toward a Theory of Motivated Political Reasoning

The human understanding, when it has once adopted an opinion … draws all things else to support and agree with it. Though there may be (more) instances to be found on the other side, yet these it either neglects or despises, or else by some distinction sets aside and rejects. –Francis Bacon, New Organon (1621) In this essay we propose a theory of motivated reasoning that can account for why both ordinary citizens and political sophisticates are prone to follow Bacons dictum. Three subtheories – hot cognition, online processing, and a “how-do-I-feel?” heuristic – working together, provide a three-step mechanism for how we believe citizens think and reason about political leaders, groups, and issues. This tripartite theory of motivated reasoning starts with the notion that all social concepts are affect laden; all social information is affectively charged (Bargh 1994, 1997; Fazio, Sanbonmatsu, Powell, and Kardes 1986; Fazio and Williams 1986; Lodge and Stroh 1993; Taber, Lodge, and Glathar 2000). This is the hot cognition hypothesis (Abelson 1963). Specific to politics, all political leaders, groups, issues, and ideas you have thought about and evaluated in the past are now affectively charged – positively or negatively, strongly or weakly – and this affective tag is stored directly with the concept in long-term memory. On-line processing (Anderson and Hubert 1963; Lodge, Steenbergen, and Brau 1995; Park and Pennington 1986) is a mechanism for updating the value of affective tags attached to concepts in memory.

MOTIVATED SKEPTICISM IN THE EVALUATION OF POLITICAL BELIEFS (2006)

Abstract We propose a model of motivated skepticism that helps explain when and why citizens are biased information processors. Two experimental studies explore how citizens evaluate arguments about affirmative action and gun control, finding strong evidence of a prior attitude effect such that attitudinally congruent arguments are evaluated as stronger than attitudinally incongruent arguments. When reading pro and con arguments, participants (Ps) counterargue the contrary arguments and uncritically accept supporting arguments, evidence of a disconfirmation bias. We also find a confirmation bias—the seeking out of confirmatory evidence—when Ps are free to self-select the source of the arguments they read. Both the confirmation and disconfirmation biases lead to attitude polarization—the strengthening of t2 over t1 attitudes—especially among those with the strongest priors and highest levels of political sophistication. We conclude with a discussion of the normative implications of these findings for rational behavior in a democracy.

Citizens and Politics: The Motivated Construction of Political Judgments

By one point of view, little remains to be said about voting behavior. A variety of models predict electoral behavior rather accurately, both at the individual and at the aggregate level. But from another point of view, forecasting vote outcomes is not enough. Twenty-seven years ago, Kelley and Mirer (1974: 572) observed that “Our ability to predict how voters will vote is far more solidly based than our ability to explain why they vote as they do.” This deficiency, which remains true today, stems from the black box nature of virtually all models of electoral choice, which are based on some form of information processing but are silent about the mechanisms that turn inputs into outputs. Political psychologists have taken note of the deficiencies of black box models of electoral choice, focusing instead on providing plausible explanations of voting behavior (Boynton and Lodge, 1994; Einhorn, Komorita, and Rosen, 1972; Herstein, 1981; Lodge, McGraw, and Stroh, 1989; Ottati and Wyer, 1990; Rahn, Aldrich, Borgida, and Sullivan, 1990; Taber and Steenbergen, 1995). But this work, our own included, has two key weaknesses. First, we have focused too heavily on the content and structure of beliefs and have paid too little attention to cognitive process. We treat people as passive receptors of information rather than as active, motivated reasoners who interpret information, make inferences, and often choose suboptimally. Second, we have established much too strong a dichotomy between affect and cognition.

Why People “Don’t Trust the Evidence” Motivated Reasoning and Scientific Beliefs

In this commentary, we embed the volume’s contributions on public beliefs about science in a broader theoretical discussion of motivated political reasoning. The studies presented in the preceding section of the volume consistently find evidence for hyperskepticism toward scientific evidence among ideologues, no matter the domain or context—and this skepticism seems to be stronger among conservatives than liberals. Here, we show that these patterns can be understood as part of a general tendency among individuals to defend their prior attitudes and actively challenge attitudinally incongruent arguments, a tendency that appears to be evident among liberals and conservatives alike. We integrate the empirical results reported in this volume into a broader theoretical discussion of the John Q. Public model of information processing and motivated reasoning, which posits that both affective and cognitive reactions to events are triggered unconsciously. We find that the work in this volume is largely consistent with our theories of affect-driven motivated reasoning and biased attitude formation.

Motivated Reasoning and Public Opinion

Citizens, especially those who are knowledgeable and care the most about politics, are motivated to defend their beliefs and attitudes in the face of discrepant information. These motivated biases strongly influence the way people think about health care policies and the politicians and parties that propose or attack these contentious policies. Three cognitive mechanisms are identified: a prior belief effect, confirmation bias, and disconfirmation bias. Together, these information processes conspire to produce persistence and polarization of opinion on health care policies.

First Steps Toward a Dual-Process Accessibility Model of Political Beliefs, Attitudes, and Behavior

With the advent of the political behavior movement in political science in the 1950s, in particular with the publication of The American Voter in 1960, beliefs, feelings, and behavioral dispositions were brought to center stage in the prediction and explanation of political behavior. In line with an implicit assumption of human rationality, the social sciences commonly presumed that thoughts, feelings, and behavioral intentions coming to mind consciously determine the lion’s share of behavior. Congruent with this assumption of conscious considerations arbitrating the expression of beliefs and emotions, political scientists commonly ask people to voice their beliefs, report their likes and dislikes, recount feelings and past behaviors, and foretell their intended actions. Because of this reliance on introspection, much of what we know about public opinion and electoral behavior and how we model the expression of political attitudes, beliefs, and behavioral dispositions is based almost exclusively on what respondents say when asked for their present, past, or future beliefs, intentions, and behavior.

THE SCOPE AND GENERALITY OF AUTOMATIC AFFECTIVE BIASES IN POLITICAL THINKING: REPLY TO THE SYMPOSIUM

Abstract Our response to this symposium on our 2006 paper centers on three questions. First, what motivations exist in the political wild, and do our experimental manipulations realistically capture them? We agree that strong accuracy motivations are likely (but not certain) to reduce biases, but we are not at all confident that the real world supplies stronger accuracy motivations than our subjects received. Second, how can we square our findings of stubbornly persistent beliefs and attitudes with the well-established literatures on framing and persuasion, which find political opinion to be more malleable? We argue that our John Q. Public theory of political information accounts for both persistence and persuasion and explains when we should expect one or the other. Citizens will be more responsive to contextual information or persuasive appeals when prior feelings are weak, knowledge is sparse, and information is encountered outside of awareness. Resistance to information or arguments is most likely when prior feelings are strong, attitudes are embedded within dense knowledge networks, and appeals are consciously perceived. Third, is belief persistence driven by feelings and emotions, as we claim, or is it more the result of a conviction that ones priors are accurate? We suggest that the distinctions that stand behind this question are suspect. We see in some of our commentators a tendency to align accuracy motivation, central processing, and cognition on one side of the dual-processing framework, with directional motivation, peripheral processing, and feelings on the other side. On the contrary, we argue that beliefs in the truth of ones priors often result from feelings and wishes and are themselves sources of motivated bias in processing new information. Moreover, directional motivations are also often a source of central processing, so it is not easy to disentangle these processes.

Beyond Simplicity: Focused Realism and Computational Modeling in International Relations

International relations faces an important choice: whether to employ formal or empirical methods that simplify our analysis or to rely on descriptive realism at the expense of focus. Facing this dilemma, many scholars have opted for focused simplicity. When this leads to over simplification, however, particularly when it leads us to ignore underlying processes and concentrate exclusively on predicting outcomes, the cost is great. Computational modeling enjoys many of the advantages of other formal approaches, while relaxing some of the technical restrictions. Thus, it allows focused realism —the representation of complex structures and processes without losing all-important analytic focus. This essay describes and illustrates the computational approach, reviews the existing international relations applications divided into three categories (dynamic simulations, knowledge-based models, and machine learning), and evaluates the capacity of computational models to achieve focused realism.

The Policy Arguer: The Architecture of an Expert System

The Policy Arguer ( POLI) is a working expert system model of U.S. foreign policy belief systems toward Asia. Expert systems—computer programs that imitate the behavior of human experts in constrained domains—have become a useful mode of inquiry for research problems in a number of the social sciences. POLI reproduces U.S. foreign policy outputs and the underlying debates and arguments. Foreign policy is explained in terms of the process that converts beliefs into arguments and policies. This paper describes the model in some detail, including its structure and program flow as well as the extraction of its knowledge bases. Keywords: belief systems, computational modeling, expert systems, decision making, fuzzy logic, international relations, U.S. foreign policy.

National arms acquisition as a rational competitive process

A formal dyadic model of the arms acquisition process is developed and analyzed through computer simulation. A rational model produces behavior consistent with a powerful security dilemma. A spiral model, in which images of the opponent become more aggressive, sparking an arms race, emerges from the simulation for deterrent nations. In addition to these anticipated findings, the rational arms model suggests that competitive factors have their greatest impact for rivals near parity; noncompetitive forces may drive the decision for unbalanced dyads. Also unpredicted from previous formal work, only one member of a competitive dyad needs to favor disarmanent for both nations to disarm.