• Corpus ID: 17473580

The Impact of Accountability on Cognitive Bias

@inproceedings{Lerner2004TheIO,
  title={The Impact of Accountability on Cognitive Bias},
  author={Jennifer Susan Lerner},
  year={2004}
}

• J. Lerner
• Published 2004
• Psychology

Research on accountability takes an unusual approach to the study of judgment and decision making. By situating decision makers within particular accountability conditions, it has begun to bridge individual, interpersonal, and institutional levels of analysis. We propose that this multi-level approach can enhance both the study of judgment and choice and the application of such research to real-world settings. To illustrate the multi-level approach, we present a review of accountability research, organized around an enduring question in the literature: Under what conditions will accountability improve judgment and decision making? After considering the shortcomings of two seemingly straightforward answers to this question, we propose a multi-factor framework for predicting when accountability attenuates "bias," when it has no effect, and when it makes matters even worse. Key factors in this framework draw from multiple levels of analysis. Content Introduction 3 Accountability As A Universal Feature Of Social Life 4 The Effects of Accountability on Putative Biases: Examining Support for Two Hypotheses 7 Toward a Multi-Factor Framework for the Effects of Accountability 12 Conclusions: Benefits of Accountability Research for the Study of Judgment and Choice 20 Bridging Levels of Analysis: 4 Bridging individual, interpersonal, and institutional approaches to judgment and choice: The impact of accountability on cognitive bias Psychological theories of judgment and choice have tended to view decision makers as isolated individuals. While elucidating cognitive processes that occur within the individual, they have tended to overlook the impact of interpersonal and institutional settings on thought and action. By contrast, organizational and political theories have tended to view decision makers as the instruments of institutional norms, rules, and constraints. While elucidating social and political processes that occur outside of the individual, they have tended to overlook the impact of intrapsychic processes. We argue that the study of accountability -pressure to justify one's views to another -can bridge these historically distinct approaches. It can do so by providing a natural link between individual decision makers and the relationships within which decision makers work and live. Bridging these approaches not only has the potential to improve the applicability of decision-making research to real-world settings, it also has the potential to improve basic theories of judgment and decision making. In section one of this chapter, we introduce accountability as a universal feature of social life and discuss the social foundations of accountability. In section two, we review the effects of accountability on putative “biases” identified in the judgment and decision making literature. In particular, we consider two overarching hypotheses about the effects of accountability. According to the first hypothesis, accountability will attenuate judgment and decision-making biases to the extent that accountability increases cognitive effort. According to the second hypothesis, accountability will amplify the dominant responses to judgment and decision-making problems -thereby attenuating bias on "easy" problems and amplifying bias on "difficult" problems. Finding that neither hypothesis receives substantial empirical support, we identify two factors that play an especially important role in determining when accountability will attenuate "bias," when it will have no effect and when it will make matters even worse. We then review evidence for these factors and propose incorporating them into a multi-factor framework Bridging Levels of Analysis: 5 of accountability effects. Finally, in the concluding section we identify specific ways in which accountability research can enhance both the study of judgment and choice and the application of judgment and choice research to real-world settings. Accountability As A Universal Feature Of Social Life Social Functions Of Accountability Many theorists -from political philosophers (Hobbes, 1660/1968) to organizational behaviorists (Katz & Kahn, 1978; March & Olsen, 1995), to social psychologists (Schlenker, 1980; Scott & Lyman, 1968; Semin & Manstead, 1983; Tetlock, 1992) -have viewed accountability as a prerequisite for social order. Although accountability ground rules obviously vary dramatically across cultures and history (Bersoff & Miller, 1993; Hong & Chiu, 1992), the underlying functions of accountability are strikingly similar. Accountability systems represent sociocultural adaptations to the perennial problem of how to coordinate relationships among individuals. Whatever the ideological or cultural value system underlying the social system -be it a decentralized market or command economy -accountability guidelines prescribe the norms and guidelines in a collectivity (Lerner & Tetlock, 1994; Tetlock, 1998) and how to deal with those who deviate from them (Stenning, 1995). "Accountability serves as a critical rule and norm enforcement mechanism -the social psychological link between individual decisionmakers on the one hand and social systems on the other (Tetlock, 1992 p. 337)." Because the transaction costs of relying on purely external forms of accountability would be prohibitive, most social control takes the form of internalized accountability. People monitor their own judgments and decisions by considering the justifiability of alternative courses of action. Indeed, this idea is a cornerstone of the symbolic interactionist approach to thought. George Herbert Mead maintained that “the very process of thinking is, of course, simply an inner conversation that goes on...He thinks it out, and perhaps writes it in the form of a book; but it is still part of social intercourse in which one is addressing other persons and at the same time addressing one’s self, and in which one controls the address to other persons by the response made to one’s own gesture (1934, p. 141).” Bridging Levels of Analysis: 6 Modeling Thought as Internalized Dialogue As long as people are concerned about maintaining their identities as moral, competent beings, a central function of thought is making sure that one acts in ways that can be persuasively justified or excused to observers. Indeed, the process of considering the justifiability of one's choices may be so prevalent that decision makers not only search for convincing reasons to make a choice when they must explain that choice to others; they search for reasons to convince themselves that they have made the "right" choice (cf. Shafir, Simonson, & Tversky, 1993). As Kuhn (1992) argues, the ability to generate persuasive justifications for beliefs may be "the most significant way in which higher-order thinking and reasoning figure in the lives of most people ( p. 155)." A number of researchers have made the case that concerns about the justifiability of action loom large in all judgments and choices (Gonzales, Kovera, Sullivan, & Chanley, 1995; Hsee, 1995; McGraw, Best, & Timpone, 1995; Schlenker, 1980, 1985; Scott & Lyman, 1968). For example, Shafir and colleagues (1993) demonstrated in a series of studies that the justifiability of reasons figured into participants' choices even when participants neither expected to explain their judgments nor even to interact with anyone. When participants were only presented with two options, it should not have mattered whether the experimenter asked them to select the option "they preferred" or the option "they would reject." Nevertheless, participants' selections varied as a function of elicitation procedures; they relied on positive features of each option when they were told to choose an option and negative features when they were told to reject an option (Shafir, 1993). The researchers explained that although this phenomenon defies a value maximization perspective, it readily fits a reason-based choice perspective: “reasons for choosing are more compelling when we choose than when we reject, and reasons for rejecting matter more when we reject than when we choose (Shafir et al., 1993, p. 18)." To recap, both theoretical and empirical arguments suggest that thought may be usefully modeled as dialogue -even when decision makers are not explicitly held accountable. Defining Accountability in the Judgment and Choice Literature Bridging Levels of Analysis: 7 Although implicit accountability may be an inevitable feature of real-world decision environments, accountability here refers to an explicit expectation that one will be called upon to justify one's beliefs, feelings, or actions to others (Scott & Lyman, 1968; Semin & Manstead, 1983; Tetlock, 1992). Additionally, accountability implies that positive or negative consequences hinge on the acceptability of one’s justification. In a few field studies, the acceptability of one's justification carries such concrete consequences as performance-contingent career advances or setbacks. More typically, however, the acceptability of one's justification carries only intangible consequences (e.g., approval or disapproval from the audience). Specifically, most laboratory studies create a situation in which people expect to explain their actions to someone they have never met before and never expect to meet again. What is remarkable about this literature is that -despite the prevalence of these minimalist manipulations -participants still reliably respond as if audience approval matters. And they do so even when the decisions at hand require them to express deeply held moral/ethical beliefs (see Brief, Dukerich, & Doran, 1991; Pennington & Schlenker, 1999). Two implications follow from the fact that researchers achieve these effects with such weak manipulations. First, it implies support for the social interactionist view that concerns about how the self relates to others -even temporarily significant others -drive cognition. Second, it implies that much more substantial effects