Rational model and justification model in ‘outcome bias’

Abstract

The authors of this study suggest that the harm-punishment link ('outcome bias') can be explained by the activation of different judgment processes depending on the outcome severity of an offense: (1) a rational model for mild outcomes in which punishment is necessarily linked to responsibility of the perpetrator; (2) a justification model for severe outcomes in which punishment and responsibility are linked only when assessment order allows the latter to rationalize the former. Participants (126 university students) considered an unintentional road accident with mild or severe outcomes and made judgments of responsibility, punishment, and perceived seriousness of the offense. The results support the authors' hypothesis. In the discussion, the authors suggest different motives of punishment (preventive or compensative justice) which explain why responsibility and punishment are not necessarily linked.The aim of this article is to examine the links between responsibility and punishment in the case of reprehensible actions leading to objective outcomes of greater or lesser severity. It attempts to demonstrate that according to the severity of the consequences of an offense, two types of judgment models are activated: a rational model in cases of mild outcomes and a justification model in cases of severe outcomes. In a rational model, the attribution of punishment to the perpetrator of an offense is the result of a series of decision stages (Fincham & Jaspars, 1980;Shaver, 1985;Weiner, 1995). In this model, the first stage is to verify the causal link between the offense and the noted effects. Second, the motive of the perpetrator is evaluated in order to establish his or her degree of responsibility. Finally, punishment is attributed according to the degree of responsibility. In this perspective, the a priori contributions of the perpetrator (causality and intentionality) should direct the judgment of responsibility and the attribution of punishment, and this evaluation should not be influenced by the a posteriori involuntary or unforeseen outcomes of the action.Certain influential factors cannot be well integrated into this rational schema, notably the 'outcome bias effect.' This bias, which different studies have confirmed (Burger

The Adaptive Logic of Moral Luck

“…Other work corroborates this, finding that outcomes are viewed as more predictable in retrospect ("a reasonable person should have expected that"), which leads to greater condemnation when the outcome is bad ( Baron and Hershey 1988;Tostain and Lebreuilly 2008). Specifically, she found that adults judged an agent whose careless action lead to a worse outcome to be more responsible than an agent who caused a mildly bad outcome.…”

“…This explanation is often referred to as hindsight bias, because it attributes clairvoyant caution with the benefit of hindsight (Baron and Hershey 1988;Tostain and Lebreuilly 2008; see also Alicke and Davis 1989;Mazzocco, Alicke, and Davis 2004;Young, Nichols, and Saxe 2010). Here, we focus in particular on resultant luck, or luck in the way things turned out.…”

Why we forgive what can’t be controlled

“…Beyond their implications for our understanding of behavioral control as a criterion for moral judgment, our results speak to a larger debate over the mechanisms underlying moral luck and its influence on punishment. Previous psychological research has characterized moral luck in terms of outcome bias (Alicke & Davis, 1989;Carlsmith, Darley, & Robinson, 2002;Darley, Carlsmith, & Robinson, 2000) or hindsight bias (Baron & Hershey, 1988;Tostain & Lebreuilly, 2008;Young et al, 2010). According to the outcome bias model, the mere presence of a bad outcome generates negative affect in the perceiver, which subsequently biases moral judgment.…”

“…Indeed, prior work on hindsight bias has shown such an influence (Baron & Hershey, 1988;Tostain & Lebreuilly, 2008;Young, Nichols, & Saxe, 2010 Figure 2). As intended by our manipulation, we found that participants attributed greater control to the agent who chose between two available options compared with the agent who was forced to -choose‖ the only available option.…”

Looking back on the London Olympics: Independent outcome and hindsight effects in decision evaluation

“…In our example above and starting with the indirect effect (via hindsight bias), such an effect would be present if the outcome feedback first led to exaggerated impressions of the foreseeability of a fatal outcome and these in turn to negative evaluations of the decision to perform the operation and of the surgeon's competence. A direct effect of outcome knowledge (outcome bias) would be present if observers let their evaluations be influenced by the outcome irrespective of any associated foreseeability impressions, for example in an effort to restore a sense of justice by ‘punishing’ the surgeon for the fatal outcome (Alicke, ; Tostain & Lebreuilly, ). The precise nature of the reasons, justifications, or more generally processes involved in direct outcome effects depends of course on the specific context of a case, but there are two essential requirements that distinguish them from the previously discussed indirect effects: (1) that they bypass any hindsight‐related cognitions and (2) that they are driven by outcome valence , not by outcome knowledge per se .…”