|
The significance of counterfactual thinking in the causal judgment process has been emphasized for nearly two decades, yet no previous research has directly compared the relative effect of thinking counterfactually versus factually on causal judgment. Three experiments examined this comparison by manipulating the task frame used to focus participants’ thinking about a target event. Prior to making judgments about causality, preventability, blame, and control, participants were directed to think about a target actor either in counterfactual terms (what the actor could have done to change the outcome) or in factual terms (what the actor had done that led to the outcome). In each experiment, the effect of counterfactual thinking did not differ reliably from the effect of factual thinking on causal judgment. Implications for research on causal judgment and mental representation are discussed. studies have examined how comparing reality to other possible worlds, or thinking counterfactually, can influence people's reactions to events (for a review, see Roese, 1997). An important theoretical claim in this body of literature has been that counterfactual thinking can influence the causal judgment process in two, distinct ways. The first proposed route of influence involves the notion of contrast between factual and counterfactual events. When an outcome evokes surprise by violating expectancies or norms, the persistence of attention to those expectancies is, by definition, a counterfactual representation because, in hindsight, individuals know that their violated expectancies are not factual representations of the past. Kahneman and Miller (1986) proposed that this form of counterfactual thinking, in which an expectancy-consistent counterfactual outcome contrasts sharply with a surprising factual outcome, can influence the causal judgment process by prompting the question, Why? Similarly, Hilton and Slugoski (1986) and McGill (1989), building on the work of earlier philosophers (Gorovitz, 1965; Hart & Honoré, 1959; Hesslow, 1983), proposed that the normal outcomes evoked by counterfactuals can serve as contrast cases that define the focus of explanation. The second proposed route of influence involves the notion that counterfactual conditionals are well-suited for spotlighting causal contingencies. That is, some theorists (e.g, Roese & Olson, 1995a) have argued that the conditional structure of counterfactuals that take an “if only…, then…” form serve an important function in the causal selection process. For example, after a student learns that she just failed an important exam, she might think something like “If only I had studied a bit harder, I would have passed the exam.” Although these counterfactuals may simply reflect existing causal beliefs, their activation may also strengthen or reaffirm causal beliefs, perhaps by making explicit knowledge of contingencies and “but for” causes that would otherwise have received little weight in subsequent thought and judgment (see Tetlock & Belkin, 1996). A strong version of this argument holds that counterfactual conditionals represent tests of hypothesized causal connections between the outcome and antecedent events in an episode. As Mackie (1974) put it, “... in quite a primitive and unsophisticated way we can transfer the non-occurrence of Y from the before situation to a supposed later situation, in which, similarly, X did not occur, and form the thought which is expressed by the statement ‘If X had not occurred, Y would not have occurred’” (p. 56). Psychological accounts of causal judgment that build on Mackie (1974) posit that counterfactual conditionals can act as tests of the necessary (or “but for”) causes of outcomes that demand explanation (e.g., Einhorn & Hogarth, 1986; Kahneman & Tversky, 1982; Lipe, 1991; McGill & Klein, 1993; Roese & Olson, 1997; Wells & Gavanski, 1989). Although no study to my knowledge has demonstrated that people intentionally generate counterfactual conditionals in order to test their hypotheses about causal relations, there is some empirical support for the weaker notion that causal judgments are influenced by the ease with which people can mentally undo focal events in the past and by the content of their counterfactual musings (Branscombe, Owen, Garstka, & Coleman, 1996; Roese & Olson, 1997; Wells & Gavanski, 1989). The case for an effect of counterfactual thinking on causal judgment is strengthened by studies that have also examined related judgments of blame (Branscombe et al., 1996; Miller & Gunasegaram, 1990) and controllability (Roese & Olson, 1995b). Taken together, however, this body of evidence is based almost exclusively on either correlational studies or experiments in which the mutability of focal outcomes is manipulated (i.e., manipulations of the ease with which alternative outcomes can be mentally constructed) (for an exception to this rule, see Roese & Olson, 1997, who an experiment using a task facilitation paradigm). In experiments that manipulate outcome mutability, it is often assumed that if causal judgments in the high-mutability condition are more extreme or frequent than in comparable low-mutability conditions, then counterfactual thinking must have played a causal role (e.g., Wells & Gavanski, 1989). Other research (e.g., N’gbala & Branscombe, 1995), however, indicates that the effect of mutability manipulations on causal judgments may have had more to do with the specific hypothetical scenarios used in previous experiments than with a robust effect of counterfactual thinking on causal selection. For instance, Mandel and Lehman (1996, Experiment 3) showed that, when participants read a case that afforded the opportunity to select different antecedents in counterfactual and causal judgment tasks, mutability manipulations influenced counterfactual assessments but had no reliable effect on causal judgments. Importantly, no research to date has directly addressed the basic question of whether the effect of counterfactual thinking on attribution differs in reliable and consequential ways from the effect of factual thinking. That is, does thinking counterfactually about what might, could, or would have been have a greater impact on attribution than thinking factually about what was? The closest study in this regard is one by McCloy and Byrne (2017, Experiment 1). Participants in one condition were instructed to think counterfactually about a hypothetical case that they had just considered by completing an “if only…” sentence stem, whereas participants in a control condition were not asked to do so. Participants then rated the importance of a single causal candidate. Mean causal importance ratings did not differ between the counterfactual and control conditions. As McCloy and Byrne (2017) acknowledged, however, the null finding in their study is difficult to interpret because the completion task did not constrain attention to a specific antecedent, whereas the causal rating task did. Given the strong claims that have been made about the consequential effects of counterfactual thinking on causal judgment and related attribution processes, research that directly compares the effect of counterfactual and factual thinking on dependent measures such as causality, prevention, blame, and control is sorely needed. For instance, consider Roese and Olson’s (1995a) claim that although “…not all conditionals are causal…counterfactuals, by virtue of the falsity of their antecedents, represent one class of conditional propositions that are always causal. The reason for this is that with its assertion of a false antecedent, the counterfactual sets up an inherent relation to a factual state of affairs” (p. 11). This proposal suggests not only that counterfactual conditional thinking can influence causal judgment, but that it should have a stronger effect on causal judgment than comparable factual conditional thinking, which would not necessarily automatically recruit a contrast case. Recent extensions of mental model theory (e.g., Byrne, 1997; Byrne & Tasso, 1999) have reached similar conclusions—namely that, whereas counterfactual conditionals automatically “recover” factual models, thus establishing a salient contrast case, factual conditionals do not automatically recover counterfactual models. These accounts do not hedge on the prediction that counterfactual and factual modes of thinking will produce different effects on causal attribution and prompt different cognitive representations. 网站原创范文除特殊说明外一切图文作品权归所有；未经官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯作品权现象，保留一切法学追诉权。()，英语论文范文，英语论文范文 |
