Counterfactual thinking about controllable events.

When people think about what might have been, they mentally undo controllable rather than uncontrollable events. We report the results of two experiments in which we examined this controllability effect in counterfactual thinking. The experiments show that the mutability of controllable events is influenced by the perceived appropriateness or inappropriateness of the events. The first experiment shows that people change inappropriate controllable actions more than appropriate controllable ones. The second experiment shows that people mutate inappropriate controllable events whether the outcome is exceptional or normal with respect to intrapersonal habitual norms, and whether the outcome is positive or negative. We discuss the implications for alternative theories of counterfactual thinking.

Counterfactual thinking about actions and failures to act.

When people think counterfactually about how a situation could have turned out differently, they mentally undo events in regular ways (e.g., they focus on actions not failures to act). Four experiments examine the recent discovery that the focus on actions in the short term switches to inactions in the long term. The experiments show that this temporal switch occurs only for particular sorts of situations. Experiment 1 showed no temporal pattern to the agency effect when 112 participants judged emotional impact and frequency of "if-only" thoughts from both short- and long-term perspectives for an investment scenario. Experiment 2 showed no temporal pattern when 190 participants considered a college choice scenario with a good outcome. Experiment 3 showed no temporal pattern when 131 participants considered an investment scenario even when the situation for the actor and nonactor was bad from the outset. Experiment 4, with 113 participants, showed a focus on actions even when the investment loss was equal for both the actor and nonactor. The implications of the results are discussed in terms of what is explicitly available in the mental representation of actions and inactions.

The temporality effect in counterfactual thinking about what might have been.

When people think about what might have been, they undo an outcome by changing events in regular ways. Suppose two contestants could win 1,000 Pounds if they picked the same color card; the first picks black, the second red, and they lose. The temporality effect refers to the tendency to think they would have won if the second player had picked black. Individuals also think that the second player will experience more guilt and be blamed more by the first. We report the results of five experiments that examine the nature of this effect. The first three experiments examine the temporality effect in scenarios in which the game is stopped after the first contestant's card selection because of a technical hitch, and then is restarted. When the first player picks a different card, the temporality effect is eliminated, for scenarios based on implicit and explicit negation and for good outcomes. When the first player picks the same card, the temporality effect occurs in each of these situations. The second two experiments show that it depends on the order of events in the world, not their descriptive order. It occurs for scenarios without preconceptions about normal descriptive order; it occurs whether the second event is mentioned in second place or first. The results are consistent with the idea that the temporality effect arises because the first event is presupposed and so it is immutable; and the elimination of the temporality effect arises because the availability of a counterfactual alternative to the first event creates an opposing tendency to mutate it. We sketch a putative account of these effects based on characteristics of the mental models people construct when they think counterfactually.

Deductive reasoning with factual, possible, and counterfactual conditionals.

We compared reasoners' inferences from conditionals based on possibilities in the present or the past (e.g., "If Linda had been in Dublin then Cathy would have been in Galway") with their inferences based on facts in the present or the past (e.g., "If Linda was in Dublin then Cathy was in Galway"). We propose that people construct a richer representation of conditionals that deal with possibilities rather than facts: Their models make explicit not only the suppositional case, in which Linda is in Dublin and Cathy is in Galway, but also the presupposed case, in which Linda is not in Dublin and Cathy is not in Galway. We report the results of four experiments that corroborate this model theory. The experiments show that reasoners make more inferences from conditionals based on possibilities rather than on facts when the inferences depend on the presupposed case. The results also show that reasoners generate different situations to verify and falsify conditionals based on possibilities and facts.

Reasoning strategies for suppositional deductions.

Deductive reasoning shares with other forms of thinking a reliance on strategies, as shown by the results of three experiments on the nature and development of control strategies to solve suppositional deductions. These puzzles are based on assertors who may or may not be telling the truth, and their assertions about their status as truthtellers and liars. The first experiment shows that reasoners make backward inferences as well as forward inferences, to short-cut their way through the alternatives, and the generation of suppositions is a source of difficulty. The second experiment establishes that the elimination of the suppositional status of an individual does not render problems easier. The third experiment shows that reasoners can improve their reasoning accuracy and speed spontaneously, without feedback, and it clarifies the transfer of strategies and their development. We discuss the implications of these data for alternative theories of suppositional deduction and for the relationship between reasoning and other forms of thinking such as problem solving.

Propositional reasoning by model.

This article describes a new theory of propositional reasoning, that is, deductions depending on if, or, and, and not. The theory proposes that reasoning is a semantic process based on mental models. It assumes that people are able to maintain models of only a limited number of alternative states of affairs, and they accordingly use models representing as much information as possible in an implicit way. They represent a disjunctive proposition, such as "There is a circle or there is a triangle," by imagining initially 2 alternative possibilities: one in which there is a circle and the other in which there is a triangle. This representation can, if necessary, be fleshed out to yield an explicit representation of an exclusive or an inclusive disjunction. The theory elucidates all the robust phenomena of propositional reasoning. It also makes several novel predictions, which were corroborated by the results of 4 experiments.

Modal reasoning, models, and Manktelow and Over.

Manktelow and Over (1991) argue that their studies of Wason's selection task favor explanations of deontic reasoning based on mental models, but that such theories need to incorporate utilities. This theoretical note proposes a simpler explanation of the phenomena: subjects in the selection task consider only those cards that are explicitly represented in their models of the conditional, and so insight into the task depends on constructing fully explicit models. Such models for modal conditionals of the form, If p occurs then q may occur are: [formula: see text] Each line denotes a separate model, and the models represent either what is possible, or, in the deontic interpretation, what is permissible. A deontic rule is accordingly violated by the contingency: [symbol: see text] p and q, for example the rule, "If you spend more than 100 pounds, then you may take a free gift" is violated by taking the free gift (q) but not spending more than 100 pounds ([symbol: see text] p). If the rule is interpreted as a bi-conditional, then the second of the models, p and [symbol: see text] q, is also now a violation, for example spending more than 100 pounds (p) but not getting the free gift ([symbol: see text] q). Manktelow and Over's instructions lead subjects to focus on one or other of the two sorts of violations of the rule. There is accordingly no need to introduce utilities into models in order to explain the phenomena.

Suppressing valid inferences with conditionals.

Three experiments are reported which show that in certain contexts subjects reject instances of the valid modus ponens and modus tollens inference form in conditional arguments. For example, when a conditional premise, such as: If she meets her friend then she will go to a play, is accompanied by a conditional containing an additional requirement: If she has enough money then she will go to a play, subjects reject the inference from the categorical premise: She meets her friend, to the conclusion: She will go to a play. Other contexts suppress the conditional fallacies. The first experiment demonstrates the effects of context on conditional reasoning. The second experiment shows that the inference suppression disappears when the categorical premise refers to both of the antecedents, such as: She meets her friend and she has enough money. In this case, subjects make both the valid inferences and the fallacies, regardless of the contextual information. The third experiment establishes that when subjects are given general information about the duration of a situation in which a conditional inducement was uttered, such as: If you shout then I will shoot you, they reject both the valid inferences and the fallacies. The results suggest that the interpretation of premises plays an even more central role in reasoning than has previously been admitted.