Counterfactuals Matter: A Reply to Weisberg & Gopnik

Abstract

Weisberg and Gopnik and I agree that understanding the development of counterfactual thinking is important. We are also in agreement that very young children's causal reasoning is not underpinned by explicit real-world counterfactual thinking, and that it is likely that there are shared psychological processes that are involved in all types of general counterfactual thinking. Where we diverge is that Weisberg and Gopnik emphasize the unity of counterfactual thoughts, whereas I emphasize the differences between types. Their point that “the space of counterfactuals is continuous” may be true, but this does not mean that exactly the same psychological processes are employed to think about all counterfactuals within this space. My argument is that real-world counterfactuals do not simply differ from other types of counterfactuals by a matter of degree (e.g., by making greater inhibitory control demands), but in their relationship with reality. Critically, real-world counterfactuals are not necessarily close to reality, in the sense that they are very similar in content, but they are constrained by making the fewest changes to the real world having made a counterfactual move. For example, the possible worlds in which (a) I wore a different dress today or (b) a pandemic wiped out the majority of human beings last month can both be real-world counterfactuals, but the second will be much less similar to the current state of affairs, while still being a nearest possible world (Lewis, 1973). This is not to say that the similarity of the counterfactual world to the real world is unimportant; as Weisberg and Gopnik rightly point out, this can influence children's behavior (e.g., in the realm of pretense). I think that this particular relationship that real-world counterfactuals have with reality makes specific psychological demands. These might be competence rather than performance executive function demands (i.e., the demands are integral to reasoning success. See Russell, 1996). Another possibility is that real-world counterfactuals require conceptual change in the young child's understanding of time: that time is event-independent rather than event-dependent (McCormack, 2014). It will also be important in the future to clarify what are the shared psychological processes that support general counterfactual thinking. Is this a specific ability dedicated to counterfactuals or might it overlap with other forms of mental simulation, such as constructing memories of past events or imagining things you believe to be true but lack direct experience of? To conclude, more research, especially with young children, will identify the similarities and differences between types of counterfactuals as well as other imaginative abilities, and indicate where there are quantitative changes and where qualitative. Further constructive discussion, of the type we have had here, is important to advance our understanding of the development of counterfactual thinking.