How Does Explanatory Virtue Determine Probability Estimation?—Empirical Discussion on Effect of Instruction

It is important to reveal how humans evaluate an explanation of the recent development of explainable artificial intelligence. So, what makes people feel that one explanation is more likely than another? In the present study, we examine how explanatory virtues affect the process of estimating subjective posterior probability. Through systematically manipulating two virtues, Simplicity—the number of causes used to explain effects—and Scope—the number of effects predicted by causes—in three different conditions, we clarified two points in Experiment 1: (i) that Scope's effect is greater than Simplicity's; and (ii) that these virtues affect the outcome independently. In Experiment 2, we found that instruction about the explanatory structure increased the impact of both virtues' effects but especially that of Simplicity. These results suggest that Scope predominantly affects the estimation of subjective posterior probability, but that, if perspective on the explanatory structure is provided, Simplicity can also affect probability estimation.

Property Causation

If one set of properties supervenes upon another, then the former have causal powers if: first, the supervenience base properties minimally metaphysically necessitate the supervening properties; second, part of the minimal supervenience base causes the target effect; third, instances of the supervening properties would all cause certain target effects in the right kind of circumstances, as a result of this. When these conditions are met, the causal relationship holds not only in virtue of the supervenience base properties but also the supervening ones. Two further explanatory virtues that citing property causes may display are: when a property has a distinctive causal profile (when it makes a causal contribution that no other property would) and when an instance of the property supplies the precise contribution required for a certain effect. Pragmatic appeal to the second explanatory virtue explains away our tendency to hear certain explicitly contrastive statements mistakenly as true.

Bayesianism and IBE

Different views have been proposed about how Inference to the Best Explanation (IBE) and Bayesianism might be compatible with one another. One is a hybrid view, according to which explanatory considerations play a role in driving the Bayesian machinery. Another is an ‘emergent compatibilist’ view, in which an independently motivated Bayesian model of IBE is provided, so that explanatory considerations emerge from the Bayesian machinery rather than driving it. Specific scientific arguments can serve as test cases for these general views. The chapter argues that the case of Williams’ argument against group selection, discussed by Elliott Sober, is better understood with the emergent compatibilist picture, than with the hybrid view. This analysis of the case challenges Elliott Sober’s claim that the epistemic significance of appeals to the explanatory virtue of parsimony is highly case-specific. Instead, a more unified picture of IBE and its connection to Bayesianism is suggested.

Inference to the Best Explanation, Cleaned Up and Made Respectable

Despite decades of focused philosophical investigation, inference to the best explanation still lacks a precise articulation and compelling defense. The primary reason for this is that it is not at all clear what it means for a hypothesis to be the best available explanation of the evidence. This chapter first seeks to rectify this problem by developing a formal explication of the explanatory virtue of power. A resulting account of inference to the best explanation is then evaluated as a form of uncertain inference. Overall, the chapter offers a precise account and novel defense of one important version of inference to the best explanation.