Confidence and Gradation in Causal Judgment

Abstract

When asking if lightning caused the forest fire, one might think that the lightning is more of a cause than the dry climate (i.e., it is a graded cause) or they might instead think that the lightning strike completely caused the fire while the dry conditions did not cause it at all (i.e., it is a binary cause). Psychologists and philosophers have long debated whether such judgments are graded. To address this debate, we started by reanalyzing data from four recent studies. In this context, we provide novel evidence that causal judgments are actually multimodal: although most causal judgements were binary, there was also some gradation. We then tested two competing explanations for the gradation we observed: the confidence explanation, which states that gradation distinguishes between certain and uncertain causes, and the strength explanation, which states that gradation distinguishes between strong and weak causes. Experiment 1 tested the confidence explanation and showed that gradation in causal judgments was moderated by confidence. People tended to make graded causal judgments when they were less confident, but they tended to make discrete causal judgments when they were more confident. Experiment 2 tested the causal strength explanation and showed that although causal judgments varied with factors associated with causal strength, confidence ratings were unchanged. Overall, we found that causal judgments are multimodal and that observed gradation reflects independent effects of confidence and causal strength on causal judgments.

Certainty and Singular Causal Judgment

“…This is a desirable property for Lewis (1974), since there should be a fact of the matter for any given causal claim. But since people appear to have uncertainty and be able to assess their uncertainty in causal claims, a psychological theory of causal judgment should account for that uncertainty (Collins & Shanks, 2006;Liljeholm, 2015;Liljeholm & Cheng, 2009;O'Neill et al, 2021;Perales & Shanks, 2003;Shanks, 1987).…”

“…It should not be terribly surprising that the Lewis-inspired account does not suffice as a psychological account of certainty in causal judgment; after all, Lewis (1974) sought to describe what causation is, not how we reason causally. But recent extensions of the general idea that counterfactuals underlie causal judgment have gained much popularity for their ability to explain how causal judgments are sensitive to dependence between events (Cheng & Novick, 1990;Davis & Rehder, 2020;Pearl, 2009), normality (Gerstenberg & Icard, 2020;Henne et al, 2021;Hitchcock & Knobe, 2009;Icard et al, 2017;Kirfel & Lagnado, 2019;Knobe & Fraser, 2008;Kominsky & Phillips, 2019;Sytsma, 2019), whether the events are actions or omissions (Henne, Bello, et al, 2019;Henne, Niemi, et al, 2019;Henne et al, 2017), the presence of alternative causes (Kominsky et al, 2015;Lu et al, 2008;Morris et al, 2019;O'Neill et al, 2021), and the perceived effectiveness of an intervention on the cause (Kushnir & Gopnik, 2005;Lagnado & Sloman, 2004;Morris et al, 2018;Sobel & Kushnir, 2006;Woodward, 2003) among other factors.…”

“…First, since process theories focus on exchanges of quantities, they all but fail to explain judgments of omissive causes (e.g., the lack of a firmware update caused the computer mouse to stop functioning). Process theories, then, do not readily account for the fact that people can consistently report confidence for omissive causal judgments (Bernstein, 2014;O'Neill et al, 2021). The second problem is that process theories do not have an established mechanism for explaining certainty in causal judgment.…”

“…And even if researchers discovered a way to observe causal judgments in non-human animals, it is not clear that non-human animals can make meaningful singular causal judgments, let alone report on their uncertainty in such judgments. Altogether, self-report measures of uncertainty can be (and have already been) applied to the domain of causal judgments, and may prove as useful for research on certainty in causal judgment as they have been for research on metacognition more broadly (Liljeholm, 2015;Liljeholm & Cheng, 2009;O'Neill et al, 2021).…”

“…In contrast to general causal relations, singular causal relations concern whether this C caused that E in particular, including whether this smoking caused that cancer, whether this tax break benefited that economy, and whether this out-of-date firmware caused that mouse to malfunction (as in the original example). Whereas there is some research on confidence in general causal judgments, there has been virtually no research on confidence in singular causal judgments (but see O'Neill et al, 2021). In this paper, I will focus exclusively on singular causal judgments like in the broken mouse example.…”