14 April 2023 in Proceedings of the ACM on Human-Computer Interaction
Prior work has identified a resilient phenomenon that threatens the performance of human-AI decision-making teams: overreliance, when people agree with an AI, even when it is incorrect. Surprisingly, overreliance does not reduce when the AI produces explanations for its predictions, compared to only providing predictions. Some have argued that overreliance results from cognitive biases or uncalibrated trust, attributing overreliance to an inevitability of human cognition. By contrast, our paper argues that people strategically choose whether or not to engage with an AI explanation, demonstrating empirically that there are scenarios where AI explanations reduce overreliance. To achieve this, we formalize this strategic choice in a cost-benefit framework, where the costs and benefits of engaging with the task are weighed against the costs and benefits of relying on the AI. We manipulate the costs and benefits in a maze task, where participants collaborate with a simulated AI to find the exit of a maze. Through 5 studies (N = 731), we find that costs such as task difficulty (Study 1), explanation difficulty (Study 2, 3), and benefits such as monetary compensation (Study 4) affect overreliance. Finally, Study 5 adapts the Cognitive Effort Discounting paradigm to quantify the utility of different explanations, providing further support for our framework. Our results suggest that some of the null effects found in literature could be due in part to the explanation not sufficiently reducing the costs of verifying the AI's prediction.
1 October 2021 in Psychological Review
How do people make causal judgments about physical events? We introduce the counterfactual simulation model (CSM) which predicts causal judgments in physical settings by comparing what actually happened with what would have happened in relevant counterfactual situations. The CSM postulates different aspects of causation that capture the extent to which a cause made a difference to whether and how the outcome occurred, and whether the cause was sufficient and robust. We test the CSM in several experiments in which participants make causal judgments about dynamic collision events. A preliminary study establishes a very close quantitative mapping between causal and counterfactual judgments. Experiment 1 demonstrates that counterfactuals are necessary for explaining causal judgments. Participants' judgments differed dramatically between pairs of situations in which what actually happened was identical, but where what would have happened differed. Experiment 2 features multiple candidate causes and shows that participants' judgments are sensitive to different aspects of causation. The CSM provides a better fit to participants' judgments than a heuristic model which uses features based on what actually happened. We discuss how the CSM can be used to model the semantics of different causal verbs, how it captures related concepts such as physical support, and how its predictions extend beyond the physical domain. (PsycInfo Database Record (c) 2021 APA, all rights reserved).
17 October 2017 in Psychological Science
How do people make causal judgments? What role, if any, does counterfactual simulation play? Counterfactual theories of causal judgments predict that people compare what actually happened with what would have happened if the candidate cause had been absent. Process theories predict that people focus only on what actually happened, to assess the mechanism linking candidate cause and outcome. We tracked participants’ eye movements while they judged whether one billiard ball caused another one to go through a gate or prevented it from going through. Both participants’ looking patterns and their judgments demonstrated that counterfactual simulation played a critical role. Participants simulated where the target ball would have gone if the candidate cause had been removed from the scene. The more certain participants were that the outcome would have been different, the stronger the causal judgments. These results provide the first direct evidence for spontaneous counterfactual simulation in an important domain of high-level cognition.
1 September 2018 in Cognitive Psychology
1 August 2018 in Cognition
1 February 2021 in Developmental Psychology
Young children often struggle to answer the question "what would have happened?" particularly in cases where the adult-like "correct" answer has the same outcome as the event that actually occurred. Previous work has assumed that children fail because they cannot engage in accurate counterfactual simulations. Children have trouble considering what to change and what to keep fixed when comparing counterfactual alternatives to reality. However, most developmental studies on counterfactual reasoning have relied on binary yes/no responses to counterfactual questions about complex narratives and so have only been able to document when these failures occur but not why and how. Here, we investigate counterfactual reasoning in a domain in which specific counterfactual possibilities are very concrete: simple collision interactions. In Experiment 1, we show that 5- to 10-year-old children (recruited from schools and museums in Connecticut) succeed in making predictions but struggle to answer binary counterfactual questions. In Experiment 2, we use a multiple-choice method to allow children to select a specific counterfactual possibility. We find evidence that 4- to 6-year-old children (recruited online from across the United States) do conduct counterfactual simulations, but the counterfactual possibilities younger children consider differ from adult-like reasoning in systematic ways. Experiment 3 provides further evidence that young children engage in simulation rather than using a simpler visual matching strategy. Together, these experiments show that the developmental changes in counterfactual reasoning are not simply a matter of whether children engage in counterfactual simulation but also how they do so. (PsycInfo Database Record (c) 2021 APA, all rights reserved).