Eye-Tracking Uncovers Cognitive Mechanisms Underlying High Level Human Judgments

MedicalResearch.com Interview with:
Tobias Gerstenberg, PhD


MedicalResearch.com: What is the background for this study? What are the main findings?

Response: The question of how causation is best understood has been troubling philosophers for a long time. As psychologists, we are particularly interested in understanding how people make causal judgments.

In our experiments, we showed participants video clips of colliding billiard balls. Participants were asked to say whether one ball (ball A) caused another (ball B) to go through a gate, or prevented it from going through. We used eye-tracking technology to record participants’ eye-movements as they were watching the clips.

The results showed that participants spontaneously engaged in counterfactual simulation when asked to make causal judgments. They not only looked at what actually happened, but also tried to anticipate where ball B would have gone if ball A hadn’t been present. The more certain participants were that ball B would have missed the goal if ball A hadn’t been there, the more they agreed that ball A caused ball B to go through the gate.

In a control condition we asked participants about what actually happened. In this condition, participants were much less likely to simulate where ball B would have gone. Together, these findings demonstrate a very close link between counterfactual simulation and causal judgment.

MedicalResearch.com: What should clinicians and patients take away from your report?

Response: Our experiments show that eye-tracking is a powerful tool for uncovering the cognitive mechanisms that underlie complex high-level judgments, such as judgments of causation. While previous work has argued for a close correspondence between counterfactual reasoning and causal judgments, our study is the first to show how direct this link is. Furthermore, we were able to show that there is a close relationship between participants’ eye-movements and their causal judgments. The more participants engaged in counterfactual simulation, the better their judgments were accounted for by the counterfactual simulation model we develop in the paper.

MedicalResearch.com: What recommendations do you have for future research as a result of this study?

Response: Our results demonstrate a close link between counterfactual simulation and causal judgments. In our experiment, we used a task that relied on people’s intuitive understanding of physics. Our model predicts that people’s causal judgments will be related to their ability to simulate the relevant counterfactual. Thus, it would be interesting to see whether people who have difficulty with mental simulation will show different eye-movement patterns, and reach different causal judgments from what we have seen in this study.

We are also planning on using eye-tracking to uncover the developmental timeline of counterfactual thinking. Current research suggests that explicit counterfactual thinking develops relatively late. However, if our model is correct, and causal judgments require counterfactual simulation, then this late development of counterfactual thinking is puzzling. It is plausible that correctly answering counterfactual questions develps much later than implicit counterfactual simulation does.   

MedicalResearch.com: Is there anything else you would like to add?

Response: A preprint of the paper is available here: https://osf.io/8q4jv/

The materials, data, and analysis scripts are available here: https://osf.io/du5jc/

MedicalResearch.com: Thank you for your contribution to the MedicalResearch.com community.


Eye-Tracking Causality

Tobias Gerstenberg, Matthew F. Peterson, Noah D. Goodman

David Lagnado Joshua B. Tenenbaum

Date created: 2017-10-17 09:31 AM | Last Updated: 2017-10-17 09:37 AM

Note: Content is Not intended as medical advice. Please consult your health care provider regarding your specific medical condition and questions.


[wysija_form id=”1″]






Last Updated on October 19, 2017 by Marie Benz MD FAAD