Search Papers | Poster Sessions | All Posters
Poster A52 in Poster Session A - Tuesday, August 6, 2024, 4:15 – 6:15 pm, Johnson Ice Rink
Confirmation Bias Is Generalizable Across Pain, Negative Emotion, and Cognitive Effort
Aryan Yazdanpanah1, Heejung Jung1, Alireza Soltani1, Tor D. Wager1 (); 1Dartmouth College
Expectations have strong influences on perception, cognition, and behavior. With subsequent learning that relies on prediction errors, one can flexibly update the association between expectation and experience, leading to a fine-tuned representation of the experience. However, this update could resist change, due to “confirmation bias”, i.e., when learning is strengthened by evidence that supports expectations and attenuated by evidence that contradicts them. Despite prior research on confirmation bias, their shared underlying mechanisms are unclear due to studies focusing on a single domain. To overcome this, we performed a large study on the effects of expectation on somatic pain, vicarious pain, and cognitive effort within the same participants. Using a combination of model-free and model-based approaches, we found evidence for domain-general expectancy effects. Moreover, the confirmation bias within individuals was correlated between somatic pain and cognitive effort and between vicarious pain and cognitive effort. Overall, our results provide evidence for some consistency of confirmation bias and shared mechanisms across cognitive and affective domains.
Keywords: confirmation bias domain-generality expectation learning