3 - (IOP 40) Biases in Belief Updating Are Influenced by Clinical Context but Not Clinical Symptoms

The optimism bias effect states that people update their beliefs more in response to "good news" (positive outcomes) than “bad news" (negative outcomes). However, research indicates the optimism bias is reversed in some clinical conditions, reflecting a pessimism bias instead. For example, people with depression learn more from punishment than reward, and people with social anxiety update self-perceptions more following negative feedback vs. positive feedback. Extant research on updating biases is characterized by two key limitations. First, the majority of studies ask participants to estimate probabilities of very unlikely negative outcomes (e.g., a serious car accident), constraining variability in belief updating measures. Second, some research shows that belief updating biases are strongest when a negative outcome aligns with a person’s unique fears (e.g., someone with OCD may show pessimism bias for contamination outcomes but not social embarrassment outcomes), yet prior work has not tested how transdiagnostic symptom categories influence belief updating within participants.

In an online study with a large non-clinical sample (N=302), we examined belief updating bias in five hypothetical scenarios reflecting a common fear for separate clinical conditions: generalized anxiety disorder (GAD), social anxiety disorder (SAD), obsessive-compulsive disorder (OCD), panic disorder (PD), and hoarding disorder (HD). For each scenario, participants made an initial estimate of the probability of the negative outcome and then received objective feedback on its “true” probability. To manipulate the likelihood of negative outcomes, we randomized feedback into two conditions: low true probability (15%) or high true probability (75%). After receiving true probability feedback, participants were asked to update their initial estimate of the probability of the negative outcome. Participants also completed self-report measures of GAD, SAD, OCD, PD, and HD symptom severity.

Using hierarchical linear models, we tested the magnitude of belief updating for “good news” (true probability lower than the initial estimate) vs. “bad news” (true probability higher than the initial estimate). We found evidence of pessimism bias: on average, participants updated their beliefs more for bad news vs. good news (β=7.68, p<.001). Scenario type significantly influenced belief updating bias (F=3.28, p=.011). Within scenario, significant pessimism bias was observed in the SAD (β=12.71, p=.003), OCD (β=20.46, p<.001), and PD (β=10.86, p=.009) scenarios, but no significant bias was observed in the GAD (β=0.22, p=.957) or HD scenario (β=-7.89, p=.061). Furthermore, none of the clinical symptom scores significantly influenced belief updating bias, within or across scenarios (all p’s>.136).

Many individuals with anxiety and related conditions live in fear of low-probability hypothetical scenarios that will never materialize. Our study suggests that for some scenarios, people exhibit pessimism bias, updating beliefs more for bad news vs. good news. Moreover, pessimism bias appears independent of symptom severity, potentially working to maintain fear across clinical conditions.