Lack of optimistic biases in depression and social anxiety is reflected in reduced positive self-beliefs, but distinct processing of social feedback

Abstract

Processing social feedback optimistically helps maintain a positive self-image and stable social relationships. Individuals with depression and social anxiety often lack this optimistic bias. Yet, the cognitive routes over which social feedback reinforces a negative self-image have remained largely unclear and may differ between depression and social anxiety. A reanalysis of previous studies (n = 450) and a pre-registered replication (n = 807) demonstrated that self-reported depressive symptoms and social anxiety were associated with better learning of negative social evaluations about the self, relative to positive evaluations, in a computerised social evaluation task. Transdiagnostically, this asymmetry was driven by reduced positive trait-like beliefs. Yet, in social anxiety, this bias reflected a heightened sensitivity to negative social feedback, whereas in depression it co-existed with a blunted response to social feedback. Recognising such differences in feedback processing may inform approaches to personalizing treatment.

Jumping to Attributions during Social Evaluation

“…We introduced changes to the delivery of the SELT-R to provide more data per trial and per participant (e.g., use continuous ratings instead of binary ratings) and proposed new computational models to better understand the rapid inference found in this task. Our new 'classify-refine' models, which allowed participants to 'jump to attributions' and then refine them, out-performed our previously published inference ('beta-belief') models 14,17 .…”

“…Following preliminary studies and the limited recovery of the initial bias parameter in previous studies 14,17 , SELT-R included eight blocks, a notable expansion from the original version. It also employed continuous measures (e.g., "20% likely to be dull," "80% likely to be exciting") rather than binary measures, aligning better with the nuanced nature of human decision-making and providing more bits of information per trial.…”

“…'Beta-belief' models have given a useful account of social learning in social learning tasks, including the SELT [14][15][16] . Their parameters are able to differentiate between healthy participants, and those with high levels of mood and social anxiety symptoms 11,14,17 .…”

“…However, our previous work showed a need for improvements in both the SELT and its computational models 14,17,18 , as they have required large samples and sophisticated statistical analyses to demonstrate effects 17 . Task models did not fully address the rapid initial learning of evaluations, and later progressive refinement, highlighting the need for new models.…”

Reactivating Positive Personality Traits During Sleep Promotes Positive Self-Referential Processing

“…This positive self-referential bias is commonly associated with lower levels of depressive symptoms (e.g., self-doubt, worthlessness), and is crucial for mental well-being, especially when facing self-threatening information (Sowislo et al, 2013). While the psychological benefits of positive self-referential processing is well-established (Taylor & Brown, 1988; Colombo et al, 2020; Orth et al, 2022; Weisenburger et al, 2023), a significant gap exists in understanding how to effectively enhance this process (Orth et al, 2022; Hoffmann et al, 2023). To address this gap, we integrated two procedures that may enhance positive self-referential processing: (1) wakeful cued-approach training (CAT, Schonberg et al, 2014), and (2) a sleep-based targeted memory reactivation procedure (TMR, Oudiette and Paller 2013).…”