Computational modeling of pessimistic future views in individuals with depressive symptoms: dynamics in affective forecasts

• Applicant: Keisuke Takano, Ph.D.
• Subject Area: Personality Psychology, Clinical and Medical Psychology, Methodology
• Term: from 2019 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 415624938

Negative beliefs about the future are proposed to be a key component of depressive cognition. Bias in imagining affective experiences, namely overestimating future negative and underestimating positive affect (i.e., negative affective forecasts), is thought to guide maladaptive emotion regulation. Theories and empirical evidence suggest that the negative future beliefs are not a mere symptom of, but also a vulnerability factor for depression. Despite extensive research in the field, it remains unresolved how people with depressive symptoms, or those vulnerable to depression, come to believe that they will never feel better. Although existing descriptive models propose that specific psychological constructs (e.g., schemas) prompt the negative future beliefs, these models are often too vague to formulate a psychometrically precise model. Given that the descriptive models and behavioral experiments assessing negatively biased cognition have limited predictive validity, a novel approach is warranted to clarify how biased cognition is generated and what dysfunction specifically contributes to such biased cognition and depressive symptoms. The focus of the proposed project is to (a) model process details in generating negative affective forecasts by using computational modeling and (b) systematically test individual differences in these forecasting processes linked to depressive symptoms. To this end, we will test the original hypothesis that depressive symptoms are associated with difficulty in updating beliefs about emotional changes, that is, holding old beliefs that negative affect will continue and ignoring a new observation that negative affect is decreasing. This belief-updating model is formulated in the framework of statistical filtering. We will test and validate the model on data collected under laboratory and daily life (experience sampling) settings to increase ecological validity. Additionally, we will evaluate the predictive power on depressive symptomatology at a six-month follow-up. In sum, this project aims to provide fundamental new insights into the mechanisms that generate negative future beliefs in depression at the theoretical and methodological levels. Our innovative computational approach is likely to refine existing theories of depressive cognition and increase the predictive validity of negative affective forecasts as a vulnerability factor for depression. Knowing the process details that generate negative affective forecasts is also relevant for clinical practices, as specifying the root of biased cognition is an important prerequisite for developing effective intervention and cognitive training.

• DFG Programme: Research Grants