Transdiagnostic mental health symptom dimensions predict use of flexible model-based inference in complex environments

Key findings from this study

• The study found that inattentive and neurodevelopmental symptoms predict superior performance in inferring complex predator behaviour, while externalising symptoms predict increased erroneous inferences.
• The researchers demonstrate that transdiagnostic symptom dimensions explain behavioural variation better than general psychopathology factors.
• The authors report that differential reliance on goal-directed decision-making mechanisms mediates the relationship between symptom profiles and inference task performance.

Overview

This study examines whether transdiagnostic mental health symptom dimensions predict differential use of goal-directed decision-making in complex inference tasks. A large sample completed a naturalistic threat inference task alongside measures of mental health symptoms and neurodevelopmental traits. Computational modelling assessed the degree to which individuals employed flexible, goal-directed reasoning versus alternative decision strategies.

Methods and approach

Participants (n = 1025) completed a validated threat inference task requiring prediction of predator behaviour in interactive decision scenarios. Self-report measures captured mental health symptoms and neurodevelopmental characteristics. Computational modelling quantified the extent of goal-directed decision-making used during inference tasks. Analyses examined whether specific symptom dimensions explained behavioural variability better than general factors.

Results

Inattentive and neurodevelopmental symptoms correlated with superior prediction accuracy of predator behaviour. Externalising symptoms associated with increased erroneous inferences. Transdiagnostic symptom dimensions explained behavioural variability more effectively than broader dimensional constructs.

Computational modelling revealed that differential reliance on goal-directed decision-making mediated the observed associations. Individuals higher in inattentive symptoms utilised goal-directed inference mechanisms more effectively. Those higher in externalising symptoms showed reduced deployment of flexible, goal-directed reasoning during threat inference tasks.

These patterns indicate that real-world symptom manifestations reflect alterations in the computational mechanisms underlying complex decision-making rather than simple performance deficits.

Implications

The findings suggest that mental health symptoms emerging in naturalistic environments may stem from domain-general computational dysfunctions rather than syndrome-specific pathology. Identifying which computational processes are altered across different symptom profiles could refine mechanistic models of psychopathology. This approach bridges symptomatology with formal computational mechanisms, potentially informing transdiagnostic intervention targets.

Understanding how goal-directed versus automatic inference processes differ across symptom dimensions may clarify why individuals with distinct psychiatric presentations exhibit similar real-world difficulties. Interventions targeting the flexible deployment of computational reasoning strategies could have broad applicability across diagnostic boundaries. Future research should validate these computational mechanisms in clinical populations and longitudinal designs.

Scope and limitations

This summary is based on the study abstract and available metadata. It does not include a full analysis of the complete paper, supplementary materials, or underlying datasets unless explicitly stated. Findings should be interpreted in the context of the original publication.