Effort-based decision-making in familial high-risk individuals for bipolar disorder and psychosis: a computational modeling study

Background:

Effort-based decision-making is impaired in schizophrenia spectrum disorders and bipolar disorder [1,2]. However, it is unclear whether the willingness to expend effort for rewards is altered in individuals with familial vulnerability. Familial high-risk (FHR) samples may help investigate vulnerability-related mechanisms independent of illness-related confounders. This study evaluated effort-based decision-making in individuals at familial high risk for bipolar disorder (FHR-BD) and psychosis (FHR-P) by examining how reward magnitude and reward probability influence hard-task choices and latent decision-making processes estimated with hierarchical drift diffusion modeling.

Methods:

Effort-based decision-making was compared in individuals at FHR-BD (n=70), FHR-P (n=50), and healthy controls (HC) (n=41). Participants in the FHR groups had either a first-degree relative diagnosed with bipolar disorder or schizophrenia, respectively, or more than one affected second-degree relative. None of the familial high-risk participants fulfilled criteria for an ultra-high-risk syndrome, suggesting that the sample represented familial/genetic vulnerability rather than a clinically prodromal state.

Effort-based decision-making was assessed using the Effort-Expenditure for Rewards Task, which measures choices under varying reward-magnitude and probability conditions. Behavioral data were analyzed using a mixed-model repeated-measures ANOVA with group as a between-subjects factor and probability and reward level as within-subjects factors. Hierarchical Bayesian drift diffusion modeling was used to examine latent decision-making parameters underlying choices and reaction times. The main computational parameters included evidence accumulation toward effortful choices and its modulation by reward probability and reward magnitude.

Results:

Repeated measures ANOVA showed no significant group differences in the likelihood of choosing the hard task. No significant group-by-reward or group-by-probability interaction was observed, suggesting that increasing reward magnitude and reward probability did not differentially affect hard-task choices across groups.

However, hierarchical drift diffusion modeling revealed latent differences. Compared with HC, both the FHR-BD and FHR-P groups showed reduced sensitivity to drift rate for reward probability and reward magnitude, suggesting weaker integration of reward-related information during evidence accumulation toward effortful choices. The two FHR groups showed largely similar posterior profiles and did not clearly differ from each other.

Discussion:

This study did not find overt behavioral differences in willingness to expend effort for rewards in FHR individuals, possibly because they were clinically unaffected and did not meet ultra-high-risk criteria. However, computational modeling suggested alterations in the decision processes underlying effort-based choices. This may indicate that conventional behavioral analyses are primarily output-oriented, whereas hierarchical drift-diffusion modeling provides a process-oriented account of reward and probability integration in decision-making [3]. Alterations in motivational information processing may already be detectable at the level of genetic vulnerability for bipolar disorder and psychosis, despite preserved overt behavioral performance. The overlapping computational profiles of FHR-BD and FHR-P may indicate shared transdiagnostic vulnerability mechanisms across bipolar and psychosis spectra. This cross-sectional study suggests that computational approaches may detect latent alterations in decision-making that conventional behavioral measures do not capture. Longitudinal studies are needed to determine the clinical and functional significance of these alterations.

Funding:

This study was supported by The Scientific and Technological Research Council of Türkiye, TÜBİTAK, grant number 1192112.

References

1-Barch, D. M., Culbreth, A. J., Zeev, D. Ben, Campbell, A., & Moran, E. K. (2023). Dissociation of Cognitive Effort – Based Decision Making and Its Associations With Symptoms , Cognition , and Everyday Life Function Across Schizophrenia , Bipolar Disorder , and Depression. Biological Psychiatry, 94(6), 501–510. https://doi.org/10.1016/j.biopsych.2023.04.007 2-Blouzard, E., Pouchon, A., Polosan, M., Bastin, J., & Dondé, C. (2023). Effort-Cost Decision-making Among Individuals With Schizophrenia: A Systematic Review and Meta-analysis. JAMA Psychiatry, 80(6), 548–557. https://doi.org/10.1001/jamapsychiatry.2023.0553 3-Kuhn, M., Palermo, E. H., Pagnier, G., Blank, J. M., Steinberger, D. C., Long, Y., Nowicki, G., Cooper, J. A., Treadway, M. T., Frank, M. J., & Pizzagalli, D. A. (2025). Computational Phenotyping of Effort-Based Decision Making in Unmedicated Adults With Remitted Depression. Biological psychiatry. Cognitive neuroscience and neuroimaging, 10(6), 607–615. https://doi.org/10.1016/j.bpsc.2025.02.006