Adam J. Culbreth, PhD

• Culbreth, A.J., Kasanova, K., Ross, T.J., Salmeron, B.J., Gold, J.M., Stein, E.A., & Waltz, J.A. (2021). Schizophrenia Patients Show Largely Intact Salience Signaling Compared to Healthy Controls in an Observational Task Environment. Brain Sciences, 11(21), 1610.
• Wu, Q., Huang, H., Culbreth, A.J., Waltz, J.A., Hong, L.E., Chen, S. (2021). Extracting Brain Disease-Related Connectome Subgraphs by Adaptive Dense Subgraph Discovery. 1– 13. https://doi.org/10.1111/biom.13537
• Culbreth, A.J., Wu, Q., Chen, S., Adhikari, B.M., Hong, L.E., Gold, J.M., & Waltz, J.A. (2021). Temporal-Thalamic and Cingulo-Opercular Connectivity in People with Schizophrenia. Neuroimage: Clinical, 29, 102531.

1. Effort-cost decision-making as a marker of motivational impairment in schizophrenia.

While a large body of research has examined experience and learning of reward in schizophrenia, considerable less work has been conducted regarding experience and learning of effort. Using a variety of task paradigms, I have conducted a series of studies demonstrating that individuals with schizophrenia show a decreased willingness, compared to controls, to exert cognitive and physical effort for monetary rewards on experimental tasks. Critically, this deficit is most pronounced for people with the most severe motivational impairment. Further, I have conducted a preliminary study examining the neural correlates of this behavioral deficit showing that activation of the ventral striatum during effort-cost decision-making is associated with the severity of motivational symptoms, linking effort impairment to a specific neural target.

2. Relating experimental measures to emotionality and motivation in daily life using mobile technology.

Little work to date has attempted to establish the ecological validity of experimental paradigms by examining relationships between experimental task performance and measures of functioning collected in daily life. Using mobile assessment methods, I, along with my close collaborator Dr. Erin Moran, have collected several samples showing that reports of interest and enjoyment in daily activities correlate strongly with clinician-rated interviews indexing motivational impairment, providing strong validity for these methods. Further, we have shown that working memory capacity moderated this relationship suggesting that the increased working memory load required to successfully recall and evaluate activities in clinical interviews render such reports potentially less reliable. Experimentally, we demonstrated that self-reported interest and enjoyment with activities in daily life show robust relationships to experimental tasks assessing reward learning and effort-based decision-making, suggesting that laboratory tasks are associated with daily motivational experience. Further, we have integrated smartphone symptom assessment, behavioral experimental tasks, and neuroimaging to link potential neural mechanisms of motivational impairment (e.g., reduced ventral striatal activation during reward learning) to motivational experience in daily life.

3. Clarifying reward learning impairment in schizophrenia.

Many previous research studies have suggested that motivational impairment in schizophrenia might arise, in part, due to maladaptive reward learning. My work has attempted to clarify the particular mechanisms involved in this deficit, as well as their neural correlates. In an initial study, we demonstrated that reward-learning dysfunction in schizophrenia was related to alterations in a fronto-parietal network of brain regions, providing a specific neural target to a behavioral deficit. However, reward learning is reliant on many sub-processes (e.g., working memory, reward responsiveness) leaving the specific mechanisms for such deficits unclear. Thus, I conducted a second study using a computational framework to further specify the sub-processes involved in this deficit. This framework postulates that two systems drive reward learning. The first, model-free, learns more habitually/retrospectively by drawing on previous reward information, and is driven primarily by striatal regions. The second, model-based, learns in a more goal-directed/prospective manner, taking into consideration factors such as future consequences and the likelihood of obtaining certain outcomes. This system is thought to be driven by frontal circuits. Utilizing a task previously validated to assess these forms of learning, we found evidence for reductions in model-based learning, while model-free learning appeared intact, suggesting that reward-learning deficits in schizophrenia may be limited to more goal-directed/prospective processing streams.

2021-2026       Computational Mechanisms of Effort-Cost Decision-Making in Schizophrenia

                        Institute: National Institute of Mental Health  

                        Grant #: K23 MH126986

                        Role: Principle Investigator