Decoding and Modulating the Mechanisms of Human Reasoning to Improve Clinical Outcomes
When beginning your next investigator-initiated application, consider the following NIH highlighted topic. The area of science described below is of interest to the listed NIH Institutes, Centers, and Offices (ICOs). This is not a notice of funding opportunity (NOFO).
Apply through an appropriate NIH Parent Funding Announcement or another broad NIH opportunity available on Grants.gov. Learn how to interpret and use Highlighted Topics.
Topic Description
Post Date: November 6, 2026
Expiration Date: November 6, 2028
Background
Reasoning is the mental ability to dynamically combine or transform information (e.g., quantity, number, spatial locations, social hierarchies) to derive new knowledge internally. The logical operations of reasoning subconstructs (e.g., transitive inference, abstraction/rule learning, generalization/discrimination, hierarchical planning) may be generated via distinct neural algorithms. Reasoning is critical for healthy functioning and well-being from early life through aging and is disrupted in many chronic brain disorders (e.g., neurodevelopmental disorders, Alzheimer’s disease and other dementias, thought/psychotic disorders). Deficits in reasoning can affect risk appraisal, health literacy, adherence, and other health behaviors. This negatively impacts prevention, diagnosis, treatment, and care coordination across diseases and disorders.
Purpose
This topic encourages research to decode and modulate the neurobehavioral mechanisms of reasoning in humans, with the goal of translating basic science findings into clinical interventions across the lifespan. Objectives are guided by the 2025 NIH workshop, “Reasoning Algorithms Across Species, Diagnoses, and Development: Theoretical Frameworks Informing Causal Manipulations.”
Multidisciplinary teams could address one or more objectives:
- Develop and validate computational models of human reasoning
- Develop neurocognitively plausible models based on data from animal studies, novel alternative methods (NAMs), or existing human (in vitro, in vivo, or large-scale) datasets
- Validate models using newly collected human data that include both behavioral and neural measures, and that incorporate best practices (e.g., standardized paradigms, preregistered protocols, adequate statistical power, open materials, and data use safeguards)
- Test predictions from computational models in humans, animals, or NAMs
- Identify causal mechanisms in well-validated research models, from NAMs to in vivo animals or humans
- Use perturbations or interventions in vitro or in vivo, guided by computational models, to identify specific neurocomputational mechanisms of reasoning subconstructs and their development across the lifespan
- Adopt iterative feedback loops where computational models inform in vivo studies and vice versa, with justified rationale
- Causally test proposed neural mechanisms using neuromodulatory perturbations in healthy participants, relevant patient populations, or both
- Link biomarkers (behavioral or biological) to clinical actionability. For advanced translational projects with appropriate resources (e.g., sufficient power) for biomarker development:
- Deliver transparent, computationally derived, clinically actionable biomarkers for defined reasoning subconstructs
- Clearly define the target patient populations for each model with pre-specified biomarker Context of Use following FDA-NIH BEST terminology
- Provide evidence supporting clinical decision thresholds and rules to guide interventions
Participating ICOs
- NCI is interested in research examining how reasoning subconstructs influence decisions and behaviors across the cancer control continuum including, but not limited to, behavior change, HPV vaccination, genetic testing, screening uptake, diagnostic follow-up, treatment choices and adherence, symptom self-management, and care coordination.
- Priority areas include computational models of cancer-related reasoning; neural and behavioral studies of how people understand risk, uncertainty, screening recommendations, diagnostic results, treatment tradeoffs, and survivorship recommendations; and causal tests of mechanisms disrupted by aging, stress, fatigue, cancer-related cognitive impairment, mental health disorders, substance use disorders, and other co-morbidities.
- Translational work may develop clinically actionable biomarkers and interventions, such as decision support tools, cognitive training, or neuromodulation techniques, to improve informed decision-making and health outcomes.
Kara L. Hall, PhD
[email protected]
Luke Stoeckel, Ph.D.
[email protected]
Todd Horowitz, Ph.D.
[email protected]
Matt Sutterer, Ph.D.
[email protected]
Jessica Mollick
[email protected]
NIDCR is interested in dental, oral, and craniofacial projects that test models reflective of cognitive frameworks, potentially with concordant nervous system correlates, that support cognitive/behavioral interventions/assessments within the context of health promotion or improved systemic health.
NIDCR Division of Extramural Research
[email protected]
NIMH seeks transdiagnostic, cross-species, developmentally informed research on:
- Bidirectional coupling between dissociable reasoning sub-constructs (e.g., latent cause inference, hierarchical planning, belief updating) and other functional domains (e.g., cognitive, socio-affective), by mapping shared neurobehavioral mechanisms of reasoning.
- Development of precision stratification methods for clinical trials targeting reasoning, using causal neuromodulatory perturbations in psychiatric samples to map sub-constructs to actionable, circuit-level mechanisms.
- Formalizing reasoning interventions such as Bayesian belief revision on estimated latent cause parameters to predict treatment response and identify augmentation needs, closing gaps left by biomarkers alone.
- Creating neurocognitive AI emulations that instantiate circuit-level inference (vs. input/output-trained large language models) to expose, characterize, and correct the algorithmic basis of reasoning failures in psychiatric patients.
Mauricio Rangel-Gomez, PhD
[email protected]
Michele Ferrante, PhD
[email protected]
Kristin Brethel-Haurwitz
[email protected]
For technical issues E-mail OER Webmaster