Last webpage update: September 2026
Overview
My research programme in Computational Neuropsychology investigates the cognitive and computational mechanisms underlying individual differences in higher-order cognition and develops methods for their individualised neuropsychological characterisation. The programme integrates neuropsychological experimentation, computational modelling, machine learning, and neuro-inspired approaches, with a particular focus on executive functions, metacognition, inner speech, and consciousness across adult and developmental populations and typical, neurodivergent, neurological, and psychiatric conditions (Granato et al., 2020, 2022, 2025; Granato & Baldassarre, 2021, 2024).
This programme is articulated around a double aim: (1) advancing the theoretical and computational understanding of individual neuropsychological profiles and the mechanisms underlying them; and (2) translating this knowledge into automated and model-based neuropsychological methods and pipelines for individualised assessment and prediction (Granato et al., 2026a, 2026b).
At CNR-ISTC, I coordinate this research line within LENAI through experimental, computational, and translational projects conducted with researchers and students under my supervision and in collaboration with academic and healthcare partners.
Computational Neuropsychology of higher-order cognition
The programme comprises an experimental and computational research line investigating how multiple cognitive processes interact to support flexible, goal-directed behaviour. Rather than treating executive functions, metacognition, inner speech, and consciousness as isolated constructs, I study their interactions through neuropsychological tasks, behavioural and trial-by-trial analyses, and theory-based computational models (Granato et al., 2020, 2022, 2025; Granato & Baldassarre, 2021, 2024; Granato et al., 2026a).
This research spans adult and developmental populations and combines experimental and computational approaches to investigate both typical cognition and individual variability associated with neurodivergent, neurological, and psychiatric conditions (Granato & Baldassarre, 2021; Granato et al., 2022, 2025, 2026a).
Neuropsychological Digital Twins and computational modelling
A second major research line develops Neuropsychological Digital Twins (NDTs): participant-specific computational representations designed to connect observable neuropsychological performance with latent cognitive and metacognitive mechanisms.
I developed the Neuropsychological Digital-Twin Method, a model-based methodology in which theory-driven and neuro-inspired computational models are instantiated, quantitatively fitted to individual neuropsychological data, and used to derive interpretable model-based profiles and predictions (Granato et al., 2026a).
This work is currently being extended through the Multi-Layer Neuropsychological Digital-Twin (ML-NDT) framework, which integrates four complementary levels of participant-specific information: aggregated neuropsychological measures, trial-by-trial cognitive and metacognitive dynamics, temporal information, and computational model-derived traits and mechanisms. The framework is operationalised through modular automated pipelines supporting neuropsychological assessment, modelling, profiling, and human-readable reporting (Granato et al., 2026b).
Translational and interdisciplinary research
My current research applies these approaches to experimental and translational studies involving adult and developmental populations in collaboration with academic and healthcare institutions. A major objective is to investigate whether interpretable individual cognitive profiles can complement traditional group- and diagnosis-based approaches while supporting increasingly personalised neuropsychological research.
The programme also explores how principles and methods from Computational Neuropsychology can contribute to cognitive robotics and human-robot interaction, particularly through computational accounts of flexible cognition, internal representations, and inner speech (Granato et al., 2020; Granato & Baldassarre, 2021, 2024). These approaches are also being investigated in other individual-profiling contexts.
Theoretical and computational contributions
The current research programme builds upon a series of theoretical and computational contributions developed throughout my earlier work. These include the Three-Component Theory of flexible cognition, which conceptualises flexible goal-directed cognition through interacting executive, representational, and motivational components (Granato et al., 2020; Granato & Baldassarre, 2021); the Goal-Aligning Representations Internal Manipulation (GARIM) theory, addressing the role of inner speech and internal representations in flexible cognition and consciousness (Granato & Baldassarre, 2024); and computational work on Motivated Categorical Perception, which proposes that motivational processes can shape learned categorical representations and was developed through computational modelling of motivated categorisation (Granato, 2022; Granato et al., 2022).
These contributions provide theoretical and computational foundations for the broader investigation of individual cognitive mechanisms pursued within my current Computational Neuropsychology research.
References
- Granato, G., Borghi, A. M., & Baldassarre, G. (2020). A computational model of language functions in flexible goal-directed behaviour. Scientific Reports, 10.
- Grranato, G., & Baldassarre, G. (2021). Internal manipulation of perceptual representations in human flexible cognition: A computational model. Neural Networks, 143, 572–594.
- Grranato, G. (2022). Flexible goal-directed manipulation of representations: computational models of healthy and pathological human cognition. Doctoral dissertation, University of Plymouth.
- Grranato, G., Cartoni, E., Da Rold, F., Mattera, A., & Baldassarre, G. (2022). Integrating unsupervised and reinforcement learning in human categorical perception: A computational model. PLOS ONE, 17(5), e0267838.
- Grranato, G., Borghi, A. M., Mattera, A., & Baldassarre, G. (2022). A computational model of inner speech supporting flexible goal-directed behaviour in autism. Scientific Reports, 12, 14198.
- Grranato, G., & Baldassarre, G. (2024). Bridging flexible goal-directed cognition and consciousness: The Goal-Aligning Representation Internal Manipulation theory. Neural Networks, 106292.
- Grranato, G., Costanzo, R., Borghi, A. M., Mattera, A., Carruthers, S., Rossell, S., & Baldassarre, G. (2025). An experimental and computational investigation of executive functions and inner speech in schizophrenia spectrum disorders. Scientific Reports, 15, 5185.
- Grranato, G., Mattera, A., Cartoni, E., & Baldassarre, G. (2026a). Modeling metacognition and executive functions in the Metacognitive Wisconsin Card Sorting Test using the neuropsychological digital-twin method. Scientific Reports, 16, 7145.
- Granato, G., Di Giulio, J., Manzi, G., Puzzo, C., Mattera, A., Giocondo, F., Adriani, W., Capobianco, M., & Baldassarre, G. (2026b). An automated pipeline for adult and developmental neuropsychological assessment of executive functions and metacognition based on a multi-layer neuropsychological digital-twin framework. Manuscript under review.
