Model-Based Electroencephalography Phenotyping Uncovers Distinct Neurocomputational Mechanisms Underlying Learning...

Area of Science:

• Neuroscience
• Computational Psychiatry
• Cognitive Psychology

Background:

• Major depressive disorder (MDD), bipolar disorder (BP), and schizophrenia (SCZ) present with learning impairments.
• The underlying neurocognitive mechanisms for these learning deficits remain poorly understood.
• Differentiating these mechanisms is crucial for developing targeted therapeutic interventions.

Purpose of the Study:

• To investigate the shared and distinct neurocognitive mechanisms of learning impairments in MDD, BP, and SCZ.
• To utilize computational modeling to quantify the interplay between working memory (WM) and reinforcement learning (RL) in these disorders.
• To identify specific neural signatures associated with learning deficits in each clinical group.

Main Methods:

• An associative learning task was administered to 255 participants, including patients with MDD, BP, SCZ, and healthy controls (CTRL).
• Computational modeling was employed to analyze the interaction between latent RL and WM processes.
• Electroencephalography (EEG) data were analyzed to capture dynamic neural signatures of RL and WM.

Main Results:

• All clinical groups exhibited behavioral learning impairments compared to controls.
• Distinct patterns in the interplay between RL and WM mechanisms were identified across clinical groups.
• Schizophrenia showed reduced WM recruitment and negative feedback integration; MDD displayed impaired WM management due to RL influence; BP exhibited deficits in both WM and RL recruitment with increased WM decay.

Conclusions:

• Integrative neurocomputational modeling successfully linked seemingly similar behavioral learning impairments to distinct neurocognitive mechanisms.
• The findings offer novel insights into the differential manifestation of learning deficits across psychopathologies.
• This approach highlights the importance of understanding the specific interplay of WM and RL in psychiatric disorders.