A task-aligned multimodal machine learning framework for studying working memory dysfunction in Parkinson's disease.

Overview

1. Biomedical Engineering Department, University of North Dakota, Grand Forks, ND 58202 USA
2. Department of Psychology, North Dakota State University, Fargo, ND 58102 USA
3. Department of Neurology, Sanford Health, Fargo, ND 58103 USA
4. Department of Neurology, University of North Dakota School of Medicine and Health Science, Grand Forks, ND 58202 USA
5. Department of Psychology, Minnesota State Community and Technical College, Moorhead, MN 56560 USA

Abstract

Cognitive impairment is one of the most functionally debilitating non-motor symptoms in Parkinson's disease (PD). Yet, current diagnostic and clinical practices rely heavily on subjective assessments and burdensome behavioral testing, which lack the temporal resolution to detect subtle deficits. While EEG and pupillometry offer promising non-invasive insights into cognitive processing, more scalable and objective methods are needed to detect subtle and often elusive dysfunction in early PD. Machine learning (ML) leverages patterns in neurophysiological signals recorded during cognitive tasks to identify subtle impairments that may not be evident in standard evaluations. We recorded EEG and pupillometry data from 68 participants (35 PD, 33 healthy controls (HC)) during a visual Change Detection working memory task. Using our custom, standardized feature-extraction toolbox, we extracted 108 features and incorporated the resulting feature matrices into an ML pipeline. We applied SMOTE to balance the classes, then used principal component analysis (PCA) to reduce dimensionality. An automated elbow method identified optimal cutoffs for principal components (PCs) and original features, which guided subsequent recursive elimination (RE) for computational efficiency. We trained a support vector machine with a radial basis function kernel (SVM-RBF) to classify PD and evaluated the final model's performance on a hold-out dataset. The automated elbow method identified optimal cutoffs at 20 PCs and 22 original features. The RE revealed that a model using 14 PCs and the top 7 PCA-weighted features achieved the highest performance, with 71% accuracy and an F1 score of 0.701. The classifier's performance declined slightly when tested on held-out data, with an accuracy of 63% and an F1 score of 0.626. This study demonstrates the potential of a task-aligned ML pipeline for uncovering early cognitive markers of PD. The framework supports post-hoc feature contribution mapping, linking retained physiological features and task stages back to classification-relevant patterns, supporting mechanistic investigation of cognitive dysfunction in PD.

Supplementary Information: The online version contains supplementary material available at 10.1007/s11571-026-10526-z.

Reproduced under the paper's license (CC BY), from the paper cited above.

Code

The paper says that its authors' code is available on request: it was not published with the paper, so there is nothing to verify.

The paper's code and data availability statement is in the Data section.

Tracing map

A tracing map links a paper to the code its authors published: this paper has none (its code is available on request), so it has no map.

Data

No dataset and no data link were found in the paper.

Data availability

De-identified feature-level data, behavioral metrics, and analysis scripts are available from the corresponding author upon reasonable request. Raw EEG and pupillometry recordings are available from co-authors Dr. Jeffrey S. Johnson or Dr. Samuel A. Birkholz upon reasonable request, subject to institutional and ethical regulations governing human subject research.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 3, 28 September 2026

• Funding: added North Dakota State University

Version 1, 27 September 2026: the first record

Recorded: type, language, journal, volume, issue, pages, dates, 7 authors, 6 keywords, 36 references.

Cite

This paper

Terry, M., Birkholz, S. A., Johnson, J. S., Lou, J.-S., Heuther, A. X., Keller, J., & Vazquez, E. A. (2026). A task-aligned multimodal machine learning framework for studying working memory dysfunction in Parkinson's disease. Cognitive neurodynamics, 20(1), 158. https://doi.org/10.1007/s11571-026-10526-z

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.3390/brainsci16080781 [code]

A Comparison of Machine Learning Models for Classification of Parkinson's Disease During a Working Memory and Sustained Attention Task.

Journal: Brain sciences

In common: Parkinson's, EEG, cognitive, 9 references

[2] doi:10.1002/brb3.71563

A Hybrid Random Forest-SARSA Framework for Resting-State EEG-Based Parkinson's Disease Detection With Temporal Decision Refinement.

Journal: Brain and behavior

In common: Parkinson's, EEG, 4 references

[3] doi:10.1093/braincomms/fcag329

A meta-analysis of periodic and aperiodic electrophysiological features in Parkinson's disease.

Journal: Brain communications

In common: Parkinson's, EEG, 2 references

[4] doi:10.1186/s40708-026-00317-x

Parkinson's disease classification using optimized attention-based deep learning from EEG signals with interpretable sub-band topography.

Journal: Brain informatics

In common: Parkinson's, EEG, 2 references

[5] doi:10.1038/s41531-026-01360-5 [code]

Hippocampal atrophy in untreated de novo Parkinson's disease with obstructive sleep apnea.

Journal: NPJ Parkinson's disease

In common: Parkinson's, 2 references

[6] doi:10.1093/braincomms/fcag328 [code]

Subthalamic stimulation modulates working memory-related cortical dynamics in Parkinson's disease.

Journal: Brain communications

In common: Parkinson's, EEG, cognitive, 1 reference

[7] doi:10.1038/s42003-026-10108-z [code]

Time-resolved EEG decoding reveals altered neural dynamics of affective semantic evaluation in depression and suicidality.

Journal: Communications biology

In common: EEG, cognitive, 1 reference

[8] doi:10.7717/peerj-cs.3860

Systematic review of unveiling the potential of AI using machine learning and deep learning methods in neurodegenerative diseases.

Journal: PeerJ. Computer science

In common: Parkinson's, 1 reference

[9] doi:10.1002/mds.70355 [code]

Gray Matter Microstructure Measured Using Diffusion Imaging as a Biomarker of Severity in Lewy Body Diseases.

Journal: Movement disorders : official journal of the Movement Disorder Society

In common: Parkinson's, 1 reference

[10] doi:10.1002/mds.70265 [code]

Differential Progression of Neuroinflammation in Patients with Isolated Rapid-Eye-Movement Sleep Behavior Disorder.

Journal: Movement disorders : official journal of the Movement Disorder Society

In common: Parkinson's, 1 reference

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.