Task-state prefrontal cortex activation differences between executive dysfunction and perceptual dysfunction phenotypes of freezing of gait in Parkinson’s disease during the medication ON state: a retrospective fNIRS study
Neurological Sciences · 2026-12 · 47 (12)
DOI: 10.1007/s10072-026-09520-4 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neurological Sciences, online erschienen am 11.11.2026. Verfasst von Ling Zhang und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neurology, The First Affiliated Hospital of Dalian Medical University, Dalian, China. Hauptthemen laut MEDLINE: Parkinson Disease, Prefrontal Cortex, Gait Disorders, Neurologic, Executive Function. Die Arbeit nennt 33 Literaturangaben.
Abstract (Originalsprache)
BACKGROUND: The dual-model pathophysiological framework of freezing of gait (FOG) in Parkinson's disease proposes that, in some patients, freezing arises predominantly from impaired frontostriatal executive control (the executive dysfunction model), whereas in others it arises predominantly from impaired posterior-cortical visuospatial/perceptual processing (the perceptual dysfunction model). This framework therefore predicts opposite, task-specific patterns of impaired cortical activation in the two patient groups during executive and visuospatial tasks. OBJECTIVE: To compare prefrontal cortical (PFC) activation during seated cognitive tasks in the medication ON state between two proposed FOG phenotypes in Parkinson's disease (PD): PD-FOG-EDF, based on the executive dysfunction model, and PD-FOG-PDF, based on the perceptual dysfunction model, thereby testing these predictions. METHODS: This retrospective study included 125 participants: PD-FOG-EDF (n = 32), PD-FOG-PDF (n = 28), PD without FOG (PD-nFOG, n = 35), and healthy controls (n = 30). Participants underwent neuropsychological testing, wearable gait analysis, and functional near-infrared spectroscopy (fNIRS) during a verbal fluency task (VFT) and Benton Line Orientation Test (BLOT). Assessments were performed in the ON state 60-120 min after dopaminergic medication; freezing episodes were objectively recorded by synchronized video and inertial sensors. Analyses included covariate-adjusted models, Bonferroni and Benjamini-Hochberg false discovery rate correction, and internal validation by bootstrap, leave-one-out, and repeated 5-fold cross-validation. RESULTS: PD-nFOG showed a lower MoCA score than HC but a higher score than both FOG groups, mildly reduced VFT, TMT, and SCWT performance relative to HC, and BLOT performance comparable to HC, without selective impairment of any single cognitive domain. During the VFT, PD-FOG-EDF showed lower activation in bilateral DLPFC, VLPFC, mPFC, OFC, and SMA/PMC than the other groups (all P < 0.05). During the BLOT, PD-FOG-PDF showed the lowest VLPFC, OFC, and PPC activation (between-group main effect P < 0.001), with bilateral VLPFC activation and all four BLOT-related connections significantly lower than in PD-FOG-EDF, as well as task-specific hypoconnectivity; the direction of the between-phenotype differences reversed with task type, indicating a task-specific double dissociation pattern. Left DLPFC activation during the VFT correlated negatively with NFOGQ score (r=-0.52, P < 0.001). The combined model showed an apparent AUC of 0.945, an optimism-corrected AUC of 0.934, and a LOOCV AUC of 0.910; an fNIRS-only model retained discrimination after removal of classification-defining behavioral scores (corrected AUC = 0.920; permutation P = 0.002). Core group differences remained after full covariate adjustment and comparison with cognitively comparable PD-nFOG subgroups, whereas gait parameters and objective freezing-event counts did not differ between the two FOG phenotypes. CONCLUSION: The two proposed PD-FOG phenotypes showed distinct task-related PFC activation patterns in the medication ON state, and this task-specific double dissociation pattern was consistent with the predictions of the dual-model framework. Because fNIRS was recorded during seated cognitive tasks rather than freezing episodes, these findings should not be interpreted as direct neural mechanisms of FOG. They provide preliminary, indirect neurophysiological support for the dual-model framework but are insufficient to confirm it, and require prospective validation using walking paradigms, ON/OFF designs, and independent external samples.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Neurology, The First Affiliated Hospital of Dalian Medical University, Dalian, China.
- Department of Neurology, The First Affiliated Hospital of Dalian Medical University, Dalian, China. liangzhanhua@dmu.edu.cn.
Zitieren
APA
Zhang, L., Zhang, Z., Qu, Y., & Liang, Z. (2026). Task-state prefrontal cortex activation differences between executive dysfunction and perceptual dysfunction phenotypes of freezing of gait in Parkinson’s disease during the medication ON state: a retrospective fNIRS study. Neurological Sciences, 47(12). https://doi.org/10.1007/s10072-026-09520-4
Vancouver
Zhang L, Zhang Z, Qu Y, Liang Z. Task-state prefrontal cortex activation differences between executive dysfunction and perceptual dysfunction phenotypes of freezing of gait in Parkinson’s disease during the medication ON state: a retrospective fNIRS study. Neurological Sciences. 2026;47(12). doi:10.1007/s10072-026-09520-4
BibTeX
@article{Zhang2026,
author = {Zhang, Ling and Zhang, Zhen and Qu, Yan and Liang, Zhanhua},
title = {Task-state prefrontal cortex activation differences between executive dysfunction and perceptual dysfunction phenotypes of freezing of gait in Parkinson’s disease during the medication ON state: a retrospective fNIRS study},
journal = {Neurological Sciences},
year = {2026},
volume = {47},
number = {12},
doi = {10.1007/s10072-026-09520-4}
}
EUVIMED