<p>Parkinson’s Disease (PD) is associated with impaired temporal cognition and disrupted cortical dynamics. This study investigates EEG complexity during short (3-second) and long (7-second) interval timing tasks using a diverse set of entropy features, including Shannon, Sample, Permutation, Dispersion, Conditional, Fuzzy, Bubble, Wavelet, and Wavelet-based synchrosqueezing transform entropy (WsstEn). Task-related EEG from 74 PD patients and 37 controls were analyzed across 59 channels, focusing on differences that emerged during the RESPONSE phase but not at baseline. WsstEn consistently captured significant group differences, particularly in motor and midline regions, with strong effect sizes. Shannon entropy and Conditional entropy exhibited significant alterations, primarily in parietal and posterior regions, such as CP1, P1, CPz, and Pz, especially during the long-duration task. Notably, frequency-based entropy measures, such as WsstEn and Wavelet entropy, showed decreased values in PD patients, whereas amplitude-based entropies, including Shannon and Conditional entropy, exhibited increased values, suggesting differential alterations in signal complexity dimensions. Long-duration tasks induced broader effects across attention and timing networks, while short tasks highlighted motor preparatory areas. Exploratory classification using subject-level entropy summaries showed moderate performance, reinforcing the statistical findings and indicating the complementary role of these features. The Wavelet-based synchrosqueezing transform’s improved time-frequency localization may explain its superiority. These findings propose entropy-based EEG features as physiologically grounded biomarkers for task-evoked dysfunction in PD.</p>