The human brain's ability to perceive and process time is a complex phenomenon involving intricate interactions between various brain regions. This research focuses on understanding the neural mechanisms underlying time perception, particularly how the brain handles predictable versus unpredictable events. Traditional views of time perception have been challenged by cognitive neuroscientists, who suggest that the brain constantly generates and updates expectations for future events to effectively interact with a dynamic environment. The capacity to predict event timing is critical for appropriate responses. The study uses electroencephalography (EEG) to investigate functional connectivity between brain regions during time perception tasks, offering high temporal resolution to capture rapid brain activity changes. EEG signals provide an approximate measurement of postsynaptic activity of neuronal cells with a temporal resolution on the order of milliseconds, enabling the description of brain activity dynamics. Various methods exist to analyze functional connectivity from EEG, and this study utilizes the Phase Lag Index (PLI), chosen for its robustness to volume conduction artifacts. The specific research question explores how brain connectivity patterns differ when participants anticipate stimuli at fixed versus variable time intervals.

Previous research has highlighted the involvement of various brain regions in time perception, including the cerebellum, thalamus, posterior parietal cortex, prefrontal cortex, and motor cortex. Studies on monkeys have revealed cellular activity changes related to temporal processing in these regions. Neurological disorders impacting the cerebral cortex, such as Parkinson's disease, often lead to timing impairments, further emphasizing the cortex's role in time perception. Existing research also suggests that prefrontal cortical regions play a crucial role in controlling attention and memory, both of which are closely linked to timing processes. A prior study by the authors investigated the effect of stimulus timing prediction on sensory labeling accuracy, observing predictability effects on alpha power at posterior parietal sites during the longest delay (800 milliseconds). This study builds on this foundation by employing functional connectivity analysis to provide a more comprehensive understanding of the neural networks involved in time perception under different predictability conditions.

This study analyzed EEG data from a previous experiment involving 29 healthy male participants. EEG signals were recorded at 1000 Hz with 64 electrodes, preprocessed using EEGLAB toolbox in MATLAB. Preprocessing steps included removing DC offset, applying a high-pass filter (0.5 Hz), replacing corrupted channels using interpolation, and employing Independent Component Analysis (ICA) to remove artifacts like eye blinks and movement. A subset of 19 standard 10-20 electrodes was selected for further analysis. The experiment presented visual stimuli (clockwise or counterclockwise rotation) after variable delays (83, 150, 400, and 800 milliseconds) following a visual cue. Two conditions were tested: predictable (fixed delay for each block) and unpredictable (randomly selected delay from the four options). Functional connectivity was estimated using the Phase Lag Index (PLI) across five frequency bands (delta, theta, alpha, beta, and gamma). Statistical analysis using SPSS24, including paired t-tests and ANOVA, was conducted to compare clockwise and counterclockwise stimuli, compare delays within each condition, and compare predictable and unpredictable conditions for each delay. The False Discovery Rate (FDR) test was used to control for multiple comparisons.

No significant difference was found between clockwise and counterclockwise stimuli. ANOVA revealed significant differences between delays within both predictable and unpredictable conditions, most pronounced in the beta, theta, and gamma bands. In the delta band, the predictable condition showed increased differences between delays of 83 ms, 150 ms, and 400 ms compared to 800 ms, mostly in frontal regions; while in the unpredictable condition, the largest difference was between 150 ms and 800 ms, in posterior regions. The theta band showed the highest significant differences between 83 ms and 800 ms, and 150 ms and 800 ms in both conditions. Alpha band showed minimal differences. The beta band had greatest differences between 83 ms and 800 ms and 150 ms and 800 ms delays. Gamma band showed the greatest differences between 83 and 400 ms delays in the predictable and 150 and 400 ms delays in the unpredictable condition. Comparing predictable and unpredictable conditions for each delay revealed significant differences in all delays across all frequency bands. Figures 3-10 illustrate the network connections, showing that connections generally decrease with increasing delays in both conditions, with stronger connections between hemispheres. In the alpha band, increased connectivity was observed between posterior and anterior regions in the unpredictable condition, particularly at the 400ms delay, while the predictable condition showed connectivity even before the stimulus, suggesting anticipatory learning in the parietal region. The delta band showed central-frontal connectivity in predictable conditions, while unpredictable conditions had more connections, decreasing at 800 ms. Theta band had connections in frontal, prefrontal, and parietal regions, decreasing at 800 ms. Beta band showed prefrontal connections across delays, with more connections in the unpredictable condition. Gamma band showed connections throughout the brain in unpredictable conditions and fewer connections in predictable conditions.

The findings demonstrate that brain connectivity patterns related to time perception vary significantly depending on the predictability of events. The more pronounced differences in beta, theta, and gamma bands during predictable conditions support the hypothesis that anticipatory mechanisms are highly active when temporal information is known in advance. The increased alpha connectivity in the unpredictable condition aligns with previous research suggesting its role in attentional processing during uncertainty. The involvement of the prefrontal cortex confirms its critical role in attention and working memory for temporal processing. The decline in connectivity at the longest delay in both conditions suggests a potential limit to the brain's ability to maintain precise temporal information over extended durations. These findings provide crucial insights into the neural basis of time perception and highlight the complex interplay between predictability, attentional processes, and the allocation of cognitive resources. The study's observations of distinct connectivity patterns at various delays suggest the involvement of multiple neural circuits adapted to different temporal windows.

This study provides evidence of distinct neural network dynamics underlying time perception in predictable and unpredictable conditions. Significant differences in functional connectivity across multiple frequency bands were observed, particularly in the gamma, beta, and theta bands. The right prefrontal cortex played a prominent role. Future research could explore the effects of different stimulus modalities or investigate individual differences in time perception. Further studies could explore the interaction of specific brain networks implicated in these processes using methods like Granger causality analysis.

The study used a relatively small sample size (29 participants), limiting the generalizability of the findings. The use of only male participants may limit the results' applicability to females. The focus on visual stimuli may not fully capture the complexities of time perception across all sensory modalities. Moreover, future studies may need to consider a wider range of delays and include more trials per delay group, taking into account participant fatigue as a potential confounder.