Revisiting post-stimulus theta activity: evidence for an aperiodic rather than oscillatory origin

Area of Science:

• Neuroscience
• Cognitive Neuroscience
• Electrophysiology

Background:

• The aperiodic, 1/f-like component in electrophysiological activity is often overlooked. Conventional electroencephalography (EEG) studies often assume this component is stable, which can impact interpretations of oscillatory power changes.
• Transient changes in oscillatory power are frequently interpreted as neural signatures of cognitive processes like attention and control, but these interpretations may be flawed due to uncorrected aperiodic activity.

Purpose of the Study:

• To investigate the assumption that aperiodic activity remains stable before and after stimulus presentation in EEG data.
• To compare conventional spectral analyses with methods that explicitly model and remove the aperiodic component.
• To re-evaluate apparent changes in oscillatory power (theta, beta, gamma, alpha) after accounting for aperiodic activity.

Main Methods:

• Utilized high-density EEG recordings from typically developing children during audiovisual reaction-time and visual oddball tasks.
• Compared standard spectral analyses with analyses that modeled and removed the aperiodic component in pre- and post-stimulus windows.
• Examined changes in aperiodic exponent and offset, and their impact on oscillatory power estimates (theta, beta, gamma, alpha).

Main Results:

• Stimulus onset consistently altered the aperiodic component (exponent and offset) across tasks, contradicting the stability assumption.
• Apparent increases in theta power following stimulus onset were largely abolished after correcting for aperiodic activity.
• Conventional analyses overestimated beta desynchronization and misinterpreted gamma desynchronization as suppression, when it was an artifact of spectral slope changes.
• Alpha desynchronization remained robust and was even enhanced after aperiodic correction, suggesting a genuine oscillatory suppression.

Conclusions:

• A significant portion of commonly observed time-frequency effects in EEG, particularly theta synchronization, may stem from changes in aperiodic activity rather than true neural oscillations.
• The findings challenge fundamental assumptions in conventional time-frequency analyses of EEG data.
• Accounting for the dynamic aperiodic component is crucial for accurate interpretation of neural activity and cognitive processes.