eLife Assessment: Cross-modal interaction of Alpha Activity does not reflect inhibition of early sensory processing: A frequency tagging study using EEG and MEG

Iris I. A. GroenOpenalex Unknown 2025 · DOI · OpenAlex

Abstract

Selective attention is fundamental to human cognition, allowing us to prioritize relevant sensory inputs while filtering out distractions. A dominant theory in neuroscience, the "alpha inhibition hypothesis," posits that oscillatory alpha-band activity (~10 Hz) serves as a functional mechanism to inhibit task-irrelevant sensory regions. In this study, we challenged this notion using a cross-modal attention paradigm combined with frequency-tagging in both EEG and MEG. While we observed the expected increase in posterior alpha activity during auditory attention, this increase did not suppress early visual processing. Instead, both visual and auditory frequency-tagged responses were amplified, correlating positively with alpha amplitude. These findings suggest that alpha oscillations do not act as a simple "kill switch" for sensory input but rather orchestrate the transmission of prioritized information to downstream processing stages.

Introduction: The 'Brakes' of the Brain

For decades, alpha-band oscillations have been characterized as the "brakes" of the brain. According to the alpha inhibition hypothesis, when an individual focuses on one sensory modality (e.g., listening to a conversation), the brain actively suppresses other modalities (e.g., vision) by increasing alpha wave power in the corresponding sensory cortices. This "sensory gain control" is thought to prevent distracting information from interfering with the primary task. However, the direct link between increased alpha power and reduced sensory responsiveness has remained controversial. This study aimed to determine whether alpha oscillations truly inhibit early sensory processing or if they serve a more nuanced role in neural communication.

Methods: Frequency Tagging and Cross-Modal Attention

To test the inhibition hypothesis, researchers employed a sophisticated "frequency tagging" technique. Participants were presented with a fixation cross flickering at 36 Hz (visual tag) and amplitude-modulated white noise at 40 Hz (auditory tag). By monitoring the brain's rhythmic response at these specific frequencies, the team could measure the real-time responsiveness of the visual and auditory systems.

The experimental task required participants to attend to either a visual target (rotating Gabor patches) or an auditory target (pitch discrimination) based on a preceding cue. The study was first conducted using EEG (n=24) to identify core effects and subsequently replicated using MEG (n=28) to provide high-spatial-resolution source localization.

Figure 1: Theoretical framework and experimental paradigm. Panel A contrasts the traditional alpha inhibition theory with the authors' revised model. Panel B details the experimental timeline for both EEG and MEG studies.

Results: Challenging the Inhibition Model

The results provided a striking contradiction to traditional theories. As expected, when participants prepared for an auditory target, alpha power in the visual cortex increased. However, rather than being suppressed, the visual frequency-tagged response (36 Hz) actually increased in amplitude.

Key findings include:

  • Positive Correlation: On a trial-by-trial basis, higher alpha power was positively correlated with stronger frequency-tagged responses. If alpha were inhibitory, this correlation should have been negative.
  • Source Localization: MEG data localized the cue-induced alpha increase to the early visual areas, specifically the right lingual gyrus. Conversely, the enhanced 36 Hz visual response was localized to later stages of the visual stream, including the precuneus and medial occipital cortex.
  • Cross-Modal Integration: The researchers detected a 4 Hz intermodulation response (the difference between the 40 Hz auditory and 36 Hz visual tags). This signature suggests that the brain was actively integrating or comparing signals across modalities rather than shutting one down.

Discussion: A Revised Role for Alpha Oscillations

The authors propose a "revision" to the alpha inhibition theory. Instead of blocking early sensory input (like a roadblock at V1), alpha oscillations likely regulate downstream transmission. The study suggests that alpha might "package" sensory information into prioritized bundles for higher-level processing.

Furthermore, the simultaneous boost in both visual and auditory responsiveness during the difficult auditory task suggests a "vigilance effect." When a task is demanding, the brain may enter a state of heightened general sensitivity to ensure no critical information is missed, even in the non-attended modality.

Limitations and Future Directions

While the findings are robust across EEG and MEG, some caveats remain. The auditory task was notably more difficult than the visual task, which may have influenced the observed vigilance effects. Future research should investigate whether these non-inhibitory alpha effects persist when shifting attention between two tasks of equal difficulty or within the same sensory modality. Additionally, exploring how alpha "pulses" might temporally order signal transmission could provide deeper insights into how the brain manages information overflow.

Summary updated

Abstract

Selective attention involves prioritizing relevant sensory input while suppressing irrelevant stimuli. It has been proposed that oscillatory alpha-band activity (∼10 Hz) aids this process by functionally inhibiting early sensory regions. However, recent studies have challenged this notion. Our EEG and MEG studies aimed to investigate whether alpha oscillations serve as a ‘gatekeeper’ for downstream signal transmission. We first observed these effects in an EEG study and then replicated them using MEG, which allowed us to localize the sources.We employed a cross-modal paradigm where visual cues indicated whether upcoming targets required visual or auditory discrimination. To assess inhibition, we utilized frequency-tagging, simultaneously flickering the fixation cross at 36 Hz and playing amplitude-modulated white noise at 40 Hz during the cue-to-target interval.Consistent with prior research, we observed an increase in posterior alpha activity following cues signalling auditory targets. However, remarkably, both visual and auditory frequency tagged responses amplified in anticipation of auditory targets, correlating with alpha activity amplitude. Our findings suggest that when attention shifts to auditory processing, the visual stream remains responsive and is not hindered by occipital alpha activity. This implies that alpha modulation does not solely regulate ‘gain control’ in early sensory areas but rather orchestrates signal transmission to later stages of the processing stream.
BibTeX
@inproceedings{groen2025elife,
  author = {Iris I. A. Groen and Iris Groen},
  title = {{eLife Assessment: Cross-modal interaction of Alpha Activity does not reflect inhibition of early sensory processing: A frequency tagging study using EEG and MEG}},
  booktitle = {Openalex Unknown},
  year = {2025},
  doi = {10.7554/elife.106050.1.sa4},
}
Figure