Differential Contributions of High-Frequency Activity and C1 ERP Component to Visual Processing: A Combined EEG-MEG Study

Academic Background

In the field of neuroscience, understanding how visual information is processed in the brain is a central question. Visual information processing is typically divided into two main processes: feedforward and feedback. The feedforward process refers to the transmission of information from the retina to the primary visual cortex (V1), while the feedback process involves the return of information from higher visual regions back to V1. Although both processes play crucial roles in visual processing, their specific mechanisms and temporal dynamics remain incompletely understood.

High-Frequency Activity (HFA), which refers to neural oscillations in the range of 80-150 Hz, is generally considered a marker of local neuronal population activity. HFA in human intracranial recordings shows differential modulation in response to behaviorally relevant stimuli, suggesting its involvement in feedforward processes. However, HFA also exhibits a significant dependence on superficial cortical layers and peaks within 200 milliseconds after stimulus onset, implying that it may be more related to feedback signals. Therefore, the specific role of HFA in feedforward and feedback processes remains controversial.

To clarify this issue, this study compared the response characteristics of HFA and the C1 event-related potential (ERP) component by simultaneously recording electroencephalography (EEG) and magnetoencephalography (MEG). The C1 component, the earliest visual ERP component in EEG, is considered a marker of feedforward input to V1. By comparing the temporal dynamics and modulation properties of C1 and HFA, this study aimed to reveal whether HFA primarily represents feedforward information or also includes feedback signals.

Source of the Paper

The authors of this study include Paul Schmid, Christoph Reichert, Robert T. Knight, and Stefan Dürschmid. They are affiliated with the Leibniz Institute for Neurobiology in Germany and the University of California Berkeley in the United States. The study was first published on November 26, 2024, in the Journal of Neurophysiology, with the DOI 10.1152/jn.00292.2024.

Research Process

1. Participants and Experimental Design

The study recruited 25 healthy participants (12 females, aged 20-33 years). All participants reported normal or corrected-to-normal vision and no history of neurological or psychiatric disorders. The experiment was conducted at the Department of Neurology at the University of Magdeburg, Germany, and was approved by the local ethics committee. Ultimately, 19 participants completed the experiment, while the remaining six were excluded due to eye movements or motion artifacts.

The experiment employed a visual detection paradigm in which participants were required to respond as quickly as possible to a black arrow pointing either left or right on the screen. The arrow was accompanied by task-irrelevant bilateral high-contrast black-and-white checkerboard stimuli, presented in either the upper visual field (UVF) or lower visual field (LVF) to elicit the C1 response. The contrast of the checkerboard stimuli varied between 60% and 90%, divided into four blocks, with each block containing 272 trials, totaling 1088 trials.

2. Data Acquisition and Preprocessing

EEG and MEG data were recorded simultaneously. EEG was recorded using 30 passive electrodes, while MEG was recorded using the Elekta Neuromag Triux system, which includes 102 magnetometers and 204 planar gradiometers. The data sampling rate was 1000 Hz. Preprocessing steps included Maxwell filtering to reduce external noise, downsampling the data to 500 Hz, and filtering using zero-phase-shift IIR filters. EEG data were filtered between 1-40 Hz, while MEG data were filtered between 80-150 Hz to extract HFA.

3. Data Analysis

The study first compared behavioral performance (reaction times and accuracy) across different contrast levels, then analyzed the response characteristics of C1 and HFA. The C1 component was identified by comparing ERP waveforms elicited by UVF and LVF stimulation, while HFA was extracted using the Hilbert transform to obtain its envelope amplitude. Finally, the study compared the latencies and amplitude modulation of C1 and HFA.

Key Findings

1. Behavioral Results

Participants’ reaction times and accuracy did not significantly differ across contrast levels, indicating that task-irrelevant contrast variations did not affect behavioral performance.

2. C1 Response

The C1 component peaked at posterior EEG electrodes PO3 and PO4, with UVF and LVF stimulation eliciting negative and positive C1 responses, respectively. The amplitude of C1 was significantly higher than baseline between 44 and 88 milliseconds and increased significantly with stimulus contrast.

3. HFA Response

HFA showed significant amplitude modulation between 76 and 266 milliseconds, but its amplitude did not vary significantly with contrast. The peak of HFA occurred at 172 milliseconds, significantly later than the peak of C1 (68 milliseconds).

4. Comparison of C1 and HFA Latencies

The onset and peak times of C1 were significantly earlier than those of HFA, and there was no significant correlation between the onset times, peak times, or peak amplitudes of C1 and HFA, suggesting that HFA is not a mere reverberation of C1.

Conclusions and Significance

By simultaneously recording EEG and MEG, this study revealed the distinct roles of C1 and HFA in visual processing. C1, as a marker of feedforward input, is sensitive to task-irrelevant contrast variations, while HFA exhibits a more sustained response and is unaffected by contrast changes. The delayed peak of HFA suggests that it encompasses not only feedforward information but also feedback processes. These findings provide new insights into the temporal dynamics of visual information processing and highlight the different functional roles of C1 and HFA in the visual cortex.

Research Highlights

1. Revealing Temporal Dynamics: The temporal dissociation between C1 and HFA indicates their distinct roles in visual processing, with C1 representing feedforward input and HFA potentially involving feedback processes.
2. Contrast Modulation Differences: C1 is sensitive to task-irrelevant contrast variations, while HFA is unaffected, further supporting the view of C1 as a marker of feedforward input.
3. Innovative Application of Non-Invasive Methods: By combining EEG and MEG recordings, this study provides a novel approach for non-invasively investigating visual processing.