Comparing aperiodic brain activity between eyes open rest and dynamic visual input using magnetoencephalography

This magnetoencephalography study reveals that viewing the "Inscapes" animation and eyes-open rest produce distinct aperiodic neural activity profiles, particularly showing widespread differences in frontoparietal and occipital regions with a specific reversal in the primary visual cortex, indicating that these two conditions are not interchangeable for analyzing broadband spectral dynamics or excitation/inhibition balance.

The Gist

Imagine your brain is like a busy radio station. It's always broadcasting, but the signal isn't just one clear song; it's a mix of specific melodies (like a steady drumbeat or a humming tune) and a constant "hiss" of background static.

In neuroscience, scientists have recently realized that this background static (called "aperiodic activity") is actually very important. It's not just noise; it's a clue about how the brain balances its "gas pedal" (excitation) and its "brakes" (inhibition).

This paper asks a simple question: Does it matter if you sit in a quiet room with your eyes open, or if you watch a calming, abstract video?

For years, researchers have used a special video called "Inscapes" (a slow-moving, abstract animation with piano music) to help people, especially children or those with medical conditions, sit still during brain scans. The idea was that the video keeps them awake and still, but their brain is still in a "resting" state, just like when they stare at a blank wall.

The researchers in this study wanted to know: Is the brain's "background static" actually the same in both situations?

The Experiment: The Quiet Room vs. The Calming Video

The team put 54 healthy adults in a giant, shielded helmet (a MEG machine) that listens to the brain's magnetic whispers. They asked the participants to do two things:

1. Eyes Open Rest: Sit still and stare at a tiny dot on a screen, letting their minds wander.
2. Inscapes: Watch the abstract, slow-moving shapes on the screen (without the music).

The Discovery: The Brain's "Volume Knob" is Different

The researchers found that the brain's "background static" was not the same in both conditions.

• The Analogy: Imagine the brain's background static as the "volume" of a radio.
  • During the Quiet Rest (staring at the dot): The volume was higher. The "static" was louder and steeper. This suggests the brain was in a state of higher "braking" or inhibition—calmer, more internal, and less reactive to the outside world.
  • During the Video (Inscapes): The volume was lower. The static flattened out. This suggests the brain was slightly more "excited" or ready to react because it was processing visual information, even though the video was boring and abstract.

The "Visual Cortex" Exception:
There was one tiny part of the brain, the Primary Visual Cortex (the part right at the back that processes what you see), that acted differently. While the rest of the brain got "quieter" during the video, this specific spot got "louder." It makes sense: this is the part of the brain actually doing the work of watching the shapes, so it was more active than when just staring at a dot.

Why This Matters

Think of it like this: If you are trying to measure the "temperature" of a room, you wouldn't want to measure it while someone is holding a hot cup of coffee right next to the thermometer. The reading would be accurate for that moment, but it wouldn't tell you the true temperature of the room itself.

The study shows that Inscapes and "Eyes Open Rest" are not interchangeable.

• If a scientist wants to study the brain's natural "resting state" (the baseline), using the video changes the baseline.
• The video introduces a subtle but measurable shift in how the brain balances its energy.

The Takeaway

The "Inscapes" video is a fantastic tool. It helps people stay still and awake, which is great for getting clear brain scans. However, scientists can no longer assume that the brain's "background noise" is identical to when a person is just staring at a blank wall.

If you are studying how the brain balances its energy (excitation vs. inhibition), you have to know which "radio station" the brain was tuned to: the quiet static of rest, or the slightly different static of watching a movie. They are two different channels, even if they sound similar at first glance.

Technical Summary

1. Problem Statement

In neuroimaging, particularly with pediatric and clinical populations prone to head motion, the "Inscapes" paradigm (a low-demand abstract animation) is frequently used as an alternative to traditional "eyes open" (EO) resting-state conditions. Previous validation studies established that Inscapes preserves functional connectivity patterns similar to rest and reduces motion artifacts.

However, a critical gap exists: no study has examined whether Inscapes and EO rest differ in aperiodic neural activity. Aperiodic activity (the $1/f$ component of the power spectrum) is distinct from periodic oscillations (like alpha/beta bands) and is increasingly used as a proxy for the neural excitation/inhibition (E/I) balance. If Inscapes modulates the aperiodic component differently than rest, it cannot be considered a direct substitute for rest in studies analyzing broadband spectral dynamics or E/I balance.

2. Methodology

Participants:

• Sample: 54 healthy adults (mean age 26.6 years; 29 females).
• Exclusion: 2 participants excluded due to insufficient trials after artifact rejection.

Experimental Design:

• Conditions: Two 7-minute conditions:
  1. Eyes Open Rest (EO): Participants viewed a central fixation cross and let thoughts wander.
  2. Inscapes: Participants passively viewed a low-demand abstract animation (visual component only; audio was excluded).
• Task: Maintain wakefulness and minimize head motion.

Data Acquisition:

• Modality: Magnetoencephalography (MEG) using a 306-channel Triux system (102 magnetometers, 204 gradiometers).
• Sampling: 1,000 Hz, bandpass filtered 0.03–330 Hz.
• Structural Imaging: High-resolution T1-weighted MRI (3T Siemens Prisma) for source reconstruction.

Preprocessing:

• Artifact Removal: MaxFilter (tSSS) for interference suppression; Independent Component Analysis (ICA) to remove cardiac and ocular artifacts.
• Epoching: Data cropped to the final 300 seconds, segmented into 4-second epochs.
• Cleaning: Amplitude-based rejection and Signal Space Projection (SSP) for residual artifacts.

Spectral Analysis:

• Primary Method: SpecParam algorithm (Donoghue et al., 2020) to decompose the Power Spectral Density (PSD) into:
  • Aperiodic Exponent ($\beta$): Slope of the $1/f$ distribution.
  • Aperiodic Offset: Broadband power amplitude.
  • Periodic Components: Alpha band (8–13 Hz) power and center frequency.
• Secondary Method: IRASA (Irregular-Resampling Auto-Spectral Analysis) was used as a robustness check to separate fractal and oscillatory components without parametric peak modeling.
• Source Reconstruction: Dynamic Statistical Parametric Mapping (dSPM) on the fsaverage6 template using a Desikan-Killiany atlas for parcellation and vertex-level analysis.

Statistical Analysis:

• Tests: Cluster-based permutation paired t-tests (5,000 permutations) for sensor and vertex levels; paired t-tests with False Discovery Rate (FDR) correction for parcellation levels.
• Contrast: Rest minus Inscapes.

3. Key Contributions

1. First Direct Comparison of Aperiodic Activity: This study is the first to directly compare the aperiodic spectral profiles of Inscapes and EO rest in healthy adults using MEG.
2. Demonstration of Non-Interchangeability: It provides empirical evidence that while Inscapes mimics rest in functional connectivity, it induces distinct broadband spectral dynamics.
3. Regional Specificity: It identifies a unique reversal in the primary visual cortex (pericalcarine), suggesting that the direction of aperiodic modulation depends on whether a region is directly engaged by the visual stimulus.
4. Methodological Robustness: The findings were validated using two distinct spectral decomposition algorithms (SpecParam and IRASA) and different high-pass filter settings, confirming the results are not methodological artifacts.

4. Results

Aperiodic Exponent and Offset:

• Global Pattern: Both the aperiodic exponent and offset were significantly higher during EO rest than during Inscapes across widespread frontoparietal, sensorimotor, and occipital regions (both magnetometers and gradiometers).
  • Interpretation: A steeper slope (higher exponent) during rest suggests a shift toward greater inhibition, whereas the flatter slope during Inscapes suggests increased cortical excitability due to sustained visual stimulation.
• Source Level: Parcellation and vertex analyses confirmed these widespread differences (e.g., 72.5% of cortical vertices showed higher exponents during rest).
• The Exception (Primary Visual Cortex): In the pericalcarine cortex (V1), the trend was reversed (though not statistically significant after FDR correction): exponents and offsets were numerically higher during Inscapes than rest. This suggests enhanced feedforward inhibitory processing specifically in the region directly processing the visual stimulus.

Alpha Power:

• As expected, alpha power was significantly higher during rest than during Inscapes, particularly over posterior regions. This aligns with the well-known phenomenon of alpha desynchronization during visual processing.
• The spectral parameterization approach successfully dissociated these periodic alpha changes from the concurrent aperiodic changes.

Robustness Checks:

• IRASA Validation: The IRASA method replicated the primary aperiodic findings (higher exponents/offsets during rest) across nearly all sensors, confirming the results are not dependent on the specific peak-fitting algorithm of SpecParam.
• Filter Sensitivity: Re-analyzing data with a stricter high-pass filter (0.5 Hz instead of 0.1 Hz) to rule out head-motion artifacts yielded nearly identical results.

5. Significance and Implications

• Paradigm Selection: Researchers using Inscapes as a "resting-state" proxy must recognize that it is not interchangeable with eyes-open rest for analyses involving broadband spectral power, spectral slope, or E/I balance estimation.
• Interpretation of E/I Balance: The flatter aperiodic slopes observed during Inscapes likely reflect a state of increased cortical excitability driven by visual engagement, distinct from the inhibitory tone of quiet rest.
• Clinical and Developmental Research: Since Inscapes is widely used in pediatric and clinical populations (e.g., autism, ADHD) to reduce motion, these findings suggest that group differences in aperiodic activity observed in such studies may be confounded by the specific visual nature of the stimulus rather than purely by pathology or development.
• Future Directions: The study highlights the need to report condition-specific spectral baselines or to account for the known spectral shifts when comparing studies using different resting-state paradigms. The regional reversal in V1 also opens avenues for investigating how stimulus engagement modulates local E/I balance differently than global resting states.