Coupling of pupil- and neuronal population dynamics reveals diverse influences of arousal on cortical processing

Thomas Pfeffer, Christian Keitel, Daniel S. Kluger, Anne Keitel, Alena Russmann, Gregor Thut, Tobias H. Donner, Joachim Groß

TL;DR

This study reveals that spontaneous fluctuations in arousal, measured by pupil size, have diverse and region-specific effects on brain activity across the human cortex. By combining MEG and pupillometry in a large cohort, the researchers found that arousal influences different frequency bands and even exhibits non-linear relationships with neural activity, challenging the idea of arousal as a simple, uniform modulator.

What Problem Does This Paper Address?

For decades, arousal has been understood as a global "broadcast signal" from subcortical structures like the locus coeruleus (LC) and basal forebrain (BF) that uniformly influences cortical processing. This view suggests that arousal levels simply "turn up" or "down" the brain's activity. However, recent evidence points to a more nuanced picture: these arousal systems have complex projection patterns, and neuromodulator receptors are unevenly distributed across the cortex. This suggests that arousal might have more specific and varied effects than previously thought. The problem this paper addresses is the lack of a comprehensive understanding of how these arousal fluctuations specifically impact different aspects of neural population activity across the entire human brain, and whether these effects are uniform or region-specific.

Methods

The researchers employed a multi-site, multi-laboratory approach combining magnetoencephalography (MEG) with pupillometry.

  • MEG: This non-invasive technique measures magnetic fields produced by electrical activity in the brain, offering good temporal resolution to capture neural dynamics.
  • Pupillometry: Pupil diameter is a well-established peripheral marker of central arousal, reflecting the activity of neuromodulatory systems like the LC (noradrenaline) and BF (acetylcholine).
  • Concurrent Recording: By recording MEG and pupil size simultaneously in resting participants, the study could directly investigate the coupling between arousal fluctuations and brain activity.
  • Data Analysis:
    • Mutual Information (MI): Used to quantify the statistical dependence between pupil size and MEG power across various frequency bands (2-128 Hz), capturing both monotonic and non-monotonic relationships.
  • Cross-correlation Analysis: Employed to determine the temporal lag between pupil fluctuations and cortical activity, revealing frequency-dependent timing differences.
  • Source Reconstruction: MEG data was projected into source space to localize the origin of these couplings within specific brain regions.
  • Spectral Parameterization: The aperiodic component (1/f slope) of the power spectrum was analyzed, as it's thought to reflect the excitation-inhibition (E/I) balance in neural circuits.
  • Non-linear Modeling: Polynomial fitting was used to identify inverted U-shaped relationships between arousal and neural activity, akin to the Yerkes-Dodson law.

Datasets

The study utilized data from 81 healthy human participants collected across three different MEG laboratories (Glasgow, UK; Hamburg & Münster, Germany). The data consisted of resting-state MEG and simultaneous pupillometry recordings. No new benchmarks were introduced; the focus was on analyzing existing data with novel analytical approaches.

Key Insights and Ideas

  • Arousal is Not a Monolith: The most striking insight is that arousal doesn't uniformly affect the brain. Instead, its influence is highly specific to the frequency band of neural activity and the cortical region involved.
  • Arousal-Triggered Cascade: The temporal analysis suggests a sequential cascade of effects following pupil dilations, hinting at a routine "monitoring sweep" in the resting brain.
  • Non-linear Arousal Effects: The discovery of inverted U-shaped relationships in specific frequency bands (alpha/beta) and regions challenges the linear view of arousal's impact, suggesting optimal arousal levels for certain neural processes.
  • E/I Balance and Arousal: The correlation between pupil size and the spectral exponent of the aperiodic component provides a potential neural marker for the excitation-inhibition balance across the brain, modulated by arousal.

Key Results

  • Frequency-Specific Effects:
    • Low-frequency (2-8 Hz) activity in temporal and lateral frontal cortex decreased with pupil dilation.
  • High-frequency (>64 Hz) activity in mid-frontal regions increased.
  • Intermediate frequencies (8-32 Hz) showed complex, sometimes non-linear (inverted U-shaped) relationships, particularly in occipito-parietal regions.
  • Temporal Dynamics:
    • Low and high-frequency power fluctuations preceded pupil dilations by approximately 832 ms and 562 ms, respectively.
  • The 8-16 Hz range showed a closer temporal relationship (lag of ~210 ms).
  • Spatial Specificity:
    • Arousal effects varied significantly across cortical regions. For instance, visual cortex showed positive correlations in the 8-32 Hz range, while auditory cortex showed negative correlations in the low-frequency range.
  • Anterior cingulate cortex exhibited a spectral pattern similar to auditory cortex.
  • Aperiodic Component: A widespread negative correlation was found between pupil diameter and the spectral exponent, suggesting that increased arousal flattens the power spectrum, potentially indicating increased excitation relative to inhibition (E/I) in most cortical regions.
  • Non-linear Relationships: Inverted U-shaped relationships between pupil diameter and band-limited power were observed specifically in the alpha/beta frequency band (8-32 Hz) and in parieto-occipital cortices, suggesting optimal arousal levels for certain neural dynamics.

Limitations and Future Work

  • Resting State Focus: The study primarily examined arousal effects during resting state. Future work should investigate how these relationships change during cognitive tasks, which involve different arousal demands and behavioral contexts.
  • Neuromodulator Specificity: While pupil size is linked to noradrenaline and acetylcholine, the study acknowledges that other neuromodulators (serotonin, dopamine) also influence pupil size and cortical activity. Future research could disentangle these contributions.
  • Temporal Resolution of Spectral Estimates: The authors note that the wavelet analysis used for spectral decomposition has varying temporal resolution across frequencies, which can pose a challenge when comparing peak latencies.
  • Visual Cortex Exception: The absence of the E/I balance effect in some visual cortices warrants further investigation, especially considering how task performance might alter E/I in these regions.
  • Future Directions: The authors suggest exploring how these arousal influences change with different cognitive task demands and behaviors, and incorporating neuromodulator receptor distribution data to further elucidate differential contributions. They also propose including arousal as a covariate in future cognitive function studies.

Summary updated

Abstract

Fluctuations in arousal, controlled by subcortical neuromodulatory systems, continuously shape cortical state, with profound consequences for information processing. Yet, how arousal signals influence cortical population activity in detail has so far only been characterized for a few selected brain regions. Traditional accounts conceptualize arousal as a homogeneous modulator of neural population activity across the cerebral cortex. Recent insights, however, point to a higher specificity of arousal effects on different components of neural activity and across cortical regions. Here, we provide a comprehensive account of the relationships between fluctuations in arousal and neuronal population activity across the human brain. Exploiting the established link between pupil size and central arousal systems, we performed concurrent magnetoencephalographic (MEG) and pupillographic recordings in a large number of participants, pooled across three laboratories. We found a cascade of effects relative to the peak timing of spontaneous pupil dilations: Decreases in low-frequency (2-8 Hz) activity in temporal and lateral frontal cortex, followed by increased high-frequency (>64 Hz) activity in mid-frontal regions, followed by monotonic and inverted U relationships with intermediate frequency-range activity (8-32 Hz) in occipito-parietal regions. Pupil-linked arousal also coincided with widespread changes in the structure of the aperiodic component of cortical population activity, indicative of changes in the excitation-inhibition balance in underlying microcircuits. Our results provide a novel basis for studying the arousal modulation of cognitive computations in cortical circuits.
BibTeX
@inproceedings{pfeffer2022coupling,
  author = {Thomas Pfeffer and Christian Keitel and Daniel S. Kluger and Anne Keitel and Alena Russmann and Gregor Thut and Tobias H. Donner and Joachim Groß},
  title = {{Coupling of pupil- and neuronal population dynamics reveals diverse influences of arousal on cortical processing}},
  booktitle = {eLife},
  year = {2022},
  doi = {10.7554/elife.71890},
}
Figure