Abstract
Perception of sensory information is determined by both stimulus features and instantaneous neural states, such as excitability. It is commonly assumed that higher evoked activity leads to a stronger percept. This study tested this assumption in a somatosensory discrimination task, assessing single-trial excitatory post-synaptic currents from short-latency somatosensory evoked potentials (SEP), pre-stimulus alpha oscillations, and peripheral nerve measures. Findings reveal that fluctuations of neural excitability shape perceived intensity during the very first cortical response (~20 ms), but with an opposite neural signature compared to the effect of stimulus intensity. This discrepancy is explained by a framework based on electro-chemical membrane gradients, suggesting a need to reconsider conventional interpretations of brain potential magnitudes in stimulus encoding.
Introduction
In neuroscience, a simple linear relationship is often assumed: a larger electrical brain response to a stimulus implies a more intense perception. While this holds true when increasing physical stimulus strength, it overlooks the "internal weather" of the brain—fluctuating states of neural excitability. This research investigates whether these internal fluctuations bias perception at the earliest stage of cortical processing and if this bias manifests in the EEG similarly to physical intensity.
Methods and Data
The study involved 32 healthy male participants performing a somatosensory discrimination task. Participants received approximately 1,000 electrical pulses to the median nerve at two "supra-threshold" intensities that were physically similar and difficult to distinguish.
To analyze the data, the researchers employed:
- Single-Trial SEP Extraction: Canonical Correlation Analysis (CCA) was used to extract the "N20" component—the first cortical response—on a trial-by-trial basis.
- Peripheral Controls: Nerve signals (CNAP) and muscle twitches (CMAP) were recorded to ensure excitability fluctuations were central, not peripheral.
- Structural Equation Modeling (SEM): A two-level SEM disentangled the relationships between pre-stimulus alpha oscillations, the N20 response, peripheral activity, and the participant's final decision.
Key Insights: The Excitability Paradox
The study's central discovery is the "Opposing Signatures" effect. Increasing the physical intensity of a stimulus makes the N20 response larger (more negative). Conversely, an increase in internal excitability (indicated by low pre-stimulus alpha power) makes the N20 response smaller (less negative). Crucially, both lead to the same behavioral outcome: the participant perceives the stimulus as "stronger."
The authors propose a biophysical explanation based on electrochemical gradients. In a state of high excitability, neuronal membranes are already slightly depolarized (closer to the firing threshold). This reduced gradient means there is less "driving force" for ions to flow into the cell upon stimulation, resulting in a smaller post-synaptic current and a smaller EEG peak, despite the neuron being closer to firing.
Results: Perception is Biased at the Gate
Analysis of the 31,347 single trials revealed several key findings:
- Behavioral Bias: Higher pre-stimulus alpha power (low excitability) was significantly associated with a "weak" perception bias.
- N20 vs. Perception: Paradoxically, smaller (less negative) N20 amplitudes were associated with a higher likelihood of rating a stimulus as "strong."
- Coding Logic Shift: While the early N20 showed this inverse relationship, later components like the N140 followed the traditional rule where larger amplitudes correlate with stronger perception.
Discussion and Limitations
These results demonstrate that subjective experience is biased at the very first gateway of the cortex, roughly 20 milliseconds after a stimulus occurs. However, the study has limitations: the cohort was exclusively male, and while source reconstruction pointed to Brodmann area 3b, EEG's spatial resolution is limited. Future research should explore whether these findings generalize across genders and other sensory modalities like vision or hearing, and whether these excitability fluctuations are local or brain-wide.
Summary updated