State-Dependent Representation of Amplitude-Modulated Noise Stimuli in Rat Auditory Cortex

Stephan Lawrence Marguet, Kenneth D. HarrisJournal of Neuroscience 2011 · DOI

Abstract

Cortical responses can vary greatly between repeated presentations of an identical stimulus. Here we report that both trial-to-trial variability and faithfulness of auditory cortical stimulus representations depend critically on brain state. A frozen amplitude-modulated white noise stimulus was repeatedly presented while recording neuronal populations and local field potentials (LFPs) in auditory cortex of urethane-anesthetized rats. An information-theoretic measure was used to predict neuronal spiking activity from either the stimulus envelope or simultaneously recorded LFP. Evoked LFPs and spiking more faithfully followed high-frequency temporal modulations when the cortex was in a desynchronized state. In the synchronized state, neural activity was poorly predictable from the stimulus envelope, but the spiking of individual neurons could still be predicted from the ongoing LFP. Our results suggest that although auditory cortical activity remains coordinated as a population in the synchronized state, the ability of continuous auditory stimuli to control this activity is greatly diminished.

Introduction

Neuroscientists have long been puzzled by "trial-to-trial variability"—the phenomenon where the exact same sensory input triggers wildly different neural responses at different times. A primary suspect in this variability is the internal "brain state." Cortical activity naturally fluctuates between synchronized states, characterized by slow, high-amplitude oscillations similar to deep sleep, and desynchronized states, marked by low-amplitude, high-frequency activity typical of alertness or REM sleep.

While it has been established that these states change the background noise of the brain, it remained unclear how they affected the faithfulness of sensory representations. This study investigates whether a "sleepy" (synchronized) brain simply possesses a noisier signal or if it fundamentally alters the encoding of external information.

Experimental Design and Methodology

The researchers recorded neuronal populations and local field potentials (LFPs) from the auditory cortex of urethane-anesthetized rats. Urethane is a unique anesthetic because the brain spontaneously transitions between synchronized and desynchronized states, mimicking natural sleep cycles. To measure reliability, they used "frozen" amplitude-modulated (AM) white noise—a 50-second noise snippet played repeatedly to observe how neurons responded to identical input across different states.

Figure 1: Cortical state transitions and neural population responses. (A) Spectrogram of LFP power showing the shift from low-frequency synchronized activity to high-frequency desynchronized activity following a tail pinch. (B) Raster plots and LFP traces during silence and AM noise, highlighting the bursty 'Up/Down' states in synchronized epochs versus the continuous tracking in desynchronized epochs.

To quantify the "faithfulness" of the representation, the team employed information-theoretic measures to predict a neuron's spiking activity from two sources: the external stimulus envelope and the internal "chatter" of the local population (the LFP). They utilized both traditional linear Spike-Triggered Average (STA) and complex 2D nonlinear STA models to account for both sound amplitude and its derivative.

The Internal Tug-of-War: State as a Gatekeeper

The study revealed that in the synchronized state, the cortex is preoccupied with its own internal dynamics—specifically "Up" and "Down" states where neurons fire in bursts and then go silent in unison. This internal coordination is so powerful that it effectively "drowns out" the continuous modulations of external sound.

This leads to a predictability paradox: in the synchronized state, a neuron’s firing is almost impossible to predict from the sound stimulus, yet it remains highly predictable from the LFP. This indicates that the neuron is still listening intently to its neighbors, but has effectively decoupled from external sensory input. Conversely, desynchronization acts as a high-fidelity switch, suppressing internal oscillations and allowing the cortex to track high-frequency temporal modulations up to 100 Hz.

Quantitative Results

The transition to a desynchronized state significantly improved the reliability and information rate of neural firing. The Fano factor, a standard measure of variability, was found to be $<1$ in desynchronized states (indicating higher reliability than a random Poisson process) but $>1$ in synchronized states.

Figure 2: Statistical measures of stimulus representation. (A) Comparison of LFP and spiking activity across states. (B) Coherence between the AM stimulus envelope and LFP, showing significantly higher tracking in the desynchronized state up to 100 Hz. (C) Fano factors for individual neurons; values above the diagonal indicate lower variability in the desynchronized state.

Furthermore, the prediction quality (measured in bits per spike) from the stimulus envelope was markedly higher during desynchronized epochs. While the LFP remained a strong predictor in both states, the desynchronized state allowed the LFP to follow the stimulus envelope at much higher frequencies compared to the synchronized state, where tracking was severely limited.

Discussion

These findings suggest that brain state acts as a gatekeeper for sensory information. While urethane-induced states mimic natural ones, future work in awake animals performing complex tasks is needed to see how active attention further refines these representations. Additionally, the authors suggest this "filtering" might originate in the thalamus, where neurons switch between burst and tonic modes. Understanding the causal links—perhaps through optogenetic manipulation of neuromodulatory systems like acetylcholine—remains a critical next step in deciphering how the brain chooses what to hear.

Reference: Marguet, S. L., & Harris, K. D. (2011). State-Dependent Representation of Amplitude-Modulated Noise Stimuli in Rat Auditory Cortex. The Journal of Neuroscience, 31(17), 6414–6420.

Summary updated

Abstract

Cortical responses can vary greatly between repeated presentations of an identical stimulus. Here we report that both trial-to-trial variability and faithfulness of auditory cortical stimulus representations depend critically on brain state. A frozen amplitude-modulated white noise stimulus was repeatedly presented while recording neuronal populations and local field potentials (LFPs) in auditory cortex of urethane-anesthetized rats. An information-theoretic measure was used to predict neuronal spiking activity from either the stimulus envelope or simultaneously recorded LFP. Evoked LFPs and spiking more faithfully followed high-frequency temporal modulations when the cortex was in a desynchronized state. In the synchronized state, neural activity was poorly predictable from the stimulus envelope, but the spiking of individual neurons could still be predicted from the ongoing LFP. Our results suggest that although auditory cortical activity remains coordinated as a population in the synchronized state, the ability of continuous auditory stimuli to control this activity is greatly diminished.
BibTeX
@article{marguet2011state,
  author = {Stephan Lawrence Marguet and Kenneth D. Harris},
  title = {{State-Dependent Representation of Amplitude-Modulated Noise Stimuli in Rat Auditory Cortex}},
  journal = {Journal of Neuroscience},
  year = {2011},
  doi = {10.1523/jneurosci.5773-10.2011},
}
Figure