Gating of Sensory Input by Spontaneous Cortical Activity

Artur Luczak, Péter Barthó, Kenneth D. HarrisJournal of Neuroscience 2013 · DOI

Abstract

The activity of neural populations is determined not only by sensory inputs but also by internally generated patterns. During quiet wakefulness, the brain produces spontaneous firing events that can spread over large areas of cortex and have been suggested to underlie processes such as memory recall and consolidation. Here we demonstrate a different role for spontaneous activity in sensory cortex: gating of sensory inputs. We show that population activity in rat auditory cortex is composed of transient 50-100 ms packets of spiking activity that occur irregularly during silence and sustained tone stimuli, but reliably at tone onset. Population activity within these packets had broadly consistent spatiotemporal structure, but the rate and also precise relative timing of action potentials varied between stimuli. Packet frequency varied with cortical state, with desynchronized state activity consistent with superposition of multiple overlapping packets. We suggest that such packets reflect the sporadic opening of a "gate" that allows auditory cortex to broadcast a representation of external sounds to other brain regions.

Introduction

Neuroscientists have long observed that the brain remains intensely active even in the absence of external stimuli. This "spontaneous activity" often mirrors the patterns evoked by actual sensory inputs, leading to theories that it represents neural "noise" or the "replay" of memories for consolidation. However, a fundamental question remains: if the brain is constantly firing in these patterns spontaneously, how does it distinguish between internal activity and real-world sensory input? Luczak et al. investigated how these internal patterns interact with sustained sensory stimuli to understand how the cortex "gates" and represents the external environment.

Methods

The researchers recorded from the deep layers (primarily Layer V) of the primary auditory cortex in four unanesthetized, head-fixed rats.

  • Recording Technology: Silicon microelectrodes (tetrodes) captured simultaneous activity from populations of 17 to 45 neurons per animal.
  • Stimuli: 1-second pure tones (3.3 to 24 kHz) were presented.
  • Analysis Metrics: Multiunit activity (MUA) and Local Field Potentials (LFP) were analyzed. A novel metric, $\mu_{cc}$ (Center of Mass), quantified the relative timing of spikes within population bursts. Linear Discriminant Analysis (LDA) assessed the information content regarding tone frequency.

The Activity Packet: A Fundamental Unit of Cortical Processing

A central finding of this study is that cortical activity is not a continuous stream but is organized into discrete "packets" lasting 50–100 ms. These packets occur reliably at the onset of a sound but also appear irregularly during silence and throughout long, sustained sounds.

Figure 1: Trial-triggered responses versus actual population dynamics. (A) Trial-averaged responses show a clear onset peak. (B) Shuffling neurons between trials destroys the temporal structure. (C) Raw data reveals that activity is composed of discrete packets (MUA) synchronized with LFP fluctuations, occurring both during silence and tone presentation.

Interestingly, the "burst" of activity seen at sound onset is not inherently more intense than a spontaneous packet; it is simply more predictable. In trial-averaged views (PSTHs), this onset response appears as a massive peak because it is perfectly aligned across trials, whereas spontaneous packets are averaged out.

The Gating Hypothesis

The authors propose that these packets act as a "gate" for sensory information. In a "synchronized" brain state (typical of quiet wakefulness or slow-wave sleep), the gate is mostly closed, opening only sporadically. When a packet occurs, it provides a window of opportunity for the cortex to "broadcast" its current sensory representation to other brain regions.

Figure 2: Population rate dynamics and packet statistics. (A) Average population rate showing the onset peak and sustained/spontaneous periods. (B) The maximum amplitude (z-score) of packets is similar across onset, sustained, and spontaneous conditions. (C) The probability of a packet occurring is significantly higher at tone onset.

As shown in Figure 2, while the probability of a packet is highest at onset (C), the actual amplitude of these packets (B) is remarkably consistent across spontaneous, onset, and sustained periods. This suggests that the onset response is a "reliable" version of the same fundamental packet structure seen spontaneously.

Key Results and Information Gating

  • Temporal Stereotypy: The internal sequential timing of neurons within a packet ($\mu_{cc}$) was highly correlated between spontaneous and evoked activity ($r \approx 0.75$), suggesting a conserved internal structure.
  • Information Content: Using LDA, the researchers found that bins with higher population firing rates (during packets) contained significantly more information about stimulus frequency than low-activity bins. This supports the idea that packets are the primary carriers of sensory information.
  • State Modulation: In desynchronized states (associated with high alertness), packets occur so frequently that they overlap, making activity appear continuous. However, the fine-scale temporal relationships between neurons remain preserved.
  • LFP Correlation: Packet timing was strongly locked to the phase of the Local Field Potential (LFP) at frequencies below 20 Hz, particularly around 10 Hz.

Discussion and Limitations

While these findings provide a compelling model for cortical gating, several questions remain. The recordings were focused on deep layers; how these packets propagate across different cortical layers requires further study. Additionally, while the rats were awake, the influence of prior anesthesia and the specific top-down mechanisms (like attention) that might "open the gate" by inducing desynchronization are key areas for future research. The study suggests that spontaneous activity is not merely noise but a structured framework that regulates the flow of sensory information throughout the brain.

Summary updated

Abstract

The activity of neural populations is determined not only by sensory inputs but also by internally generated patterns. During quiet wakefulness, the brain produces spontaneous firing events that can spread over large areas of cortex and have been suggested to underlie processes such as memory recall and consolidation. Here we demonstrate a different role for spontaneous activity in sensory cortex: gating of sensory inputs. We show that population activity in rat auditory cortex is composed of transient 50-100 ms packets of spiking activity that occur irregularly during silence and sustained tone stimuli, but reliably at tone onset. Population activity within these packets had broadly consistent spatiotemporal structure, but the rate and also precise relative timing of action potentials varied between stimuli. Packet frequency varied with cortical state, with desynchronized state activity consistent with superposition of multiple overlapping packets. We suggest that such packets reflect the sporadic opening of a "gate" that allows auditory cortex to broadcast a representation of external sounds to other brain regions.
BibTeX
@article{luczak2013gating,
  author = {Artur Luczak and Péter Barthó and Kenneth D. Harris},
  title = {{Gating of Sensory Input by Spontaneous Cortical Activity}},
  journal = {Journal of Neuroscience},
  year = {2013},
  doi = {10.1523/jneurosci.2928-12.2013},
}
Figure