Intrinsic Arousal Dynamics Alternately Support Sensory Encoding and Hippocampal Ripples

2558 

Abstract Submission 

Yifan Yang¹, David Leopold², Jeff Duyn³, Grayson Sipe⁴, Xiao Liu⁴

¹The Pennsylvania State University, State College, PA, ²National Institute of Mental Health, Bethesda, MD, ³National Institutes of Health, Bethesda, MD, ⁴The Pennsylvania State University, STATE COLLEGE, PA

Yifan Yang  
The Pennsylvania State University
State College, PA

David Leopold  
National Institute of Mental Health
Bethesda, MD

Jeff Duyn  
National Institutes of Health
Bethesda, MD

Grayson Sipe  
The Pennsylvania State University
STATE COLLEGE, PA

Xiao Liu  
The Pennsylvania State University
STATE COLLEGE, PA

The variable nature of the brain's response to identical stimuli has long intrigued researchers. The state of arousal appears to be a critical factor given the enhancement of sensory responses during active behaviors, such as whisking or locomotion (1, 2), and through manipulation of the noradrenergic and cholinergic systems (3, 4). However, in absence of overt active behaviors or external arousal modulators, response variability persists, even over time frames of a few seconds (5–7). The nature and source of this seconds-scale fluctuation in the brain's response remains unclear. Recent studies revealed that ongoing spiking activity during rest is organized as a highly structured cascade dynamic that entrains ~70% of neurons across various regions into a temporal sequence of activations spanning multiple seconds (8). Importantly, this cascade dynamic is phase coupled to rapid, seconds-scale arousal modulations, which may in turn contribute to the seconds-scale variance of neural responses. By analyzing large-scale neuronal recordings from mice, we investigated whether the cascade dynamic persists during periods of continuous external sensory stimulation and, if so, whether the dynamic affects how the brain processes sensory inputs.

We analyzed large-scale neuronal recordings in the Allen Visual Coding dataset (9), focusing on spiking activity of ~22,000 neurons recorded from 32 mice (Fig. 1A) across 4 experimental sessions, including three involving natural scene visual stimuli and one spontaneous session (Fig. 1B). We adopted the delay-profile decomposition method (8, 10) to detect cascade events and formulate negative- and positive-delay neuron groups. To measure the efficacy of visual sensory processing, we trained and evaluated support vector machine (SVM) decoders to predict the image identity based on the evoked spiking activity (Fig. 2A).

When sorted by their principal delay profile, the spiking activity of the population of recorded neurons showed clear cascade instances during immobile periods of visual stimulation that closely resembled cascades observed during rest (Fig. 1C and 1D). A state index, capturing the temporal population cascade dynamics using relative activation level of the negative- and positive-delay neurons, displayed a bimodal distribution, suggesting the existence of two distinct brain arousal states (Fig. 1E). The two states were associated with distinct pupil sizes (Fig. 1F) and correspond well to distinct phases of the cascades (Fig. 1C).
The precision of visual information encoding, which was measured by the decoding accuracy of the trained SVM, showed a striking difference between the stationary high- and low-arousal states (Fig. 2B). In fact, state index and decoding accuracy displayed a strong and highly reproducible linear association (Fig. 2C). These associations appeared to be driven by the cascade dynamic. The decoding accuracy averaged over the cascades demonstrated systematic changes across the cycle and was tightly locked to the modulation of the state index (Fig. 2D). Further, the hippocampal sharp wave ripples (SPW-Rs), recorded simultaneously, were modulated in completely inverse way across the cascade cycle and peaked when sensory encoding was at its minimum (Fig. 2E).

Here we show that the mouse brain, during stationary periods of passive visual stimulation, undergoes quasi-periodic state fluctuations indicated by stereotyped population dynamics spanning several seconds. Importantly, this cascade dynamic modulates sensory coding and hippocampal ripples in completely opposite ways. Thus, the brain alternates the brain between two functional modes that are optimized for external information processing and internal processes of memory relevance. These findings offer novel insights into the basis of neural variability over the timescale of seconds and the mechanisms by which the brain juggles competing demands on its computational resources.

Learning and Memory Other ²

Classification and Predictive Modeling

Perception: Visual ¹

Computational Neuroscience

Cortex

ELECTROPHYSIOLOGY

Machine Learning

Memory

Perception

Other - Ongoing global brain dynamics