Perception and propagation of activity through the cortical hierarchy is determined by neural variability

Abstract

The brains of higher organisms are composed of anatomically and functionally distinct regions performing specialised tasks; but regions do not operate in isolation. Orchestration of complex behaviours requires communication between brain regions, but how neural activity dynamics are organised to facilitate reliable transmission is not well understood. We studied this process directly by generating neural activity that propagates between brain regions and drives behaviour, allowing us to assess how populations of neurons in sensory cortex cooperate to transmit information. We achieved this by imaging two hierarchically organised and densely interconnected regions, the primary and secondary somatosensory cortex (S1 and S2) in mice while performing two-photon photostimulation of S1 neurons and assigning behavioural salience to the photostimulation. We found that the probability of perception is determined not only by the strength of the photostimulation signal, but also by the variability of S1 neural activity. Therefore, maximising the signal-to-noise ratio of the stimulus representation in cortex is critical to its continued propagation downstream. Further, we show that propagated, behaviourally salient activity elicits balanced, persistent, and generalised activation of the downstream region. Hence, our work adds to existing understanding of cortical function by identifying how population activity is formatted to ensure robust transmission of information, allowing specialised brain regions to communicate and coordinate behaviour.

What does the mean mean? A simple test for neuroscience

“…To test these two assumptions across a range of experimental conditions, we examined two complementary data sets featuring neuronal recordings in behaving mice. Dataset 1 consists of two-photon recordings in primary and secondary somatosensory cortex (S1 and S2) as mice detected a small optogenetic stimulus in S1 [33] (Fig 1B). Mice were trained to lick for reward in response to the optogenetic activation of 5 to 50 randomly selected S1 neurons.…”

Tackling the subsampling problem to infer collective properties from limited data

“…Second, after a perturbation by stopping sensory input, neural networks transiently decrease their branching parameter m but then recover a value close to unity, and thus longer intrinsic timescales, within few days, presumably via homeostasis [104]. Third, cortical stimulation is detected better, when the network shows lower instantaneous recurrency before the stimulus is applied, indicating that the larger signal-to-noise ratio may facilitate detection of such stimuli [105]. All of these results in neuroscience explicitly or implicitly rely on the fact that autocorrelation timescales or estimated recurrency are invariant under observation of a small part of the system.…”

“…A gold standard in studying experimental systems are the application of causal interventions [105,163]. Combining causal interventions at the microscopic scale with recordings at multiple scales may shed light on their interaction mechanisms.…”

Role of interneuron subtypes in controlling trial-by-trial output variability in the neocortex

“…Instead of being just noise, trial-by-trial variability varies (typically gets reduced) during the stimulus presentation (Churchland et al, 2010; Oram, 2011), due to attentional shifts (Kanashiro et al, 2017) or external stimulation (De Luna et al, 2017). The presence and change in trial-by-trial variability are not merely a statistical property of neuronal activity as it is necessary for behavior (Waschke et al, 2021) and affects the stimulus-response (Arieli et al, 1996) and behavioral performance (Arazi et al, 2017; Rowland et al, 2021). Given the noisy inputs, stochastic neurons, random connectivity, and unreliable synapses, trial-by-trial variability is not surprising.…”