Neuronal activity patterns and sensory responses depend strongly on behavioral state. Active behavioral states are associated with enhanced gain, the presence of fast cortical dynamics, and a reduction in spontaneous activity. These effects depend strongly on the activity of specific GABAergic interneurons. Inactive behavioral states like sleep are associated with enhanced spontaneous activity, reduced response gain, and slow cortical dynamics that are temporally highly structured. Spontaneous activity patterns are strongly implicated in memory consolidation processes.


Introduction

The Greek philospher Heraclitus said – “No man ever steps in the same river twice, for it's not the same river and he's not the same man.” While neuroscientists have made great progress in describing the average tendency of neuronal activity in response to external stimuli, there is enormous variability in the neuronal reponses to the same sensory inputs, and the brain continuously shows profound fluctuations in behavioral state, from multiple sleep stages, to relaxed wake states, to attentive or aroused wake states[1]. The functional implications and mechanisms of brain states and neuronal variability remain poorly understood. Man-build intelligent systems like computers and robots respond in a highly predictable way to the same input, and are essentially always in the same state (on or off). The brain’s responses to the same sensory inputs are highly variable and dependent on the behavioral state[1]. Furthermore, when the organism is resting, e.g. during sleep, it exhibits spontaneous activity patterns that are highly structured, and during which neurons can be more active as compared to aroused states[2,3].


Average pupil diameters is shown as a function of time around air puff onset. The air puff arouses the animal without inducing locomotor activity. Cortical dynamics (LFP signal) change with arousal and show slow waves in low-arousal state. Spontaneous activity is suppressed during arousal period[2].

What is state?

Global behavioral states – like sleep and arousal – are characterized by several aspects[1]: 1) They are associated with particular neuronal dynamics, which can be measured both at the network level and through intracellular recordings. 2) They are accompanied by specific neuromodulatory tone; 3) The neurons respond in characteristic ways to sensory stimuli; 4) They are associated with specific behavioral performance levels and biases.  

Neuronal dynamics

In general, the more active wake states are characterized by the presence of faster waves (beta and gamma, 15-90 Hz) in the neocortex, which are linked conscious processing and attention, while inactive states are characterized by large-amplitude slow waves[1]. It is thought that these slower waves play a role in memory consolidation[4]. Fast gamma activity in sensory areas is positively modulated by arousal, even in the absence of locomotor activity[2], which depends strongly on the activity of GABAergic interneurons[6,1].



We compare here V1 activity of control animals, and animals in which there is a developmental disorder in GABAergic VIP interneurons, which compose about 1% of cells[6]. Animals in which there is a dysfunction of VIP interneurons do now show any modulation of firing activity by behavioral state (locomotor activity).[5]

Neuromodulators

Active states are general associated with an increase in noradrenergic and cholinergic tone. These neuromodulators are released through long-range projections of neurons in specific nuclei in the brain stem like the locus coeruleus and the nucleus basalis, and these nuclei are in turn activated by other nuclei of the ascending arousal system[1]. Because the activity of neuromodulatory projection systems correlates very well with pupil diameter, we can use pupil diameter as a readout for arousal level during wakefulness[1,2]. To understand the mechanisms underlying state modulation of cortical activity patterns, it is critical to advance our understanding of the impact of these neuromodulatory subsystems. Neuromodulators can affect pyramidal neurons directly, but also have highly specific effects on the GABAergic interneurons. We have shown that state modulation depends critically on the developmental integrity of VIP interneurons, which are strongly affected by neuromodulators [1,5].

How does state influence sensory processing

The way in which neurons respond to sensory stimuli depends strongly on arousal level. While spontaneous activity is decreased by cortical arousal, the gain of sensory responses increases [1,2]. Furthermore, noise correlations, which can be detrimental to sensory coding, are reduced with to arousal[2]. The dynamic regime in which the cortex operates – from fast fluctuations in the beta and gamma range, or slow fluctuations – might make an important contribution to these changes in gain[1,6]. We are currently investigating whether state makes a more important contribution to the response gain for stimuli that enhance fast cortical dynamics. We are furthermore investigating whether modulating oscillations without changing the neuromodulatory tone (through optogenetic manipulations) alone can change the gain of neuronal responses to sensory inputs, and the behavioral output.  

State and memory formation

We are studying how the activity during sleep resembles that of waking activity. Our hypothesis is that spontaneous activity in neocortex depends in a precise way on previous waking activity and contributes to memory consolidation[4]. With newly developed mathematical algorithms in our group, we can perform powerful unsupervised clustering of temporal patterns [7]. We are examining the structure and dimensionality of temporal patterns during spontaneous activity, and how these patterns depend on experience.


[1] McGinley MJ, Vinck M, Reimer J, Batista-Brito R, Zagha E, Cadwell CR, Tolias AS, Cardin JA, McCormick DA. Waking State: Rapid Variations Modulate Neural and
Behavioral Responses. Neuron. 2015 Sep 23;87(6):1143-61. doi: 10.1016/j.neuron.2015.09.012. Review. PubMed PMID: 26402600; PubMed Central PMCID: PMC4718218.

[2] Vinck M, Batista-Brito R, Knoblich U, Cardin JA. Arousal and locomotion make  distinct contributions to cortical activity patterns and visual encoding. Neuron. 2015 May 6;86(3):740-54. doi: 10.1016/j.neuron.2015.03.028. Epub 2015 Apr 16. PubMed PMID: 25892300; PubMed Central PMCID: PMC4425590.

[3] Steriade M, Timofeev I, Grenier F. Natural waking and sleep states: a view from inside neocortical neurons. J Neurophysiol. 2001 May;85(5):1969-85. PubMed PMID: 11353014.

[4] Genzel L, Kroes MC, Dresler M, Battaglia FP. Light sleep versus slow wave sleep in memory consolidation: a question of global versus local processes? Trends Neurosci. 2014 Jan;37(1):10-9. doi: 10.1016/j.tins.2013.10.002. Epub 2013  Nov 7. Review. PubMed PMID: 24210928.

[5] Batista-Brito R, Vinck M, Ferguson KA, Chang JT, Laubender D, Lur G, Mossner JM, Hernandez VG, Ramakrishnan C, Deisseroth K, Higley MJ, Cardin JA. Developmental Dysfunction of VIP Interneurons Impairs Cortical Circuits. Neuron.  2017 Aug 16;95(4):884-895.e9. doi: 10.1016/j.neuron.2017.07.034. PubMed PMID:
28817803; PubMed Central PMCID: PMC5595250.

[6] Fries P. Rhythms for Cognition: Communication through Coherence. Neuron. 2015  Oct 7;88(1):220-35. doi: 10.1016/j.neuron.2015.09.034. Review. PubMed PMID: 26447583; PubMed Central PMCID: PMC4605134.

[7] Grossberger, L., Battaglia, F, Vinck, M.
Unsupervised clustering of temporal patterns in high-dimensional neuronal ensembles using a novel dissimilarity measure. 2018. Biorxiv.