Repertoires of Spike Avalanches Are Modulated by Behavior and Novelty

Abstract

Neuronal avalanches measured as consecutive bouts of thresholded field potentials represent a statistical signature that the brain operates near a critical point. In theory, criticality optimizes stimulus sensitivity, information transmission, computational capability and mnemonic repertoires size. Field potential avalanches recorded via multielectrode arrays from cortical slice cultures are repeatable spatiotemporal activity patterns. It remains unclear whether avalanches of action potentials observed in forebrain regions of freely-behaving rats also form recursive repertoires, and whether these have any behavioral relevance. Here, we show that spike avalanches, recorded from hippocampus (HP) and sensory neocortex of freely-behaving rats, constitute distinct families of recursive spatiotemporal patterns. A significant number of those patterns were specific to a behavioral state. Although avalanches produced during sleep were mostly similar to others that occurred during waking, the repertoire of patterns recruited during sleep differed significantly from that of waking. More importantly, exposure to novel objects increased the rate at which new patterns arose, also leading to changes in post-exposure repertoires, which were significantly different from those before the exposure. A significant number of families occurred exclusively during periods of whisker contact with objects, but few were associated with specific objects. Altogether, the results provide original evidence linking behavior and criticality at the spike level: spike avalanches form repertoires that emerge in waking, recur during sleep, are diversified by novelty and contribute to object representation.

Keywords: avalanches; criticality; memory; novelty; patterns; sleep; spikes.

Figure 1

Avalanches in freely-behaving animals comprise families and recur above chance. (A) Avalanches represented as binary vectors in which colors indicate spike activity. Depicted are representative examples from four different samples of similar avalanche pairs separated by several minutes (interval of recurrence indicated on the right). (B) Similarity matrix for 200 representative avalanches from one sample, before and after clustering (left and right panels, respectively). Black lines indicate 10 families formed by similar avalanches. (C) Contrast function for one sample (colors represent different avalanche durations), with the maximum for one duration indicated by the dashed line. (D) Number of families of each size for one sample. Gray indicates all families detected (left Y axis); black indicates significant families (right Y axis). Note that the Y axes are scaled by a factor of 1000. (E) The recurrence over time of highly similar avalanches defines avalanche families. For the two representative families shown, the occurrence time is relative to the first observed avalanche.

Figure 2

A significant fraction of the avalanche repertoire occurs exclusively during waking. (A) To assess the significance of the results, significant families were subjected to the shuffling of family labels. This procedure preserves the distribution of family sizes (see Figure 1D), the number of avalanches in each behavioral state, and the times of occurrence of each avalanche. (B) Fraction of avalanches occurring during waking (WK), slow-wave sleep (SWS) and rapid-eye-movement sleep (REM) for each of the 361 significant families of one sample (ranked by WK and then SWS prevalence). The dashed line indicates the boundary that separates waking-specific families from other families. (C) Same as in (B), but for families obtained from one single label-shuffled set. Note that many shuffled families comprising multiple states fall to the left of the dashed line, indicating that the number of the non-shuffled families that are waking-specific is larger. (D) For each data sample (animal/brain region), the fraction of state-specific families is compared to 1000 shuffled sets. Asterisk indicates significant results (p = 0.05). While the state specificity of families that occur during sleep is limited (3/9 samples in SWS and 2/9 in REM), waking-specific families occur significantly in 8/9 samples.

Figure 3

Family repertoire is behavior-dependent. (A) Setlists are defined as binary vectors composed of 1s (or 0s) for each significant family observed (or absent) in a given 1 min time window. (B) Mean similarity between WK and SWS setlists (label-shuffled data shown in black). Significant cases (p = 0.025) are indicated by asterisks on top (similar setlists) or bottom (dissimilar setlists) of the plot. All data samples represented; symbols indicate brain region, colors indicate condition. (C) Same as in (B), now comparing WK setlists through different experiment stages (PRE, EXP and POST). Gray areas indicate comparisons within the same stage.

Figure 4

Objects exploration affects repertoire growth. (A) Rate of new significant families observed as function of time. Original data for one sample (blue) as well as the threshold for significance obtained through 1000 label-shuffled sets (black) are shown, with PRE, EXP and POST periods indicated. Red areas represent periods of significant generation rate (p = 0.05). (B) Total number of families that appeared first in each experiment stage, for each data sample (same order as in Figures 2, 3; open symbols: original data, solid circles: label-shuffled sets). Significant cases (p = 0.025) are indicated by asterisks on top (high rate) or bottom (low rate) of the plot.

Figure 5

Contact specificity is observed in avalanche patterns. (A) Fraction of avalanches during EXP that occurred while the animal was in contact with each object (coded by colors). Note that the original data (top) and the label-shuffled dataset (bottom) profiles are similar, but the total number of families differs. Dashed lines and colored arrows emphasize families that are specific to a particular object. (B) Time occurrence of avalanches from families which are specific to one object. In detail above, the color-coded bar shows contact windows for each object. (C) Number of object-specific families observed (see “Results” Section). All samples are shown (same order as in previous figures, in colors), together with 1000 label-shuffled sets (in black). Significant cases (p = 0.025) are marked by asterisks on top (high number) or bottom (low number) of the plot. (D) Same as in (C), but for contact families or, in other words, families exclusively composed of avalanches that occurred during contact with any object.