Correlated activity supports efficient cortical processing

Abstract

Visual recognition is a computational challenge that is thought to occur via efficient coding. An important concept is sparseness, a measure of coding efficiency. The prevailing view is that sparseness supports efficiency by minimizing redundancy and correlations in spiking populations. Yet, we recently reported that “choristers”, neurons that behave more similarly (have correlated stimulus preferences and spontaneous coincident spiking), carry more generalizable object information than uncorrelated neurons (“soloists”) in macaque inferior temporal (IT) cortex. The rarity of choristers (as low as 6% of IT neurons) indicates that they were likely missed in previous studies. Here, we report that correlation strength is distinct from sparseness (choristers are not simply broadly tuned neurons), that choristers are located in non-granular output layers, and that correlated activity predicts human visual search efficiency. These counterintuitive results suggest that a redundant correlational structure supports efficient processing and behavior.

Electrophysiological and firing properties of neurons: Categorizing soloists and choristers in primary visual cortex

“…In addition, noise correlation (R sc ) is higher for choristers than soloists in IT cortex [18]. Anatomical differences were also observed since the cortical laminar distribution for these types of units is distinct [18]. In mouse V1, population coupling varies between the narrow-spiking and regular-spiking neurons; however, this difference was reported to be smaller than the class-variability [27].…”

“…It has been suggested that soloists are distinct from choristers in many aspects such as the synaptic connectivity [27] and the laminar distribution [18]. We examined the relationship between the waveform type, the firing rate and the correlation-strength of the cells' firing with the population activity.…”

“…For instance, highly correlated neurons may receive stronger synaptic drive from neighboring cells [27]. In addition, noise correlation (R sc ) is higher for choristers than soloists in IT cortex [18]. Anatomical differences were also observed since the cortical laminar distribution for these types of units is distinct [18].…”

Functional investigation of neural circuits in larval zebrafish

“…In this study, core cells differentially combine with cells from a larger pool of inconsistently responding neurons to a given stimulus presentation. 'Core' cells of an assembly have only recently been described in the literature, but only in the context of cells most representative of, or having the highest correlation to, a particular responding cluster (Carrillo-Reid et al, 2015;Hung et al, 2015;Lin et al, 2014). To the best of our knowledge, this study provides the first description of a spatially localised population P ag e 138 of core cells in tectal assemblies that respond consistently and are more selective for their preferred stimulus features.…”

“…Using eight-pin, multi-site electrodes, two recent studies have been able to describe the presence of 'core' cells in the macaque inferior temporal cortex (Hung et al, 2015;Lin et al, 2014). More recently, the presence of core cells in the mouse visual cortex have also been described from calcium dye recordings (Carrillo-Reid et al, 2015).…”