Visual information is broadcast among cortical areas in discrete channels

Abstract

11Neural circuitry represents sensory input with patterns of spiking activity. Across brain regions, initial 12representations are transformed to ultimately drive adaptive behavior. In mammalian neocortex, visual 13 information is processed by primary visual cortex (V1) and multiple higher visual areas (HVAs). The 14interconnections of these brain regions, over which transformations can occur, span millimeters or more. 15Shared variability in spiking responses between neurons, called "noise correlations" (NCs), can be due 16to shared input and/or direct or indirect connectivity. Thus, NCs provide insight into the functional 17 connectivity of neuronal circuits. In this study, we used subcellular resolution, mesoscale field-of-view 18 two-photon calcium imaging to systematically characterize the NCs for pairs of layer 2/3 neurons across 19 V1 and four HVAs (areas LM, LI, AL and PM) of mice. The average NCs for pairs of neurons within or 20 across cortical areas were orders of magnitude larger than trial-shuffled control values. We characterized 21 the modulation of NCs by neuron distance, tuning similarity, receptive field overlap, and stimulus type 22 over millimeter scale distances in mouse visual cortex, within and across V1 and multiple HVAs. NCs 23were positively correlated with shared tuning and receptive field overlap, even across cortical areas and 24 millimeter length scales. We compared the structure of these NCs to that of hypothetical networks to 25 determine what network types can account for the results. We found that to reproduce the NC networks, 26 neuron connectivity was regulated by both feature similarities and hub mechanism. Overall, these results 27revealed principles for the functional organization and correlation structure at the individual neuron level 28 across multiple cortical areas, which can inform and constrain computational theories of cortical networks. 30Navigating the environment, detecting predators, and recognizing mates are natural behaviors of 31 mammals that can depend upon sophisticated visual perception. Visual information is encoded and 32 abstracted through multiple interconnected brain areas, which span millimeters in the mouse 1 . These 33interconnections, both within and between brain areas, are critical components underlying brain functions. 34Anatomical tracing studies of flies 2 and mice 3 , have provided insights into basic neuronal wiring principles. 35However, anatomical connectivity is insufficient for understanding the circuit interactions of an active 36 brain, and the relationships between function and connectivity. Thus, in this study we investigated the 37 functional circuit connectivity structure at millimeter scale. 38Neuronal responses vary on nominally identical trials. The trial-to-trial variability of responses is 39 correlated in a neuronal population. This shared variability between pairs of neurons can be quantified, 40and is called noise correlation (NC) 4 . The structure of NCs is rooted in the architecture of the neuronal 41circuit, e.g., shared...

Mouse higher visual areas provide both distributed and discrete contributions to visually guided behaviors

“…We used the same approach to calculate hit rate, detection threshold and FA rate, and compare those between control and area suppression conditions ( Figures 5B-C and Supplementary item 5 ). Since neurons in AL and PM have distinct speed preference profiles [18,19,37–40], and therefore may be important for speed change detection, we focused on the role of these two areas in this task.…”

Identifying signal and noise structure in neural population activity with Gaussian process factor models

“…Gratings had 8 different orientations which persisted for 4 seconds each. For more information, see [24,25] and the supplemental materials.…”

Diesel2p mesoscope with dual independent scan engines for flexible capture of dynamics in distributed neural circuitry

“…TITL-GCaMP6s mice were kindly provided by Allen institute. Mice craniotomy was performed as previously described 22 . Briefly, mice were deeply anesthetized using isoflurane (1.5–2%) augmented with acepromazine (2 mg/kg body weight) during craniotomy surgery.…”

“…Calcium imaging was processed following our previously established protocol 22 . Ca 2+ signals were analyzed using custom software in MATLAB (Mathworks).…”