Specific excitatory connectivity for feature integration in mouse primary visual cortex

Abstract

Local excitatory connections in mouse primary visual cortex (V1) are stronger and more prevalent between neurons that share similar functional response features. However, the details of how functional rules for local connectivity shape neuronal responses in V1 remain unknown. We hypothesised that complex responses to visual stimuli may arise as a consequence of rules for selective excitatory connectivity within the local network in the superficial layers of mouse V1. In mouse V1 many neurons respond to overlapping grating stimuli (plaid stimuli) with highly selective and facilitatory responses, which are not simply predicted by responses to single gratings presented alone. This complexity is surprising, since excitatory neurons in V1 are considered to be mainly tuned to single preferred orientations. Here we examined the consequences for visual processing of two alternative connectivity schemes: in the first case, local connections are aligned with visual properties inherited from feedforward input (a 'like-to-like' scheme specifically connecting neurons that share similar preferred orientations); in the second case, local connections group neurons into excitatory subnetworks that combine and amplify multiple feedforward visual properties (a 'feature binding' scheme). By comparing predictions from large scale computational models with in vivo recordings of visual representations in mouse V1, we found that responses to plaid stimuli were best explained by assuming feature binding connectivity. Unlike under the like-tolike scheme, selective amplification within feature-binding excitatory subnetworks replicated experimentally observed facilitatory responses to plaid stimuli; explained selective plaid responses not predicted by grating selectivity; and was consistent with broad anatomical selectivity observed in mouse V1. Our results show that visual feature binding can occur through local recurrent mechanisms without requiring feedforward convergence, and that such a mechanism is consistent with visual responses and cortical anatomy in mouse V1. Author summaryThe brain is a highly complex structure, with abundant connectivity between nearby neurons in the neocortex, the outermost and evolutionarily most recent part of the brain. Although the network architecture of the neocortex can appear disordered, connections between neurons seem to follow certain rules. These rules most likely determine how information flows through the neural circuits of the brain, but the relationship between particular connectivity rules and the function of the cortical network is not known. We built models of visual cortex in the mouse, assuming distinct rules for connectivity, and examined how the various rules changed the way the models responded to visual stimuli.We also recorded responses to visual stimuli of populations of neurons in anesthetized mice, and compared these responses with our model predictions. We found that connections in neocortex probably follow a connectivity rule that groups together neurons that differ in ...

Feedforward mechanisms of masking in mouse V1

“…While the diversity of orientation tuning is largest in mouse V1, primate V1 does contain neurons with low orientation selectivity (Ringach et al, 2002;Tan et al, 2014), whereas neurons in cat V1 are uniformly sharply tuned (Scholl et al 2013). Thus, this explains the frequency of observations of facilitation being highest in the mouse (Juavinett & Callaway, 2015;Muir et al, 2015;Muir et al, 2017;Palagina et al, 2017), present in the primate (Guan et al, 2020), and absent in the cat (DeAngelis et al 1992, Priebe & Ferster 2006. Finally, our work resolves that differences in the prevalence of facilitation and suppression across mouse studies lies in the definition of the test stimulus.…”

“…To measure the interactions between superimposed oriented gratings in mouse V1, we measured GCAMP6 calcium signals from genetically identified pyramidal cells using a twophoton microscope while presenting single drifting gratings of the preferred direction and superimposed orthogonal direction (plaids, Figure 1A). As previously reported, neurons varied widely in their responses to plaid stimuli (Juavinett & Callaway, 2015;Muir et al, 2015;Muir et al, 2017, Palagina et al, 2017Ringach et al, 2020). At one end of the spectrum were neurons whose responses resembled those seen in cats, exhibiting robust decreases in response amplitude to the plaid stimulus compared to their preferred grating alone (Figure 1B, left panel).…”

“…Furthermore, experiments which measure responses to plaids in mouse V1 must consider the effect of spatial phase. For example, plaids have been used to measure selectivity to complex motion patterns in mouse V1 (Juavinett & Callaway, 2015;Muir et al, 2015;Muir et al, 2017;Palagina et al, 2017) and the reported effects might be reconsidered in light of our results. We have likely underestimated the total range of phaseinduced modulations across the population as we employed a limited number of spatial phases which may not match the peaks or troughs of the phase interactions (Figure 6 A,B).…”

“…These responses are then transformed by both thalamic and cortical nonlinearities, resulting in suppression (Priebe & Ferster 2006, Li et al, 2006Koelling et al, 2008). While cross-orientation suppression is pervasive in cat V1 (DeAngelis et al, 1992;Priebe & Ferster, 2006), the impact of orthogonal stimuli is more heterogeneous in mice and primates (Juavinett & Callaway, 2015;Muir et al, 2015;Muir et al, 2017, Palagina et al, 2017Ringach et al, 2020;Guan et al, 2020). Although many mouse and primate V1 neurons exhibit robust cross-orientation suppression, other neurons exhibit increased responses when the mask is overlaid.…”