Responses to conflicting binocular stimuli in mouse primary visual cortex

Background:

Binocular vision enables organisms to perceive a unified, three-dimensional view of the world by integrating two slightly different images from each eye. Under normal conditions, the visual system effectively fuses modest spatial disparities between retinal images to support depth perception, or stereopsis (Parker, 2007). This ability depends on the brain’s capacity to match corresponding visual features across the two eyes within a certain disparity range.

However, when disparities exceed this range either due to natural variance or pathological conditions such as strabismus, the visual system fails to achieve fusion, resulting in phenomena like diplopia (double vision) or binocular rivalry (Harrad et al., 1996; Blake & Logothetis, 2002). The neural mechanisms that underlie the transition from binocular integration to perceptual suppression under conditions of interocular conflict remain incompletely understood.

The authors utilized the mouse visual system to investigate how circuits in the primary visual cortex (V1) respond to qualitatively different types of binocular conflict. V1 is the first cortical stage where inputs from the two eyes converge, making it a critical site for binocular integration and disparity processing. They employed visual evoked potentials (VEPs), extracellular recordings, and two-photon calcium imaging to examine neural responses to two forms of interocular disparity: one in which the stimuli maintain the same orientation across both eyes but differ in phase, and another where the stimuli differ in orientation between the eyes, mimicking the conditions commonly used to study binocular rivalry. VEPs capture the summed electrical activity of neuronal populations in response to visual input, providing a population-level readout of how cortical responsiveness changes under different binocular conditions.

Key findings:

1. Binocular conflict suppresses cortical responses

When the two eyes viewed stimuli differing in phase or orientation, visually evoked potentials (VEPs) recorded from binocular V1 were markedly reduced, indicating that cortical responses are actively suppressed under interocular conflict. This suppression was evident even at low contrast levels, suggesting that V1 neurons detect and down-weight discordant binocular input early in the visual processing hierarchy.

 

 

 

 

Figure 1. Binocular stimuli with an interocular phase or orientation disparity elicit smaller visually evoked potentials. (A) Schematic of mouse brain showing placement of a chronic recording electrode in binocular V1 (V1b). (B) Mice viewed phase reversing sinusoidal grating stimuli on a dichoptic display. (C) Phase offset experimental design. The contralateral (Contra) eye viewed full-contrast stimuli, while the ipsilateral (Ipsi) eye viewed stimuli ranging from 0% to 50% contrast, presented either in phase (Concordant) or out of phase (Phase Offset) with the contra eye stimulus. (E) Orthogonal orientation experimental design. Preprint figures placed in the Public Domain. 

2. High sensitivity to subtle mismatches

Systematic variation of interocular orientation revealed that a 10° orientation difference between the eyes was sufficient to significantly attenuate VEP amplitude. This demonstrates that binocular V1 circuits are highly sensitive to even minor disparities, consistent with the role of early visual cortex in establishing precise interocular correspondence required for depth perception and binocular fusion.

3. Distinct neural mechanisms for different conflict types

Phase and orientation disparities recruited different temporal components of the VEP and distinct spiking dynamics. Phase-offset stimuli selectively suppressed the early negative VEP component and reduced early cortical spiking, whereas orthogonal stimuli diminished the later positive component but elicited prolonged activity across layers. These dissociable signatures indicate that V1 engages separate circuit mechanisms to process phase and orientation mismatches.

4. Layer- and cell-type–specific modulation

Laminar recordings revealed that conflict-related suppression and enhancement patterns depend on both cortical layer and neuronal subtype. Regular-spiking (excitatory) neurons showed early response suppression for phase-offset stimuli in layers 2/3, 5, and 6 but increased late activity under orthogonal stimulation. Fast-spiking (inhibitory) neurons exhibited extended activation primarily in layer 2/3 during orthogonal conflict. These results demonstrate structured, layer-specific reorganization of intracortical dynamics in response to binocular disparity.

Figure 2. Fast-spiking single unit responses to different stimulus conditions. Preprint figure placed in the Public Domain. 

Z-scored raster plots of regular-spiking (RS) single units for Monocular (A), Concordant (B), Phase Offset (C), and Orthogonal (D) stimuli. Units were assigned to L2/3, L4, L5, or L6 based on the location of the electrode contact with the maximal single unit waveform, then sorted based on their average activity between 40-80ms in the Monocular condition. 

5. Disinhibition drives sustained activity under orthogonal conflict

Two-photon calcium imaging identified a disinhibitory circuit mechanism underlying the prolonged activation observed during orthogonal stimulation. Activity of somatostatin-positive (SOM⁺) interneurons, which normally suppress excitatory and PV⁺ cells, was strongly reduced during orthogonal conflict, leading to sustained excitation in both populations. This finding links the suppression of SOM⁺ activity to persistent cortical firing, providing a cellular explanation for how interocular orientation conflict can maintain extended activity associated with binocular rivalry.

What I like about the preprint?

To me, what is most compelling about these findings is how a single cortical network, mouse V1, flexibly reorganizes its activity depending on the nature of binocular conflict. The same neurons that typically fuse signals from both eyes can, under conflicting conditions, shift into distinct response modes either suppressing early input or sustaining prolonged activation. This adaptability reveals a remarkable computational versatility within the primary sensory cortex, showing that early visual areas are not merely feedforward filters but dynamic processors sensitive to context and stimulus congruence.

Equally striking is the ability to trace these divergent network states to specific interneuron mechanisms, particularly the unexpected disinhibition of SOM⁺ interneurons during orientation conflict. Even minor binocular mismatches can reconfigure inhibitory dynamics, revealing the fine-tuned adaptability of cortical networks. It is fascinating that such a localized circuit adjustment could underlie processes like binocular rivalry linking the dynamics of single interneuron classes to the broader question of how the brain resolves disagreement between the eyes.

Questions to the authors:

• Is interocular suppression in mouse V1 sufficient to drive perceptual exclusion?
• What are the developmental and plasticity mechanisms shaping binocular conflict responses?
• Are there distinct behavioral consequences associated with each form of binocular conflict in mice?

Future directions:

While the preprint in itself covers a wide range of questions regarding biocularity, it would be interesting to see what is the role of VIP+ interneurons in interocular conflict resolution? Given that vasoactive intestinal peptide-expressing (VIP+) interneurons disinhibit SOM+ cells, they are strong candidates for top-down or lateral modulation in binocular conflict. Their role remains unexplored in this context.

Tags: binocular, cortex, two-photon, vision

doi: https://doi.org/10.1242/prelights.41880

Read preprint (1 votes)