Context-dependent selectivity to natural images in the retina

Stimulus invariant aspects of the retinal code drive discriminability of natural scenes

Everything that the brain sees must first be encoded by the retina, which maintains a reliable representation of the visual world in many different, complex natural scenes while also adapting to stimulus changes. This study quantifies whether and how the brain selectively encodes stimulus features about scene identity in complex naturalistic environments. While a wealth of previous work has dug into the static and dynamic features of the population code in retinal ganglion cells, less is known about how populations form both flexible and reliable encoding in natural moving scenes. We record from the larval salamander retina responding to five different natural movies, over many repeats, and use these data to characterize the population code in terms of single-cell fluctuations in rate and pairwise couplings between cells. Decomposing the population code into independent and cell-cell interactions reveals how broad scene structure is encoded in the retinal output. while the single-cell activity adapts to different stimuli, the population structure captured in the sparse, strong couplings is consistent across natural movies as well as synthetic stimuli. We show that these interactions contribute to encoding scene identity. We also demonstrate that this structure likely arises in part from shared bipolar cell input as well as from gap junctions between retinal ganglion cells and amacrine cells.

Biophysical neural adaptation mechanisms enable artificial neural networks to capture dynamic retinal computation

Adaptation is a universal aspect of neural systems that changes circuit computations to match prevailing inputs. These changes facilitate efficient encoding of sensory inputs while avoiding saturation. Conventional artificial neural networks (ANNs) have limited adaptive capabilities, hindering their ability to reliably predict neural output under dynamic input conditions. Can embedding neural adaptive mechanisms in ANNs improve their performance? To answer this question, we develop a new deep learning model of the retina that incorporates the biophysics of photoreceptor adaptation at the front-end of conventional convolutional neural networks (CNNs). These conventional CNNs build on 'Deep Retina,' a previously developed model of retinal ganglion cell (RGC) activity. CNNs that include this new photoreceptor layer outperform conventional CNN models at predicting male and female primate and rat RGC responses to naturalistic stimuli that include dynamic local intensity changes and large changes in the ambient illumination. These improved predictions result directly from adaptation within the phototransduction cascade. This research underscores the potential of embedding models of neural adaptation in ANNs and using them to determine how neural circuits manage the complexities of encoding natural inputs that are dynamic and span a large range of light levels.

Pattern completion and disruption characterize contextual modulation in the visual cortex

Vision is fundamentally context-dependent, with neuronal responses influenced not just by local features but also by surrounding contextual information. In the visual cortex, studies using simple grating stimuli indicate that congruent stimuli - where the center and surround share the same orientation - are more inhibitory than when orientations are orthogonal, potentially serving redundancy reduction and predictive coding. Understanding these center-surround interactions in relation to natural image statistics is challenging due to the high dimensionality of the stimulus space, yet crucial for deciphering the neuronal code of real-world sensory processing. Utilizing large-scale recordings from mouse V1, we trained convolutional neural networks (CNNs) to predict and synthesize surround patterns that either optimally suppressed or enhanced responses to center stimuli, confirmed by in vivo experiments. Contrary to the notion that congruent stimuli are suppressive, we found that surrounds that completed patterns based on natural image statistics were facilitatory, while disruptive surrounds were suppressive. Applying our CNN image synthesis method in macaque V1, we discovered that pattern completion within the near surround occurred more frequently with excitatory than with inhibitory surrounds, suggesting that our results in mice are conserved in macaques. Further, experiments and model analyses confirmed previous studies reporting the opposite effect with grating stimuli in both species. Using the MICrONS functional connectomics dataset, we observed that neurons with similar feature selectivity formed excitatory connections regardless of their receptive field overlap, aligning with the pattern completion phenomenon observed for excitatory surrounds. Finally, our empirical results emerged in a normative model of perception implementing Bayesian inference, where neuronal responses are modulated by prior knowledge of natural scene statistics. In summary, our findings identify a novel relationship between contextual information and natural scene statistics and provide evidence for a role of contextual modulation in hierarchical inference.

An actor-model framework for visual sensory encoding

A fundamental challenge in neuroengineering is determining a proper artificial input to a sensory system that yields the desired perception. In neuroprosthetics, this process is known as artificial sensory encoding, and it holds a crucial role in prosthetic devices restoring sensory perception in individuals with disabilities. For example, in visual prostheses, one key aspect of artificial image encoding is to downsample images captured by a camera to a size matching the number of inputs and resolution of the prosthesis. Here, we show that downsampling an image using the inherent computation of the retinal network yields better performance compared to learning-free downsampling methods. We have validated a learning-based approach (actor-model framework) that exploits the signal transformation from photoreceptors to retinal ganglion cells measured in explanted mouse retinas. The actor-model framework generates downsampled images eliciting a neuronal response in-silico and ex-vivo with higher neuronal reliability than the one produced by a learning-free approach. During the learning process, the actor network learns to optimize contrast and the kernel's weights. This methodological approach might guide future artificial image encoding strategies for visual prostheses. Ultimately, this framework could be applicable for encoding strategies in other sensory prostheses such as cochlear or limb.