Constrained inference in sparse coding reproduces contextual effects and predicts laminar neural dynamics

Abstract

A central goal in visual neuroscience is to understand computational mechanisms and to identify neural structures responsible for integrating local visual features into global representations. When probed with complex stimuli that extend beyond their classical receptive field, neurons display non-linear behaviours indicative of such integration processes already in early stages of visual processing. Recently some progress has been made in explaining these effects from first principles by sparse coding models with a neurophysiologically realistic inference dynamics. They reproduce some of the complex response Author summaryAn influential hypothesis about how the brain processes visual information posits that each given stimulus should be efficiently encoded using only a small number of cells. This idea led to the development of a class of models that provided a functional explanation for various response properties of visual neurons, including the non-linear modulations observed when localized stimuli are placed in a broader spatial context. However, it remains to be clarified through which anatomical structures and neural connectivities a network in the cortex could perform the computations that these models require. In this paper we propose a model for encoding spatially extended visual scenes. Imposing the constraint that neurons in visual cortex have direct access only to small portions of the visual field we derive a simple yet realistic neural population dynamics. Connectivities optimized for natural scenes conform with anatomical findings and the resulting model reproduces a broad set of physiological observations, while exposing the neural mechanisms relevant for spatio-temporal information integration.

Back-Propagation Learning in Deep Spike-By-Spike Networks

“…These so called inference populations iteratively perform inference on the potential causes of their inputs from each stochastic spike impinging on the population (Ernst et al, 2007). While the corresponding dynamics might appear artificial, it captures the essence of models with more biologically realistic neurons (Rozell et al, 2008; Moreno-Bote and Drugowitsch, 2015; Zhu and Rozell, 2015) that perform sparse efficient coding, which is a leading hypothesis for understanding coding in the brain (Olshausen and Field, 2006; Spanne and Jörntell, 2015; Zhu and Rozell, 2015; Capparelli et al, 2019).…”

Sparse long-range connections in visual cortex for cost-efficient small-world networks

“…From their extraordinarily long wiring (up to 2-3 mm), LRCs are distinguished from local connections of a short lateral spread (up to 0.5 mm) 22–26 (Fig 1c). Previous studies have suggested that LRCs may play a particular role in visual processing, compensating for their large wiring cost, and possibly contributing to layer-wide functions such as contextual modulation at early stages 27–29 . However, the exact functions of LRCs for visual processing are still unknown.…”