A critical evaluation of computational mechanisms of binocular disparity processing

The past decades of research in visual neuroscience have generated a large and disparate body of literature on the computation of binocular disparity in the primary visual cortex. Models have been proposed to describe specific phenomena, yet we lack a theoretical framework which is grounded in neurophysiology and also explains the effectiveness of disparity computation. Here, we examine neural circuits that are thought to play an important role in the computation of binocular disparity. Starting with the binocular energy model (Ohzawa et al. 1990), we consider plausible extensions which include suppressive mechanisms from units tuned to different phase disparities (Tanabe et al. 2011), which is formerly theorized to perform false disparity detection (Read & Cumming 2007) as well as coarse-to-fine (Menz & Freeman 2004a,b) and recurrent processing (Samonds et al. 2013). We rigorously cross-examine the consistency of these circuits with neurophysiology data including ocular dominance and binocular modulation (Ohzawa & Freeman 1990), spike-triggered analysis and temporal dynamics of disparity tuning (Tanabe et al. 2011) and attenuation to anti-correlated stimuli (Cumming & Parker 1997; Tanabe et al. 2011). We further evaluate the ability of the resulting computational models to recover depth, both theoretically and experimentally, using a dataset of natural and synthetic images. Overall, we find that a computational model which combines suppressive mechanisms by units with non-zero phase disparity, contrast normalization as well as lateral interaction between units tuned to specific combinations of phase and position disparities, seems consistent with all of the available V1 neurophysiology data and achieves the highest accuracy in real-world depth computation