Here are additional materials that may be useful for understanding our papers on the neural basis of stereopsis. Available items are: visually intuitive presentations of binocular receptive field of a complex cell in the 3D space (X,Y,Z), Mathematica notebooks showing details of derivations and creation of plots presented in the papers, and some slide files for prsentations.

Visualizations

• 3-D stereoscopic visualization of binocular receptive field of complex cells
  
  A binocular receptive field model of complex cells is presented using a 3-D rendered stereogram. See a neural mechanism of stereopsis using your own stereo vision.

Mathematica Notebooks

• Prediction of disparity tuning curves from receptive fields of simple cells
  Mathematica V3.0 Notebook (552 kB) ... Mathematica V2.x Notebook (525 kB)
  Acrobat file (170 kB) ... Compressed PostScript file (160 kB, uncompressed 630 kB)
  
  This Mathematica notebook shows how a disparity tuning curve may be obtained from left and right receptive field profiles of simple cells. Predicted disparity tuning curves are shown in Perception 1995 (Figs. 2 and 11), and NeuroReport 1997 (Fig. 1).

• Disambiguating disparity in the joint orientation-phase-spatial frequency domain
  Mathematica V3.0 Notebook (184 kB) ... Mathematica V2.x Notebook (157 kB)
  Acrobat file (60 kB) ... Compressed PostScript file (60 kB, uncompressed 180 kB)
  
  This notebook shows the derivation of a constant-disparity surface in the orientation-phase-spatial frequency domain. By combining information from multiple neurons with tuning properties lying on this surface, the disparity of a target may be unambiguously determined. This is Fig. 19 of Ohzawa, et al., J. Neurophysiol. 1996.

• Disparity energy model for complex cells
  Mathematica V3.0 Notebook (296 kB) Mathematica V2.x Notebook (257 kB)
  Acrobat file (128 kB) ... Compressed PostScript file (92 kB, uncompressed 305 kB)
  
  This Mathematica notebook explores various aspects of disparity energy models as presented in Ohzawa et al., Science 1990, and Ohzawa et al., J. Neurophysiol. 1997. The opponent energy model of Bowne, McKee and Tyler, ARVO 1990 is also examined.

• Disparity energy model II: phase, position and decomposition of response
  Mathematica V3.0 Notebook (256 kB) ... Mathematica V2.x Notebook (224 kB)
  Acrobat file (110 kB) ... Compressed PostScript file (125 kB, uncompressed 420 kB)
  
  This Mathematica Notebook contains computations and derivations used to create plots in Figs. 8, 9 and 10 of our paper Ohzawa, et al. J. Neurophysiol. 1997.

Slides

• Left and right space-time receptive fields of a simple cell
  (Cover photo for NeuroReport vol. 8, no 3, 1997)
  
  Click the image for the full size version (792x528, 49 kB)
  
  Most neurons in the visual cortex can be activated by stimulation through either eye. These binocular neurons are thought to serve as the first stage of processing of stereoscopic depth perception. Left and right eye receptive fields are illustrated here for a simple cell in the primary visual cortex of the cat. Each receptive field is defined in three dimensions, two of space (X, Y) and one of time (T), as represented by a cube. To the upper-left of each cube, spatial cross-sections are shown for four different time values (front to back: T=50, 100, 150, and 200 msec). Each of these profiles is shown as a density plot, in which green and red indicate regions that are responsive to a bright bar or a dark bar, respectively. Brighter regions are more responsive. Integrating the three-dimensional data sets along the Y axis, which in each case is parallel to the axis of preferred orientation of the cell, yields simplified spatiotemporal receptive field profiles, or X-T plots. The gradual changes in spatial receptive field structure as a function of time are related to motion selectivity.
  Text and figures of the article: Ohzawa et al., NeuroReport 8 (3): R3-R12, 1997.