A simple model of binocular luster

The effect of stimulus size on binocular luster and its implications for the interocular conflict model

Using a series of dichoptic center-ring-surround stimuli in a psychophysical experiment, we examined how the phenomenon of binocular luster is affected by the size of the central target area. Generally, we found that the lustrous sensation continuously decreases with increasing patch size. However, this effect also depends on further stimulus features such as the interocular contrast polarity pairings and the width of the ring element. To account for these complex influences on the magnitude of the lustrous response, significant extensions had to be made to our interocular conflict model. We present two improved versions of our model, a summation and an averaging model, which differ in the way the local conflict values are spatially integrated. Both versions have a very high predictive power. At least for the summation model, we show that the modifications are in good agreement with physiological processes. In particular, we provide evidence that the influence of stimulus size on the lustrous effect can be explained by the combination of two mechanisms: An increasing receptive field size of contrast detector cells and at the same time a decrease of the local density of these cells with increasing retinal eccentricity.

Binocular luster elicited by isoluminant chromatic stimuli relies on mechanisms similar to those in the achromatic case

The phenomenon of binocular luster can be evoked by simple dichoptic center-surround stimuli showing a luminance contrast difference between the eyes. Previous findings support the idea that this phenomenon is mediated by a low-level conflict mechanism that integrates the monocular signals from different types of contrast detector cells. Also, isoluminant stimuli with different chromatic contrasts between eyes can trigger sensations of luster. Here, we investigate whether the lustrous impression in such purely chromatic stimuli depends on interocular contrast differences and in particular on interocular contrast polarity pairings in a similar way as in the achromatic case. In our experiments, we measured the magnitude of the lustrous response using a series of isoluminant dichoptic center-ring-surround stimuli with varying ring width whose chromatic properties were varied along the red-green and blue-yellow cardinal directions. The trends in the data were very similar to those of our former study with achromatic stimuli, indicating similar mechanisms in both cases. The empirical luster data could also be predicted fairly well by a chromatic version of our interocular conflict model (with overall R2 values between 0.577 and 0.639), for which two different receptive field models were used, simulating the behavior of color-sensitive double-opponent cells in V1.