The presentation of the results is organized into three sections. First we review the precision of the asymmetric color-matching task by comparing the repeated matches made by observers. The precision of repeated matches is important because we use the covariance matrix derived from these matches to measure the size of the pattern effects on color and to evaluate the precision of the model fits to the data.
The empirical properties of contrast-homogeneity and pattern-superposition are reviewed in the second and third sections. All linear models of the asymmetric matches imply that these two general properties should hold. Hence, we evaluate these properties prior to considering specific linear models in the next section.