The CODE Theory of Visual Attention (CTVA) is a mathematical modelexplaining the effects of grouping by proximity and distance upon reactiontimes and accuracy of response with regard to elements in the visual display.The predictions of the theory agree quite acceptably in one and twodimensions (CTVA-2D) with the experimental results (reaction times andaccuracy of response). The difference between reaction-times for thecompatible and incompatible responses, known as the responsecompatibilityeffect, is also acceptably predicted, except at small distancesand high number of distractors. Further results using the same paradigm ateven smaller distances have been now obtained, showing greaterdiscrepancies. Then, we have introduced a method to evaluate the strength ofsensory evidence (eta parameter), which takes grouping by similarity intoaccount and minimizes these discrepancies