Blocking of stimulus control was studied in a simultaneous discrimination task. During Phase I of the experiment, stimulus Dimension A was relevant to solution for Experimentals and stimulus Dimension C for Controls; during Phase II, Dimensions A and B were redundant and relevant for both Groups; during Phase III, both Groups received unreinforced test trials on Dimension B. In Experiment 1, Experimental subjects demonstrated significantly greater increases in response latency during Phase III than Controls. In Experiment 2, Experimental subjects demonstrated greater increases when testing on B followed immediately after Phase II, but not when test trials on A preceded testing on B.