Evaluative learning is said to differ from Pavlovian associative learning in that it reflects stimulus contiguity, not contingency. Thus, evaluative learning should not be subject to stimulus competition, a proposal tested in the current experiments. Participants were presented in elemental and compound training phases with pictures of shapes as CSs. Each shape/pair of shapes was followed by a picture of a happy or an angry face as the US. In Experiments 1 and 2, evaluative ratings were collected before and after the experiment, and, in Experiment 3, participants provided evaluations online. Stimulus competition was evident in all experiments confirming that evaluative learning is sensitive to stimulus contingencies. (Contains 2 figures and 1 table.)