Kniveton & Stephenson (1970) showed that middle‐class children, when permitted to play with toys before presentation of a filmed model, will later imitate the model less than those who do not have this pre‐experience. An experiment is described which tests the same hypothesis for working‐class children. The hypothesis is rejected. This apparent interaction between social class and pre‐experience is confirmed in a second experiment. A further hypothesis was tested that the class difference in response to pre‐experience will vary inversely with the magnitude of pre‐experience. This was not supported.