Abstract: | In 2 experiments, a total of 160 nonprogrammers (undergraduates) learned a computer programming language and took a multileveled transfer posttest. Ss who were given pretraining with a concrete model of the computer before learning performed better on novel transfer and worse on near transfer relative to no-pretraining Ss including Ss who were given posttraining with the same model after learning. Similar differences in the pattern of transfer were noted for Ss who controlled the order of presentation of frames relative to E-controlled Ss. Results suggest that availability and activation of assimilative sets influence the structure of learning outcomes. (PsycINFO Database Record (c) 2010 APA, all rights reserved) |