Rats trained on an insoluble problem in a Lashley jumping apparatus built up fixated responses which persisted during 200 trials of a soluble problem situation. When the animals were trained to walk to the correct window they learned without difficulty, although this response did not transfer to trials in which jumping to the correct window was required. "These results were interpreted to mean that one cannot account for fixations on the basis of a rat's having no reason to believe that alternative responses would be better than its stereotyped responses." (PsycINFO Database Record (c) 2006 APA, all rights reserved)