The assimilation and accommodation was tested. Both assimilation and accommodation were hypothesized to be processes that produce changes in cognitive structure--internal representations used to organize behavior. The hypothesis is that, in contrast to assimilation, the changes in structure caused by accommodation are only partially reversible. The experiment was a multidimensional scaling procedure repeated three times to reproduce walking distances between buildings on a familiar college campus. The second time, the students in the assimilation condition estimated distances between additional buildings, whereas the students in the accommodation condition were asked to estimate distances as though an imaginary barrier blocked the middle of the campus. The third time, the students were asked to reproduce their original estimates. The results showed that the accommodation condition had a statistically significant effect on the judgments made on the third trial, thus supporting the hypothesis that accommodation produces changes in cognitive structures that are not fully reversible.