Osgood's transfer surface: Extension and test

Abstract

Summary The purpose of the experiment was to determine transfer effects for (a) Osgood's Surface as presently defined, and (b) additional points on the enlarged Surface resulting from extension of interlist stimulus similarity from identical, similar, and unrelated to include opposed and antonymous. Subjects learned a final common list after learning one of twenty-five different first lists having one of five categories of interlist stimulus similarity (identical, similar, unrelated, opposed, or antonymous) combined factorially with one of the same five conditions of interlist response similarity. Results suggested (acceptable levels of statistical significance were hardly ever achieved) were (a) when stimuli on both lists are identical or similar, positive transfer occurs for opposed and antonymous response relations as well as for those of identity and similarity; (b) when responses are identical on both lists, only high degrees of stimulus similarity produce positive transfer; (c) when stimuli are meaningfully opposed or antonymous, positive transfer occurs only when responses are also meaningfully opposed or antonymous; and (d) simultaneous variation of interlist stimulus and response similarity tends to produce increasing positive transfer as relation between lists, either similarity or opposition, increases.