Symmetric matching to sample in 2-year-old children

The emergence of symmetric matching to sample was studied in fourteen e-yea r-old children. Three two-choice, arbitrary matching-to-sample tasks were a rranged successively. Each task was followed by a test of symmetric respond ing. The behavior required by the tasks was taught with sample stimulus-con trol shaping (Zygmont, Lazar, Dube, & McIlvane, 1992). Five children did no t learn identity matching, which is required for sample stimulus-control sh aping. Two others showed a deterioration of baseline responding when the fi rst test was given. The participation of these seven children was stopped. The others showed symmetric responding on all tests. These results demonstr ate generalized symmetric matching to sample (one aspect of stimulus equiva lence) in 2-year-old children.