DA-023

Interval between acquisition and extinction (A-E interval) was varied in order to investigate effects of regularly alternated partial reinforcement on resistance to extinction and to assess the notion of pattern memory which was suggested on the basis of single alternation study. Random alternation (RA) and double alternation (DA) groups were trained with extensive trials and then extinguished under one-day's (RA1, DA1) or 23-days' (RA23, DA23) A-E interval. RA1 was more resistant to extinction than DA1, but DA23 did not differ from RA23. Moreover, extinction performance of DA schedule was retained well in spite of the extension of A-E interval, unlike that of RA schedule. These extinction results cannot be interpreted consistently even by the N-length variables. The persistence of DA schedule shown in delayed extinction was discussed in terms of the pattern memory assumption.