Residual effects in recall after a stimulus suffix

Abstract

Two experiments investigated serial recall with eight-word lists in which the frequency rating of the terminal word was manipulated. The effect on recall of two kinds of verbal stimulus suffix as well as a control noise suffix was also tested. Recall for the terminal items in the lists was analysed. Experiment I showed that interference with a verbal suffix, irrespective of the nature of the suffix, was at least 40 per cent compared with a noise suffix for homogeneous lists in which the terminal list-word was a member of the same standard recall set as the previous words in the list. This suffix interference was considerably reduced with heterogeneous lists in which a new terminal list-word was introduced. However, the reduction in interference was less when the verbal suffix itself was also a new word compared with a standard verbal suffix. Experiment II introduced an additional end-of-list cue before the suffix, and demonstrated that this cue did not affect the pattern of findings of the previous experiment. In the heterogeneous lists of both experiments, new words of high frequency were not differentiated in recall from new words of low frequency, but the latter condition was characterized by an interesting increase in recall error at earlier serial positions in the list.