Interserial and intraserial interference processes in morning recall of mental sleep experience.

This study investigated ways in which predictions from interference theory might explain partial or total failure in the morning recall of mental sleep experience (MSE) episodes reported after provoked night awakening. Interference could take place either between the contents of one MSE episode (intraserial interference) or between the contents of different MSE episodes (interserial interference) and would be more likely to affect morning recall where similar (i.e., interrelated) contents in MSEs are involved. If interference were mainly intraserial, in the morning there would be no change in the difficulty of retrieving interrelated contents, whereas, if it were mainly interserial, there would be increased difficulty in the retrieval of such contents. All the possible pairs of reports obtained by awakening 8 subjects four times in REM sleep during each of four experimental nights spaced at weekly intervals were scored for interrelated contents, using Clark's (1970) system of analysis of paired units, which operates by matching linguistic features. Results showed that interrelations in morning report pairs are as frequent as in night report pairs, maintaining the same linguistic characteristics and with little transfer of contents from one MSE to another. Overall these findings are compatible with the hypothesis that interference is primarily intraserial rather than interserial. It is consequently suggested that morning recall failure may be due to temporary blocking of retrieval.