Transfer of new learning in memory-impaired patients

Previous research has produced conflicting evidence concerning transfer of new learning by amnesic patients. The present experiment investigated the hypothesis that different numbers of learning trials account for differences in transfer, such that the greater the number of repetitions of material in identical stimulus contexts the poorer the transfer. Six memory-impaired patients and six control subjects attempted to learn the names of business-related documents in response to descriptive definitions. Learning continued until one of the following criteria was reached: 50% correct, 100% correct, 100% correct plus 10 trials. In a transfer task, subjects were then asked to produce the target responses to altered definitional cues. The results of the experiment demonstrated that, contrary to prediction, transfer improved with numbers of learning trials. Results are consistent with the view that continued study of information allows better integration of new learning with prior knowledge and correspondingly higher levels of transfer. The theoretical implications of the findings are discussed in terms of the declarative/procedural and the episodic/semantic memory distinction. It is suggested that memory-impaired patients are capable of acquiring new semantic information although not at a normal rate. Implications for memory rehabilitation are also outlined.