Towards a Theory of Learning for Naming Rehabilitation: Retrieval Practice, Retrieval Effort, and Spacing Effects

• ¹ Moss Rehabilitation Research Institute, United States

Background. Rehabilitation practices can benefit from learning theories that have the potential illuminate the mechanism of action of conventional treatments. This study sought to establish an empirical foundation for a theory of learning for naming rehabilitation that is centered on retrieval practice, retrieval effort, and spacing. Psychological research has shown that retrieval practice (the act of retrieval from long-term memory) confers powerful and persistent learning, particularly when retrieval is effortful. The spacing effect refers to the ubiquitous advantage to learning when repeated training opportunities for individual items are distributed over time, rather than massed. In our prior work [1], we found that retrieval practice improved common object naming in people with chronic aphasia (PWA). Building on this, the goals of the present study were to: (1) investigate the impact of retrieval practice in a new performance domain (proper noun naming) in PWA by comparing retrieval practice training to an “errorless” condition where presentation of the name for repetition averted the need for retrieval from long-term memory; (2) determine whether greater retrieval effort during retrieval practice is associated with greater retention of training benefits; (3) determine whether spacing confers longer-lasting learning than massing. Methods. Four PWA with naming impairment named and gave familiarity ratings to a corpus of 700 pictures of proper noun entities twice over two weeks. For each participant, we selected items the participant knew recognized but could not consistently name for assignment into the conditions, with a minimum of 36 (max=72) items per condition across participants. The design involved a 2-level factor of type of training (retrieval practice versus errorless learning, i.e., repetition) and a factor of spacing, which included a massed condition (lag 1) and three spaced conditions (lags 5, 15, and 30). Lag corresponded to the number of training trials for other items that intervened between three presentations of an item for retrieval practice or repetition training. On a repetition trial, the name was presented (seen/heard) and the participant repeated the name at picture onset. On a naming trial, only the picture was presented. All trials ended in feedback (i.e., the name was presented). Primary outcome was naming performance on a retention test administered 1-day following training, with a 1-week follow-up test administered to measure persistence of the effects. Results & Conclusions. Mixed regression analyses revealed that the naming condition was associated with superior performance over repetition, observed both at the retention test (p=.001) and follow-up (p=.01; Figure 1, left panel). Also, spaced training conferred superior benefits compared to massed, both at retention test (p<.001) and follow-up (p=.006; Figure 1, right panel). An analysis of the spaced lags in the naming condition revealed that though increasing lag made retrieval practice more effortful (i.e., error-prone) during training, increasing lag conferred more powerful learning at retention test. The present study provides definitive evidence of the relevance of retrieval practice, retrieval effort, and spacing for optimizing existing treatments, their explanatory power, and their importance in driving future research.