You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster C53, Thursday, November 9, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Incremental learning and lexical access: Evidence from aphasia

Julia Schuchard1, Erica L. Middleton1;1Moss Rehabilitation Research Institute

Introduction. Incremental learning accounts of lexical access for speech production posit that each instance of word retrieval (e.g., naming an object) adjusts the strength of the word’s semantic-lexical connections, facilitating its future retrieval. Here we hypothesized that a use-dependent, incremental learning mechanism operates at each of two stages of lexical access that involve retrieval of a word from semantics (“Stage 1”), followed by retrieval of its constituent phonemes (“Stage 2”). To examine this possibility, we investigated two naming treatment techniques expected to differentially strengthen the two stages of lexical access: retrieval practice (i.e., practice retrieving names from long-term memory) and errorless learning (i.e., patient hears the name and orally repeats it). Because producing a name during retrieval practice requires both Stage 1 and Stage 2, this technique should use and strengthen item-specific connections in both stages. Because errorless learning involves oral repetition of names, which circumvents semantically-driven retrieval, this technique should primarily use and strengthen item-specific connections in Stage 2. Method. In two experiments, we examined how naming training involving retrieval practice versus errorless learning impacted post-training naming performance. Using a case comparison design, Experiment 1 studied two individuals with aphasia (S1 and S2) who were selected based on evidence that S1’s naming impairment primarily resulted from difficulty at Stage 1, and S2’s naming impairment primarily resulted from difficulty at Stage 2. Experiment 2 tested the effects of the training conditions on items that elicited difficulty at Stage 1 or Stage 2 for each of ten additional individuals with aphasia. In both experiments, pictures of common objects that elicited naming errors in pretesting for each participant were divided in matched fashion to the retrieval practice and errorless learning conditions. Following initial familiarization of each picture with its associated name, opportunities were provided to retrieve the name (retrieval practice) or repeat the name (errorless learning), with correct-answer feedback following all trials. Retention tests of naming for the trained items were administered without feedback one day and one week after training. Naming accuracy on these tests was the primary outcome variable, analyzed with logistic regression. Results. The results showed significant interaction effects between the type of training (retrieval practice versus errorless learning) and the stage of lexical access at which items presented difficulty (Stage 1 versus Stage 2). In Experiment 1, retrieval practice resulted in significantly higher retention test accuracy than errorless learning for S1 (Stage 1-type impairment), whereas no significant difference between the two conditions was observed for S2 (Stage 2-type impairment). In Experiment 2, retrieval practice resulted in higher retention test accuracy compared to errorless learning for Stage 1 items, whereas the opposite pattern was observed for Stage 2 items. Discussion. The results suggest that retrieval practice strengthens item-specific connections in Stage 1 to a greater extent than errorless learning, whereas errorless learning primarily strengthens Stage 2 connections. These findings provide empirical evidence to motivate the integration of use-dependent, incremental learning mechanisms into a two-stage framework of lexical access.

Topic Area: Meaning: Lexical Semantics

Back to Poster Schedule