Improving Production of Treated and Untreated Verbs in Aphasia: A Meta-Analysis

Abstract

Background: Demographic and clinical predictors of aphasia recovery have been identified in the literature. However, little attention has been devoted to identifying and distinguishing predictors of improvement for different outcomes, e.g., production of treated vs. untreated materials. These outcomes may rely on different mechanisms, and therefore be predicted by different variables. Furthermore, treatment features are not typically accounted for when studying predictors of aphasia recovery. This is partly due to the small numbers of cases reported in studies, but also to limitations of data analysis techniques usually employed.Method: We reviewed the literature on predictors of aphasia recovery, and conducted a meta-analysis of single-case studies designed to assess the efficacy of treatments for verb production. The contribution of demographic, clinical, and treatment-related variables was assessed by means of Random Forests (a machine-learning technique used in classification and regression). Two outcomes were investigated: production of treated (for 142 patients) and untreated verbs (for 166 patients).Results: Improved production of treated verbs was predicted by a three-way interaction of pre-treatment scores on tests for verb comprehension and word repetition, and the frequency of treatment sessions. Improvement in production of untreated verbs was predicted by an interaction including the use of morphological cues, presence of grammatical impairment, pre-treatment scores on a test for noun comprehension, and frequency of treatment sessions.Conclusion: Improvement in the production of treated verbs occurs frequently. It may depend on restoring access to and/or knowledge of lexeme representations, and requires relative sparing of semantic knowledge (as measured by verb comprehension) and phonological output abilities (including working memory, as measured by word repetition). Improvement in the production of untreated verbs has not been reported very often. It may depend on the nature of impaired language representations, and the type of knowledge engaged by treatment: it is more likely to occur where abstract features (semantic and/or grammatical) are damaged and treated.

‘What's cooking?’ A comparison of an activity‐oriented and a table‐top programme of therapy on the language performance of people with aphasia

“…This result is generally in line with studies where verb naming has been targeted within a sentence context (Thompson et al 2013). It has been suggested that sentence level treatment may be more effective at encouraging generalization, because in addition to rebuilding specific knowledge of lexemes it trains the semantic and syntactic features associated with the verb (de Aguiar et al 2016). However, unlike other sentence level treatments (e.g., Thompson et al 2013), our therapy did not involve direct training of argument structure or morphology.…”

“…It has been suggested that sentence level treatment may be more effective at encouraging generalization, because in addition to rebuilding specific knowledge of lexemes it trains the semantic and syntactic features associated with the verb (de Aguiar et al . ). However, unlike other sentence level treatments (e.g., Thompson et al .…”

Identifying the components of a successful spoken naming therapy: a meta-analysis of word-finding interventions for adults with aphasia

“…In comparison, de Aguiar et al (2016) is distinctive. It applied a robust technique (random forest) that is novel for the field, and included clinically-relevant variables for simultaneous evaluation.…”

“…It is these ingredients that speech and language therapists are most interested in, when designing a therapy to treat a client. With the exception of de Aguiar et al (2016), most reviews tended to either frame their discussions of therapies using broad terminologies like phonological or semantic approaches without examining the specific components; or they may simply narrow their focus and manipulate just one component (e.g., number of treatment items by , but in doing so, forced to commit the non-realistic assumption of ceteris paribus. Consequently, outstanding issues remain unanswered.…”

“…Random forest is deemed suitable for our purposes because: Firstly, it has the versatility to handle data from different designs, including single case studies (see de Aguiar et al, 2016, for an example). On the other hand, classical meta-analytic approaches do not lend themselves easily to single case research: Their typical formulae set out to summarise effect sizes extracted from group studies, not single case studies (specifically, they required group mean differences and standard deviations at the group level to be entered as input, see Borenstein, Hedges, Higgins, & Rothstein, 2009; also see Shadish, Rindskopf & Hedges, 2008, for a detailed discussion on related challenges).…”