Frontiers in Psychology (Apr 2014)
A Link Between Lexical Competition and Fluency in Aphasia
Abstract
Background: There are at least three distinct accounts of fluency deficits in aphasia. The traditional view is that fluency deficits are specific to language production. The reduced lexical activation theory proposes that non-fluent language production in aphasia is a consequence of reduced overall activation of the lexicon, making it difficult for word representations to reach sufficient activation to drive production. The cognitive control view proposes that fluency deficits in aphasia arise from deficits of resolving competition. These three views of fluency deficits make three distinct predictions regarding inhibitory effects of phonological neighborhood density on spoken word recognition: (1) If fluency deficits are specific to language production, then fluency should not modulate neighborhood effects. (2) If fluency deficits are due to reduced lexical activation, then they should be associated with smaller effects of neighborhood density because competing neighbors will be more weakly active. (3) If fluency deficits are due to impaired competition resolution (cognitive control), then they should be associated with larger neighborhood effects due to difficulty in resolving competition between neighbors. Methods: Fourteen participants with aphasia completed the Western Aphasia Battery (WAB) measures of fluency and severity (Aphasia Quotient, AQ), the Peabody Picture Vocabulary Test (PPVT) measure of receptive vocabulary, and a spoken-to-written word matching task (which was also completed by 15 neurologically-intact controls). The target words were either high or low in phonological neighborhood density and were matched on word frequency and length. Spoken words were presented in 62 dB of white noise to make the spoken-to-written word matching task more challenging. Data were analyzed using multilevel regression with crossed random effects of participants and items. Results: Neurologically-intact controls exhibited the standard inhibitory effect of phonological neighborhood density: faster responses to words from low-density than high-density phonological neighborhoods (Estimate = 103.05, SE = 42.05, p = 0.014). For participants with aphasia, the effect of phonological neighborhood density was larger for those with lower fluency (χ2(1) = 3.95, p = 0.047; Figure 1, left panel), but was not modulated by overall aphasia severity (χ2(1) = 1.64, p = 0.2). Receptive vocabulary size also modulated the neighborhood density effect (χ2(1) = 7.93, p = 0.005), which was larger in participants with aphasia who had larger receptive vocabularies (Figure 1, right panel). The correlation between WAB Fluency and PPVT scores was not significant (r = -0.325, p = 0.257), indicating that the opposite effects of fluency and receptive vocabulary were not the same underlying pattern. Conclusions: These results are consistent with the view that fluency deficits in aphasia are a consequence of difficulty selecting among completing candidates – both in comprehension and in production. Furthermore, the fact that vocabulary size and fluency had opposite effects on sensitivity to neighborhood density indicates that the effect of fluency is not an effect of lexicon size.
Keywords
