Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams

A Simple Scale for Measuring the Lexical-Sublexical Speech Impairment Dissociation in Individuals with Aphasia

Poster E64 in Poster Session E, Saturday, October 8, 3:15 - 5:00 pm EDT, Millennium Hall

Grant Walker¹, Julius Fridriksson², Gregory Hickok¹; ¹University of California, Irvine, ²University of South Carolina

The Walker et al. (2018) cognitive psychometric model of picture naming suggests that independent cognitive processes can be disproportionately challenged by different target words, opening the possibility to detect disproportionate impairments based on targeted assessments. Specifically, selection among lexical representations based on meaning can be challenged separately from the formulation and production of the sublexical speech segments that comprise words. Using response type data from 365 participants with left hemisphere stroke and aphasia who named 175 unique items, we created two 20-item lists that predominately challenged lexical or sublexical production processes, respectively. In a new group of 91 participants with left hemisphere stroke, we examined differences in accuracy between these lists to determine how much information can be gained about the neurocognitive locus of impairment, over and above what can be learned from overall naming accuracy. We examined four predictions: 1) There should exist participants who perform better on one test than the other and participants who exhibit the reverse pattern. 2) The difference between lexical and sublexical naming test scores should be directly related to the difference between the Word Finding and Repetition subscores of the Western Aphasia Battery. 3) Disproportionately sublexical naming impairment should indicate damage to dorsal speech motor regions of the brain; disproportionately lexical naming impairment should indicate sparing of dorsal speech motor regions. 4) Disproportionately sublexical naming impairment should indicate increased effort and associated increased fMRI activation of speech motor regions of the brain during naming; disproportionately lexical naming impairment should indicate increased effort and associated increased fMRI activation in ventral temporal lobe regions during naming. All predictions were confirmed to some extent, along with some novel findings. A Barnard’s exact test (p < .2) identified 13 participants who performed significantly better on the lexical than the sublexical naming test and 10 participants with the reverse pattern. An interaction with overall severity was observed, with predominately sublexical impairment being associated with more severe overall impairment and predominately lexical impairment being associated with less severe overall impairment. The difference between naming test scores was correlated with the difference between WAB Word Finding and Repetition subscores (r = .47, p = .0000026). Participants with better Repetition subscores (>10%) performed better on sublexical naming (p = .031); participants with better Word Finding subscores (>10%) performed better on lexical naming (p = .014). Partial correlations predicting lesion load within each region of the AICHA atlas from the difference in naming test scores while controlling for external lesion volume revealed significant associations in left perisylvian and speech motor regions for predominately sublexical impairment and in left cingulate cortex for predominately lexical impairment. Partial correlations predicting fMRI activation during naming within each region of the AICHA atlas from the difference in naming test scores while controlling for overall naming accuracy revealed significant associations in bilateral speech motor regions for predominately sublexical impairment and in bilateral middle and inferior temporal cortex and the temporal poles for predominately lexical impairment. Implications for clinical assessment and neurobiological theories of language will be discussed.

Topic Areas: Disorders: Acquired, Language Production