The anterior temporal lobe (ATL) sustains a degree of damage during surgery for temporal lobe epilepsy (TLE), and although classical language models implicate the ATL in language function, and naming difficulties are commonly reported by TLE patients post-surgically, the role of the ATL in language is not well understood. The present study aimed to examine the role of the ATL in language function, and methods for evaluating pre- and post-surgical language function of the ATL in patients with epilepsy using two approaches. The first study employed neuropsychological testing of pre-and post-surgical TLE patients on 3 conventional tests of language function, the Boston Naming Test (BNT), Controlled Oral Word Association Test (COWAT) and Animal Fluency, and one novel test: The Category Specific Names Test (CSNT). The CSNT was selected in an attempt to compensate for the problems of heterogeneous items, and the low ceiling in the BNT. Results of the CSNT had not been previously validated with TLE patients, and had not been routinely used in Australia. The second study looked at ATL activation in healthy controls and TLE patients using two new functional magnetic resonance imaging (fMRI) tasks. For the neuropsychological study, Bayesian analysis showed that the BNT was effective in differentiating left from right TLE in a sample of 42 patients with intractable epilepsy (LR+ = 8.37). The COWAT, Animal Fluency and the CSNT were not effective in differentiating left from right TLE. Very small likelihood ratios indicated only modest changes from pre-test odds. For the fMRI component of the study, although the two new fMRI tasks (Famous Faces naming, and Sentence Reading) resulted in the expected activation of language areas in group averaged data, they did not elicit consistent ATL activation across individuals, and subsequently did not meet the iii criteria recommended for fMRI protocol development recommended by Schwartz, Devinsky, Doyle and Perrine. (1997). Their inclusion in an fMRI protocol was not supported. Findings of the present study have important implications for pre-surgical evaluation of ATL language function in patients with epilepsy. Preliminary evidence was not found to support the use of the CSNT in patients with epilepsy. Results suggest that the BNT alone should be used for detecting naming impairment in patients with left-sided epilepsy. Benefits of using Bayesian analysis to examine the clinical applicability of research results, rather than conventional Null Hypothesis Significance Testing (parameter statistical) techniques are discussed.
Identifer | oai:union.ndltd.org:ADTP/256853 |
Date | January 2007 |
Creators | Salton, Nancy |
Source Sets | Australiasian Digital Theses Program |
Detected Language | English |
Page generated in 0.0017 seconds