Abstracts

Rationale: Assessment of naming is a core component of most comprehensive neuropsychological evaluations, particularly in individuals with epilepsy. However, the most widely used naming test, the Boston Naming Test (BNT), has frequently been criticized for its poor psychometric properties, inadequate standardization, and insufficient norms. Clinically, it has also shown poor sensitivity and specificity in detecting language abnormalities due to the significant influence of verbal intelligence, level of formal education, and vocabulary knowledge. In fact, some authors have suggested that the BNT norms should be reported and interpreted in conjunction with vocabulary scores to control for the large amount of shared variance. However, less is known about the relationship between vocabulary and naming in individuals who learned English as a second language. Based on clinical observations, we hypothesized that non-native English speakers would exhibit weaker BNT scores relative to native English speakers of comparable age, education, and vocabulary level.Methods: Participants included 259 adults with epilepsy who underwent neuropsychological evaluations and obtained normal range scores on the WAIS-R/III/IV Vocabulary subtest (scaled score ≥ 8). Of those, 203 were native English speakers (Vocab M = 10.98, SD = 2.16), and 56 learned English as a second language (ESL; Vocab M, = 10.36, SD = 2.11). Independent samples t-tests and chi-squared analyses were used to compare groups on demographic factors and test performances, including the BNT, Auditory & Visual Naming Test, word reading, verbal fluency, and full scale IQ.Results: Groups did not differ with respect to age, education, full-scale IQ, vocabulary, seizure lateralization, or age of seizure onset (all p’s > .05). However, the mean BNT total score for native English speakers (M = 51.35, SD = 6.07) was significantly higher than that for ESL participants (M = 43.98, SD = 9.60; t(67.60) = 5.45, p < .001). Significant differences were also observed on other naming measures, including auditory (t(38.72) = 3.73, p = .001) and visual naming (t(37.55) = 2.71, p = .01), with the ESL group achieving lower scores on both. Importantly, the two groups did not differ on tests of word reading, phonemic fluency, or semantic fluency (all p’s > .05). Among ESL participants, Pearson correlations revealed an inverse relationship between age at which they learned English and BNT scores (r = -.435, p = .001).Conclusions: These findings are consistent with data from published literature describing the effects of linguistic background, ethnicity, and acculturation on cognitive test performance in general. Moreover, these findings specifically bring to light the complexity of language assessment in individuals with epilepsy who learned English as a second language. Despite normal or above average scores on measures of vocabulary, reading, and verbal fluency, these results suggest that naming (i.e., targeted word-retrieval) results should be interpreted differently for native versus non-native English speakers. NIH R01 NS35140(MH)