Naming or word-finding tasks are a mainstay of the typical neuropsychological evaluation, particularly with older adults. However, many older adults have significant visual impairment and there are currently no such word-finding tasks developed for use with older visually impaired populations. This study presents a verbal, non-visual measure of word-finding for use in the evaluation of older adults with possible dysnomia. Stimuli were chosen based on their frequency of usage in everyday spoken language. A 60-item scale was created and given to 131 older Veterans. Rasch analyses were conducted and differential item functioning assessed to eliminate poorly-performing items. The final 55-item scale had a coefficient alpha of 0.84 and correlated with the Neuropsychological Assessment Battery Naming test, r = 0.84, p < .01, Delis-Kaplan Executive Function System (D-KEFS) Category Fluency, r = 0.45, p < .01, and the D-KEFS Letter Fluency, r = 0.40, p < .01. ROC analyses found the measure to have sensitivity of 79% and specificity of 85% for detecting dysnomia. Patients with dysnomia performed worse on the measure than patients with intact word-finding, t(84) = 8.2, p < .001. Patients with no cognitive impairment performed significantly better than patients with mild cognitive impairment, who performed significantly better than patients with dementia. This new measure shows promise in the neuropsychological evaluation of word-finding ability in older adults with or without visual impairment. Future directions include the development of a shorter version and the generation of additional normative data. (JINS, 2015, 21, 239-248)

Naming, or word-finding ability, is typically assessed using measures that require a patient to name a pictured object. Words used less frequently tend to be more difficult to find when speaking; thus word frequency can be used as a measure of item difficulty on such tests. However, frequency data for words on naming measures has either not been used in the creation of these tests or has been derived from data on how frequently words have been used in written materials. The present study determined how frequently words on the Boston Naming Test (BNT), the Neuropsychological Assessment Battery (NAB) naming subtest, the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) naming subtest, and auditory naming measures developed by Hamberger and Seidel (2003, Journal of the International Neuropsychological Society, 9, 479) and Brandt et al. (2010, The Clinical Neuropsychologist, 24, 1326) are used in spoken language. Items on the auditory naming measures had the highest mean frequency, and the BNT items 30-60 had the lowest mean frequency. Furthermore, item frequency on the full BNT, NAB naming forms 1 and 2, and RBANS forms A correlated with item number, indicating items increase in difficulty on these tests, with trends in the same direction found for RBANS form B and Hamberger and Seidel's auditory naming measure. Finally, differences in mean word frequency between tests underscore how interpretation of change in naming ability based on different measures should be made with caution.

No abstract

No abstract

No abstract