Slide Slam D9
A novel Brief Executive Language Screen (BELS) for propositional language
Gail A Robinson1, Lucy Shi1, Amelia Ceslis1; 1The University of Queensland
Conversational speech occurs at the interface between language and other cognitive processes including executive functions. Aphasia assessment tools have typically focused on core language skills like naming and classical aphasia type and severity. We develop a novel Brief Executive Language Screen (BELS) that is designed to assess language skills and executive functions critical for spontaneous speech. The design is based on theoretical models that broadly agree on three stages of spoken language (conceptualization, linguistic formulation, articulation). The BELS comprises 11 subtests that assess articulation, core nominal language skills (repetition, naming and comprehension) and propositional language (connected speech and phonemic/semantic word fluency), and also incidental verbal memory. Uniquely, the BELS incorporates a sentence completion task that measures executive functions (verbal initiation, inhibition, selection and strategic processes). One-hundred and fifty-eight healthy controls and 195 acute stroke patients were recruited for the BELS development and validation studies. Construct validity was determined using confirmatory factor analysis and discriminative validity was evaluated with logistic regression. This informed the final version of the BELS, which was assessed via comparison to standard neuropsychological tests to ascertain convergent and divergent validity. The three-factor structure of the BELS was confirmed; that is, the articulation, nominal language and propositional language components comprised of uniquely different constructs. The BELS also demonstrated practical utility as a sensitive measure of aphasia as it was able to discriminate between the performance of healthy controls and stroke patients on almost all subtests. Thus, 30-50% of stroke patients were impaired on propositional language or executive function subtests. Surprisingly, the BELS incidental memory subtest revealed impairment in ~40% of stroke patients. Based on these initial findings, two of the 11 subtests were adjusted and the final version of the BELS was validated against select neuropsychological tests. The BELS is a brief, sensitive and novel tool that uniquely assesses the executive components critical for the spontaneous expression of language. In addition to assessing propositional language, the BELS is a valid and clinically useful tool for detecting articulation and nominal language deficits, along with verbal memory.