Validity of an eye-tracking method to index working memory in people with and without aphasia

Background: Working memory (WM) is essential to auditory comprehension; thus understanding of the nature of WM is vital to research and clinical practice to support people with aphasia. A key challenge in assessing WM in people with aphasia is related to the myriad deficits prevalent in aphasia, including deficits in attention, hearing, vision, speech, and motor control of the limbs. Eye-tracking methods augur well for developing alternative WM tasks and measures in that they enable researchers to address many of the potential confounds inherent in tasks traditionally used to study WM. Additionally, eye-tracking tasks allow investigation of trade-off patterns between storage and processing in complex span tasks, and provide on-line response measures. Aims: The goal of the study was to establish concurrent and discriminative validity of a novel eye movement WM task in individuals with and without aphasia. Additionally we aimed to explore the relationship between WM and general language measures, and determine whether trade-off between storage and processing is captured via eye-tracking measures. Methods & Procedures: Participants with (n = 28) and without (n = 32) aphasia completed a novel eye movement WM task. This task, incorporating natural response requirements, was designed to circumvent potential confounds due to concomitant speech, motor, and attention deficits. The task consisted of a verbal processing component intermixed with presentation of colours and symbols for later recall. Performance on this task was indexed solely via eye movements. Additionally, participants completed a modified listening span task that served to establish concurrent validity of the eye-tracking WM task. Outcomes & Results: Performance measures of the novel eye movement WM task demonstrated concurrent validity with another established measure of WM capacity: the modified listening span task. Performance on the eye-tracking task discriminated effectively between participants with and without aphasia. No consistent relationship was observed between WM scores and Western Aphasia Battery aphasia quotient and subtest scores for people with aphasia. Additionally, eye-tracking measures yielded no trade-off between processing and storage for either group of participants. Conclusions: Results support the feasibility and validity of employing a novel eye-tracking method to index WM capacity in participants with and without aphasia. Further research is required to determine the nature of the relationship between WM, as indexed through this method, and specific aspects of language impairments in aphasia.