Using eye movement responses to index auditory comprehension: An adaptation of the Revised Token Test

Abstract

Background: Tracking spontaneous eye movement responses may improve the accuracy of comprehension assessment in patients with neurological impairments. Eye movement methods provide an on-line response mode that does not require an overt planned motoric response, tax participants' understanding of instructions, or interrupt the comprehension process with intervening instructions or prompts. Aims: The purpose of this study was to examine an adaptation of a standardised test of auditory comprehension for adults with neurological impairments, using eye movement responses. Individuals with normal language comprehension should demonstrate a consistent pattern of eye movement responses indicating comprehension. Methods & Procedures: Nineteen adults with no history of neurological involvement were presented with auditory comprehension stimuli from the Revised Token Test (RTT) (McNeil & Prescott, 1978) in each of three conditions: traditional, pointing, and eye movement. In the traditional condition, the standardised version of the RTT was administered. In the pointing condition, RTT stimuli were manipulated to fit a multiple-choice format; participants responded to modified verbal stimuli for each item by pointing to one of four images shown in a printed test manual. In the eye movement condition, participants viewed the same stimuli presented in the multiple-choice pointing version, shown on a computer monitor rather than in a test manual. A remote pupil centre/corneal reflection system was used to monitor eye movements. Participants were not instructed to look at anything in particular, and were not told that there were target images within the displays. The eye movement condition was presented first to avoid participants/ perceptions that they were expected to select a target image with their eyes. Dependent variables included the proportion of total viewing time that participants fixated on target images and non-target foils. Outcomes & Results: Traditional RTT scores indicated normal comprehension for all participants, according to published norms. Likewise, pointing condition scores indicated good comprehension. For each of the subtests in the eye movement condition, the proportional amount of time that fixations were allocated to target images significantly exceeded chance expectations.