Interactive Voice Response Systems and Older Adults: Examination of the Cognitive Factors Related to Successful IVR Interaction and Proof-of-Principle of IVR Administration and Scoring of Neuropsychological Tests

Abstract

The main goal of this project was to enhance the use and usability of Interactive Voice Response Systems (IVR) for older people. The objective of study one was to examine older people’s perceptions of the technology and identify the most common difficulties experienced by older people when interactive with IVR using focus groups. Twenty-six people aged 65 and older took part in the study. Data were analyzed using frequency and chi square analyses. The results revealed negative attitudes towards the technology. Long menus, frustration about not being able to reach an operator and absence of shortcuts were some of the most common difficulties reported by participants. Study two examined the cognitive factors predicting successful IVR interaction in four commercially available IVR systems in a sample of 185 older adults. Linear regressions were performed on the data. Results indicated that working memory and auditory memory were the best predictors of successful IVR interaction. Using the same sample of participants as study two, study three examined older adults’ attitudes towards the four IVR systems in relation to their success in interacting with the technology. The study also evaluated the impact of gender on success and attitudes towards IVR. There was a significant positive correlation between success with IVR and favorable attitudes towards the technology. No gender differences emerged in both performance on IVR tasks and attitudes towards the technology. Study four evaluated the feasibility of using a voice-activated IVR to administer and score three short neuropsychological tests using a sub sample of the original sample of 185 older adults involved in study two and three. One hundred and fifty eight participants took part in the study. Results showed high correlations between the IVR and clinician scoring of the three tests. Nevertheless, a number of discrepancies and technical issues were discovered.