Hal, do I have memory problems? A pilot study investigating a voice recognition cognitive screening tool for detection of neuropsychological changes

Background The prevalence of mild, age‐advanced changes in cognition, defined as mild cognitive impairment, is spiking as the baby boomer population ages. Early and accurate detection of age‐advanced changes in thinking abilities leads to proper care, efficient use of resources, and promotion of research. Indeed, government based health care systems and insurance programs incentivize routine neurocognitive assessment. Brief, paper and pencil cognitive screeners such as the MMSE and MOCA are widely utilized in clinical settings. Such screeners lack sensitivity to subtle cognitive change and specificity related to the cause of decline. Moreover, administration and interpretation of these screeners is time consuming in a busy, clinical setting. Prior research has determined that electronic‐based testing is effective in detecting cognitive change (Zygouris & Tsolaki, 2014). This pilot study aimed to investigate a convenient, platform agnostic, voice‐recognition screening tool for detection of mild, age‐advanced neurocognitive problems. Method Participants (n=16) were recruited from a large community‐based neurology practice, who underwent a comprehensive neuropsychological (NP) evaluation as part of their standard of care. Participants were administered an auditory list learning/recall test and a verbal fluency test (F/A/S and animals) via a voice recognition cognitive screening tool. Pearson correlations were conducted to examine the concurrent validity between the standard NP measures and those obtained using the current screening tool. Result Preliminary data suggests a positive correlation between standard NP measures and those obtained using voice recording and speech recognition technologies; language: F/A/S (r = .716, p < .005), animals (r .810, p < .005); memory: delayed recall (r = .515, p < .05). Conclusion This pilot study demonstrated concurrent validity between standard NP measures of memory and verbal fluency and those obtained using a novel, voice recording and speech recognition screening tool. These initial data provide support for the efficacy of a computerized based screening tool for older adults. Such a screening too has the potential to maximize utilization across multiple settings (e.g., medical office, hospital, in‐home) and improve early detection of cognitive decline.