Improving Upper-limb Prosthesis Usability: Cognitive Workload Measures Quantify Task Difficulty

Abstract

AbstractProviding user-focused, objective, and quantified metrics for prosthesis usability may help reduce the high (up to 50%) abandonment rates and accelerate the clinical adoption and cost reimbursement for new and improved prosthetic systems. We comparatively evaluated several physiological, behavioral, and subjective cognitive workload measures applied to upper-limb neuroprosthesis use.Users controlled a virtual prosthetic arm via surface electromyography (sEMG) and completed a virtual target control task at easy and hard levels of difficulty (with large and small targets, respectively). As indices of cognitive workload, we took behavioral (Detection Response Task; DRT) and electroencephalographic (EEG; parietal alpha and frontal theta power, and the P3 event-related potential) measures for one group (n = 1 amputee participant, n = 10 non-amputee participants), and electrocardiographic (ECG; low/high frequency heart-rate variability ratio) and pupillometric (task-evoked pupillary response) measures for another group (n = 1 amputee participant, n = 10 non-amputee participants), because all measures could not reasonably be recorded simultaneously. Participants of both groups also completed the subjective NASA Task-Load Index (TLX) survey.Ease of use, setup, piloting, and analysis complexity varied among measures. The DRT required minimal piloting, was simple to set up, and used straightforward analyses. ECG measures required moderate piloting, were simple to set up, and had somewhat complex analyses. Pupillometric measures required extensive piloting but were simple to set up and relatively simple to analyze. EEG measures required extensive piloting, extensive setup and equipment, careful monitoring, and moderately complex analyses.Across subjects, the DRT, low/high frequency heart-rate variability ratio, task-evoked pupillary response, and NASA TLX significantly differentiated between the easy and hard tasks, whereas EEG measures (alpha power, theta power, and P3 event-related potential) did not. Aside from the NASA TLX, the DRT was the easiest to use and most sensitive to cognitive load across and within subjects. Among physiological measures, we recommend ECG, pupillometry, and EEG/ERPs, in that order.This study provides the first evaluation of multiple objective and quantified cognitive workload measures during the same task with prosthesis use. User-focused cognitive workload assessments may increase our understanding of human interactions with advanced upper-limb neuroprostheses and facilitate their improvements and translation to real-world use.Significance StatementThe human arm is dexterous and able to sense objects it contacts. Restoring sensory and motor function to a person with limb loss presents multiple challenges and requires improvements in robotics, biological interfaces, decoding biological signals for prosthesis movement, and sensory restoration. The scientific and engineering communities have made progress toward restoring arm function through advanced neuroprostheses. However, most studies focus solely on task performance, and they typically employ artificial experimental paradigms in which the user can devote full attention to the task, which is often unrealistic for use in everyday activities. To develop neuroprostheses capable of restoring intuitive arm function, engineers and scientists must also consider the difficulty of use, or cognitive burden, of using the neuroprosthesis. Although many measures of cognitive workload have been developed, few studies directly interrogate cognitive workload during neuroprosthesis use. An engineer or scientist seeking to employ cognitive workload measures during neuroprosthesis use will likely wonder, as we did, which measures are most suitable for their needs. To address this question, we empirically assess the practical and functional merits and limitations of several physiological, behavioral, and subjective techniques to measure cognitive workload during use of an advanced prosthesis. We anticipate that these findings may influence other medical and consumer areas of human-computer interaction, such as virtual reality or exoskeleton use.