Evaluations of haptic feedback in myoelectric prostheses are generally limited to task performance outcomes, which while necessary, fail to capture the mental effort of the user operating the prosthesis. Cognitive load is usually investigated with reaction time metrics and secondary task accuracy, which are indirect, and may not capture the time-varying nature of mental effort. Here, we propose wearable, wireless functional near infrared spectroscopy (fNIRS) neuroimaging to provide a continuous direct assessment of operator mental effort during use of a prosthesis. Utilizing fNIRS in a two-alternative forced-choice stiffness discrimination task, we asked participants to differentiate objects using their natural hand, a (traditional) myoelectric prosthesis without sensory feedback, and a myoelectric prosthesis with haptic (vibrotactile) feedback of grip force. Results showed that discrimination accuracy and mental effort are optimal with the natural hand, followed by the prosthesis featuring haptic feedback, and then the traditional prosthesis, particularly for objects whose stiffness were difficult to differentiate. This experiment highlights the utility of haptic feedback in improving task performance and lowering cognitive load for prosthesis use, and demonstrates the potential for fNIRS to provide a robust measure of cognitive effort for other human-in-the-loop systems.

中文翻译：

---

肌电假肢触觉反馈的神经生理学评估


肌电假肢触觉反馈的评估通常仅限于任务执行结果，虽然这是必要的，但无法捕捉操作假肢的用户的精神努力。认知负荷通常通过反应时间指标和次要任务准确性进行研究，这些指标是间接的，并且可能无法捕捉脑力劳动随时间变化的性质。在这里，我们提出了可穿戴、无线功能性近红外光谱 (fNIRS) 神经成像技术，可以对操作者在使用假肢期间的脑力活动进行连续的直接评估。在两种选择的强制选择刚度辨别任务中利用 fNIRS，我们要求参与者使用他们的自然手、没有感觉反馈的（传统）肌电假肢和具有握力触觉（振动触觉）反馈的肌电假肢来区分物体。结果表明，自然手的辨别准确性和脑力劳动最佳，其次是具有触觉反馈的假肢，然后是传统假肢，特别是对于难以区分刚度的物体。该实验强调了触觉反馈在提高任务绩效和降低假肢使用认知负荷方面的效用，并证明了 fNIRS 为其他人机交互系统提供稳健的认知努力测量的潜力。