HCC: Human-Computer Interaction and Brain Measurement Using fNIR Spectroscopy

在脑机接口(BCI)的一般领域已经有了很多以前的研究，主要目的是帮助严重运动障碍的人通过将他们的大脑活动转化为特定的设备控制信号来与他们的环境互动。完全瘫痪(“锁定”)或缺乏肌肉控制的用户(例如，脑瘫或中风患者)可以使用BCI回答简单的问题，控制他们的环境，以及进行文字处理。但到目前为止，已尝试用于脑机接口的各种脑成像技术都存在严重缺陷，因此目前通过此类系统进行交流既耗时又费力。公开的研究挑战涉及BCI的准确性(系统经常曲解用户的意图)，以及此类系统的信息传输率(在现实世界环境中使用通常太慢)。在这个项目中，PI将基于他们在设计、实施和评估非命令、自适应用户界面(例如，基于眼球运动)方面的经验，通过利用新兴技术来测量大脑活动，并结合机器学习来分析人机交互中的用户数据，从而促进BCI。虽然FNIR技术仍处于初级阶段，但PI希望将其用作自适应接口的实时输入，以开辟新的天地。PIS相信，这种结合还将导致新的、更客观的方法来评估下一代交互风格的一般人机界面。PI的方法是对基于现实的交互概念的自然扩展，并将利用他们之前与基于现实的交互概念有关的工作，该概念侧重于新的交互方式如何更多地利用用户从现实世界中预先存在的技能和期望，而不是经过培训的计算机技能，同时帮助区分致力于界面相关或句法方面的智力努力与致力于基本任务或语义方面的精神努力。广泛影响：该项目将开发对肢体残疾用户直接受益的技术。此外，这项工作将开辟人机交互研究的新领域(即与能够实时动态适应用户认知负荷的界面相关)，并将从总体上推进交互风格的理论和评估。

There has been much previous research in the general area of brain-computer interfaces (BCI), which has primarily aimed to help people with severe motor disabilities interact with their environment by translating their brain activity into specific device control signals. Users who are completely paralyzed ("locked-in") or those who lack muscle control (e.g., people with cerebral palsy or stroke victims) can use BCIs to answer simple questions, to control their environment, and for word processing. But the variety of brain imaging techniques that have been tried to date for BCIs have all suffered from serious drawbacks, so that communicating through such systems is currently time consuming and mentally demanding. Open research challenges concern the accuracy of BCIs (systems often misinterpret a user's intentions), and the information transfer rate of such systems (which is often too slow for use in real world settings). In this project, the PIs will build on their experience in designing, implementing, and evaluating non-command, adaptive user interfaces (e.g., based on eye movement), to advance BCI by bringing to bear emerging technology for measuring brain activity using functional near infrared (fNIR) spectroscopy coupled with machine learning to analyze user data in human-computer interaction. While fNIR technology is still in its infancy, the PIs expect its use as a real-time input to an adaptive interface to break new ground. The PIs believes that this combination will also lead to new, more objective methods for evaluating next generation interaction styles for HCI in general. The PIs' approach is a natural extension of and will exploit their prior work relating to the concept of reality-based interaction, which focuses on the ways that new interaction styles exploit the user's pre-existing skills and expectations from the real world more than trained computer skills, while at the same time helping to differentiate mental effort devoted to interface-related or syntactic aspects from that devoted to the underlying task or semantic aspects.Broader Impacts: This project will develop technologies of immediate benefit to users with motor disabilities. Additionally, the work will open up a new area of HCI research (namely, relating to interfaces that can dynamically adapt in real time to the user's cognitive load), and will advance the theory and evaluation of interaction styles in general.