What we can and cannot (yet) do with functional near infrared spectroscopy.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Functional near infrared spectroscopy (NIRS) is an emerging neuroimaging technique.
• NIRS complements electroencephalography (EEG) in brain-computer interface (BCI) development.
• NIRS has shown promise in detecting cognitive and affective states for adaptive human-computer interaction (HCI).

Purpose of the Study:

• To review current NIRS applications in measuring user states and passive BCIs.
• To identify key challenges for NIRS-BCIs in realistic HCI settings, including signal processing and modeling.
• To quantitatively compare NIRS models to assess the impact of challenges on signal reliability.

Main Methods:

• Literature review of NIRS-based cognitive and affective state detection.
• Discussion of challenges such as detection latencies and motion artifacts in NIRS-BCIs.
• Quantitative comparison of three NIRS models to evaluate signal reliability under challenging conditions.

Main Results:

• NIRS demonstrates capability in detecting cognitive and affective states for HCI applications.
• Significant challenges, including detection latencies and motion artifacts, impede reliable NIRS-BCI deployment.
• The study quantitatively analyzes the impact of these challenges on NIRS signal reliability across different models.

Conclusions:

• NIRS holds promise for advancing brain-computer interfaces (BCIs).
• Overcoming signal processing and modeling challenges is essential for robust NIRS-BCI systems in real-world HCI.
• Further research is needed to enhance NIRS robustness and utility for the BCI community.