Abstract
This study evaluates a low-cost, single-channel fNIRS device in cognitive neuroscience, aiming to overcome the financial barriers of commercial systems by testing its efficacy in tasks of varying complexity. Twenty-six participants engaged in motor control (finger-tapping), working memory (n-back), and creativity (AUT) tasks while their prefrontal cortex activity was monitored using the device, with behavioral and cerebral blood flow changes recorded. Results showed the device's capability to detect significant blood flow variations across different tasks, thereby supporting its use in cognitive research. The study confirms the potential of single-channel fNIRS as a cost-effective tool for diverse cognitive assessments, from simple motor actions to complex creative thinking.
Introduction
Functional Near-Infrared Spectroscopy (fNIRS) is a non-invasive neuroimaging technique that measures changes in blood oxygenation to monitor brain activity. While powerful, commercial fNIRS systems often cost tens of thousands of dollars, creating a significant financial barrier for many researchers and educational institutions. This study investigates the validity of a low-cost, open-source alternative—a modified HEGduino V2—to determine if "budget" hardware can reliably capture hemodynamic responses across a range of cognitive tasks.
Methods and Materials
The researchers utilized a modified HEGduino V2 platform. Key hardware modifications included replacing the standard 940nm LED with an 880nm LED to better align with hemoglobin absorption peaks. A single-channel sensor was positioned at the F3 location to target the left dorsolateral prefrontal cortex (dlPFC), a region critical for executive function.
The study involved 26 healthy participants (mean age ~20.6 years) who performed three tasks of varying complexity: a motor control baseline (finger-tapping), a working memory test (N-back), and a divergent thinking task (Alternate Uses Task). The signal processing pipeline involved converting raw light intensity to Optical Density, correcting motion artifacts via spline interpolation, and applying a 0.01–0.1 Hz bandpass filter to mitigate physiological noise.
Key Results
The device demonstrated high signal fidelity across tasks. The mean Signal-to-Noise Ratio (SNR) was 30.307 for the motor task and 23.883 for the creativity task.
Motor and Cognitive Performance
Significant increases in cerebral blood flow were observed during the finger-tapping task compared to the resting state ($p = 0.020$).
In the working memory domain, the device successfully distinguished between 1-back and 2-back conditions ($p = 0.028$). Notably, the creativity task (AUT) showed a robust increase in blood flow compared to rest ($p < 0.001$), with a large effect size (Cohen’s $d = 0.728$), proving the device's sensitivity to high-level cognitive processes.
Discussion and Limitations
The findings suggest that low-cost, single-channel fNIRS can reliably detect hemodynamic shifts during both simple motor and complex creative tasks. This democratizes access to neuroimaging tools, particularly for under-resourced labs.
However, several limitations remain. The single-channel design lacks spatial mapping capabilities and "short-distance channels" used in high-end systems to filter out scalp-level noise. Additionally, the current setup only measures oxyhemoglobin (ΔHbO). Future iterations aim to integrate multi-modal data (e.g., EEG) and include short-distance channels to improve signal specificity.
Conclusion
This validation study confirms that modified open-source hardware can produce scientifically valid hemodynamic data. By providing a cost-effective blueprint for functional neuroimaging, this research paves the way for broader participation in cognitive neuroscience.
Summary updated