Characterization of forehead blood flow bias on NIRS signals during neural activation with a verbal fluency task.

The major problem of near-infrared spectroscopy (NIRS) for brain activity measurement during verbal fluency task is the overlapping forehead scalp blood flow (FBF) on the target cerebral blood flow (CBF). There could be among-individual differences in the influence of FBF on CBF. We investigated effects of FBF on CBF by comparing signals obtained through a laser Doppler flowmeter (LDF) and NIRS using the modified Beer-Lambert Law (MBLL). Among 25 healthy individuals, 7 participants showed a strong correlation between LDF and NIRS signals (r >0.500). There were no significant differences according to age or sex. Subsequently, we applied the hemodynamic separation method to the values calculated using the MBLL (Δ[oxy-Hb]): to separate the concentration of oxygenated hemoglobin in the forehead (Δ[oxy-Hb]) and cerebral cortex (Δ[oxy-Hb]). First, we found that the influence of Δ[oxy-Hb] on Δ[oxy-Hb] in the high r group was almost twice as large as that in the low r group. Second, presence of sex and age differences in the influence of Δ[oxy-Hb] on Δ[oxy-Hb] were suggested. Based on the results, we discuss the factors affecting FBF and the resulting variations in NIRS signals.