Elsevier

NeuroImage

Volume 61, Issue 1, 15 May 2012, Pages 70-81
NeuroImage

The physiological origin of task-evoked systemic artefacts in functional near infrared spectroscopy

https://doi.org/10.1016/j.neuroimage.2012.02.074Get rights and content
Under a Creative Commons license
open access

Abstract

A major methodological challenge of functional near-infrared spectroscopy (fNIRS) is its high sensitivity to haemodynamic fluctuations in the scalp. Superficial fluctuations contribute on the one hand to the physiological noise of fNIRS, impairing the signal-to-noise ratio, and may on the other hand be erroneously attributed to cerebral changes, leading to false positives in fNIRS experiments. Here we explore the localisation, time course and physiological origin of task-evoked superficial signals in fNIRS and present a method to separate them from cortical signals. We used complementary fNIRS, fMRI, MR-angiography and peripheral physiological measurements (blood pressure, heart rate, skin conductance and skin blood flow) to study activation in the frontal lobe during a continuous performance task. The General Linear Model (GLM) was applied to analyse the fNIRS data, which included an additional predictor to account for systemic changes in the skin.

We found that skin blood volume strongly depends on the cognitive state and that sources of task-evoked systemic signals in fNIRS are co-localized with veins draining the scalp. Task-evoked superficial artefacts were mainly observed in concentration changes of oxygenated haemoglobin and could be effectively separated from cerebral signals by GLM analysis. Based on temporal correlation of fNIRS and fMRI signals with peripheral physiological measurements we conclude that the physiological origin of the systemic artefact is a task-evoked sympathetic arterial vasoconstriction followed by a decrease in venous volume.

Since changes in sympathetic outflow accompany almost any cognitive and emotional process, we expect scalp vessel artefacts to be present in a wide range of fNIRS settings used in neurocognitive research. Therefore a careful separation of fNIRS signals originating from activated brain and from scalp is a necessary precondition for unbiased fNIRS brain activation maps.

Highlights

► Task-evoked changes in the skin blood volume induce artefacts in fNIRS signals. ► Artefacts result from a venous volume decrease due to sympathetic vasoconstriction. ► This phenomenon might induce false positives in fNIRS experiments on the forehead.

Keywords

Time-domain optical brain imaging
Systemic changes
Skin blood flow
Periphery physiology
fMRI
Frontal lobe

Cited by (0)