# Concurrent Validity, Test-Retest Reliability, and Sensitivity to Change of Functional Near-Infrared Spectroscopy for Measuring Language-Related Brain Activity in Post-Stroke Aphasia

> **NIH VA IK2** · VETERANS HEALTH ADMINISTRATION · 2022 · —

## Abstract

More than 2.5 million people in the U.S. have aphasia, a language disorder most often caused by stroke that
dramatically affects an individual’s functional independence and quality of life. For several decades, fMRI has
critically advanced aphasia research, but it has notable limitations. For example, individuals with conditions
that affect many Veterans (e.g., implanted medical devices or a history of injury involving metal) cannot safely
be scanned. These individuals have therefore been systematically excluded from fMRI studies of aphasia.
Additionally, fMRI is conducted in a noisy, restrictive environment, necessitating the use of artificial
experimental paradigms that may not engage the full range of anatomical and physiological mechanisms
supporting real-world language and communicative function. In contrast to fMRI, functional near-infrared
spectroscopy (fNIRS) has few contraindications, is silent, inexpensive, and can be administered in a standard
clinic room with naturalistic language paradigms. Despite these advantages, fNIRS has rarely been used to
study aphasia, likely due to a lack of data validating it as compared to fMRI in this specific population. This
study will address critical gaps in the evidence that would support fNIRS’ use by investigating its correlation
with fMRI for language mapping (Aim 1), its test-retest reliability (Aim 2), and its sensitivity to changes in
brain function due to behavioral language treatment (Aim 3) in people with chronic post-stroke aphasia.
 In part A of this study, 24 people with chronic aphasia (PWA) will be scanned twice over a 3-week interval
with both fNIRS and fMRI while performing three language tasks: picture naming and semantic and
phonological matching. Concurrent validity will be established by correlating language activation measured by
fNIRS with that of fMRI. fNIRS’ test-retest reliability will be assessed by computing intraclass correlation
coefficients across the first and second fNIRS scans, with comparable comparisons of the fMRI scans. Twenty-
four healthy older adults will complete the same protocols to inform interpretations of the results in PWA. In
part B, 12 PWA will complete fNIRS and fMRI scans during language tasks before and after three weeks of
intensive language therapy, administered as part of a separate clinical trial of treatment for word-retrieval
impairments. Comparisons between pre- and post-treatment fNIRS and fMRI scans will be used to assess
fNIRS’ sensitivity to change after intervention, relative to that of fMRI. Behavioral research activities (including
intensive language therapy) will be conducted at the VA Pittsburgh Healthcare System (VAPHS). fMRI and
fNIRS will be performed at the CMU-Pitt Brain Imaging Data Generation & Education Center and the Brain
and Auditory Sciences Research Initiative at the University of Pittsburgh, respectively.
 The results of this study will help determine the viability of fNIRS for measuring language functions in
post-strok...

## Key facts

- **NIH application ID:** 10538100
- **Project number:** 1IK2RX004252-01
- **Recipient organization:** VETERANS HEALTH ADMINISTRATION
- **Principal Investigator:** Jeffrey P Johnson
- **Activity code:** IK2 (R01, R21, SBIR, etc.)
- **Funding institute:** VA
- **Fiscal year:** 2022
- **Award amount:** —
- **Award type:** 1
- **Project period:** 2022-10-01 → 2027-09-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10538100

## Citation

> US National Institutes of Health, RePORTER application 10538100, Concurrent Validity, Test-Retest Reliability, and Sensitivity to Change of Functional Near-Infrared Spectroscopy for Measuring Language-Related Brain Activity in Post-Stroke Aphasia (1IK2RX004252-01). Retrieved via AI Analytics 2026-05-29 from https://api.ai-analytics.org/grant/nih/10538100. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*