Light-Evoked Neural Activation Patterns in Gulf War Illness: Associations with Symptom Burden
Jonghoon Chang; Alex Hattenhauer; Luisa Saccaro; Shyamal Raolji; David Valdes Arias; Loralei Ann Parchejo; Mireya Hernandez; Brandon Chou; Brinna Desai; Robby Mattes; Elizabeth R Felix; Eric Moulton; Anat Galor
Web | Investigative Ophthalmology & Visual Science | Abstract
Jonghoon Chang; Alex Hattenhauer; Luisa Saccaro; Shyamal Raolji; David Valdes Arias; Loralei Ann Parchejo; Mireya Hernandez; Brandon Chou; Brinna Desai; Robby Mattes; Elizabeth R Felix; Eric Moulton; Anat Galor
Purpose
Gulf War Illness (GWI) is a chronic, multi-symptom condition affecting veterans of the 1990–1991 Gulf War with reported symptoms spanning across six domains: neurologic/cognitive/mood, fatigue/sleep, pain, gastrointestinal, respiratory, and skin. Prior studies have found light sensitivity in veterans with GWI compared to controls. As the pathophysiology of GWI symptoms and light sensitivity remains unclear, this study investigates symptoms of GWI using the eye as a window to the brain, specifically examining photosensitivity as a potential biomarker of central nervous system dysfunction using light-evoked functional magnetic resonance imaging (fMRI) as a proxy.
Methods
GWI status was determined using the Kansas case definition, in which “Kansas-positive” (Kansas+) is defined as having one or more moderate or severe symptoms in ≥3 domains on the Kansas questionnaire. Of 23 total subjects, 13 participants were classified as Kansas+ and 10 as Kansas-negative (Kansas–). fMRI data from all 23 participants were analyzed, leveraging blood oxygen level-dependent (BOLD)-based responses to infer neural activity. A bright-screen versus black-screen paradigm was used to elicit light-evoked activation. Participants viewed alternating black (~0.5 lux) and white (~65 lux) screens, with sixteen 6-second white-screen periods presented and separated by variable intervals of 51–66 seconds. Group-level fixed-effects analyses were performed in FSL using a generalized linear model. The voxel wise significance threshold was Z > 2.3 (uncorrected for multiple comparisons).
Results
Study participants had a mean age of 60.4 ± 4.8 years, and 78% were male. The Kansas+ group showed greater activation, compared to the Kansas- group, across multiple cortical regions, including the left occipital pole (Zmax = 5.83, p = 0.001). Visual inspection of Kansas+ > Kansas- activation also revealed increased activity in the primary somatosensory cortex BA1 L (Z = 2.75), right insula (Z = 3.00), cingulate gyrus (Z = 3.11), left caudate (Z = 3.05) and left ventrolateral prefrontal cortex (Z = 3.25).
Conclusions
Kansas+ veterans demonstrated greater light-evoked activation in visual cortex and in regions involved in sensory–affective pain processing compared to Kansas– veterans. These exploratory findings suggest that light-evoked fMRI may capture central nervous system alterations relevant to symptom burden in Gulf War Illness.
This abstract was presented at the 2026 ARVO Annual Meeting, held in Denver, CO, May 3-7, 2026.
Gulf War Illness (GWI) is a chronic, multi-symptom condition affecting veterans of the 1990–1991 Gulf War with reported symptoms spanning across six domains: neurologic/cognitive/mood, fatigue/sleep, pain, gastrointestinal, respiratory, and skin. Prior studies have found light sensitivity in veterans with GWI compared to controls. As the pathophysiology of GWI symptoms and light sensitivity remains unclear, this study investigates symptoms of GWI using the eye as a window to the brain, specifically examining photosensitivity as a potential biomarker of central nervous system dysfunction using light-evoked functional magnetic resonance imaging (fMRI) as a proxy.
Methods
GWI status was determined using the Kansas case definition, in which “Kansas-positive” (Kansas+) is defined as having one or more moderate or severe symptoms in ≥3 domains on the Kansas questionnaire. Of 23 total subjects, 13 participants were classified as Kansas+ and 10 as Kansas-negative (Kansas–). fMRI data from all 23 participants were analyzed, leveraging blood oxygen level-dependent (BOLD)-based responses to infer neural activity. A bright-screen versus black-screen paradigm was used to elicit light-evoked activation. Participants viewed alternating black (~0.5 lux) and white (~65 lux) screens, with sixteen 6-second white-screen periods presented and separated by variable intervals of 51–66 seconds. Group-level fixed-effects analyses were performed in FSL using a generalized linear model. The voxel wise significance threshold was Z > 2.3 (uncorrected for multiple comparisons).
Results
Study participants had a mean age of 60.4 ± 4.8 years, and 78% were male. The Kansas+ group showed greater activation, compared to the Kansas- group, across multiple cortical regions, including the left occipital pole (Zmax = 5.83, p = 0.001). Visual inspection of Kansas+ > Kansas- activation also revealed increased activity in the primary somatosensory cortex BA1 L (Z = 2.75), right insula (Z = 3.00), cingulate gyrus (Z = 3.11), left caudate (Z = 3.05) and left ventrolateral prefrontal cortex (Z = 3.25).
Conclusions
Kansas+ veterans demonstrated greater light-evoked activation in visual cortex and in regions involved in sensory–affective pain processing compared to Kansas– veterans. These exploratory findings suggest that light-evoked fMRI may capture central nervous system alterations relevant to symptom burden in Gulf War Illness.
This abstract was presented at the 2026 ARVO Annual Meeting, held in Denver, CO, May 3-7, 2026.