The work stands somewhat apart: its closest neighbors study white matter connections, the packaging of visual images, and age-related brain maps, but no one links pupillometry to neurochemical systems across the lifespan. Pupillometry has long been attractive as an accessible tool for assessing the activity of the locus coeruleus, the first site where tau protein accumulates in Alzheimer's disease. However, the pupil's link to the locus coeruleus is unstable and context-dependent, and age-related changes in the pupil's periphery can mask this link. So a robust characterization of pupillary dynamics and their associations with neurotransmitter systems in adults of different ages was needed.
89 participants aged 19–78 performed a rare-target detection task inside an MRI scanner while an infrared camera sampling at 1000 Hz recorded pupil diameter. Brain activity was measured with multiecho fMRI at 2 mm resolution. The locus coeruleus was localized in two ways: a Bayesian algorithm and manual tracing. Then correlation maps were built between pupil time series and the BOLD signal in each voxel, and these maps were compared with 19 PET maps of receptor and transporter density across nine neurotransmitter systems.
Rapid pupil changes (the derivative) correlated with activity across extensive regions of the cortex, cerebellum, and brainstem, and this map correlated positively with noradrenaline transporter (NET) density. Absolute pupil size, after removing the contribution of changes, correlated with the density of mu-opioid and cannabinoid CB1 receptors. The link between pupil changes and the brain was weaker in older adults: the number of significant voxels declined with age, and the correlation with NET diminished. At the same time, the link between pupil size and opioid and cannabinoid receptors did not change with age.
The work differs from its neighbors in that it is the first to split the pupillary signal into two independent components, size and change, and tie each to different neurochemical systems. Neighbors study structural or cognitive aspects of aging, but they do not use the pupil as a bridge to neuromodulators. This is the first test of the hypothesis that rapid pupil changes reflect noradrenaline rather than the cholinergic system, in an adult sample spanning the entire lifespan.
The topic could move toward longitudinal studies in which pupillometry is combined with PET scanning of the noradrenaline transporter in older adults, to check whether the weakening link really does reflect degradation of the locus coeruleus. Studying the opioid and cannabinoid systems as modulators of tonic pupil size in clinical populations is also promising.
The results will influence neuroimaging of aging, clinical pupillometry, and the development of noninvasive markers for Alzheimer's disease.
Conduct a PET study of NET in older adults and compare it with pupillometry. Develop protocols that separate the contribution of noradrenaline from that of the opioid system in pupil dynamics.
The work connects pupillometry to a fundamental problem in the neuroscience of aging: how neuromodulation changes with age and whether it can be measured noninvasively. It also touches on the question of how specific pupillary markers are for different neurotransmitter systems, which remains open.
The locus coeruleus occupies on average just 5 voxels at 2 mm resolution, making it one of the smallest objects that can be studied with fMRI.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.