Table 1.

Key findings and study characteristics of studies on air pollution exposure.

First author (study year)	Toxicant	N	Age at toxicant measure	Toxicant measure	Age at MRI	MRI measure	Exposure main effect
Alemany et al. (2018)	PAH, EC, and NO2	163	7-11 years	High volume sampler	7-11 years	Basal ganglia vol. (caudate, putamen, globus pallidus)	Greater exposure to PAH and NO2 associated with smaller caudate vol. only.
Beckwith et al. (2020)	ECAT	135	First year of life	Land use regression	11 – 14.71 years	Whole brain CT, Whole brain VBM	High ECAT group had reduced GMV in bilateral cerebellum and left parietal lobe, and thinner parietal and frontal cortex. No other associations.
Brunst et al. (2019)	ECAT	127	At birth, average from birth to MRI, average past year	Land use regression	At age 12 MRI visit	MRS voxel in bilateral ACC (metabolites : mI, NAA, Cr, Cho, Glu, GLX, GSH).	Elevated past year ECAT associated with elevated mI in bilateral ACC. No associations with other time periods or metabolites detected.
Burnor et al. (2021)	PM2.5	7602	9-10 years	Hybrid spatiotempor al model	9-10 years	rND, hD, rNO, MD, FA of all white matter fibers in each hemisphere and of ATR, CGC, CGH, CC, CST, FX, UF, IFO, ILF, and SLF in each hemisphere.	Hemisphere-stratified models revealed positive associations between PM2.5 and rNO in L hemisphere tracts (CGH, UF, FX, SLF) and R hemisphere tracts (UF, FX). Negative associations between PM2.5 and MD globally, in L hemisphere tracts (ATR, CGH, FX, SLF, UF, IFO, ILF), and R hemisphere tracts (ILF, UF, CGH, FX). No associations with FA, rND, or hD.
Calderon-Garciduenas et al. (2008)	Air pollution (general)	36	N/A	Lifetime residency	M=10.71 years	WMH	7.6% of Polotitlan youth had WMH, vs. 56.5% from Mexico City.
Calderon-Garciduenas et al. (2011)	Air pollution (general)	30	N/A	Lifetime residency	M= 6.95 years	WMH, cortical GMV and cortical WMV of lobes, CSF, vol. of hippocampus, caudate, putamen, globus pallidus, amygdala	WMH+ children (vs. WMH−) had greater WMV in R parietal lobe. Polotitlan children had greater WMV in bilateral temporal and R parietal, lobes (compared to Mexico City youth). No associations with any other ROI.
Calderon-Garciduenas et al. (2012)	Air pollution (general)	30	N/A	Lifetime residency	M=7.13 years	WMH, combined GMV and WMV of lobes, vol. of frontal cingulate, parietal cingulate, cerebellum, CC, PFC, insula, hippocampus, caudate, putamen, globus pallidus, amygdala	Between baseline and follow-up, WMH+ children had greater volumetric brain growth (GMV and WMV) in temporal, parietal, and frontal regions, compared to WMH-children. WMH+ and Polotitlan children did not differ. No associations with other ROIs.
Calderon-Garciduenas et al. (2015)	Air pollution (general)	57	N/A	Lifetime residency	M = 12.45 years	MRS voxels in R and L hippocampus (metabolites : NAA, Cho, Cr, mI)	Children with APOE e4 genotype from Mexico City, compared to children from Polotitlan, had reduced NAA/Cr ratio in R hippocampus.
Cserbik et al. (2020)	PM2.5	9697	9-10 years	Hybrid spatiotemporal model	9-10 years	31 SA regions, 27 CT regions, 8 subcortical vols, total SA, total CT, TBV, cortical GMV, subcortical GMV, cortical WMV, CSF, ICV, ventricle vols.	Greater PM2.5 exposure associated with: (1) thinner L frontal, R temporal lobes; (2) thicker R frontal, L temporal lobes, (3) thicker and thinner bilateral cingulate, (4) reduced SA in R frontal pole, L cuneus, (5) greater SA in R lateral OFC, (6) greater R thalamus, R pallidum, L accumbens vol, (7) reduced L putamen, L pallidum vol. No other associations detected.
Guxens et al. (2018)	NO2, particulate matter (coarse and fine particles), and absorbance of fine particles	783	Prenatal	Land use regression	6-10 years	Whole brain CT, TBV, cortical GMV, cortical WMV, subcortical GMV, vol. of caudate, putamen, pallidum, accumbens, hippocampu s, amygdala, thalamus, ventricles	Higher fine particle exposure associated with thinner R frontal, R parietal, and L occipital cortex. Higher coarse particle exposure associated with thinner R lateral OFC. Absorbance of fine particles associated with thinner L fusiform. No associations with any other measure.
Lubczynska et al. (2021)	NO2, NOx, PM10, PM2.5, PMcoarse, PM2.5 absorbance, PAH, BAP, OC, Cu, Fe, K, Si, Zn, OP, UFP	3133	Prenatal, average from birth to MRI	Land use regression	9-12 years	TBV, cortical GMV, subcortical GMV, cortical WMV, vol. of CC, cerebellum, thalamus, caudate, putamen, pallidum, hippocampus, amygdala, accumbens. Whole brain CT and SA.	Both thinner and thicker cortex associated with exposure, prenatally and across childhood, depending on region. Across childhood exposure associated with less PFC SA, but greater SA in other regions. No associations between pregnancy exposure and SA reported. Both pregnancy and across childhood exposure associated with smaller CC, smaller hippocampus , larger amygdala, larger cerebellum. No associations with other measures.
Lubczynska et al. (2020)	NO2, NOx, PM10, PM2.5, PMcoarse, PM2.5 absorbance, PAH, BAP, OC, Cu, Fe, K, Si, Zn, OP, UFP	2,954	Prenatal, average from birth to MRI	Land use regression	9-12 years	Global FA, global MD, follow-up analyses on individual white matter tracts	In single pollutant analyses, prenatal and average childhood exposure predicted lower FA and higher MD for many air pollutants assessed.
Mortamais et al. (2017)	Total sum of PAH levels, BAP	242	8-12 years	High volume sampler	8-12 years	Basal ganglia vol. (caudate, putamen, globus pallidus), brain parenchyma l fraction	Outdoor and indoor BAP predicted smaller caudate vol. Effect of total outdoor PAH on caudate vol. was similar but smaller than the effect of outdoor BAP. No other associations found.
Mortamais et al. (2019)	PM2.5	186	Prenatal, average across first 2 years of life, time of MRI	Land use regression	8-11.7 years	Cortical GMV, cortical WMV, vol. of lateral ventricles and CC sub-regions	Greater 3rd trimester exposure linked to decreased vol. of anterior and body of CC. Associations did not survive correction for multiple comparisons. No associations with other ROIs or time periods.
Perez-Crespo et al. (2022)	NOx, NO2, PM2.5, PM2.5 absorbance, PM10	2197	Prenatal, 0-3, 3 – 6, 6 to time of MRI	Land use regression	9-12 years	Resting state functional connectivity of 6 brain networks	NO2 from 0-3 and NOx from 3-6 both associated with higher visual - task positive network connectivity. PM2.5 absorbance from 0-3 associated with higher connectivity between visual and auditory, task positive, and task negative networks as well as between auditory and task positive and task negative networks, and between task positive and task negative networks. PM2.5 absorbance from 0-3 also associated with greater connectivity within the visual, motor, task positive, and task negative networks. No other associations with other air pollutants or time periods.
Peterson et al. (2015)	PAH	40	Prenatal, at age 5	Personal backpack monitor for 48 hours; spot urine sample from child	7-9 years	Whole brain CT, surface morphology	Prenatal exposure to PAH correlated with reduced vol. in bilateral frontal, superior temporal, and parietal lobes and in L rostro-caudal extent of mesial surface—associations driven by changes in white matter. PAH at age 5 correlated with reduced WMV in bilateral dorsal PFC, especially SFG. No association with CT.
Peterson et al. (2022)	PM2.5, PAH	332	Prenatal	Personal backpack monitor for 48 hours; spatiotempor al model	6-14 years	Whole brain CT, FA, ADC, MRS (metabolites: NAA, Cr, Cho, GLX), rCBF, surface morphology	PAH & PM2.5 associated with (1) thicker lateral temporal, posterior inferior, and mesial wall cortices, (2) thinner dorsal parietal cortices, (3) smaller white matter vols in the inferior parietal lobes and cingulate gyrus, (4) larger white matter vols in dorsal convexity, SFG, and posterior inferior brain surface, (5) higher FA in basal ganglia and ACC, (6) lower rCBF in frontal gray matter, (7) higher NAA and Cho in cingulate. ADC alterations different for PM2.5 vs. PAH.
Pujol, Fenoll et al. (2016)	Copper	263	8-12 years	High volume sampler	8-12 years	Whole brain CT,Whole brain VBM, FA, resting state functional connectivity with seeds in PFC and caudate	Greater Cu associated with higher gray matter concentratio n in the caudate, greater FA in white matter in and around the caudate, diffusion changes in the caudate, reduced resting state functional connectivity between the caudate and frontal lobe operculum, and increased CT in L SMA.
Pujol, Martinez-Vilavella et al. 2016	Average of EC, NO2	239	8-12 years	High volume sampler (EC), passive dosimeter (NO2)	8-12 years	Whole brain CT, Whole brain VBM, FA, brain activation during sensory task, resting state functional connectivity with seeds in medial frontal, dorsal frontal, PCC, and SMA, MRS voxel in L frontal lobe (metabolite:Cho/Cr ratio)	In the resting state analysis, higher exposure to air pollution was associated with weaker connectivity within DMN and stronger connectivity between the DMN and the lateral boundary of the DMN. In the task analysis, greater exposure to air pollution was associated with lower deactivations in the SMA and somatosenso ry cortex. No association with sMRI, FA, or MRS measures.

Note. See notes following Table 9 for table notes and a comprehensive list of abbreviations.