Table 2:

Overview of prospective observation studies investigating level change or correlated change associations of GM and cognitive abilities.

Brain measure/author, study	Sample size at baseline (age range in years)	Measurement occasions	Time intervals in years	Healthy cognitive aging: screening instruments at baseline	Statistical method (change measure)	Covariates	Neuroanatomical measure (cf. Table 1)	Brain regions	Cognitive variables	Brain-cognition relations (+ = positive, − = negative)
GM
Cohen et al. (2001)	25 female (~50–70a)	2	2	Noneb	Correlation (change ratiobrain/cog)	None	GM volume	Hippocampus	Episodic memory Buschke-fuld test (free and delayed recall, consistency) Wechsler memory scale-revised (verbal, visual, general, attention, and delay index) Verbal fluency Letter, category, supermarket items Visuospatial processing Rey figure test (copy, recall). Block design test	Level-change Not reported Correlated change No effect
Raz et al. (2007)	46 (45–77)	2	5–6	MMSE ≥ 26, BIMC ≥ 85% correct, CES-D ≤ 15	Repeated-measures regression analysisc,d	Age	GM volume, adjusted for ICV	Average cortical volumee	Fluid intelligence CFIT (subtest 3)	Level-change Not reported Correlated change No effect
Raz et al. (2008)	87 (20–77)	2	5–6	See Raz et al. (2007)	Bivariate LCS (latent change scorebrain/cog)	Age Sex VR	GM volume, adjusted for ICV	LPFC OFC Inferior temporal gyrus Fusiform gyrus VC Hippocampus Entorhinal cortex Caudate Cerebellum	Fluid intelligence CFIT (subtests 1 and 3)	Level-change: Fluid intelligence T1/T2 + Δentorhinal cortexf Correlated change Not reportedg
Leow et al. (2009), ADNI	40 normal controls (75.27 ± 5.33)	2	1	MMSE ≥ 28, CDR = 0	Regression analysis (structural change mapbrain)h	None	GM volume	Temporal lobe	Episodic memory Logical memory (immediate, delayed recall) AVLT (delayed recognition)	Level-change No effect Correlated change No effect
Persson et al. (2012), Betula	26/15 with VR (55–79)	2	6brain/10cog	MMSE ≥ 24	Regression analysis (difference scorebrain/cog)	Age APOE-ε4	GM volume, adjusted for body size	Hippocampus	Episodic memory Composite of five tasks (immediate/delayed recall of commands with and without enactment, auditory word list)	Level-change ΔEpisodic memory + right hippocampus T2 Correlated change ΔRight hippocampus + Δepisodic memory
Fjell et al. (2014), ADNI	94 normal controls (55–90)i	2	1	MMSE ≥ 24 CDR = 0 Logical Memory Test ≥ 3–9	Correlation (change ratiobrain/cog)	Nonej	Cortical thickness	Entorhinal cortex	Episodic memory AVLT (learning, delayed recognition)	Level-change ΔEntorhinal + AVLT delayed T2 Correlated change ΔEntorhinal + ΔAVLT learning
Daugherty et al. (2015)	125 (19–77)	2	2	MMSE ≥ 26 CES-D < 16	Bivariate LCSbrain/cog	Age Iron content (baseline and change)	GM volume, adjusted for ICV	Hippocampus Caudate Putamen	Episodic memory Logical memory (immediate, delayed recall) Working memory Listening span Verbal N-back	Level-change Not reported Correlated change No effect
Mak et al. (2015b)	33 normal controls (76.7 ± 5.3)	2	1	No history of dementia or CAMCOG score < 80	Correlation (change ratiobrain)h	Age Sex	Cortical thickness	Whole brain	Episodic memory Brief BVMT Global cognition CAMCOG	Level-change Not reported Correlated change No effect
Ritchie et al. (2015b), Lothian Birth Cohort	657 (Mean age 73)	2	3	MMSE ≥ 24	Multivariate LCSkbrain/cog	Age Sex	GM volume	-	Fluid intelligence Matrix reasoning Block design Digit span backward Letter-number sequencing Episodic memory Logical memory (immediate, delayed recall) Verbal paired associates, Spatial span (forward, backward) Processing speed Digit-symbol Symbol search Simple-/four-choice reaction time task Inspection time task	Level-change No effect Correlated change ΔGM volume + Δspeed, Δfluid intelligence
Fjell et al. (2016), Cognition and plasticity through the lifespan	56 (63–86)	2	3.1	MMSE ≥ 26, BDI ≤ 16 IQ ≥ 85	Stepwise regression analysis (difference scorebrain/change ratiocog)	Age Sex Head motion time interval Δfunctional connectivity (caudate, hippocampus)	GM volume	Hippocampus Caudate	Episodic memory CVLT immediate recall	Level-change Not reported Correlated change No effects
Möller et al. (2016)	34 normal controls (60.6 ± 6)	2	2	Noneb-	Correlation (change ratiobrain/cog)	None	Cortical thickness	Left and right lobar compartments (frontal, parietal, temporal, occipital), Cingulate Insula	Episodic memory AVLT (immediate, delayed recall, recognition) Visual association test A EF Digit span backwards TMT B Letter fluency Stroop color-word Attention TMT A Digit span forward Letter digit substitution Visuospatial functioning Number location Language Picture naming Category fluency	Level-change Left and right occipital T1 + Δepisodic memory Right occipital T1 + Δlanguage Correlated change ΔRight cingulate + Δepisodic memory ΔRight occipital − ΔEF
Persson et al. (2016)	167 (19–79)	2	2	MMSE ≥ 26, CES-D ≤ 15	Bivariate LCSbrain/cog	Age SES VR AP0E-ε4	GM volume, adjusted for ICV	LPFC, Hippocampus Parahippocampal gyrus Cerebellar hemispheres VC	Fluid intelligence CFIT (four subtests) Episodic memory Name-picture associative memory, story recall (both immediate/delayed recall) Vocabulary two vocabulary lists (selecting synonyms)	Level-change Hippocampus T1, parahippocampal gyrus T1, cerebellar hemispheres T1 + Δfluid intelligence Episodic memory T1 + ΔLPFC Correlated Change No effectl
Anblagan et al. (2018), Lothian Birth Cohort	655 (72.5 ± 0.7)	2	3	MMSE ≥ 24	Multivariate LCSbrain/cog	Age Sex	GM volume FA Diffusivity	Hippocampus (volume unadjusted, adjusted for ICV, FA, MD, longitudinal relaxation time T1)	Working memory Spatial span (forward, backward) Digit span backward Letter-n umber sequencing Verbal memory Logical memory (immediate, delayed recall) Verbal paired associates, Processing speed Digit-symbol substitution Symbol search Four-choice reaction time task Inspection time task	Level-change MD T1 − Δverbal memory, Δspeed, Δworking memory Correlated change ΔMD + Δworking memory
Gorbach et al. (2017), Betula	264 (55–80)	2brain/3–4cog	4brain/15cog	Noneb	Correlation (change ratiobrain/regression slope estimatescog)	Age VR	GM volume	LPFC (two subregions) Rostral anterior cingulate cortex Hippocampus Parahippocampal gyrus Inferior temporal cortex Superior parietal cortex Precuneus Putamen Lateral occipital cortex	Fluid intelligence Block design Episodic memory Composite of five tasks Ommediate/delayed recall of commands with and without enactment, auditory word list) Processing speed Letter-digit substitution Letter comparison Figure comparison Verbal fluency Letter fluency task (three conditions)	Level-change Not reported Correlated change ΔHippocampus + Δepisodic memory
Hohman et al. (2017), ADNI	379 normal controls (73.9 ± 5.9)	> 2m	4	MMSE ≥ 24 CDR = 0 Logical memory test ≥ 3–9	MLM analysisn	Age CSF biomarkers (Tau, Ab-42)	GM volume, adjusted for ICV	Default mode network ROIs (DMN) Frontal cortex (FC) Medial temporal cortex (MTL) Sensorimotor areas (SENS)	Episodic memory AVLT (learning, delayed recognition), MMSE Logical memory Alzheimer’s disease assessment scale (cognitive subscale) EF TMT (A and B) Digit span backward Digit symbol Category fluency (animals/vegetables) Clock drawing	Level-change Hippocampus T1 + ΔEF, Δepisodic memory Correlated Change No effecto
Leong et al. (2017), Singapore Longitudinal Aging Brain Study	111 (56.1–83.1)	2–4	1.5–2	MMSE ≥ 26	Correlation (MLM slope estimatesbrain/cog)	Age Sex Education VR BMI ICV	GM volume	Total GM Lobar GM (frontal, parietal, temporal, occipital) Hippocampus	Global cognition Composite of all abilities below EF Category fluency (three categories) Design fluency TMT B Processing speed TMT A Symbol search Symbol-digit modalities Visuospatial memory Visual paired associates test (immediate, delayed recall) Verbal memory AVLT (immediate, delayed recall) Attention Digit span Spatial span (forward and backward)	Level-change Not reported Correlated change ΔTotal GM, Δlobar GM (frontal, parietal, temporal), Δhippocampus) + Δglobal cognition ΔOccipital GM − Δglobal cognition ΔHippocampus + Δverbal memory, ΔEF
Sala-Llonch et al. (2017)	89 (64.74–89.79)	2	2	Noneb	Regression analysis (change ratiobrain/cog)	Age Sex	Cortical thickness	-	Episodic memory Word list memory task Verbal Fluency Animal Naming Test (ANT) EF TMT B	Level-change Not reported Correlated change ΔCortical thickness (right hemispheric clusters covering supramarginal, postcentral and inferior parietal cortices) + ΔANT
Yuan et al. (2018)	73 (51–85)	1–4	~1–5 (max 8 years in the study)	MMSE ≥ 26, CES-D < 16	MLM analysis	Age	Cortical thickness GM volume	Total cortical thicknessp Cortical GM volumep	Fluid intelligence CFIT (four subtests) Crystallized intelligence Vocabulary Tests (two versions)	Level-change Fluid intelligence T1/T2 + Δcortical GM volume Cristallized intelligence T1/T2 − Δcortical GM volume, Δtotal cortical thickness Correlated change Not reportedg

brain, brain measurements; cog, cognitive measurements; ADNI, Alzheimer’s disease neuroimaging initiative; MMSE, mini-mental-state-examination; BIMC, blessed information-memory-concentration test; CES-D, Center for Epidemiologic Studies Depression Scale; CDR, clinical dementia rating; LCS, latent change score model; MLM, multilevel model; change ratio, a measure of change proportionalized on the baseline level and in some cases also on the time interval (for exact formula consult original publications); difference score, value at T1 – value at T2; ICV, intracranial volume; SES, socioeconomic status; VR, vascular risk (for details consult original publications); APOE-ε4, apolipoprotein E allele 4; BMI, body mass index; ROIs, regions of interest; LPFC, lateral prefrontal cortex; OFC, orbitofrontal cortex; VC, primary visual cortex; TMT, trail making test; AVLT, Rey auditory verbal learning test; CFIT, Catell Culture Fair Test; CVLT, California verbal learning test; EF, executive functions; T, measurement occasion (e.g. T1 = measurement occasion 1); Δ, change; GM, gray matter.

^aAge reported separately for two subgroups of APOE-ε4 negative (n = 9, M age = 60.6 ± 10.2 years) and APOE-ε4 positive (n = 16, M age = 55.1 ± 6.0 years) participants.

^bNo details provided on screening instruments and/or cut-offs used (for description of healthy aging criteria consult original publication).

^cTotal WMH volume was included as a predictor in the same model (cf. Table 3).

^dChange measure was not clearly described.

^eUnclear, which cortical regions were included in this variable.

^fNo longer significant after sex and VR were included.

^gA lack of interindividual differences in intraindividual change in the cognitive variables precluded correlated change analyses.

^hChange measure for the cognitive variables not clearly describe.

ⁱAge range of the full sample, as age range of the normal subsample was not reported.

^jAge had no significant relation with any of the brain/cognitive performance measures, therefore it was not included as a covariate.

^kAll cognitive variables were included in the same statistical model.

^lOf the cognitive measures, only intelligence showed significant interindividual differences in intraindividual change and was used for the correlated change analyses.

^mChange was modeled over age (instead of time; consult original publication for details).

ⁿNumber of measurement occasions not clearly indicated.

^oThe correlated change analysis was conducted using a latent measure of global brain changes that comprised the single composites.

^pMeasures of interest are modeled across 19 regions, using region as crossed random effect (for details consult original publication).