Table 1 Remote and unsupervised digital cognitive assessments for use in preclinical AD samples in alphabetical order

From: A scoping review of remote and unsupervised digital cognitive assessments in preclinical Alzheimer’s disease

Tool	Report included	N included in analyses or ongoing study	Cognitive domain (task)	Reliability	Analytical and clinical validity according to associations with:
Tool	Report included	N included in analyses or ongoing study	Cognitive domain (task)	Reliability	Neuropsychological tests	Aβ/tau
Altoida	Muurling et al., 2023¹⁴	56 (29 PET or CSF Aβ−, 27 Aβ+) + 65^b	Episodic memory (Augmented reality tasks) High-dimensional multi-modal data	Test-retest reliability: ICC = 0.48	Corr. with a cognitive composite (ρ = .56)	Lower scores in Aβ+ than Aβ− (β ≈ −0.40)⁹ⁱ; distin. Aβ− vs. Aβ+ (AUC = 0.76)
ARC	Wilks et al., 2021¹³	169 (81 CSF pTau:Aβ42−, 32 pTau:Aβ42+)	Episodic memory (Prices) Processing speed (Symbols) Working memory (Grids) EMA-based	Between-person reliability: ICCs > 0.80		Sundowning in CSF p-tau:Aβ42+ group (Cohen’s d = 0.19)
ARC	Nicosia et al., 2023³⁷	290 (212 with PET; 146 with CSF)		Between-person reliability: ICCs > 0.81; test-retest reliability: ICCs > 0.85	Corr. with a cognitive composite (r = −0.53)^h	Corr. with Aβ (r = 0.26) and tau PET (0.11); CSF Aβ42 (−0.23), t-tau (0.28), p-tau181 (0.25)^h
BRANCH	Papp et al., 2021⁶²	234 (144 with PiB; 129 with FTP)	Episodic memory (Categories, Face-Name-Occupation, Groceries, Signs) Processing Speed (Digit-Signs) Learning curves	Test-retest reliability: r = 0.81	Corr. with PACC (r = 0.62)	Corr. with cortical Aβ (r = −0.21) and entorhinal tau PET (−0.18)
	Papp et al., 2024¹¹	164 (128 PET Aβ−, 36 Aβ+)		Test-retest reliability: ICC = 0.94²¹	Corr. with subsequent decline on PACC (r = 0.54)	Reduced learning curves in Aβ+ (Cohen’s d = 0.49)
	Jutten et al., 2025¹⁶	167 (106 PET Aβ−/tau−, 46 Aβ+/tau−, 15 Aβ+/tau+)				Reduced learning curves in Aβ+/tau+ (βs = −0.47 to −0.61), and all Aβ+ (−0.24 to −0.58)
CANTAB	Weiner, Aaronson, et al., 2023⁶³	Brain Health Registry	Episodic memory (Delayed Matching to Sample, Paired Associates Learning^d, Pattern Recognition Memory, Verbal Paired Associates, Verbal Recognition Memory) Working memory (Spatial Working Memory, Digit Span, N-Back, Spatial Span) Executive function (Cambridge Gambling Task, Digit Symbol Substitution Task, Intra-Extra Dimensional Set Shift, Multitasking Test, One-Touch Stockings of Cambridge, Stockings of Cambridge, Stop Signal Task, Match to Sample Visual Search, Rapid Visual Information Processing) Psychomotor (Motor Screening Task, Reaction Time, Adaptive Tracking Task) Emotion and social (Emotional Bias Task, Emotion Recognition Task)		Awaiting validation in preclinical AD
CANTAB	Malzbender et al., 2024⁶⁴	AD-RIDDLE			Awaiting validation in preclinical AD
CBB	Langford et al., 2020⁷	4486 (3163 PET Aβ−, 1323 Aβ+)	Episodic memory (One Card Learning) Working memory (One-Back) Executive function (Identification) Psychomotor (Detection)	Test-retest reliability: ICCs = 0.90 to 0.95⁹⁸		Distin. Aβ− vs. Aβ+ (AUCs = 0.60 to 0.73)
	Kaye et al., 2021⁶⁵	TRC-PAD			Awaiting further validation in preclinical AD
	Weiner, Aaronson, et al., 2023⁶³	Brain Health Registry			Awaiting further validation in preclinical AD
cCOG	Malzbender et al., 2024⁶⁴	AD-RIDDLE	Episodic memory (Learning task, Recall task, Recognition task) Processing speed (Trail Making Tests) Psychomotor (Reaction tests)		Awaiting validation in preclinical AD
Cognitron	Leuzy et al., 2024⁶⁶	REAL AD	Episodic memory (Card Pairs, Mallas Memory Short, Paired Associates Learning) Working memory (Digit Span, Reverse Digit Span, Number Location Pairs, Picture Completion, Spatial Span) Processing speed (Blocks, 2D Manipulations, Choice Reaction Time, Trail Making Test) Executive function (Selective Attention, Target Detection, Switching Stroop, Stop Change Task, Tower of London) Fluid intelligence (Verbal Reasoning, Information Sampling, Odd One Out) Psychomotor (Motor Control, Simple Reaction Time) Mathematics (Balloons)		Awaiting validation in preclinical AD
C3 FNAME	Samaroo et al., 2020¹⁰^,a	94 (69 PET Aβ−, 25 Aβ+; 84 with FTP)	Episodic memory (Face Name Associative Memory Exam) Learning curves	Parallel-forms reliability: Cronbach’s α > 0.85²⁰		Reduced learning curves in Aβ+ (Cohen’s d = 0.66), corr. with tau (r = −0.22)
C3 FNAME	Jutten et al., 2022⁷⁵	114 (81 PET Aβ−, 33 Aβ+; 105 with FTP)		Parallel-forms reliability: Cronbach’s α > 0.85²⁰	Corr. with concurrent PACC5 decline (r = 0.69); distin. PACC5 decliners (AUC = 0.91)	Corr. with global Aβ (r = −0.20), entorhinal tau (−0.38), and inf.-temp. tau PET (−0.23)
ki:e SB-C	Gregory et al., 2022⁶⁷	SPeAk	Structured speech-based tasks	Test-retest reliability: rs = 0.57 to 0.72⁸⁴	Awaiting validation in preclinical AD
ki:e SB-C	König et al., 2023⁶⁸	PROSPECT-AD	Structured speech-based tasks	Test-retest reliability: rs = 0.57 to 0.72⁸⁴	Awaiting validation in preclinical AD
MemTrax Memory Test	Weiner, Aaronson, et al., 2023⁶³	Brain Health Registry	Episodic memory		Awaiting validation in preclinical AD
Mezurio	Muurling et al., 2021³⁰	RADAR-AD	Episodic memory (Gallery Game) Executive function (Tilt Task) Structured speech-based task (Story Time)		Awaiting validation in preclinical AD
MTD	Stricker et al., 2024⁶⁹	353 (228 PET Aβ−, 125 Aβ+; 250 with FTP)	Episodic memory (Stricker Learning Span) Processing speed (Symbols)	Test-retest reliability: ICCs > 0.71⁸⁵	Corr. with in-person NPTs of same construct (rs = \|0.26\| to \|0.51\|)⁸⁵	Distin. Aβ− vs. Aβ+ (AUCs = 0.63 to 0.77), Aβ−/tau− vs. Aβ+/tau+ (0.67 to 0.83)
MTD	Boots et al., 2024⁷⁰	684 (670 with PiB; 667 with FTP)		Test-retest reliability: ICCs > 0.71⁸⁵	Corr. with PACC (r = 0.68); corr. with in-person NPTs of same construct (rs = \|0.59\| to \|0.61\|)	Corr. with Aβ (ρ = −0.24), entorhinal tau (−0.23), and global tau PET (−0.21)
M2C2	Thompson et al., 2023⁷¹	69 (44 PET Aβ−, 25 Aβ+)	Episodic memory (Prices) Processing speed (Symbol Match) Working memory (Color Shapes) EMA-based	Between-person reliability: ICCs = 0.25 to 0.97⁸³^,e	Corr. with in-person NPTs of same construct (βs = \|0.22\| to \|0.44\|)⁸³	Distin. Aβ− vs. Aβ+ (AUCs = 0.73 to 0.77)
neotiv	Berron, Olsson, et al., 2024⁴⁶	100 (58 PET Aβ−, 42 Aβ+; 100 with FTM)	Episodic memory (Mnemonic Discrimination Task, Object-in-Room Recall, Complex Scene Recognition)	Test-retest reliability: ICCs = 0.65 to 0.83	Corr. with PACC (rs = 0.62 to 0.70); pred. concurrent PACC decline (βs ≈ 0.41 to 0.59)ⁱ	Pred. Aβ in precuneus (βs ≈ 0.23 to 0.38) and tau in the MTL (0.20 to 0.40)ⁱ
	Berron, Olsson, et al., 2024⁴⁶	BioFINDER-2			Awaiting further validation in preclinical AD
	Berron, Glanz, et al., 2024⁴⁴	DELCODE, WRAP			Awaiting further validation in preclinical AD
	Malzbender et al., 2024⁶⁴	AD-RIDDLE			Awaiting further validation in preclinical AD
	Leuzy et al., 2024⁶⁶	REAL AD			Awaiting further validation in preclinical AD
NeuroVocalix	Malzbender et al., 2024⁶⁴	AD-RIDDLE	Speech-based episodic memory		Awaiting validation in preclinical AD
NIH Mobile Toolbox	Weiner, Aaronson, et al., 2023⁶³	Brain Health Registry	Episodic memory (Faces and Names with Delay, Arranging Pictures) Working memory (Sequences) Processing speed (Number-Symbol Match) Executive function (Arrow Matching, Shape-Color Sorting) Language (Spelling, Word Meaning)		Awaiting validation in preclinical AD
Novoic	Fristed et al., 2022⁷⁶	115 (59 PET or CSF Aβ− , 56 Aβ+)	Speech-based storytelling	Parallel-forms reliability: ρs = 0.39 to .85⁴⁷^,f	Predicted PACC corr. with measured PACC (r = 0.74)	Distin. Aβ− vs. Aβ+ (AUC = 0.74)
Novoic	Weiner, Veitch, et al., 2023⁷⁷	ADNI4	Speech-based storytelling		Awaiting further validation in preclinical AD
OCTAL	Toniolo et al., 2024⁷²	99 (99 with plasma) + 352^c	Episodic memory (Object-in-Scene Memory Task, Rey-Osterrieth Complex Figure) Working memory (Oxford Memory Test, Freestyle Corsi Block Task) Processing speed (Digit Symbol Substitution Test, Trail Making Test)			Corr. with plasma p-tau181 (r = 0.34); pred. plasma p-tau181^j
ORCA-LLT	Lim et al., 2020¹²^,a	80 (42 PET Aβ−, 38 Aβ+)	Learning curves		Pred. baseline EM (β = 0.26); pred. retrospective EM change (β = 0.27)	Reduced learning curves in Aβ+ (Cohen’s d = 2.22); pred. Aβ PET (β = −0.23)
RWLRT	Weiner, Aaronson, et al., 2023⁶³	Brain Health Registry	Speech-based episodic memory		Awaiting validation in preclinical AD
SIDE-AD Platform	Saunders et al., 2024⁷⁸	SIDE-AD	Unstructured speech-based tasks		Awaiting validation in preclinical AD
TapTalk	Alty et al., 2024⁷⁹	Validation studies of TapTalk	Speech- and motor-based task		Awaiting validation in preclinical AD
WLA	van den Berg et al., 2024¹⁵	50 (27 PET Aβ−, 23 Aβ+)	Structured and unstructured speech tasks	Test-retest reliability: ICCs = −0.06 to 0.97 ^g		More pauses in Aβ+ than Aβ− (βs ≈ 0.30 to 0.34)ⁱ

ARC Ambulatory Research in Cognition, BRANCH Boston Remote Assessment for Neurocognitive Health, C3 FNAME Computerized Cognitive Composite Face-Name Associative Memory Exam, CANTAB Cambridge Neuropsychological Test Automated Battery, CBB Cogstate Brief Battery, ki:e SB-C ki:elements Speech Biomarker for Cognition, M2C2 Mobile Monitoring of Cognitive Change, MTD Mayo Test Drive, NIH National Institutes of Health, OCTAL Oxford Cognitive Testing Portal, ORCA-LLT Online Repeatable Cognitive Assessment—Language Learning Test, RWLRT ReVeRe Word List Recall Test, SIDE-AD Speech for Intelligent cognition change tracking and DEtection of Alzheimer’s disease, WLA Winterlight Assessment; PET positron emission tomography, CSF cerebrospinal fluid, PiB Pittsburgh compound B, FTP flortaucipir, FTM flutemetamol, EMA ecological momentary assessment, PACC Preclinical Alzheimer’s Cognitive Composite, NPTs neuropsychological tests, corr. correlated, pred. predicted, distin. distinguished.
^aIncluded in ref. ⁹
^bn = 65 did not complete remote testing.
^cn = 352 were included as an online sample with no biomarkers.
^dOnly Paired Associates Learning was announced to be used in association with the Brain Health Registry.
^eBetween-person reliability was greatly improved for the M2C2 when aggregating multiple measurements (≥0.76).
^fNovoic tested 153 correlations across test versions, and average ρs reached 0.73.
^gWLA tested 186 ICCs; reliability reached ICCs of ≥0.50 across two or more assessments.
^hHigher ARC score indicates worse performance, thus the correlations reported here are in the expected direction.
ⁱUnstandardized estimates originally reported were standardized using the standard errors or 95% confidence intervals and Ns reported.
^jOnly p values were reported for the models, including individual metrics. The best model to predict p-tau181 combined remote and in-person metrics, as well as Aβ42/40, adjusted R² = 0.50.
For samples including groups other than preclinical AD, effect sizes are reported for the findings pertaining to the preclinical AD sub-sample.

Back to article page

Search

Quick links