Table 1: Types of COT applied to MCI patients.

StudyDesign studySampleCOTCognitive domains trainedCCIIntervention formatOutcome measuresfMRI protocolIntervention fMRI results on MCIMCI COT clinical results

Rosen et al. [23]RCT double-blind12 MCIs
Age/education: 70.7/16.7
Age/education: 70.0/13.4 Petersen criteria
Computer-based auditory processing training by Posit Science CorporationAuditory, accuracy, processing speed, immediate and working memoryAudio books, online newspaper Myst game.Home-based computer-assisted for COT & CCI
24 sessions
5 times/week 90–100 min/session
Task-related hippocampal functional activity changes; performance at RBANs, and relationship with brain activityAuditory verbal task with discrimination of concrete from abstract wordsIncreased activation on L hippocampus after COT and decreased activation in this area after CCI. L hippocampal activation was associated with immediate memory on RBANs.COT MCI showed better gain in immediate memory performance on RBANs compared to CCI MCI.

Belleville et al. [17]CGT Single–blind15 COT HC Age/education: 70.0/13.4
15 COT MCI Age/education: 70.1/13.7
Petersen criteria
Visual imagery Method of loci
Face-name association
Hierarchical and semantic organization
Episodic memory and executive functionsNoGroup sessions
6 sessions
1 times/week
120 min/session
Task-related whole-brain functional activity changes; performance at fMRI task, and off-scan word list recallVerbal memory encoding and retrieval of a word listMCI COT increased activation in R inferior parietal lobule, frontal gyrus, R cerebellum, and basal ganglia on encoding and increased activation in R/L superior temporal gyrus, L frontal, and parietal cortex on retrieval. Activation in R parietal lobule was associated with better performance on delayed word recall.Improvement in performance on the retrieval scan task and on word list recall. Better performance on immediate recall compared with delayed recall

Hampstead et al. [18]Case series6 COT MCI Age/education: 73.5/15.7
Petersen criteria
Explicit memory strategy: visually identify a facial feature and then link it to a phonological cue to recall the associated nameAssociative episodic memoryNoIndividual sessions
5 sessions
2 weeks
Task-related whole-brain functional activity and connectivity; performance on face-name recognition off-scan taskAssociative encoding of face name pairsCOT-specific activations in L temporoparietal junction, L frontal operculum, L temporal cortex, and R/L medial areas of frontal, parietal, and occipital cortices. Increased connectivity in lateral frontal and parietal regionsImprovement on memory performance for TS and US stimulus after COT. Better performance and fast reaction times in recognition of TS

Hampstead et al. [20]RCT single blind16 HC vs. 18 MCI 8 COT HC
Age/education: 72.1/15.8
8 CCI HC Age/education: 72.1/16.5
9 COT MCI Age/education: 71.7/17.4
9 CCI MCI Age/education: 70.8/16.8
Petersen criteria
Explicit memory strategy: identify a salient feature within the room near the object, use a verbally based “reason” that related the object to the specific feature, and then form a corresponding mental imageAssociative episodic memoryRepeated exposition of stimuliIndividual sessions for COT & CCI
5 sessions
2 weeks
Task-related hippocampal functional activity; performance on-scan retrieval taskAssociative encoding and retrieval of object location association taskCOT aMCI showed increased activation in L hippocampal body on encoding of TS and US. Increased activation in R/L hippocampal body and tail on retrieval of TS and increased activation in R hippocampus for USCOT groups showed better performance on-scan retrieval task for TS compared to CCI groups.

Balardin et al. [24]CGT single blind18 COT HC Age/education: 68.25/11.19
Age/education: 69.5/9.2
17 COT MCI Age/education: 69.5/9.2
Petersen criteria
Semantic strategic training: organization of word lists into categories during encoding and retrieval according to their categoryEpisodic memory and executive functionsNoIndividual session
1 session
30 min/session
Task-related whole-brain functional activity; performance at SR and UR word lists, recall, and relationship with brain activityVerbal memory encoding of SR and UR word listsCOT MCI showed increased activation in frontoparietal network regions including L DLPFC and L VLPFC and decreased activations on the occipital cortex. Increased activation on R superior frontal gyrus and vmPFC cortex related to semantic strategy implementation. Better performance on strategic clustering was predictive of increased activation on OFC cortex.Improvement on performance on word list recall and on semantic strategy implementation after COT. Better recall on the SR list compared to UR list.

Lin et al. [15]RCT single blind21 MCI
10 COT MCI Age/education: 1 72.9/1
11 CCI MCI Age/education: 73.1/5
Petersen criteria
Vision-based speed of processing training online program by Posit Science CorporationVisual accuracy, processing speed attention, and working memoryComputer-based activities: online crosswords, sudoku, and solitaire games24 sessions
6 weeks
4 times/week 60 min/session
Home-based computer-assisted
Connectivity in DMN and CEN; performance at visual processing speed, working memory, verbal fluency, and IADL tasksResting stateIncreased connectivity on CEN after COT. COT MCI showed increased connectivity on DMN compared to CCI MCI.Improvement on performance on visual processing speed and attention test, and also on transfer domains tests related to working memory and IADLs after training

De Marco et al. [16]RCT single blind37 MCI
23 COT MCI Age/education: 73.74/8.7
14 CCI MCI Age/education: 73.74/10.5
Petersen criteria
Cognitive training packageSemantic processing, reasoning, executive functions, and episodic memory retrievalSocial interaction activities: volunteering, tour guiding, attending a club, or gardeningHome-based computer-assisted
20 sessions
5 times/week 60–90 min/session
Connectivity in DMN - performance at CCRI and relationship with brain activityResting StateIncreased connectivity on Precuneus – Cuneus after COT. Enhances on CCRI were associated with increased connectivity on parietal DMNNo improvement verified on neuropsychological measures applied

RCT: randomized controlled trial; CGT: controlled group trial; COT: cognitive oriented treatment; CCI: control condition intervention; MCI: mild cognitive impairment; HC: healthy controls; fMRI: functional magnetic resonance imaging; TS: trained stimulus; US: untrained stimulus; SR: semantic related word list; UR: unrelated word list; R: right; L: left; DLPFC: dorsolateral prefrontal cortex; VLPFC: ventrolateral prefrontal cortex; vmPFC: ventromedial prefrontal cortex; OFC: orbitofrontal cortex; CEN: central executive network; DMN: default mode network; IADLs: instrumental activities of daily living; CCRI: composite cognitive change ratio index.