This study was stopped due to time and budgetary constraints.

Mild cognitive impairment (MCI) is the prodrome of the cognitive function declining before Alzheimer's disease or other dementia showed up, the impairments of language, visuospatial relationship, attention, and memory included and instrumental activities of daily living (IADL) influenced. MCI is considered as a transitional stage between normal aging and mild dementia, and the patients with MCI has differently fluctuated cognitive functions in a period of time, such as from normal cognition to MCI or developing to dementia. The annual conversion rate (ACR) of older adults with normal cognition developed to MCI is 30%, and 5% in clinical setting, and community, respectively. Not all of patients with MCI develop to Alzheimer's disease, the reversion of patients with MCI to normal cognition exists. However, MCI is a significant risk factor. The ACR of older adults with normal cognition or MCI developed to dementia is 1-2%, and 5-15%, respectively; moreover, about half of patients with MCI developed to dementia in 5 years.

Cognitive training (CT) improves cognitive functions with repetitive practicing standardized cognitive tasks of specific cognitive functions, such as memory, attention, or problem solving. CT has widely defined including strategy training, in which contained cognitive exercise, strategy indicating and practicing to reducing cognitive impairments and improving performances. CT is more effective for MCI. Recently, computer-based CT (CCT) with many advantages gradually replaced the traditional paper-pencil form. Brief systematic review showed that the computer-based intervention had positive effects on behavioral symptoms, such as depression and anxiety, in patients with MCI and/or dementia. Previous studies demonstrated that computer-based intervention exhibited moderate treatment effects on overall cognitive functions in patients with MCI, and also had positive effects on learning, short-term memory, and behavioral symptoms.

Older people with cognitive impairments is expected to increase by global aging. It is important for improving or maintaining cognitive functions of older adults with MCI. The efficacy of the CCT on cognitive functions, neuropsychiatric symptoms, daily functions, and brain activated imaging of the magnetoencephalography (MEG) of in older adults with MCI is worth to explore for busy clinical practice.

The study design was a prospective and single-blinded randomized controlled trial. 36 participants with MCI were recruited and demographic data (age, gender, education level /years of education, marriage status etc.) were also collected. The MCI participants underwent the comprehensive review at baseline including neuropsychological assessment and Magnetic Resonance Imaging or Computerized Tomogram. Randomization treatment assignment will be generated by the random number table and assign the patient's intervention group accordingly. Sealed opaque envelopes containing the CCT group, or the dosage-matched control group sheets will be prepared and given to the therapists.

Based on the t-test effect size of index d 1.00 indicates, an estimated 17 participants in each group will be required for a power of 0.80 with a two-sided type I error of 0.05. Considering the 5% (q) drop rate, we will need to recruit 18 participants (N/1-q) for each group. The participants were randomly allocated to either the CCT group (18 participants) or the dosage-matched control group (18 participants) with individualized intervention for 30 minutes a day, 3 times a week for 4 consecutive weeks. Clinical outcome measures, and the imaging of the MEG were administered at pre- treatment, post-treatment and 1-month follow-up for further analysis.

Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS) version 19.0. In statistical testing, two-sided p value ≤ 0.05 was considered statistically significant. The distributional properties of continuous variables were expressed by mean ± standard deviation (SD), categorical variables were presented by frequency and percentage. The differences in the distributions of continuous variables, categorical variables between the treatment and control groups were examined using two-sample t-test, Wilcoxon rank-sum test (or Mann-Whitney U test), and chi-square test. In addition, the minimum norm estimates (MNE), source-based time-frequency analysis, cross-frequency coupling, and functional connectivity were used to explore the differences of the activation of brain functions in participants between different treatment groups. The t-test was used to explore the differences of reaction time, rate of correction, and physical signals in stimulus reaction test of the MEG between in participants between different treatment groups. The correlation statistics was used to explore relationships between the scores of outcome measurements and the physical signals.

Inclusion Criteria:

• (1) were aged ≥65 years
• (2) amnestic type MCI was made using National Institute on Aging-Alzheimer's Association workgroups diagnostic guidelines for Alzheimer's disease with the clinical dementia rating scale (CDR) global scores of 0.5
• (3) could follow command, understand the content of the assessments, and cooperate with treatment interventions through verbal communication

Exclusion Criteria:

• (1) had the score of Geriatric Depression Scale-Short Form (GDS-SF) > 7, indicating depression status
• (2) had the score of Barthel Index (BI)≠100, indicating dependent basic daily living of activities
• (3) were diagnosed with other MCI subtypes, including frontotemporal dementia or Lewy Body those present typically different MCI syndromes
• (4) had other neuropsychotic diseases
• (5) could not administrate with MEG
• (6) could not participate due to severe health problem

Data are available from the Institutional Review Board of the Taipei Hospital, Ministry of Health and Welfare for researchers who meet the criteria for access to confidential data.