search for


The Effect of Brain-computer Interface-based Cognitive Training in Patients with Dementia
J Korean Soc Phys Med 2018;13(4):59-65
Published online November 30, 2018;
© 2018 Journal of The Korean Society of Physical Medicine.

Se-Jung Oh, and Jeon-Nam Ryu

Department of Physical Therapy, Yeoju Institute of Technology
Jeon-Nam Ryu
Received August 21, 2018; Revised August 23, 2018; Accepted September 11, 2018.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.


The purpose of the present study is to investigate the changes in the cognitive function of elderly dementia patients residing in a residential care facility, following six weeks of brain-computer interface (BCI)-based cognitive training and to determine whether BCI-based cognitive training effectively improves their cognitive functions.


Thirty subjects diagnosed with dementia were randomly assigned to either the experimental or control group. Pre- and post-test cognitive function assessments were conducted using the mini mental state examination-Korean (MMSE-K) and Korean-dementia rating scale (K-DRS). The experimental group received BCI-based cognitive training, which consisted of games such as flying a ball and exploding a bomb, while the control group participated in music listening activities and National Health Gymnastics. Both groups engaged in a total of 18 sessions (3 times per week for 6 weeks, for 40 minutes per session).


After 6 weeks of intervention, the experimental group had significantly increased MMSE-K scores (19.53±1.30 to 22.20±1.15; p<.0011) and total K-DRS scores (87.20±4.16 to 99.33±1.15; p<.0011). In addition, the experimental group showed greater cognitive improvements than the control group.


The results of this study suggest that BCI-based cognitive training is a positive intervention tool for improving the cognitive function of dementia patients.

Keywords : Brain-computer interface, Cognition, Dementia
I. Introduction

In Alzheimer’s disease or vascular diseases, dementia is characterized by neurodegenerative changes that lead to cognitive impairment and cause emotional pain or distress to the patients and their families, as it restricts daily activities, reduces participation in social activities, and causes depression. Interventions that can restore the cognitive function and delay cognitive impairment have helped improve the quality of life and financial situation of these elderly patients (Zhu et al., 2013; Deary et al., 2013; Plassman et al., 2011).

Cognitive training involves a series of cognitive tasks usually related to memory, concentration, and problem solving. Cognitive training has been observed to increase brain activities associated with memory in elderly people with mild cognitive impairment (Belleville et al., 2011). Computerized cognitive training (CCT) has been shown to decelerate the loss of cognitive function in healthy elderly persons and those with cognitive impairments by activating cognitive function (Almeida et al., 2012). A recent review of the available literature reported that CCT improved the cognitive function of healthy elderly subjects and had some effect on the cognitive function and anxiety in patients with mild cognitive impairment; however, the effect size of CCT was small, and there were not enough high-quality studies in the review (Reijnders et al., 2013; García-Casal et al., 2017). CCT usually requires the help of a therapist or a caregiver and cannot be performed by healthy elderly individuals or those with cognitive impairment on their own.

The brain-computer interface (BCI) refers to a technique in which the human brain is connected to a computer that analyzes the signals from cranial nerves in real time and transmits and communicates information without muscles or external devices (Broetz et al., 2010). The BCI has mainly been used to help quadriplegic patients or physically disabled patients by using real-time visual and auditory feedbacks (De Massari et al., 2013). Because active participation and practice are necessary to enable patients to use the BCI, the use of BCI has been limited to patients who could fulfil this requirement. However, a study that used BCI-based games showed that elderly persons or patients with cognitive impairment could participate in cognitive training using the BCI (Liberati et al., 2012). Lee et al. (2013) reported that BCI-based game programs can improve the memory and attention span of health elderly persons. However, dementia patients with mild cognitive impairment who reside in residential care facilities, such as senior care facilities, have limited opportunities to participate in cognitive training and cannot perform cognitive training without help from a therapist.

The objective of this study was to investigate changes in the cognitive function of elderly dementia patients with mild cognitive impairment, residing in a residential care facility, following six weeks of BCI-based cognitive training and to determine whether BCI-based cognitive training effectively improved their cognitive functions.

II. Methods

1. Participants

Elderly persons aged 60 years or more who resided in senior care facilities, received long-term care insurance from the government, were diagnosed with Alzheimer’s type dementia and vascular dementia were included. Thirty subjects who understood the content and purpose of this study signed the consent form (IRB No. MC13FISI0089). Patients with dementia who were unable to understand the content of the agreement received the consent of the family, and the rest of the subjects also received guidance and consent from their families.

Inclusion criteria were: 1) subjects who were diagnosed with dementia at least six months ago, 2) elderly persons who had MMSE scores of 18–23 points (Kramer et al., 1985), and 3) ability to sit independently on a chair for at least 30 minutes. Exclusion criteria were: 1) subjects who were undergoing other cognitive training, 2) previous history of epileptic seizure, and 3) visual or hearing impairment (Bae et al., 2012).

2. Experimental procedures

All subjects were subjected to pre-evaluation before study participation using the Korean-Mini-Mental Status Evaluation (MMSE-K) and the Korean version of Dementia Rating Scale (K-DRS). Subsequently, the 30 subjects were randomly divided into two groups using a number table. The experimental group was subjected to cognitive training using BCI-based games for six weeks, and the control group was subjected to music listening and National Health Gymnastics. All subjects participated in the training three times per week, 40 minutes per session, for six weeks. The subjects were re-evaluated after each training session.

The Mindset device (Mindset, Neurosky) was used during the BCI-based cognitive training. This device detects changes in brainwaves in the frontal lobe through a sensor attached to the subject’s forehead and calculates the subject’s concentration. The signals are converted to integers and sent to the computer via Bluetooth. A subject participating in the training sat comfortably on a chair using a headset and stared at a monitor screen 50 cm away. The subject played BCI-based cognitive training games that involved activities such as exploding bombs and floating balls (Kang et al., 2009). The bomb explosion game was played by a virtual avatar that burns objects (cars) from the hand and, with continued concentration, eventually explodes the objects. The bomb explosion game induced tension to activate the beta waves in charge of the conscious activities of the brain. The next game was the ball floating game. The ball floating game helped the subject focus his/her consciousness and reach a stable state to induce alpha waves. In this game, when the participant did not relax or concentrate on a floating ball, the ball slowly floated to the floor. The difficulty of the game gradually progressed from a light ball to a heavy ball. A bar graph and sounds were used to give real-time feedback on brain wave changes to help the subject focus on the training. The game difficulty was sequentially increased as the subject cleared each stage (Sung et al., 2012). To help the subjects understand how to participate in the training, simple feedback was provided during preliminary education and practice games. If subjects complained of tired eyes or fatigue, training was ceased and resumed after a sufficient break.

The control group participated in music listening activities and National Health Gymnastics according to the same schedule and for the same duration as the BCI training group. The titles and musical contents and how to perform the National Health Gymnastics were all explained to the subjects.

3. Measurements and statistical analysis

The MMSE was developed in 1975. The Korean version, developed in 1989, was used to evaluate the cognitive impairment of elderly subjects and identify those with dementia (Kwon and Park, 1989). The MMSE is a cognitive function assessment tool used to primarily identify and diagnose elderly persons who show clinical signs of cognitive impairment. The diagnostic reliability of this tool for dementia as a cognitive function assessment tool is .93 (Kim et al., 2003), and its reliability is .90 (Kim et al., 1998).

The K-DRS is the Korean version of the Dementia Rating Scale developed by Mattis that was refined and standardized, according to age and education level in Korea, to diagnose and evaluate the progress of Korean patients with dementia (Mattis, 1988). The K-DRS evaluates five domains of cognitive function that are important for diagnosing dementia. The domains include attention, the ability to receptively receive and process information and to concentrate; initiation and preservation, the ability to effectively execute an intended action; construction, spatiotemporal ability and motor reactions; conceptualization, the ability to abstractly process concrete concepts or events; and memory, the ability to remember, recall, or recognize information. The K-DRS consists of 36 tasks with a maximum total score of 144. Higher scores indicate higher levels of cognitive function. The validity of this composition was .82, and the reliability of evaluationreevaluation was .96 (Suk et al., 2010).

All measurements were analyzed using SPSS version 18.0 (SPSS Inc., Chicago, USA) for Windows. Descriptive statistics were used for the general characteristics of the subjects. An independent t-test was used to test the homogeneity of the variables. The Kolmogorov-Smirnov test was used to confirm the normality of the data. The data were found to satisfy the normality. A paired t-test was performed to compare the values obtained before and after the experiment. An independent t-test was performed to test the homogeneity and the difference values between the two groups. The level of alpha was set at p<.05.

III. Results

The general characteristics of the participants are shown in Table 1. The homogeneity test showed no significant difference between the two groups (p>.05).

General characteristics of subjects

Control group (n=15) Experimental group (n=15) p
Age (years) 73.40±4.95 71.00±3.63 .14
Gender (male/female) 6 / 9 6 / 9
Education (years) 2.60±2.61 3.40±2.44 .39
Hospitalization (month) 2.33±1.05 2.00±.93 .36
Duration of illness (years) 2.93±1.33 3.40±1.64 .40
MMSE-K 19.33±1.35 19.53±1.30 .68
K-DRS 87.33±4.37 87.20±4.16 .93

Values are presented as mean±standard deviation

MMSE-K : MiniMentalStateExamination-Koreanversion

K-DRS : Korean version of the Dementia Rating Scale

The BCI-based cognitive training significantly increased the MMSE-K score from 19.53±1.30 to 22.20±1.15 (p<.05). The MMSE-K scores changed from 19.33±1.35 to 19.00±1.31 in the control group; this change was insignificant. Comparison of the groups showed a greater increase in the BCI-based cognitive training group than in the control group (p<.05)(Table 2.).

The Comparison of MMSE-K Scores between and Within Two Groups

Control group (n=15) Experimental group (n=15) p
Pre-test 19.33±1.35 19.53±1.30 <.001**
Post-test 19.00±1.31 22.20±1.15
p .09 <.001*

MMSE-K : Mini Mental State Examination-Korean version

p<.05 (within each group);

p<.05 (betweenexperimentalandcontrolgroups)

Comparison within the groups showed a significant increase in attention scores from 29.00±.93 points to 31.40±1.59 in the BCI-based cognitive training group (p<.05). Initiation and preservation scores showed a statistically significant increase from 20.13±22.07 to 22.07±1.62 points, construction points showed a statistically significant increase from 3.60±.51 to 4.07±.46 points, conceptualization points showed a significant increase from 20.00±1.25 to 22.27±1.87 points, and memory points showed a significant increase from 14.47±1.77 to 19.53±2.33 points (p<.05). Comparison between the two groups showed a more significant increase in all these scores in the BCI-based training group than in the control group (p<.05)(Table 3.).

The Comparison of K-DRS Scores between and Within Groups

Control group (n=15) Experimental group (n=15) t p

Pre Post p Pre Post p
Total scores 87.33±4.37 86.87±3.89 .20 87.20±4.16 99.33±3.70 <.0011* 14.110 <.001**
Initiation and Perseveration 20.13±1.46 20.13±1.46 1.00 20.13±22.07 22.07±1.62 <.0011* -4.610 <.001**
Construction 3.47±.52 3.33±.49 .16 3.60±0.51 4.07±.46 .014* -2.824 <.001**
Conceptualization 19.73±1.28 19.60±1.40 .16 20.00±1.25 22.27±1.87 <.0011* -4.900 <.001**
Memory 14.87±1.64 14.73±1.22 .55 14.47±1.77 19.53±2.33 <.0011* -8.718 <.001**
Attention 29.13±1.13 28.93±.96 .27 29.00±.93 31.40±1.59 <.0011* -5.392 <.001**

p<.05 (within each group);

p<.05 (between experimental and control groups)

IV. Discussion

We investigated whether BCI-based cognitive training could effectively improve the cognitive function of elderly persons in senior care facilities. The cognitive training group who played BCI-based games for six weeks showed a more significant improvement in cognitive function than the group that listened to music and performed the National Health Gymnastics; therefore, the BCI-based cognitive training was found to be helpful for patients who suffer from dementia.

Cognitive function assessment and cognitive training for cognitive impairment that increases with age are important for dementia prevention. Existing CCT for patients with Alzheimer’s disease (Cipriani et al., 2006; Jelcic et al., 2014; Lee et al., 2013) and mild cognitive impairment (Talassi et al., 2007) improved the MMSE scores and brain activities more effectively than other cognitive training programs and showed promising improvements in cognitive function. The individualized cognitive function assessments and the real-time feedback are strengths of CCT. Although CCT has been widely used in hospitals, minimal help from therapists is necessary for subjects who are not used to using computers. Both cognitive training with existing computers as well as BCI-based cognitive training have been shown to improve MMSE scores and high-level cognitive functions, such as attention improvement in dementia patients.

BCI-based cognitive training can be considered cognitive processing that mainly processes visual information and integrates previous experiences and sensory information from various domains of the central nervous system, like CCT (Park and Park, 2015). BCI-based training significantly improved memory, attention span, and the spatio-temporal ability of healthy elderly subjects. The dementia patients received feedback as they processed visual and auditory information during the BCI-based cognitive training and improved their visual perception through repeated training (Kim et al., 2013). In addition, we believe that the bomb explosion game in the BCI-based cognitive training caused tension to increase the subjects’ conscious behaviors, and the ball floating game increased their attention in a relaxed state (Lee and Park, 2016). Therefore, the K-DRS results demonstrated that BCI-based cognitive training improved the subjects’ attention skills, which allow the subjects to process information and focus on the training without being distracted. It also improved their ability to store, recall, and use memory.

Because of the small number of participants in the study, it is difficult to generalize the effect of the BCI-based cognitive training in dementia patients. Although the MMSE-K and K-DRS scores increased after six weeks of cognitive training using BCI-based games, the subjects were still classified as dementia patients who could have minimal cognitive impairment. Therefore, future studies should verify the effects of BCI-based cognitive training tools by including more subjects, analyzing the effects of these tools according to different levels of dementia, and investigating long-term and follow-up effects.

V. Conclusion

The dementia patients who participated in the BCI-based cognitive training showed significantly greater improvements in the MMSE-K and K-DRS scores compared with the control group who listened to music and performed the National Health Gymnastics. Like the existing CCT, the BCI-based cognitive training can be used as a basic tool to help improve the cognitive ability of dementia patients. Competitive advantages of the BCI-based cognitive training tools are its accessibility and ease of use compared to existing cognitive training tools. Therefore, BCI-based cognitive training may be used to prevent dementia among patients in care facilities who have less access to professional medical benefits.

  1. Almeida OP, Yeap BB, and Alfonso H et al. Older men who use computers have lower risk of dementia. PLOS One 2012;7:e44239.
    Pubmed KoreaMed CrossRef
  2. Bae SJ, Jeong WS, and Lee HG et al. Effect of tDCS Stimulation for Improving Working Memory on Stroke Patients' EEG Variation. J korea Con Assoc 2012;12:261-72.
  3. Belleville S, Clement F, and Mellah S et al. Training-related brain plasticity in subjects at risk of developing Alzheimer's disease. Brain 2011;34:1623-34.
    Pubmed CrossRef
  4. Broetz D, Braun C, and Weber C et al. Combination of braincomputer interface training and goal-directed physical therapy in chronic stroke: a case report. Neurorehabil Neural Repair 2010;24:674-9.
    Pubmed CrossRef
  5. Cipriani G, Bianchetti A, and Trabucchi M. Outcomes of a computer-based cognitive rehabilitation program on Alzheimer''s disease patients compared with those on patients affected by mild cognitive impairment. Arch Gerontol Geriatr 2006;43:327-35.
    Pubmed CrossRef
  6. Deary IJ, Corley J, and Gow AJ et al. Age-associated cognitive decline. Br Med Bull 2013;92:135-52.
    Pubmed CrossRef
  7. De Massari D, Ruf CA, and Furdea A et al. Brain communication in the locked-in state. Brain 2013;136:1989-2000.
    Pubmed CrossRef
  8. Reijnders J, van Heugten C, and van Boxtel M. Cognitive interventions in healthy older adults and people with mild cognitive impairment: a systematic review. Ageing Res Rev 2013;12:263-75.
    Pubmed CrossRef
  9. García-Casal JA, Loizeau A, and Csipke E et al. Computer-based cognitive interventions for people living with dementia: a systematic literature review and metaanalysis. Aging &mental health 2017;21:454-67.
    Pubmed CrossRef
  10. Jelcic N, Agostini M, and Meneghello F et al. Feasibility and efficacy of cognitive telerehabilitation in early Alzheimer''s disease: A pilot study. Clin Interv Aging 2014;9:1605-11.
  11. Kang WK, Kang MH, and Yang JL et al. The comparison of EEG activity by computer assisted cognitive rehabilitation program in the normal elderly and the elderly with dementia. J Korean Soc Occup Ther 2009;17:1-13.
  12. Kim JM, Shin LS, and Yoon JS et al. Comparison of Diagnostic Validities between MMSE-K and K-MMSE for Screening of Dementia. J Korean Neuropsychiatr Assoc 2003;42:124-30.
  13. Kim MK, Lee SS, and Lee YH et al. An Investigation to the Diagnostic Validity of MMSE-K and the Psychopathology in a Group of Urban Patients with Dementia of Alzheimer Type. J Korean Neuropsychiatr Assoc 1998;37:1277-90.
  14. Kim SG, Kim CS, and Kim K et al. The Effect of EEG through Proprioceptive Exercise and Computerized Cognitive Therapy on Stroke. J Korean Soc Phys Med 2013;8:505-12.
  15. Kramer M, German PS, and Anthony JC et al. Patterns of mental disorders among the elderly residents of eastern Baltimore. J Am Geriatr Soc 1985;33:236-45.
    Pubmed CrossRef
  16. Kwon YC, and Park JH. Standardization of Korean of the Mini-Mental State Examination (MMSE-K) for use in the elderly. J Korean Neuropsychiatr Assoc 1989;28:125-3.
  17. Lee GY, Yip CC, and Yu EC et al. Evaluation of a computer assisted errorless learning-based memory training program for patients with early Alzheimer''s disease in Hong Kong: A pilot study. Clin Interv Aging 2013;8:623-33.
    Pubmed KoreaMed CrossRef
  18. Lee TS, Goh SJ, and Quek SY et al. A brain-computer interface based cognitive training system for healthy elderly: A randomized control pilot study for usability and preliminary efficacy. PLoS ONE 2013;8:e79419.
    Pubmed CrossRef
  19. Lee HS, and Park YJ. The Effect of Physical Activity Program for Elderly with Dementia on Cognitive Function: Meta-Analysis of Studies in Korea. J Korean Soc Phys Med 2016;11:115-21.
  20. Liberati G, da Rocha JL, and Van der Heiden L et al. Toward a brain-computer interface for Alzheimer's disease patients by combining classical conditioning and brain state classification. J Alzheimers Dis 2012;31:211-20.
    Pubmed CrossRef
  21. Mattis S. Dementia Rating Scale (DRS). Florida: Psychological Assessment Resources; 1988.
  22. Park JH, and Park JH. The effects of a Korean computer-based cognitive rehabilitation program on cognitive function and visual perception ability of patients with acute stroke. J Phys Ther Sci 2015;27:2577-9.
    Pubmed CrossRef
  23. Plassman BL, Langa KM, and McCammon RJ et al. Incidence of dementia and cognitive impairment, not dementia in the United States. Ann Neurol 2011;70:418-26.
    Pubmed CrossRef
  24. Suk JS, Chey JY, and Kim HY. An Additional Normative Study of the Korean-Dementia Rating Scale. Kor J Clin Psychol 2010;29:559-72.
  25. Sung Y, Cho K, and Um K. A development architecture for serious games using BCI (brain computer interface) sensors. Sensors (Basel) 2012;12:15671-88.
    Pubmed CrossRef
  26. Talassi E, Guerreschi M, and Feriani M et al. Effectiveness of a cognitive rehabilitation program in mild dementia (MD) and mild cognitive impairment (MCI): A case control study. Arch Gerontol Geriatr 2007;44:391-9.
    Pubmed CrossRef
  27. Zhu CW, Sano M, and Ferris SH et al. Health-related resource use and costs in elderly adults with and without mild cognitive impairment. J Am Geriatr Soc 2013;61:396-402.
    Pubmed KoreaMed CrossRef

May 2019, 14 (2)
Full Text(PDF) Free

Social Network Service

Cited By Articles
  • CrossRef (0)