Table 2.
Summary of articles on BCI-based applications for students with learning, memory and attention disabilities.
| Study (year) | Sample | Study contents/method | Key findings |
|---|---|---|---|
| Arns et al., (2007) | 19 children with dyslexia average age 10.33 and 19 control children average age 10.34 | qEEG data were acquired from 28 channels: Fp1, Fp2, F7, F3, Fz, F4, F8, FC3, FCz, FC4, T3, C3, Cz, C4, T4, CP3, CPz, CP4, T5, P3, Pz, P4, T6, O1, Oz and O2. Neuropsychological assessment made using a touch screen monitor with EO. | qEEG results showed an increased (left) frontal and right temporal slow activity in the Delta and Theta bands and increased Beta 1 power at F7 in children with developmental dyslexia. No important correlations between the EEG power data and the EEG coherence data within frequency bands was found. |
|
Arns et al., (2013) Meta-Analysis |
1253 ADHD children and 517 control children. 6–18 years group and. 6–13 years Group | TBR data during EO from location Cz were investigated from children-adolescents between 6-18 years old with or without ADHD. | The grand mean ES obtained in this meta-analysis is rather misleading and was considered an overestimation. An increased TBR cannot be considered a reliable measure used for the diagnosis of ADHD at this time. The ESs obtained were 0.75 for the 6- to13-year-olds and 0.62 for the 6- to 18-year-olds. |
| Breteler et al., (2010) | Experimental Group of 10 children and a Control Group of 9 children who were diagnosed with dyslexia | QEEG data were acquired from 28 channels: Fp1, Fp2, F7, F3, Fz, F4, F8, FC3, FCz, FC4, T3, C3, Cz, C4, T4, CP3, CPz, CP4, T5, P3, Pz, P4, T6, O1, Oz, O2. Neuropsychological measurement was completed using a touch screen monitor with EO. | The main effect is a large and clinically relevant progress in spelling, whereas no progress in reading abilities was found. Cohen's 3.02 d value, implies an important enhancement in spelling of the neurofeedback group |
| Gevensleben et al., (2009) | 102 children (Neurofeedback Group N = 59, Control group N = 35, dropouts = 8) with ADHD aged 8–12 years. | Children performed either 36 sessions of NFT or a computerised attention skills training within two blocks of about four weeks each. | For parent and teacher ratings, improvements in the NF group were superior to those of the CG. A significant effect was found for the inattention subscale (t(60) = 1.94; p < .05) and a trend for the hyperactivity/impulsivity subscale (t(60) = 1.59; p < .1). |
| Heinrich et al., (2004) | 22 children with ADHD aged 7–13 years. (n = 13 Training Group, n = 9 Waiting-List Group) | 25 sessions of 50 min duration in 3 weeks. 100–120 trials of 8 s duration (2 s baseline, 6 s feedback) per session | Significant effects for the SCPs training group only: ADHD rating scale – total score: 25% decrease after training |
| Jeunet et al., (2015) | 18 participants, aged 21.5 ± 1.2, were instructed to learn to control an EEG-based MI-BCI by performing 3 MI-tasks. 2 of them were non-motor and they spanned over 6 training sessions, on 6 different days. | The EEG signals were recorded using 30 scalp electrodes (F3, Fz, F4, FT7, FC5, FC3, FCz, FC4, FC6, FT8, C5, C3, C1, Cz, C2, C4, C6, CP3, CPz, CP4, P5, P3, P1, Pz, P2, P4, P6, PO7, PO8). Each participant took part in 6 sessions, on 6 different days spread out over several weeks. | In this study, it was shown how users' profiles can influence their MI-BCI control levels. It therefore creates a path for designing new protocols that involve MI-BCI techniques, that can be adjusted to each user's specific profile. |
| Leins et al., (2007) | Two EGs groups included 19 ADHD children each, with ages between 8–13 years. | This study worked toward answering whether: (i) children manage to improve their cortical self-regulation, (ii) if treatment can work in favor of their behavioral and cognitive skills (ii) the two groups show differences between the skill amelioration outcomes. | Self regulation of SCPs: between sessions 2 + 3 and 29 + 30 (t[18] = 3.51, p = .006, ES = 1.09) as well as 2 + 3 and 32 + 33 (t[14] = 3.07, p = .016, ES = 1.05). Amplitudes of SCP in activation and deactivation tasks: sessions 2 + 3 and sessions 29 + 30 had a notable difference (t[18] = 3.67, p = .004, ES = 1.03). The same is observed between sessions 2 + 3 and sessions 32 + 33 (t[15] = 5.28, p < .001, ES = 1.07). Theta/Beta-ratios were also notably different for the two tasks, at the end of treatment (sessions 29 + 30) for the feedback (t[36] = 4.224,p < .001, ES = 1.37) and transfer condition (t[36] = 3.003, P = .010, ES = 2.25). Only the Theta/Beta group showed a substantial progress in Tests for the full scale IQ (t[17] = 3.26, p = .015, ES = .62). |
| Lévesque et al., (2006) | 20 ADHD children aged 8–12 randomly assigned to either an experimental (EG N = 15) or a control (CG N = 5) group. | Before the study began, participants in both groups were treated with methylphenidate. No participant underwent cognitive training before the experiment. Psychostimulants were not allowed over the course of the experiment. | Neutral Trials: For the EG group, this score was significantly greater (P < 0.05) at Time 2 (67%, S.D.: 18.3) than Time 1. Interference Trials: For the EXP group, this score was significantly higher (P < 0.05) at Time 2 (68%, S.D.: 13.9) than Time 1. |
| Lim et al., (2010) | 10 ADHD children aged 7–12 as EG and 10 ADHD children as controls | 20 sessions of therapy over a 10-week period. Three-channel EEG signals are recorded from the frontal (Fp1, Fp2) and parietal (Pz) positions, covering theta, alpha, beta 1, and beta 2 waves. | Effect size for parental ratings was about –0.95 SD (95% C.I. –1.92 to 0.01 SD), and that for teachers' ratings was –0.85 SD and (95% C.I. –2.14 to 0.44 SD). |
| Lim et al., (2012) | 20 unmedicated ADHD children with significant inattentive symptomatology (combined and inattentive subtypes). | A BCI-based attention training game-system tracked attention with a headband with dry electrodes for EEG sensing in order to control a feedforward game. The treatment's design included 8 weeks of training that comprised of 24 training sessions in total, with 3 follow-up booster training sessions once a month. | Results show significantly improved inattentive and significantly improved hyperactive-impulsive symptoms of ADHD respectively for inattentive and combined subtypes, according to parents' behavioural ratings. |
| Martínez et al., (2016) | The population sample includes children that were invited directly to the study as well as children that receive therapy for reasons concerning their communication and behavior. | A case-based study aiming to examine and study the attention, cognition and memory of children with ADHD. Children were invited to play the Stroop and Flanker tests and other games that require the use of memory, practicing skills related to memory, attention and reasoning. | EEG and video analysis data as well as childrens' scores during the game were used to code their affective states relevant to engagement, frustratio and excitement. |
| Nan et al., (2012) | 32 students aged 20–29 years (Neurofeedback group N = 16, control group N = 16) | 20 sessions of NFT | The increases in forward and backward digits of the NFT group were significantly larger than those of the control group (t(30) = 2.944, p < 0.005 in forward increase; t(30) = 4.091, p < 0.001 in backward increase). |
| Qian et al., (2018) | 66 boys with ADHD, combined or inattentive subtypes, were split in a random manner in two groups (ADHD-Intervention group N = 44 and ADHD-Non Intervention group N = 22) | The ADHD-I participants went through three BCI-based therapy interventions every week, throughout eight weeks. Two dry EEG sensors were placed at the frontal sites FP1 and FP2. The BCI-based attention training game included a headband with mounted dry EEG sensors that transferred EEG readings to the computer through Bluetooth-enabled protocol. | The ADHD-I group had significantly greater reduction in the ADHD-RS clinician inattention scores compared to the ADHD-NI group (p = 0.038). Intra- and inter-network FC showed significant group and time interaction effect (p < 0.05). The results show that BCI-based therapy sessions can be useful in improving the behavioral skills of children with ADHD, by modifying salience network processing. |
| Shenjie et al., (2014) | Four healthy subjects. | Attention estimated from the signals recorded from 4 EEG channels namely O1, O2, AF3, and AF4. Subjects played the game with three difficulty levels for three consecutive days. | The proposed control mechanism in the designed video game is capable of enhancing attention and brain functions including the ability to sustain the attention for a longer period. |
| Walker and Norman (2006) | 12 subjects aged 7–16 years | A QEEG and a reading difference Topograph is obtained. Next, the therapists, train down the irregularities that show an important increase and train up the ones that show a notable decrease. | Each of the 12 subjects that underwent treatment, showed an improvement by more than two grade levels after 30–35 ten-minute sessions each. |
| Wang et al., (2007) | 2 ADHD subjects aged 8 and 11 years | The 2 children were examined by an IVA – CPT, with 3 electrodes, Cz, Fp1 and Fp2 that were placed on their head in positions specified by the International 10–20 system. | The BCI–NFB-VR system can lengthen the attention span of children with ADHD. |
| Yang et al., (2017) | 10 university students | Creation of a brain controlled game, that involves motor-imagery with an original BCI and game design. EEG output is generated by the Neuroscan device with 27 different channels of electrodes. | The bandpower analysis findings showed that participants' attention level improved during the experiment. |
| Zoefel et al., (2011) | 24 students (14 in the NFT group aged 23.7 ± 2.3 years old and 10 in the control group aged 22.1 ± 3.8) | For each student, the treatment included one training session for each day from Monday to Friday. On the first and the fifth session, cognitive skilles were examined with a mental rotation test. | The UA amplitude during the first base rate showed a significantly higher amplitude than the first base rate of the first session (t(10) = 3.59, p = 0.003). Cognitive performance was significantly increased for the NFT group (t(16) = 2.21, p = .029) Independence was significant in the trained UA range between IAF and IAF+2 Hz (t(10) = 2.39, p = 0.019). |
EEG: electroencephalogram; qEEG: Quantitative electroencephalogram; EO: Eyes Open; ADHD: attention-deficit/hyperactivity disorder; TBR: Theta/Beta Ratio; ES: Effective Size; EG: experimental group; CG: control group; VR: virtual reality; NF: neurofeedback; NFT: neurofeedback training; TOVA: Test of Variables of Attention; SCPs: Slow Cortical Potentials; 3D: 3 dimensional; BCI: brain computer interface; MI: motor imagery; T/B: Theta/Beta; ADHD-I: ADHD-Intervention; ADHD-NI: ADHD-Non-Intervention; ADHD-RS: ADHD-Rating Scale; EOG: electrooculography; IVA: Integrated Visual & Auditory; CPT: Continuous Performance Test; FC: Functional Connectivity.