2.1. Participants
The study’s participants included 55 ADHD patients and 55 healthy controls. The participants in the ADHD group were outpatients of Beijing Anding Hospital. Irrespective of gender, children who a. aged 6-12 years; b. are outpatients who met the diagnostic criteria for ADHD in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) of the U.S. are recruited in the study. The patients and/or their guardians agreed to participate in the study by voluntarily signing an informed consent sheet, and participants aged 8 years old and above should also sign the informed consent form for children. Also, children who have been diagnosed to develop a. Neurodevelopmental disorders such as cerebral palsy, epilepsy, adrenoleukodystrophy, etc.; b. Serious mental illnesses such as Tourette's disorder, autism spectrum disorders, mental retardation, schizophrenia, bipolar disorder, depressive disorders, and developmental disorders of specific learning skills, etc.; c. Auditory disorders result in manifestations of child-like Attention Deficit-Hyperactivity Disorder (ADHD), and those who d. need to take other antipsychotic, antidepressant and other central nervous system drugs (including Chinese and Western medicines and health care products) at any stage of the study; e. are ineffective after systematic treatment with two different classes of ADHD drugs; f. cannot use software normally, such as those who are color-blind; and g. have a previous or current gaming addiction are excluded from the current study. The control group (CON group) was recruited from a middle school in Nanjing, China, from children with normal development corresponding to the ADHD group in terms of age and sex. The study was conducted in accordance with the World Medical Association Code of Ethics (Declaration of Helsinki), which establishes ethical principles for research involving human beings and was approved by the Ethics Committee, Institute of Psychology, Chinese Academy of Sciences (H23067). Participants and their parents signed a written informed consent form and volunteered to participate in the study.
2.2 Performance Measures
Swanson Nolan, and Pelham-IV rating scales (SNAP-IV)
The SNAP-IV rating scale [19] is a revised version of the Swanson, Nolan Pelham (SNAP), which was developed based on the description of ADHD symptoms in the Diagnostic and Statistical Manual of Mental Disorders (DSM). It consists of three subscales: attention deficit, hyperactivity-impulsivity, and oppositional defiance. The items are rated from 0-3, and the mean value of each subscale is calculated. Itis mainly used for screening, diagnostic support, and assessment of treatment efficacy and symptom improvement in children and adolescents between the ages of 6 and 18 years. The reliability was 0.68~0.84.
Strengths and Difficulties Questionnaire (SDQ)
The SDQ rating scale [20] is a brief behavioral screening questionnaire designed and developed by American psychologist Goodman R in 1997 based on the Diagnostic and Statistical Manual of Mental Disorders-IV and the Diagnostic Criteria for the Classification of Mental and Behavioral Disorders, 10th Edition (ICD-10). It is used to assess behavioral and emotional problems in children and adolescents and The reliability was 0.84~0.96.
Behavior Rating Inventory of Executive Function (BRIEF)
The BRIEF rating scale [21] was developed by Gioia et al. to assess the executive functioning of children aged 6-18 years at home and at school. The scale consists of 86 entries in two major areas: the Behavior Management Index (BMI), which includes the three subscales of Inhibition, Conversion, and Affect, and Control, Conversion and Affect, and Control subscales; and the Metacognitive Functioning Index: five subscales including Task Initiation, Working Memory, Planning, Organization, and Monitoring. It has good reliability and validity. The higher the score, the more severe the functional impairment, the reliability was 0.68~0.96.
2.3 Neurocognitive/attentional task
Continuous Performance Test (CPT)
CPT software is one of the most commonly used experimental paradigm to examine attention. It is a computer-assisted information processing test that provides objective measures independent of supervisors[22]. The results of CPT reflect the participant's ability to sustain, inhibit, and impulse sustained attention. The classic CPT paradigm consists of a series of stimuli (i.e., numbers or characters) being presented rapidly on a computerized monitor, requiring the participant to respond to a pre-specified target. The CPT paradigm adopted in this study involves a white boxed area on a black background on the calculator screen, with white squares randomly appearing above or below the white boxed area, requiring the target stimulus to be a right mouse button pressed when the square appears below it. The main measures of the CPT are reaction time, hit rate (i.e., number of times response is made to the target stimulus therein), miss rate (i.e., number of times no response is made to the target stimulus therein), false alarm rate (i.e., number of responses to the non-target stimuli, which is a measure of impulsivity), correct rejections rate (i.e., number of times no response is made to the non-target stimulus therein), d' = Z hits - Z false alarms, which response to the participant's perceptual sensitivity, and β = O hits/O false alarms, which represents the participant's propensity to respond, computed according to the principle of signal detection[23].
Eye movement analysis
Eye movement research is an important psychological research method, which has been widely used in various fields of psychology. This measure was developed based on the strong correlation between attention and eye movement[24]. To explain, the extent to which individuals concentrate on or dissipate from a task has been evidenced to affect the trajectory of their eye movements, underlining eye movement as an external indicator of attention. Different eye movement tasks also reflect different cognitive behaviors[25]. In the present study, we used the anti-saccade and delayed-saccade tasks to examine participants' cognitive functions, including inhibition and working memory[26-27]. The task descriptions are as follows:
Anti-saccade task[26]
The central gaze point cross was first presented for 2s. After, the central cross disappeared while the target green dot appeared randomly at four positions at 3°, 6° left and right sides horizontally. Each target dot was presented at each position for 1000ms, and 5 times. A total number of 20 trials are involved. When the response time is over, a 300-ms feedback point will appear in the correct position. In order to complete the anti-saccades task, the participant must do three things[28]: ① inhibit the impulse to look at the suddenly appearing stimulus; ② calculate the spatial position that is opposite to and equidistant from the stimulus, which involves working memory; ③ initiate volitional saccades to look at the opposite point where there is no stimulus.
Delayed-saccade task[27]
In this task, the central gaze point cross was also presented for 2s. After, as the cross remained on the screen, the target green dot stimulus appeared randomly at four positions (i.e., 3°, 6° left and right sides horizontally) for 100 ms. Subsequently, the gaze point was presented only at the center of the screen for 1, 3, and 5s, followed by a blank screen lasting 1.4s. Participants were required to stare at the cross intently while the green dots were presented, and they were required to recall and look at the original position of the target dots when the blank screen was presented. Lastly, a 200-ms feedback point appeared at the correct location. In the delayed-saccade tasks, participants were required to fix on the central gaze point. They were also instructed to make saccade only when the central gaze point disappears. The number of early saccades (i.e., trials in which the participants did not wait until the disappearance of the central gaze point to perform a saccade) in the delay saccade task reflected the participants’ inhibition ability. Participants also were required to recall the location of the stimulus point before making a saccade. This measures also primarily reflect participants’ spatial working memory capacity.
In the saccade task, the actual saccade refers to the real, observed eye movement that occurs when a person shifts their gaze from one point to another. The target saccade, on the other hand, represents the intended or planned eye movement from one point to another. It refers to the desired or anticipated saccadic eye movement based on the instructions or requirements of a task or experimental paradigm. In the anti-saccade task, the participants were required not only to make a saccade in the opposite direction of the stimulus but also to calculate roughly where the spatial location of an equal distance in the opposite direction of the stimulus is for working memory capacity measures. This suggests that in the anti-saccade task, the target saccade is from the central gaze point to an equal distance in the opposite direction of the stimulus. Subsequently, we calculated the difference between the actual saccade and the target saccade to represent the participants’ working memory capacity (i.e., a small difference indicates high capacity).
2.4 Safety and Satisfaction Survey
The safety incident survey was conducted with parents who reported any adverse events to the researchers by phone during the study period and recorded any adverse events spontaneously reported during the study visit. At the end of the intervention, all participants and their parents were verbally interviewed on their overall feelings about the game and the training process, and the scoring statistics were recorded by the staff.
2.5 Procedure
ADHD patients were recruited by recruitment information posted at the outpatient clinic of Anding Hospital. Participants in the ADHD patient group were screened among patients with a clear diagnosis of ADHD by a clinician, meeting the inclusion and exclusion criteria of this study. The participants were not on medication or had stopped taking medication in the last two weeks, and were enrolled after assessments on the safety of not taking the medication during the intervention period. Informed consent from participants’ family members, who agreed to the enrollment were also received. Healthy participants were also enrolled after screening and confirmation that they were normally developing children. All participants went through baseline measures on the scales, CPT tests, and eye movement tasks prior to the intervention. After a short break, the researchers explained the mechanics of the game and the training method to the children and their families, and the children were allowed to complete the first training session for 20 min, during which the researchers recorded any discomfort that the children experienced. The intervention period was 4 weeks after enrollment, and the participants were asked to perform game training at home 5 days per week, 20-25min per day. The game anti-addiction setting restricts the maximum daily game time to 30min. Participants were prompted to perform the game training according to their own ability and to adaptively adjust the difficulty of the game by switching the levels as the training progressed. The participants were followed up weekly to ensure compliance rate, and the training time was tracked in the back office. The final assessment was conducted after the 28th day, and the measures were consistent with the baseline measures conducted during the pre-intervention period.
2.6. Data Analysis
Only data from participants who completed the entire 4-week intervention were included in the statistics. The present study focused on the statistical analyses on the cognitive-behavioral measures conducted during the pre-and post-intervention visits. Firstly, descriptive statistics on the general demographic information of the ADHD and CON groups was analyzed at the baseline level. Secondly, the CPT scores of the main observational indexes of ADHD symptoms and SNAP-IV data were compared between the two groups, and a chi-square test was performed on the categorical data. The Shapiro-Wilk normality test was performed on the continuous data first, followed by the independent samples t-test for those that conformed to the normal distribution, and the Mann-Whitney test for the two groups that did not conform to the normality. Thirdly, the cognitive-behavioral outcomes before and after the intervention were analyzed separately for the ADHD and CON groups. For pre-and post-intervention evaluation data, the Shapiro-Wilk normality tests were first conducted with data on the three scales, CPT and eye movement tests results. 2-tailed paired sample t-tests were performed on normally distributed values to compare whether there were differences between pre-and post-intervention data, and the equivalent Wilcoxon signed-rank test was performed for non-normal values. The findings are considered statistically significant when p < 0.05. All analyses were performed in the open-source statistical package R (v.3.3.1).