Skip to main content

Table 1 Summary of included contributions (n = 28)

From: Features and effects of computer-based games on cognitive impairments in children with autism spectrum disorder: an evidence-based systematic literature review

Contribution

Journal/Conference name

Name/Type of

games used for intervention

Platform

Study design

Final group characteristics:

Sample size:

Groups (Sex)/Age

Procedure details

(session number, duration and frequency)

Targeted cognitive measures

Assessment time

Evaluation scales

Evaluation

Effectiveness

Main message

Alvares GA et al., 2019, Australia [42]

Autism Research

Serious Game

Android playstore

Randomized controlled trial

N = 56

EG: 28 (22 M, 6 F)

CG: 28 (26 M, 2F)

Age: 5–12 yrs

1 session, 15 min

• Social interaction skills

Baseline and post intervention

• SRS-2,

•MSEL

• WISC-IV

Pairwise comparisons confirmed increased proportion of social characters selected in the training group across all levels (ap-values < 0.001); Children in the training group significantly raised the percentage of engagements to faces relative to objects after training. The adjusted mean change is 17.24 for the training group, and adjusted mean change is − 12.87 for the control group

Statistically significant at all evaluation scales

Findings indicated that training-based paradigms can improve statistically target fundamental attentional processing of social information in ASD

Chen J et al., 2019, China [43]

Computer Assisted Surgery

FaceSay: Serious Game

PC

Before & After Clinical Trial (without control)

N = 11

EG: 11

Age: 6–12 yrs

5 sessions, 29 min, once a week

• Emotional Skills

Baseline and at the end of the last week of intervention

• SRS-2

The results show that there is a non-significant difference between SRS total score at baseline and endpoint, with Chi-sq. 0.28, p value 0.60

Not effective

In short-term therapy, there is no improvement in social cognition, social awareness and social communication

Özen A, 2015, Turkey [44]

Educational Sciences: Theory & Practice

Video game

IOS apple

Non-randomized Controlled Clinical Trial

N = 6

EG: 3 (1 M, 2 F)

CG: 3 (2 M, 1 F)

Age: 5–11 yrs

15 sessions, 40 min, twice

a week

• Social interaction skills

Baseline, the first and the second week after the intervention

• SRS-2

The analysis shows that one of the participants performed social interaction skills with 85% accuracy, another person with 91% accuracy, and the third person with 88% accuracy. The 4th, 5th, and 6th person performed the skill of responding appropriately to teaching opportunities with more than 80% accuracy in the intervention sessions

Statistically significant at all evaluation scales

The data suggest that participants can learn how to use the social interaction skills necessary for iPad gaming activities

Fantasia V et al., 2020, Italy [45]

Autism

Serious Game

Android playstore

Before & After Clinical Trial (without control)

N = 29

EG: 29 (25 M, 4 F)

Age: 6–12 yrs

2 sessions, once a week

• Short-term memory

• Long-term memory

Baseline and the first half of the study phase and after 1-week later intervention

• PPVT-R

• VABS-II

• RCPM

Results showed that memory-based recognition accuracy was high for objects studied in the active condition in both the test (active: M = 0.63, SD = 0.20; yoked: M = 0.53, SD = 0.20) and the retest (active: M = 0.54, SD = 0.22; yoked: M = 0.48, SD = 0.20)

Statistically significant at all evaluation scales

Advancements in active study status continued, especially for one week after the initial study session

Aresti-Bartolome N and Garcia-Zapirain B, 2015, Spain [46]

Bio-medical materials and engineering

Serious videogame

PC

Non-randomized Controlled Clinical Trial

N = 40

EG: 20

CG: 20

Age: 3–8 yrs

2 sessions, 12 min

• Attention

• Social interaction skills

Baseline and end of each game session

• Researchers set scales (Errors, Response time, Interaction with visual contact, Interaction without visual contact, and Gaze and pupil information

during the game)

The reaction time of the first group was lower after rehabilitation sessions (M = 4.52 s, SD = 3.40); the clinical group (M = 36, SD = 38.18); the response time decreased when the clinical group provided interaction with eye contact

Statistically significant on some evaluation scales

The results showed that the interaction involving looking at the eyes improved as the reaction time decreased, indicating that the system can help with cognitive rehabilitation

Almeida LM et al., 2019, Brazil [47]

International Journal of Computer Games Technology

ALTRIRAS

PC

Non-randomized Controlled Clinical Trial

N = 38

EG: 10

CG: 28

Age: 6–12 yrs

4 sessions, 60 min, once a week

• Facial processing skills

• Social interaction skills

Baseline and post intervention

• Researchers set scales (QuizEmotion scale)

Based on the applied statistical test, there was no significant difference between the number of correct answers of children with ASD in the pre- and post-test

Not effective

Play exposure time in children with ASD should be increased to effectively aid facial expression recognition

Fernandes FD et al.,2010, Brazil [48]

PRO-FONO: Revista de Actualizacao Cientifica

Not reported

PC

Before & After Clinical Trial (without control)

N = 23

EG: 23

Age: 3–12 yrs

10 sessions

• Language skills

Baseline and after each session

• Functional communicative

Profile /scales

Statistical analysis has not identified a significant difference (0.05) with any variables, but the number of subjects and areas with progress can be considered

Not effective

Quantitative and qualitative improvements were identified without statistical significance. This progress was observed after a shorter period than is usually applied to this type of comparison, which seems favorable

Al-Hammadi M andAbdelazim A, 2015, USA [49]

In2015 IEEE Global Engineering Education Conference

Miss.Fly: Video game

PC

Partially Randomized clinical trial

N = 98

EG: 49

CG: 49

Age: 6–7 yrs

Not reported

• Attention

• Short-term memory

• Long-term memory

End of each game session

• Researchers set scales: (Time response (seconds), Accuracy, Pay Attention, Strengthen Memorization, React to Random Processes)

The independent samples t-test was associated with a statistically significant effect since p < 0.05. Then, the autistic children’s performance does differ significantly; the independent samples t-test was associated with a statistically significant effect since p < 0.05, t (35) = 4.02, p = 0.002. Thus, the autistic children’s time response does differ significantly

Statistically significant at all evaluation scales

The designed game improves the ability of children of both groups to pay attention, remember, react to random processes, and process parallel information

Pedreschi VB et al., 2019, Peru [50]

Virtual Reality

Serious Game

Android playstore

Before & After Clinical Trial (without control)

N = 20

EG: 20

Age: 3–10 yrs

During 2-weeks

Emotional Skills

Social interaction skills

Facial processing skills

Baseline and post intervention

• Researchers set scales: (Emotion Recognition Times)

The findings demonstrate that 67% of the ASD patients enhanced their emotion recognition skills (represented in caricatured and human facial expressions)

Positive without statistical argument

The outcomes revealed a significant advancement in emotion recognition after employing the Serious Game

Bono V et al., 2016, UK [51]

Frontiers in psychiatry

GOLIAH: Serious Game

PC and Mobile app

Before & After Clinical Trial (without control)

N = 10

EG: (10 M, 0 F)

Age: 5–9 yrs

60 sessions, during 3-month, 20 min

• Attention

• Imitation skills

Baseline, after each session and post 3-month intervention

• WPPSI

• VIQ

• PIQ

The time to terminate the assignment significantly declined along sessions (p < 0.001); correct answers raised considerably with the session numbers (p = 0.005); the quality of imitation enhanced throughout the sessions

Statistically significant at all evaluation scales

The game platform of this study is useful both in the child-therapist interaction in the hospital and in the child-parent interaction at home

Kamaruzaman NN et al., 2017, Malaysia [52]

Indian Journal of Science and Technology

Quranic: Serious game

Android playstore

Before & After Clinical Trial (without control)

N = 15

EG: 15 (13 M, 2 F)

Age: 5–12 yrs

Not reported

• Learning

• Attention

• Imitation skills

Baseline and post intervention

• Researchers set scales: (Involvement Scale, Reaction Time, Facial Expression & Posture, Complexity and Creativity)

There was a significant effect of SDs of signals in the involvement scale (P-value < 0.05); the level of student engagement in learning Al-Quran improved after using the Quranic game

Statistically significant at all evaluation scales

The upshots pointed out that the involvement of children in playing games influences the level of engagement, as the analysis suggests a significant relationship between involvement and engagement level

Jeekratok K et al., 2014, Thailand [53]

International Journal of Web-Based Learning and Teaching Technologies

Video game

PC

Before & After Clinical Trial (without control)

N = 10

EG: 10

Age: 7–10 yrs

36 sessions, during 3-months, 60 min

• Attention

Imitation skills

Recognizing and differentiating

Baseline and post intervention

• S-CAT

• PECS

• Researchers set scales

Average post-test scores were higher than pre-test scores (Z = 2.81, 2.81 and 2.80 respectively); the t-values of the three variables were significantly different at p < .01

Statistically significant at all evaluation scales

There is robust evidence to support the hypothesis that web-based games and social stories can be efficacious as instruments for behavior changeover

Bernardini S et al., 2014, UK [54]

Information Sciences

ECHOES: Serious game

PC

Before & After Clinical Trial (without control)

N = 19

EG:19 (18 M, 1 F)

Age: 7- 13 yrs

several times a week over a 6-week, 10–20 min

• Social interaction skills

• Facial processing skills

• Language skills

Baseline inter and post intervention

• Researchers set scales

The mean probability of children responding to the practitioner’s bids for interaction during the pre-test was 0.66 and during the post-test was 0.71 (SD = 0.14); this slight increase in responses between the pre and post-test was not significant. Across the sessions, by some children, the results showed a slight but non-significant decrease

Statistically significant only for some children

Empirical marks of the agent's effectiveness are based on an extensive evaluation of the ECHOES platform, which shows encouraging tendencies for several children

Mairena MÁ et al., 2019, Spain [55]

Research in Autism Spectrum Disorders

Pico’s Adventure: Video game

PC

Randomized controlled trial

N = 15

EG: 7 (7 M, 0 F)

CG: 8 (8 M, 0 F)

Age: 4–7 yrs

4 sessions, 30 min

• Social interaction skills

Baseline and post intervention

• Researchers set scales: (Social Initiation, Spontaneous Gestures, Responses, Interlocutor, Given helps, Other social Interaction),

• ASEBA

• ABC

• SSRS

Children demonstrated significantly more social initiation during the videogame (M = 9.33, SD = 9.61) than during free Play (M = 4.08, SD = 3.82); t (11) = 2.438, p = 0.033. Besides, outcomes associated with repetitive actions indicate the prospect of this game to lessen repetitive behaviors

Statistically significant on some evaluation scales

There is s potential use of full-body interactive videogames as tools to foster social

initiation conducts in children with ASD

de Vries M et al., 2014, Netherlands [22]

Journal of Child Psychology and Psychiatry

Braingame Brian: Video game

PC

Randomized controlled trial

N = 90

EG1: 31

EG2: 27

CG: 32

Age: 8–12 yrs

25 sessions, one a week, 40–50 min

• Working memory

• Facial processing skills

Baseline, post intervention, and 6-week-follow-up

• BRIEF

• SART

• Corsi-BTT

Children in the WM and flexibility intervention-conditions improved significantly in sequence lengths in the WM-training tasks, and level in the flexibility-training task (p’s < .001.(

Statistically significant at all evaluation scales

All children are enhanced in working memory, flexibility, attention, executive functions, social behavior, and quality of life. However, the adaptive intervention conditions did not result in a more considerable improvement than the mock training

Saniee S et al., 2019, Iran [56]

Journal of Intellectual Disability Research

Tatka: Video game

PC

Before & After Clinical Trial (without control)

N = 13

EG: 13 (11 M, 2 F)

Age: 5–7 yrs

4 sessions, 15-min per day during 2-months

• Set-shifting ability

Baseline, post intervention, and 6-week-follow-up

• SSIT

• WCST

• MCST

According to the intervention findings; a significant difference in BFRS-R was observed between pre-training and post-training (P = 0.0001)

Statistically significant at all evaluation scales

All children improved considerably in cognitive and behavior flexibilities when they were given the SSIT

Khowaja K and Salim SS, 2019, Malaysia [57]

International journal of human–computer interaction

Serious Game

PC

Before & After Clinical Trial (without control)

N = 5

EG: 5 (5 M, 0 F)

Age: 6–10 yrs

15 sessions, 20 min, during 1-months

• Learning

Baseline, post intervention, and week one and week two following the withdrawal of intervention (follow-up)

• Researchers set scales: (Correct responses, Number of attempts)

• SSIT

The use of SGs during intervention improved the performance of participants 1, 2, 3, 4, and 5 to 97%, 94%, 81%, 96%, and 89%, respectively

Statistically significant at all evaluation scales

The results indicated that learning vocabulary entities among children with ASD enhanced after using the game

Fridenson-Hayo S et al., 2017, Israel [58]

European child & adolescent psychiatry

Emotiplay: Serious Game

PC

Phase 1: Before & After Clinical Trial (without control)

Phase 2: Randomized controlled trial with two intervention groups and controls

N (1) = 15, EG 1: 15 (11 M, 4 F)

N (2) = 74

EG (2): 34

CG (2): 40

Age: 6–9 yrs

At least 2 h per week,

over a period of 8 weeks

• Emotional Skills

• Facial processing skills

• Vocal intonation

• Body language

Baseline and post intervention

• SRS-2

• VABS-II

• Emotion Recognition Tasks

Phase 1: SG use significantly improved participants’ performance on the ER body language task (Pre: M = 14.33, Post: M = 18.73, p < .01) and the ER integrative task (Pre: M = 11.13, Post: M = 13.47, p < .05)

Phase 2: Pairwise comparisons for the time by group interaction demonstrated that considerable improvement over time was found on all ER tasks for the intervention groups (Face: Mean difference = 2.17, p < .001; Voice: Mean difference = 2.19, p < .001; Body: Mean difference = 4.63, p < .001)

Statistically significant at all evaluation scales

This game (Emotiplay’s SG) is an influential and motivating intervention, cross-culturally leading ER from faces, voices, body language, and their integration in context to children with high-functioning ASC

Spaniol MM et al., 2017, Brazil [32]

Journal of Autism and Developmental Disorders

CPAT

PC

Randomized controlled trial

N = 14

EC: 8 (6 M, 2 F)

CG: 6 (6 M, 0 F)

Age: 6–10 yrs

13 sessions, 45 min, twice a week across a 2-month period

• Attention

• Academic performance

Baseline (1 and 2 weeks

before) and post intervention

• CPM

• Academic Assessments (Maths, Reading Comprehension, Copying)

• CARS

Effects showed a significant increase in the CPM scores from pre to post-assessment for the CPAT group (pre = 86.25 ± 6.73; post = 100 ± 5.98); there was a significant improvement in math scores for the CPAT group (pre = 27.8 ± 10.14; post = 51.7 ± 9.98); there was a significant improvement in reading scores for the CPAT group (pre = 39.34 ± 6.19; post = 59.19)

Statistically significant at all evaluation scales

Progress was evident in various academic tests, including reading comprehension, copying speed, and math

Hu X et al., 2019, China [59]

Journal of Autism and Developmental Disorders

CAI

PC

Before & After Clinical Trial (without control)

N = 4

EG: 4 (3 M, 1 F)

Age: 9–11 yrs

7–13 sessions, 20 min during 5 weeks

• Matching skills

Baseline, post intervention and after 1 week, 3 weeks, and 5 weeks follow-up

• VMS

• Researchers set scales: (Social Validity Measures, Response Accuracy, Task Engagement)

During the intervention, three students maintained a relatively high level of task employment in CAI, compared to TII (Jiahua: PND = 100%; M = 95.3, in CAI; M = 72.5, range 59–82% in TII; Lingling: PND = 83.3%; M = 86.5 in CAI; M = 67.8 in TII; Zicheng: PND = 85.7%; M = 84.1 in CAI; M = 73.4 in TII); all four students required a fewer number of trials to achieve criterion in CAI than in TII; three of the four students required a lower number of prompts and shorter durations in instructional time in CAI than in TII

Statistically significant on some evaluation scales

CAI was more efficient than TII concerning the prompts supplied and the duration of instructional sessions

Macoun SJ et al., 2020, Canada [60]

Journal of Autism and Developmental Disorders

Caribbean Quest; CQ: serious

game

PC

Randomized controlled trial

N = 20

EG: 11 (9 M, 2 F)

CG: 9 (8 M, 1 F)

Age: 6–12 yrs

24 sessions,

3 times/week, 30 min per session during 8-weeks

• Attention

• Working memory

• Academic performance

• Inhibitory control

Baseline inter and post intervention

• BRIEF-P T GEC,

• Academic supports (tutoring, learning assistance, gifted programming)

• Conners-3-P T EF

• SRS-2

There was a statistically significant difference in errors (KiTAP ‘Sad/Happy Ghost and Colored Boxes) between the intervention and control groups, F (1,16) = 4.61, p < .05, partial η2 = .224; F (1,15) = 12.23, p < .01, partial η2 = .49 respectively. On the Woodcock-Johnson Math Fluency task, with posthoc analyses indicating that the intervention group made fewer errors than the control group (Mdiff = 14.08 [95% CI .544–7.61], p < .05)

Statistically significant on some evaluation scales

This game has preliminary support and potential efficacy for children with ASD

Mercado J et al.,2018, Mexico [61]

Multimedia Tools and Applications

FarmerKeeper: video game

PC

Randomized controlled trial

N = 12

EG: 6

CG: 6

Age: 4–11 yrs

20–30 min, 4-weeks

• Attention

• Anxiety

Baseline inter and post intervention

• CRSD-ant test

• ADHDT

Participants were 8% (2:00 min approx., p = 0.00288) of the time paying more attention during the full session (avg. full session = 13:05 min) when using FarmerKeeper (97.15% of full session) than with BrainCats (89.15% of full session)

Statistically significant at all evaluation scales

FarmerKeeper can improve cognitive indicators by reducing the level of anxiety and increasing the attention of children with autism during the treatment of neurofeedback training sessions

Mercado J et al., 2020, Mexico [30]

Journal on Multimodal User Interfaces

FarmerKeeper: video game

PC

Randomized controlled trial

N = 26

EG: 13

CG: 13

Age: 4–13 yrs

13 sessions, 15 min during 10-weeks

• Attention

Baseline inter and post intervention

• CRSD-ant test

• ADHDT

There is a slight difference of close to 3% (p = 0.25), which means that participants who used FarmerKeeper had a better performance than those who used Cartoons

Statistically significant at all evaluation scales

Pre- and post-assessments revealed that participants' attention, attentional control, and sustained attention were sweetened

Wagle S et al., 2021, India [62]

Scientific Reports

Basket game, Train game, Piano Game, Face game, Shape Game: Serious games

Android playstore

Before & After Clinical Trial (without control)

N = 14

EG: 14 (13 M, 1 F)

Age: 6–13 yrs

30 min per day during 4 weeks

• Working memory

• Facial processing skills

Baseline inter and post intervention

• Corsi-BT

• ATEC

No change (W = 7.5, n = 14, P = 1) was followed in the pre and post-intervention conditions; no significant change was observed in the Corsi total score (W = 27, n = 14, P = 0.22); participants did not achieve significantly from the month-long game-based training in reducing autistic symptoms or enhancing working memory

Not effective

There is no significant change in the autistic symptoms after the intervention training with the given game

Piana S et al., 2019, Italy [63]

IEEE Transactions on Affective Computing

Guess the Emotion: serious game

PC

Non-randomized Controlled Clinical Trial

N = 15

EG: 10 (9 M, 1 F)

CG: 5 (4 M, 1 F)

Age: 8–11 yrs

10 sessions, 20 min during 4 to 6 weeks

• Emotional Skills

• Facial processing skills

Baseline inter and post intervention

• Researchers set scales (Emotions in the recognition task, Response accuracy, Time of response)

There is a significant difference for the recognition, t(9) = 3:984, p = 0:003, and the expression task, t(9) = 4:439, p = 0:002, indicating that the performance at this task increase significantly from the first to the second period; the MD between the first and the second assessment was lower in the control group (9:60) than in the experimental group (21:50)

Statistically significant at all evaluation scales

During the application of this game, the accuracy of doing the task from the beginning to the end of the training sessions has increased significantly in the trained group compared to the control group

Jouen AL et al., 2017, UK [64]

Child and Adolescent Psychiatry and Mental Health

GOLIAH: Serious Game

PC and Mobile app

Non-randomized Controlled Clinical Trial

N = 24

EG: 14 (14 M, 0 F)

CG: 10 (10 M, 0 F)

Age: 5–8 yrs

100 sessions, 30 min, during 6-months

Attention

Imitation skills

Baseline inter and post intervention

• ADI-R

• VABS-II

• CBCL

• SCQ

There were significant signs of progress in ADOS scores, Vineland socialization score, Parental Stress Index total score, and Child Behavior Checklist internalizing, externalizing, and total problems (all p < 0.05, Linear Mixed Models, time effect); meaning that treatment given in both groups was favorable and practical

Statistically significant on some evaluation scales

The findings of the 6-month training indicate that both the acceptance of using the game platform and the absence of parental stress have been seen

Hayoung A. Lim et al., 2022, USA [65]

Child Language Teaching and Therapy

An Online Music-Based Speech and Language Learning Game: Online edtech

PC and Mobile app

Before & After Clinical Trial (without control)

N = 26

Age: 2–6 yrs

4–6 sessions over a 2-week

Language skills

Baseline and post intervention

• CARS

• Researchers set scales (Response accuracy, Time of response, Verbal production)

Outcomes point that the SS4Kids program is an efficacious music-based speech and language training technique for supporting target word production in children across a two-week timespan (all p < 0.05.)

Statistically significant at all evaluation scales

Emerging proof of the significance of an online evidence-based practice supports the speech and language outcomes for various children in early intervention

Chaoxin and Jun Yang, 2022, China [66]

Brain sciences

Virtual Training

PC

Randomized controlled trial

N = 100,

EG1 = 34

EG2 = 33,

CG = 33

MAge: 12.9 yrs

3 sess in a week,

30 min

Attention

Baseline, post intervention, and 6-week-follow-up

• CARS

• Researchers set scales (Visual attention)

Although none of the three participants noticed progress in the correct rate, the observations of the VT and PE groups were significant (p < 0.05) compared to the CG group in uncovering the detection rate of the probe stimulus

Statistically significant at all evaluation scales

This program has the potential to improve visual attention in children with ASD

  1. Abbreviations: SRS-2 Social Responsiveness Scale, MSEL Mullen Scales of Early Learning, WISC-IV Wechsler Intelligence Scale for Children, PPVT-R Peabody Picture Vocabulary Test-Revised, VABS-II Vineland Adaptive Behavior Scales-II, RCPM Raven’s Colored Progressive Matrices, GPS Gaze Positions Scale, WPPSI Wechsler Preschool and Primary Scale of Intelligence, VIQ Verbal Intelligent Quotient, PIQ Performance Intelligent Quotient, S-CAT Social-Communication Assessment Tool, PECS Picture Exchange Communication System, ASEBA Achenbach System of Empirically Based Assessment, ABC The Aberrant Behavior Checklist, SSRS The Social Skills Rating System, BRIEF The Behavior Rating Inventory of Executive Function, SART Sustained attention response task, Corsi-BTT The Corsi block tapping task, SSIT Set-shifting improvement tasks, WCST Wisconsin Card Sorting Test, MCST Modified Card Sorting Test, VABS-II Vineland Adaptive Behavior Scales, CPM Colored Progressive Matrices (CPM), Academic Assessments (Math, Reading Comprehension, Copying), CARS Childhood Autism Rating Scale, VMS Visual Matching Scale, BRIEF-P T GEC Behavior Rating Inventory of Executive Function Parent T Score General Executive Composite, Conners-3-P T EF Executive Functions, CRSD-ant test Attention network test, ADHDT Attention-Deficit/Hyperactivity Disorder Test (2nd ed), ATEC The Autism Treatment Evaluation Checklist, ADI-R Autism Diagnostic Interview-Revised, CBCL The Child Behavior Checklist, SCQ The Social Communication Questionnaire, SD Standard Deviation, M Mean, PC Personel Computer, ADOS Autism Diagnostic Observation Schedule, CAI Computer-Assisted Instruction, TII Teacher-Implemented Instruction