A Computerized, Adaptive Therapeutic Gaming Approach Training Visual Perceptual Skills in Children With CVI (iVision2_WP3)

A Computerized, Adaptive Therapeutic Gaming Approach Training Visual Perceptual Skills in Children With CVI

A Randomized Controlled Trial to Evaluate the Effectiveness of a Computerized, Adaptive Therapy Approach for Children With Cerebral Visual Impairment

This clinical trial aims to compare the effectiveness of our adaptive, therapeutic game to the effectiveness of the same game, without the adaptive component. In the adaptive game, the game entry level is adapted to the visual perceptual capacities of the child, as defined by the visual perceptual profile. In addition, the difficulty level of the adaptive game will adapt itself to the gaming results and behaviour of the child. The non-adaptive version of the game consists of the same set of mini-games, but the entry-level is the same for all children (basic or 0 entry-level) and gradually increased, independent of the gaming results, success and behaviour of the child. The researchers will use a double-blind, randomized controlled trial design, including children with a developmental age between 3 and 12 years old, a diagnosis of CVI, acuity >0.2, with sufficient manual coordination to control a mouse, keypad or touch screen. All children will use the gamified therapy program for three months, with a minimum of three times per week, 15- 20 minutes. A blinded evaluator will evaluate the effectiveness on the main components of the visual perceptual profile of the child (primary outcome), on eye tracking parameters, functional vision and quality of life, at the end and at three months follow-up. Enjoyment and user experience will be monitored closely during the intervention period. As usual and regular therapy of the children will not be influenced during the intervention period, we will ask the parents and/or caretaker to register all other relevant gaming and therapy activities performed during that period. It is hypothesized that children will benefit more from an individualized, adaptive training approach compared to the generic, non-adaptive version of the program.

Cerebral Visual Impairment (CVI) is the most common cause of visual disability in developed countries (one to two cases per 1000 live births). A person with CVI usually has normal eye function but processing visual information in the brain is hindered. This results in a complex variation of symptoms, ranging from problems with object and shape recognition over problems with figure ground and cluttered scenes to deficits in spatial navigation. Each child with CVI presents with a nearly unique visual perceptual profile, due to the varying nature of the underlying damage and the appearance of age related deficits. In a previous project, we developed a method to quantify the visual profile of children with CVI, enabling more individualized and targeted therapy. The researchers developed an adaptive, personalized gamified visual perceptual therapy program for children with CVI, based on this quantified visual profile, with the aim to apply a targeted, individualized approach that strengthens motivation and thereby increases effectiveness. This clinical trial aims to compare the effectiveness of our adaptive, therapeutic game to the effectiveness of the same game, without the adaptive component. In the adaptive game, the game entry level is adapted to the visual perceptual capacities of the child, as defined by the visual perceptual profile. In addition, the difficulty level of the adaptive game will adapt itself to the gaming results and behaviour of the child. The non-adaptive version of the game consists of the same set of mini-games, but the entry-level is the same for all children (basic or 0 entry-level) and gradually increased, independent of the gaming results, success and behaviour of the child. The researchers will use a double-blind, randomized controlled trial design, including children with a developmental age between 3 and 12years old, a diagnosis of CVI, acuity >0.2, with sufficient manual coordination to control a mouse, keypad or touch screen. All children will use the gamified therapy program for three months, with a minimum of three times per week, 15- 20 minutes. The software will be installed on a personal device of choice by the child and/or its parents. Children willing to participate, but not having a personal device, will receive a tablet computer from the researchers during the intervention period. A blinded evaluator will evaluate the effectiveness on the main components of the visual perceptual profile of the child (primary outcome), on eye tracking parameters, functional vision and quality of life, at the end and at three months follow-up. Enjoyment and user experience will be monitored closely during the intervention period. As usual and regular therapy of the children will not be influenced during the intervention period, we will ask the parents and/or caretaker to register all other relevant gaming and therapy activities performed during that period.

Cerebral Visual Impairment, Training, Rehabilitation, Therapeutic gaming, Adaptive therapy, Randomized Controlled Trial, Visual Perception

Participating children, their parents/caretakers as well as their guiding therapist will be blinded for allocation to the treatment condition Assessors will be blind to the treatment condition

The adaptive therapeutic game will use a game-entry level that is based on the individual visual perceptual profile of the child. Children with higher visual perceptual capacities will be able to start the game at a higher entry level compared to children with lower visual perceptual capacities. Thereby, the entry level will also be different for the different games. In addition, the adaptive game uses an in-game adaptivity system that has been developed using artificial intelligence (more specifically, reinforcement learning): this means that the game can adjust the difficulty automatically, based on the game-behaviour and success of the child. Children learning fast, will more quickly move to higher difficulty levels compared to children learning slower. Thereby, this in-game adaptivity also enables children to return to lower difficulty levels when a difficulty level is too high.

The non-adaptive therapeutic game will use the same, lowest entry level for all children. During game play, a fixed stepwise increase in difficulty will be built in, not adjusted to the gaming behaviour or success of the child. All children will follow the same, gradual approach in difficulty and a fixed number of trials is set for each difficulty level. To prevent extreme frustration however, a safety margin is integrated by preventing the difficulty level to increase further when a child has more than a predefined number of unsuccessful trials. Likewise, the stepwise increase in difficulty level will only continue after a fixed, predefined number of successful trials. A child will also never return to a lower difficulty level, once a difficulty level is reached.

All children will be asked to use the adaptive version of the therapeutic game for three months, during their free time, at home or at the special school where they stay. They will be asked to play the game minimally three times per week between 20-30 minutes per session. The researchers will install the game on a device of choice for the children. Back-up devices are available from the study team to ensure that children having no suitable device can participate in the study. The parent and/or caretaker will be asked to support the child in starting the program, but not to intervene during the actual game time. The game is designed to ensure that children can play independently. Therefore, children will be able to play at home, at school or during their free time. The program will not replace their regular therapy program, but will be provided additionally. The number of regular therapy hours will however be registered.

All children will be asked to use the non-adaptive version of the therapeutic game for three months, during their free time, at home or at the special school where they stay. They will be asked to play the game minimally three times per week between 20-30 minutes per session. The researchers will install the game on a device of choice for the children. Back-up devices are available from the study team to ensure that children having no suitable device can participate in the study. The parent and/or caretaker will be asked to support the child in starting the program, but not to intervene during the actual game time. The game is designed to ensure that children can play independently. Therefore, children will be able to play at home, at school or during their free time. The program will not replace their regular therapy program, but will be provided additionally. The number of regular therapy hours will however be registered.

Quantified visual profile as described by Ben Itzhak et al. (2021): at individual (most deviant z-score) and general level (overall visual perceptual capacity in a composite score)

Quantified visual profile as described by Ben Itzhak et al. (2021): at individual (most deviant z-score) and general level (overall visual perceptual capacity in a composite score)

Quantified visual profile as described by Ben Itzhak et al. (2021): at individual (most deviant z-score) and general level (overall visual perceptual capacity in a composite score)

- L94 Visual Perceptual Battery: object recognition battery which evaluates (degraded) object recognition, figured-ground, motion perception and global-local processing (Ortibus et al., 2015). The Beery Visual Motor Integration (Beery VMI): a standardized and norm-referenced screening tool for visual-motor deficits. The VMI helps assess to what extent children can integrate their visual and motor abilities. In addition to a copying task, the child also performs a visual perception matching task of the same constructs seen before in the copy task, and a motor coordination task, developed to assess the supplementary motor deficits. (Beery et al., 2010). The Preschool Judgement of Line Orientation (PJLO) (Stiers et al., 2005): In 24 items the orientation of one or two target lines has to be matched to 2, 4, or 11 differently orientated choice lines. Motion perception tasks: tackling structured motion, motion coherence and biological motion (Van der Zee et al., 2019).

- L94 Visual Perceptual Battery: object recognition battery which evaluates (degraded) object recognition, figured-ground, motion perception and global-local processing (Ortibus et al., 2015). The Beery Visual Motor Integration (Beery VMI): a standardized and norm-referenced screening tool for visual-motor deficits. The VMI helps assess to what extent children can integrate their visual and motor abilities. In addition to a copying task, the child also performs a visual perception matching task of the same constructs seen before in the copy task, and a motor coordination task, developed to assess the supplementary motor deficits. (Beery et al., 2010). The Preschool Judgement of Line Orientation (PJLO) (Stiers et al, 2005): In 24 items the orientation of one or two target lines has to be matched to 2, 4, or 11 differently orientated choice lines. Motion perception tasks: tackling structured motion, motion coherence and biological otion (Van der Zee et al., 2019).

- L94 Visual Perceptual Battery: object recognition battery which evaluates (degraded) object recognition, figured-ground, motion perception and global-local processing (Ortibus et al., 2015). The Beery Visual Motor Integration (Beery VMI): a standardized and norm-referenced screening tool for visual-motor deficits. The VMI helps assess to what extent children can integrate their visual and motor abilities. In addition to a copying task, the child also performs a visual perception matching task of the same constructs seen before in the copy task, and a motor coordination task, developed to assess the supplementary motor deficits. (Beery et al., 2010). The Preschool Judgement of Line Orientation (PJLO) (Stiers et al., 2005): In 24 items the orientation of one or two target lines has to be matched to 2, 4, or 11 differently orientated choice lines. Motion perception tasks: tackling structured motion, motion coherence and biological otion (Van der Zee et al., 2019).

- Test of Visual Perceptual Skills-3 (TVPS-3): includes subtasks of visual discrimination, visual memory, visual-spatial relationships, form constancy, visual sequential memory, figure ground, and visual closure (Martin et al., 2006). Beery VMI (see above) Subtasks from the Revisie Amsterdamse Kinder Intelligentie test 2 (Rakit 2), (Bleichrodt et al. (1999)): In the Hidden figures subtasks, hidden objects need to be identified in a crowded background. In the figure recognition subtask, the child has to recognize incomplete drawings from everyday objects. Subtasks from the Developmental Neuropsychological Assessment (NEPSY-II-Nl): evaluating different aspects of visuospatial functioning. In the arrows subtask, the child must determine visually which arrows will end up in the middle of a target if they were extended. In the geometric puzzles subtask, the child must find geometric figures amongst other geometric figures.

- Test of Visual Perceptual Skills-3 (TVPS-3): includes subtasks of visual discrimination, visual memory, visual-spatial relationships, form constancy, visual sequential memory, figure ground, and visual closure (Martin et al., 2006). Beery VMI (see above) Subtasks from the Revisie Amsterdamse Kinder Intelligentie test 2 (Rakit 2), (Bleichrodt et al. (1999)): In the Hidden figures subtasks, hidden objects need to be identified in a crowded background. In the figure recognition subtask, the child has to recognize incomplete drawings from everyday objects. Subtasks from the Developmental Neuropsychological Assessment (NEPSY-II-Nl): evaluating different aspects of visuospatial functioning. In the arrows subtask, the child must determine visually which arrows will end up in the middle of a target if they were extended. In the geometric puzzles subtask, the child must find geometric figures amongst other geometric figures.

- Test of Visual Perceptual Skills-3 (TVPS-3): includes subtasks of visual discrimination, visual memory, visual-spatial relationships, form constancy, visual sequential memory, figure ground, and visual closure (Martin et al., 2006). Beery VMI (see above) Subtasks from the Revisie Amsterdamse Kinder Intelligentie test 2 (Rakit 2), (Bleichrodt et al. (1999)): In the Hidden figures subtasks, hidden objects need to be identified in a crowded background. In the figure recognition subtask, the child has to recognize incomplete drawings from everyday objects. Subtasks from the Developmental Neuropsychological Assessment (NEPSY-II-Nl): evaluating different aspects of visuospatial functioning. In the arrows subtask, the child must determine visually which arrows will end up in the middle of a target if they were extended. In the geometric puzzles subtask, the child must find geometric figures amongst other geometric figures.

Questionnaires evaluating daily functioning: Insight Question Inventory, (IQI), Pediatric Quality of Life Inventory (PedsQL), the Flemish cerebral visual impairment questionnaire (FCVIQ).

Questionnaires evaluating daily functioning: Insight Question Inventory, (IQI), Pediatric Quality of Life Inventory (PedsQL), the Flemish cerebral visual impairment questionnaire (FCVIQ).

Questionnaires evaluating daily functioning: Insight Question Inventory, (IQI), Pediatric Quality of Life Inventory (PedsQL), the Flemish cerebral visual impairment questionnaire (FCVIQ).

Inclusion Criteria: Voluntary written informed consent of the participant or their legally authorized representative A confirmed diagnosis of CVI Dutch-speaking A developmental age between 3 and 12 years. Sufficient motor abilities (cerebral palsy with a Gross Motor Function Classification level <V) Sufficient fine motor function to handle a tablet or laptop (Manual Ability Classification System level <4) Able to express their experiences with the game. A visual acuity of more than 0.2 (Snellen notation) Exclusion Criteria: Any disorder, which in the Investigator's opinion might jeopardise the participant's safety or compliance with the protocol Any prior or concomitant treatment(s) that might jeopardise the participant's safety or that would compromise the integrity of the Trial Children not speaking or understanding Dutch language Children with limited gross motor function abilities (GMFCS V) Children with limited hand function (MACS 4-5) Children unable to express their experiences with the game due to serious speech disorders, deafness or autism. A visual acuity of less than 0.2 (Snellen notation) Serious behavioural problems limiting participation to the games

No data will be shared with other researchers. Only upon request, in specific conditions, when ethical approval and after additional patient consent, data will be shared.