Plastic TASTER: a switching training game for people with PWS that adapts to individual needs (year 2)

Task switching is a cognitive process important for regulating behaviour. People with PWS generally show impaired switching and this difficulty is linked to people resisting change and showing temper outbursts triggered by changes.

We are currently funded by FPWR to develop a prototype computer game, TASTER (Training Attention Switching for Temper Episode Reduction) for improving task switching in children with PWS. We have developed the prototype to satisfy the needs and preferences of a small group of children with PWS using methods that capitalised on expertise from Brain Behavioural Science, and from Software Engineering. With these methods, we have built the game so that it establishes conditions that research shows are critical for supporting the transfer of gains in switching that occur in the game, to gains in real life functioning. We have shown that even very brief engagement with the game can improve in-game switching. Further, we are about to start a controlled evaluation of how far these in-game improvements transfer to gains in other measures of switching.

In Year 2 we propose to extend the TASTER prototype so that it can train switching in any individual with PWS who can engage with the tablet based game. We will do this using a technology – applied in successful commercial games – that allows on the fly generation of new game play so that the game remains motivating and exciting for all players. We will also apply methods we have refined during the first year to iteratively develop how the game adapts itself to trainees’ ongoing performances. Importantly, this adaptation will involve game demands on switching and on non-switching cognitive processes, meaning that individuals’ relative cognitive strengths can support improvements in switching. Together, these new plastic features of the TASTER game will be tailored to establish conditions that best promote the transfer of in-game switching gains to daily life, and compliance with training across diverse participants. Research activities are designed to facilitate multiple potential avenues to further support for game development.

24-60 trainees with PWS of varied ages and from different countries, showing a strong resistance to change, along with their caregivers, will be recruited. Caregivers will be asked to keep a time efficient temper outburst diary over the course of the project. Participants will engage with preference gathering and play testing sessions at home, facilitated via Skype by a researcher. These sessions will allow the game to be tailored to preferences and needs across the full range of trainees. Following this development, the game will be evaluated for its effects on switching and behaviour by comparing a random selection of individuals asked to engage with Plastic TASTER to a random selection asked to engage with a (placebo) version of the game that places only low demands on switching.

Research Outcomes: Public Summary

In this project, we aimed to develop a cognitive training programme capable of reducing the difficulties with changes to routines, plans and expectations that individuals with PWS experience. The cognitive training programme was based on the prototype we created and tested in earlier work funded by FPWR. Fundamentally, the training is a computer game that involves moving a character around the screen to collect creatures of a particular type, but avoid creatures of other types.
1. We aimed to build the new task switching training so that it could motivate a wide range of people with PWS to play over extended periods of time.
We achieved this objective by building “The Three Crowns” game so that it has four different systems that provide players with ongoing rewards for play. All of these systems involve players collecting things, and so complement the particular affiliation for collecting things that we know many people with PWS show. One of the systems is attached to an external prize that people with PWS can choose (provided using an Amazon voucher).
27 children and adults with PWS, aged between 6 and 31 years played with a version of The Three Crowns during a 5 week evaluation period (we created an “active” version designed to train task switching and a “placebo” version that was the same in as many respects as possible but placed minimal demands on task switching). The minimum total engagement duration was 93 minutes, but more than half of the participants engaged for at least 500 minutes (including 8 of 14 participants playing the active version). 500 minutes equates to 20 minutes per day, 5 days per week over the 5 weeks, which is the duration applied in a number of previous research studies with computer mediated cognitive training. Several participants engaged for much longer than 500 minutes. The participant who engaged most played for more than 63 hours.
Thus, we succeeded in creating a game that is capable of motivating a wide variety of individuals with PWS to engage for extended periods of time. Although not all participants achieved our target gameplay duration within the 5 week period, most families reported that this was due to life commitments getting in the way rather than a reluctance to play.
2. We aimed to develop the task switching game to be highly adaptive to suit each individual player’s needs and flexibly train each individual’s task switching in the best way for them.
We achieved this objective by building multiple systems into The Three Crowns Game that allow the game to adapt the gameplay experience presented to each player on an individual basis, depending on how that player is performing in the game.
We built an override function into the game to ensure that if a player reached a point where they were unable to progress in the game, we could manually modify their gameplay experience to allow them to be successful and continue to make progress. However, demonstrating the efficacy of our systems for automatically adapting players’ gameplay experiences based on ongoing performance, we did not need to use the override function during the evaluation.
3. We aimed to develop the new task switching game so that it helps players to develop skills in cognitive capacities that appear to support task switching, including forming abstract representations, inhibition and working memory updating.
We achieved this objective by designing The Three Crowns game to systematically expose players to a very large number of new rules in a large number of different contexts. Both of these situations require the formation of abstract representations. We also created a large number of game features specifically designed to increase demands on either task switching, inhibition or working memory updating. For example, a feature that placed increased demands on working memory updating involved a cloud that appeared around the player’s character and obstructed their view, requiring the played to remember which creatures entered the space around their player, and keep updating this memory. The systems that adapted the gameplay based on players’ ongoing performance did this separately for performance with switching, inhibition and working memory updating features.
Thus, players could progress through the game in different ways depending on their own profile of strengths and weaknesses in skills that support task switching. We have monitored each individual players’ progress through the game during our evaluation. Initial inspection of these data has revealed that players did indeed take different pathways through the game and show evidence of different relative strengths and weaknesses over time in various game features that we know to be most taxing on specific cognitive capacities.
4. We aimed to evaluate our new task switching game for its capacity to mediate improvements in task switching and resistance to change.
We achieved this objective by conducting a placebo controlled evaluation of the training game over a 5 week period. The 27 participants were randomly allocated to engage with either the active or placebo versions of The Three Crowns. Participants remained blind to their allocation as did members of the research team who were liaising with participants on a regular basis. Following the 5 week test period, participating families were informed which version of the game they had been engaging with and if this was the placebo version, they were invited to engage with the active version. A follow up assessment was conducted 5 weeks following the end of the test period.
During the test period of the evaluation, the research team monitored participants’ game play and made regular contact (approximately once per week) to encourage engagement in the form of text messages referring to progress in the game, and specifically towards reaching the target required to receive the external prize. The external prize was awarded to all participants following the test period. During the follow up period, the research team made less regular contact with participating families. There was reduced engagement with The Three Crowns during the follow up period relative to during the test period. Thus, the external motivation system, including monitoring from outside of the family and the external prize appears to be important for mediating ongoing engagement. Further development of these aspects of the intervention in future may therefore be beneficial.
Outcome measures included three informant report questionnaires on resistance to change and four game-like assessments of task switching administered to the person with PWS via a computer. Total time to complete all of the assessments was less than 1 hour. However, fitting these assessments around other commitments on the multiple occasions we required was still demanding for many participating families. We have just finished collecting the final assessments from families. We will go on to perform the formal efficacy analysis, blinding ourselves to group allocation.
Anecdotal comments from families following their completion in the study suggest that some families feel that the training may have contributed to positive behaviour changes shown by individuals with PWS. However, for the most part these are subtle and not in major resistance to change behaviours.

Research Outcomes: Publications

Developing a Task Switching Training Game for Children With a Rare Genetic Syndrome Linked to Intellectual Disability. Robb N, Waller A, Woodcock KA. Simulation & Gaming, 2019.