Our paper, “Serious Game for ADHD Diagnosis,” presented at the 17th International Conference on Game and Entertainment Technologies in Budapest, proposed a different approach to ADHD assessment: an AI-powered serious game that embeds clinically informed tasks in an engaging, everyday-like environment.
The research explores how gameplay data can complement established assessment methods by capturing patterns related to attention, impulsivity, working memory, and executive function.
ADHD assessment often involves interviews, questionnaires, behavioural observation, and reports from different settings. These methods are essential, but assessment can sometimes feel stressful, abstract, or disconnected from everyday activities.
Our research explores how serious games can provide an additional behavioural channel.
Instead of asking players only to describe their attention or organisation, the game places them in realistic situations where they must act. Players manage orders, follow multi-step instructions, switch between activities, and respond to changing demands.
These tasks create an environment in which cognitive and behavioural patterns can be observed through performance.
During gameplay, the system records performance metrics related to:
For example, a player may need to remember several orders, prioritise tasks, switch between activities, and avoid irrelevant actions. These demands resemble some of the attentional and executive-function challenges people may encounter in everyday life.
The system does not interpret one mistake as a diagnosis. Instead, machine-learning models analyse patterns across multiple tasks and performance measures.
The game embeds tasks informed by the symptom domains and assessment considerations associated with the DSM framework.
The purpose is to create a structured environment in which relevant behavioural signals can be collected without requiring the player to complete only traditional forms.
The resulting data can be used to train and evaluate machine-learning models that distinguish patterns associated with ADHD-related behaviour.
These models may examine combinations of:
The important feature is the integration of task design, data collection, and machine-learning analysis within one serious-game environment.
This research underpins NeuroEscape, SeriousMind’s product for AI-supported behavioural and cognitive assessment.
NeuroEscape translates the research into an interactive experience set in a realistic kitchen environment. Players must manage orders, follow instructions, and coordinate multiple tasks while the system captures behavioural and cognitive data.
For clinicians, this type of information may provide an additional perspective alongside established assessment methods. For adults seeking clarity about possible ADHD symptoms, it may offer a more engaging and less intimidating starting point for discussion.
NeuroEscape is not intended to function as a standalone diagnostic tool. A formal ADHD diagnosis requires a comprehensive assessment by qualified professionals, taking into account clinical history, symptoms across settings, developmental information, and validated instruments.
The role of NeuroEscape is to support that process—not replace it.
One of the advantages of an everyday-like game environment is that it can make assessment feel less abstract.
A kitchen task is familiar. Managing several orders, remembering instructions, prioritising actions, and responding to interruptions are activities that resemble real-world demands.
This does not make the game equivalent to everyday life. It creates a structured simulation in which specific cognitive processes can be observed repeatedly and consistently.
The goal is to make the experience:
Serious games may also help users become more involved in the assessment process because they can see how their actions generate data.
Presenting the research at the 17th International Conference on Game and Entertainment Technologies in Budapest marked an important step in the development of this work.
Academic conferences create an opportunity to test ideas with researchers, receive feedback, compare methods, and identify the questions that need further investigation.
The journey from research paper to product requires additional work:
NeuroEscape represents the product direction that grew from this research.
AI can analyse patterns in gameplay data, but it cannot independently understand a person’s complete history or clinical context.
For that reason, any AI-supported ADHD assessment tool must be transparent about:
The scientific literature on serious games for ADHD shows promise, but it also highlights the need for validation across different ages, populations, and symptom profiles. For example, recent research on a serious game found stronger discriminative results in children than adolescents, demonstrating why age-specific validation matters.games.jmir
The research behind NeuroEscape reflects a broader vision for SeriousMind: combining artificial intelligence, psychology, neuroscience, and game design to create more engaging assessment experiences.
A serious game cannot replace a clinician.
But it can help create a new behavioural channel—one in which users complete meaningful tasks and generate structured data through interaction.
For adults and clinicians, this may support better questions, richer observations, and more informed conversations about attention, impulsivity, memory, and executive function.
The future of assessment may not be about choosing between traditional clinical methods and technology.
It may be about combining them responsibly.
© 2026 SeriousMind. All Rights Reserved. Privacy Policy