From role-play to situated feedback: re-envisioning child mental health promotion interventions with digital technology

The fellowship is a £1.2M research project that will enable myself and my team to envision and design new technology-enabled models for child mental health interventions over the next 4 years, with a potential 3 year extension.

Supported by an amazing group of mentors across KCL (Prof Sonuga-Barke), Stanford (Prof Gross), Oxford (Prof Van Kleek, Prof Jirotka, Prof Shadbolt), Northwestern (Prof David Mohr), Michigan (Prof Klasnja), UCL (Prof Deighton), Harvard (Prof Torous) and others, the work will span both fundamental HCI research into new intervention delivery mechanisms, as well as practical clinical application and large scale deployments.

The general summary and objectives of the fellowship work are below, and the Research page has information about our on-going work.

Feel free to contact me for more details or if you’d like to explore collaboration possibilities.

The aims of the fellowship are to examine how emerging technologies can fundamentally re-envision the conceptual models and mechanisms-of-delivery for existing prevention interventions in the context of child mental health. Such an innovative approach is needed to address the unprecedented mental health treatment gap faced across the UK and worldwide: more than 1 in 10 children and young people have a clinically diagnosable mental health disorder, yet only 30% have had access to appropriate intervention, and less than half of these improve from the treatment. Mental health promotion interventions are seen as one of the principled ways of addressing these issues: by developing key protective factors (such as emotion regulation or parenting techniques) for both at-risk and general populations, such interventions can improve wellbeing and reduce the incidence of mental disorders. However, even the most effective programmes are still dependent on in-person delivery techniques and intervention mechanisms available since the 60s, thus lacking scalable mechanisms to support children in the everyday settings where protective competencies are needed, and being developed.

The core vision proposed by the fellowship research agenda is that digital technologies can lead to entirely new model of prevention interventions that are fully incorporated into the lives of target populations, thus addressing the need for situated learning support. However, beyond PI’s pilot work, HCI and Prevention Science fields lack even a basic understanding of the fundamental research questions necessary to deliver such situated interventions: it is not at all clear, for example, how to design technologies that provide useful contextualised support for children and their adults around protective competencies (technical RQs), how to do so in a psychologically effective way (psychological RQs), and how to design interventions that are engaging for users and are addressing their immediate needs (socio-technical RQs).

The fellowship takes up these challenges and lays out an ambitious programme of work designed to start unpacking these broader issues at the intersection of technology, psychological theory, and human-centred design. We address such highly interdisciplinary agenda by grounding the work in specific case studies that target fundamental protective factors for child mental health—emotion regulation and positive parenting—while encompassing all of the issues outlined above. The case study interventions then serve not only as exemplar proof-of-concept of how situated interventions can be developed (and shown to be efficacious), but also as initial steps towards extrapolating the psychological mechanisms and design patterns that are generalisable to other populations and protective factors.

By focusing on carefully selected case studies—complemented with a strong interdisciplinary mentorship network across KCL, Oxford, Stanford, Harvard, Northwestern University, and University of Michigan, among others—the ambition is to not only create effective interventions that can be deployed at scale, but also influence the wider research communities, and contribute to the UK and EU policy calls to deliver new approaches for mental health promotion.

The overarching goal of this project is to show, for the first time, a technology-enabled model of ‘situated’ interventions in the context of child mental health (MH) promotion. The ambition is to trigger a long-lasting shift in the ways in which child mental health interventions are conceptualised, designed, and delivered; and position the PI at the forefront of a new interdisciplinary research area at the intersection of Human-Computer Interaction (HCI), Psychology, and Prevention Science.

The main objectives for the first four years can be summarised as follows:

1. Envision, develop, and evaluate two exemplars of situated, technology-enabled interventions that— in contrast to the current workshop and classroom based approaches—are fully embedded in the lives of children and their families; and enable contextualised skills development through support and guidance in the everyday moments that matter. This will address a key open challenge in existing programmes, which rely on in-person training and thus lack any scalable mechanisms to facilitate learning in the everyday situations where protective competencies are used and developed.

   
   

   The case study contexts (emotion regulation and positive parenting) were selected as complementary intervention exemplars that (i) target crucial protective factors for child MH and thus address important societal need; (ii) span a range of technical, psychological and socio-technical possibilities to enable exploration of this novel design space; and (iii) are grounded in the PI’s pilot work and an interdisciplinary network of mentors (across KCL, Oxford, Harvard, Stanford, Michigan, Northwestern, among others) to ensure feasibility.

   
   

2. Introduce novel evaluation methodologies (Micro-RCTs, originating in behaviour science) into the child MH promotion context, enabling iterative assessments of psychological effects of the proposed interventions throughout the design process. Traditionally, interventions are evaluated through Randomised controlled trials (RCT), which require ‘complete’ interventions and long, expensive trials; making them infeasible for the iterative design process necessary for step-wise innovation. Situated, technology-enabled interventions can overcome such limitations, by allowing a rapid collection of proximal outcome data (e.g., hourly or daily) required for more efficient evaluation methods such as Micro-RCTs. Note that such methodologies are not feasible for existing non-technological interventions, due to pragmatic and cost issues associated with data collection (e.g., requiring in-home visits by trained staff for each data point).

3. Develop open-source toolkits, synthesising the technical, psychological, and design mechanisms based on the empirical data collected in (1) and (2). Toolkits will be made freely available to academic and clinical communities. Toolkits will comprise three layers, encapsulating different types of knowledge generated: conceptual layer (intervention theory, conceptual development); tools layer (specific components such as intervention modules that can be repurposed by researchers and clinicians); and developmental layer (low-level programming primitives to enable extendability and innovation).

The second half of the fellowship (years 5-7) will build on these findings to:

4. Expand the situated interventions research agenda to other protective factors and research groups, in three interrelated directions:

   • Strengthen the empirical evidence for interventions developed in Stages 1-3 to enable their transfer into clinical practice.
   • Examine how the identified intervention designs and patterns can be applied to cognitive behavioural interventions, drawing on partnerships with clinical (Harvard; Northwestern) and commercial (Silvercloud) partners.
   • Deepen the toolkit capabilities and facilitate an interdisciplinary community around their use, drawing on the mentorship network of world-leading academics.