An Integrated Real-Time Approach to the Development of Human Problem Solving
Lead Research Organisation:
Birkbeck, University of London
Department Name: Psychological Sciences
Abstract
Problem solving is ubiquitous across every age and culture-how to navigate a cluttered environment, use a tool, and so on. As our bodies, skills, and environments change, new problems emerge and require new means to solve them. Solvers cannot perform the same habitual actions over and over by rote because real-world environments are variable, unpredictable, and full of novel situations. The overarching goal of the proposed project is to understand (and intervene on) processes that underlie the development of human problem-solving skills.
Traditionally, developmental research takes a normative, outcome-oriented approach by identifying the ages at which children succeed in solving various problems. This outcome-oriented approach established that motor problem solving begins in infancy and improves with age and experience. Yet, critical knowledge gaps exist: how do perception, cognition, and action unfold from moment to moment to enable efficient solutions? What experiences facilitate early motor problem-solving skills? And does motor problem-solving at a younger age predict subsequent high-order cognitive skills?
To answer these questions, we adopt a unique integrative approach combining perspectives and methods from developmental psychology, neuroscience, artificial intelligence, electrical engineering, and motor control. Our guiding hypothesis is that the variability in children's daily flux of a changing body in a variable world facilitates motor problem-solving skills by calibrating a real-time interactive system of perceptual, neural, and motor processes, which in turn facilitates subsequent cognitive skills.
We will test this hypothesis in a longitudinal study by using object manipulation as a model system. Aim 1 of the proposed research will determine how children's efficiency, flexibility, stability, and generalisability during problem solving rely on the timing of real-time cascade - from gathering information about the task, to neural differentiation of task-relevant information, to the recruitment of relevant muscles, and finally movement execution. Aim 2 will manipulate the variety of children's everyday experiences during object play in their home, quantify it across one year using unique custom-built toys, and determine the contribution of early variable experiences to subsequent problem-solving skills. Aim 3 will assess the developmental relations between motor problem-solving skills and subsequent high-order cognitive skills, including reasoning, planning, and executive functioning.
By uncovering the origins of motor problem solving, the proposed work has broad implications for understanding foundational learning-related processes such as planning, organisation, attention, behavioural flexibility, executive function, and self-regulation. Identifying factors that promote problem solving can support practices, programs, and policies to improve children's outcomes and make them reach their full potential, whatever their background, family income, or place of living. Moreover, this work will provide insights into how individuals solve the problem in real-time and thereby distinguish whether causes of deficits result from lack of perceptual information, neural processing, and/or motor dexterity. Thus, findings may have translational implications for understanding disorders related to deficits in problem solving and may support educational interventions. By knowing the source of deficit in individuals, interventions can be tailored to their specific needs.
Finally, the proposed work will advance international research on learning and development through open sharing of rich data, methods (experimental protocols, machine-learning algorithms, analysis scripts), findings (videos, physiological data), and engineering protocols (intelligent toys) in established repositories. The proposed project promises to pave the way for research in STEM and other disciplines to explain and promote children's outcomes.
Traditionally, developmental research takes a normative, outcome-oriented approach by identifying the ages at which children succeed in solving various problems. This outcome-oriented approach established that motor problem solving begins in infancy and improves with age and experience. Yet, critical knowledge gaps exist: how do perception, cognition, and action unfold from moment to moment to enable efficient solutions? What experiences facilitate early motor problem-solving skills? And does motor problem-solving at a younger age predict subsequent high-order cognitive skills?
To answer these questions, we adopt a unique integrative approach combining perspectives and methods from developmental psychology, neuroscience, artificial intelligence, electrical engineering, and motor control. Our guiding hypothesis is that the variability in children's daily flux of a changing body in a variable world facilitates motor problem-solving skills by calibrating a real-time interactive system of perceptual, neural, and motor processes, which in turn facilitates subsequent cognitive skills.
We will test this hypothesis in a longitudinal study by using object manipulation as a model system. Aim 1 of the proposed research will determine how children's efficiency, flexibility, stability, and generalisability during problem solving rely on the timing of real-time cascade - from gathering information about the task, to neural differentiation of task-relevant information, to the recruitment of relevant muscles, and finally movement execution. Aim 2 will manipulate the variety of children's everyday experiences during object play in their home, quantify it across one year using unique custom-built toys, and determine the contribution of early variable experiences to subsequent problem-solving skills. Aim 3 will assess the developmental relations between motor problem-solving skills and subsequent high-order cognitive skills, including reasoning, planning, and executive functioning.
By uncovering the origins of motor problem solving, the proposed work has broad implications for understanding foundational learning-related processes such as planning, organisation, attention, behavioural flexibility, executive function, and self-regulation. Identifying factors that promote problem solving can support practices, programs, and policies to improve children's outcomes and make them reach their full potential, whatever their background, family income, or place of living. Moreover, this work will provide insights into how individuals solve the problem in real-time and thereby distinguish whether causes of deficits result from lack of perceptual information, neural processing, and/or motor dexterity. Thus, findings may have translational implications for understanding disorders related to deficits in problem solving and may support educational interventions. By knowing the source of deficit in individuals, interventions can be tailored to their specific needs.
Finally, the proposed work will advance international research on learning and development through open sharing of rich data, methods (experimental protocols, machine-learning algorithms, analysis scripts), findings (videos, physiological data), and engineering protocols (intelligent toys) in established repositories. The proposed project promises to pave the way for research in STEM and other disciplines to explain and promote children's outcomes.
Publications
Dexter M
(2023)
The effects of typical ageing on cognitive control: recent advances and future directions.
in Frontiers in aging neuroscience
Golmakani S
(2025)
Children plan manual actions similarly in structured tasks and in free play.
in Journal of experimental child psychology
Grandchamp Des Raux H
(2024)
The role of action concepts in physical reasoning: insights from late childhood
in Philosophical Transactions of the Royal Society B: Biological Sciences
Hascher S
(2023)
The power of multivariate approach in identifying EEG correlates of interlimb coupling.
in Frontiers in human neuroscience
Ossmy O
(2024)
Walking and falling: Using robot simulations to model the role of errors in infant walking.
in Developmental science
| Description | Our research has centered on a new approach for understanding how children learn to solve everyday "motor problems" by playing with objects (for example, figuring out how to use a hammer or slot a handle into a hole). We developed and tested "Intelligent Toys," ordinary-looking play objects equipped with sensors and special software that can track children's actions in real time-both in the lab and at home. By capturing detailed information about how children grasp, move, and explore these toys, we can see which play patterns help them learn to plan, adapt, and solve problems more effectively. In an initial study with children and adults, we discovered that children who look carefully at an object before reaching for it (rather than rushing straight in) are more likely to choose a smarter, more efficient grip. Such findings suggest that children's problem-solving abilities emerge from a "real-time" flow of information: seeing, thinking, and then doing, all in quick succession. In other words, early steps like looking or paying attention can shape whether a child ultimately solves a problem smoothly or struggles and has to correct mistakes as they go. Taken together, these findings demonstrate that children's everyday play behaviors offer important clues about how they develop more complex skills like reasoning, planning, and creativity. By using sensor-embedded toys and analyzing the patterns of children's play over time, we can not only map their progress but also potentially catch early signs of difficulties that might benefit from extra support. We are currently working on the analyses and writing the manuscript. |
| Exploitation Route | Ultimately, the work funded by this award opens the door to a future where children's natural play is both preserved and enhanced-giving them the freedom to explore, invent, and learn, while offering researchers, educators, and caregivers deeper insights into the subtle processes that underlie a child's developing mind. We are now addressing several entities: - Researchers can build upon this work by integrating sensor technology into their own studies of child development, whether they focus on cognitive, social, or motor skills. Making sensor-based play tools available to a wider range of labs and projects can spur new insights into how children learn through hands-on exploration. - Educators and Therapists may adopt "Intelligent Toys" in classrooms or clinical settings to watch how children approach new tasks or adapt to different challenges. By receiving real-time feedback on each child's play patterns (for example, which grips they try, how they explore the toy's features), teachers or therapists could better tailor learning activities or interventions. - Parents and Caregivers could benefit from a future user-friendly version of "Intelligent Toys," which might provide gentle guidance or suggestions on which types of toys or play activities would best support their child's current developmental stage. This could help parents feel more confident about promoting healthy development at home without needing extensive technical expertise or specialized training. - Toy Manufacturers and Tech Companies interested in bridging technology and early childhood learning might collaborate to produce more sophisticated, mass-market versions of these sensor-embedded toys. By combining everyday play with data-driven insights, the toy industry could expand the range of inclusive, adaptive products available to families everywhere. We are in the process of communicating with all of them to push forward the use of intelligent toys, while we continue our studies. |
| Sectors | Digital/Communication/Information Technologies (including Software) Education Leisure Activities including Sports Recreation and Tourism Other |
| Description | Identifying Early Markers of Autistic Spectrum Disorder in Naturalistic Motor Behavior using High-Frequency Sampling |
| Amount | $499,366 (USD) |
| Organisation | Simons Foundation |
| Sector | Charity/Non Profit |
| Country | United States |
| Start | 03/2023 |
| End | 03/2025 |
| Description | Real-time mechanisms underlying short-term and long-term effects of physical activity on problem-solving skills in children with ADHD |
| Amount | £64,890 (GBP) |
| Funding ID | 917/4975 |
| Organisation | The Waterloo Foundation |
| Sector | Charity/Non Profit |
| Country | United Kingdom |
| Start | 03/2023 |
| End | 06/2025 |
| Description | The effects of multi-modal information on the development of action planning |
| Amount | £479,687 (GBP) |
| Funding ID | RPG-2022-327 |
| Organisation | The Leverhulme Trust |
| Sector | Charity/Non Profit |
| Country | United Kingdom |
| Start | 06/2023 |
| End | 01/2028 |
| Title | Intelligent Toys |
| Description | "Intelligent Toys" is an innovative technology that combines everyday child-friendly objects embedded with motion and sensory detectors-such as accelerometers, gyroscopes, force sensors, light sensors, and so on-with advanced machine-learning based software. This integration enables real-time monitoring and analysis of how children interact with toys, providing invaluable insights into their developmental progress. The technology detects atypical patterns in play, identifies which toys encourage the most variability and engagement (which are critical for child development according to research), and offers personalised recommendations for future play activities. By harnessing data-driven insights, "Intelligent Toys" revolutionises the way we understand and support child development through object play, benefiting educators, healthcare professionals, and parents alike. |
| Type Of Material | Physiological assessment or outcome measure |
| Year Produced | 2024 |
| Provided To Others? | No |
| Impact | "Intelligent Toys" is a comprehensive and pioneering system that seamlessly integrates sensor-embedded toys with advanced software algorithms to monitor and analyse children's play behaviours with unparalleled precision. The toys are designed and equipped with an array of sensors-including motion sensors like accelerometers and gyroscopes, as well as force and light sensors-that capture high-resolution data on how children manipulate and interact with objects in real-time and in high frequency. The collected data is transmitted to a software platform that employs state-of-the-art machine learning and artificial intelligence techniques to interpret the rich stream of information. The software utilises complex pattern recognition algorithms and predictive modelling to analyse play behaviours, assess developmental milestones, and detect any atypical patterns that may indicate developmental delays. It leverages deep learning neural networks to evaluate the variability and complexity of play, providing insights into cognitive and motor skill development at the individual child level. Furthermore, the system includes a recommendation engine powered by AI, which suggests new toys or activities tailored to each child's unique developmental needs, thereby promoting optimal growth and learning. This recommendation system analyses longitudinal data to adapt to the child's evolving abilities and interests, ensuring personalised and dynamic support. The next state in developing this technology is to design a highly user-friendly, accessible, and intuitive interface for parents, educators, and healthcare professionals. It will offer comprehensive reports and real-time feedback, making it an invaluable tool in educational settings, healthcare environments, and home use. The main innovation lies in the novel combination of existing toys available in the market, advanced sensor technology (that also exists in the market), and the sophisticated adaptation of machine learning algorithms specifically for monitoring and enhancing child development through play. This novel holistic approach transforms ordinary child playtime into a scientifically-informed developmental assessment and enhancement opportunity. Initial testing has been conducted on a relatively small scale involving several families and children, yielding promising results that validate the concept. As more data is collected and the system is expanded to more families, the machine learning models will continue to improve, providing increasingly accurate indications of typical development patterns and personalised recommendations for each child's next developmental steps. Intelligent Toys" addresses the critical need for objective, high-frequency, automatic, and data-driven monitoring of child development through play. Traditional methods of assessing developmental progress are often subjective, infrequent, and reliant on expert observation, which can overlook subtle but significant variations in behaviour. The technology solves several problems: • Early Detection of Atypical Development: By continuously monitoring play patterns, it can identify early signs of developmental delays or disorders, allowing for timely interventions. • Personalised Developmental Support: It provides customised recommendations for toys and activities that align with each child's unique needs, promoting more effective learning and growth. • Parental Guidance: It empowers parents with insights into their child's development, reducing uncertainty and helping them make informed decisions about play and learning resources. • Educational and Clinical Applications: For educators and healthcare professionals, it offers a valuable tool for monitoring progress, planning interventions, and engaging with children in more meaningful ways. • Standardisation of Developmental Assessment: By utilising objective data and consistent metrics, the technology reduces the variability and bias inherent in human observation, leading to more accurate assessments of a child's developmental stage. • Contribution to Research: The aggregated data from multiple users can provide a rich source of information for researchers in child development, psychology, and education, potentially leading to new insights and advancements in these fields. • Customisable Learning Paths: The technology adapts to each child's evolving abilities, creating a personalised learning trajectory that can adjust in real-time, ensuring that developmental support remains aligned with the child's growth. • Resource Optimisation for Institutions: For educational settings and healthcare facilities, the data can help optimise resource allocation by identifying common areas where children may need additional support, allowing for more efficient planning and intervention strategies. • Integration with Other Systems: The technology can be integrated with existing educational and healthcare records, providing a holistic view of a child's development and facilitating better-informed decisions by professionals involved in the child's care. • Support for Special Needs Education: For children with special educational needs, the technology can offer tailored insights and recommendations that cater to specific challenges, aiding in the development of targeted support plans. By addressing these multifaceted needs, "Intelligent Toys" represents a significant advancement in how child development is monitored and supported, leveraging technology to create a more informed, responsive, and effective developmental environment. |
| Title | Data file from The Role of Action Concepts in Physical Reasoning: Insights from Late Childhood |
| Description | Data_GameRaw_Supplemental_.csv |
| Type Of Material | Database/Collection of data |
| Year Produced | 2024 |
| Provided To Others? | Yes |
| Impact | scientific paper: Grandchamp des Raux, H., Ghilardi, T., Soderberg, C., Ossmy, O. (2024). The role of action concepts in physical reasoning: insights from late childhood. Philosophical Transactions B, 379(1911), 20230154. |
| URL | https://rs.figshare.com/articles/dataset/Data_file_from_The_Role_of_Action_Concepts_in_Physical_Reas... |
| Description | A European consortium to determine how complex, real-world environments influence brain development |
| Organisation | University of East London |
| Country | United Kingdom |
| Sector | Academic/University |
| PI Contribution | This consortium is dedicated to pioneering innovative methodologies and techniques for examining human development in naturalistic contexts. Building on the foundation of our research in this award, the postdoctoral researcher supported by the grant and I are creating specialized resources for fellow researchers to use toy-based methods to investigate how young children manipulate objects and develop problem-solving skills at their homes and nurseries. By incorporating everyday play scenarios into research protocols, these resources aim to capture more realistic expressions of children's cognitive and motor development. Our goal is to make these tools widely accessible, thereby promoting robust, ecologically valid studies of child development and supporting a deeper, more nuanced understanding of early learning processes. |
| Collaborator Contribution | Other consortium partners are applying similar methodological approaches to additional domains of child development, including language acquisition, perceptual processes, social interaction, and neural recording. By integrating these diverse areas of expertise, the consortium aims to generate a comprehensive picture of early development and uncover how different developmental processes interconnect and influence one another. Through this interdisciplinary collaboration, we are not only refining tools for studying specific skill sets in children but also identifying broader patterns and principles that cut across multiple facets of human growth and learning. |
| Impact | This is currently work in progress. We will submit several paper soon. |
| Start Year | 2024 |
| Description | Appearance in BBC documentary - Brain Hacks |
| Form Of Engagement Activity | A broadcast e.g. TV/radio/film/podcast (other than news/press) |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Public/other audiences |
| Results and Impact | Science journalist Melissa Hogenboom created a documentary to set out to understand more about the brain's capacity to respond to change, helping humans to learn and to heal. She looked at the most cutting-edge scientific research and has her own brain scanned and analysed, with intriguing results. As part of this documentary, Melissa interviewed Ori Ossmy, the PI on this grant who talked about his research, including the one that is covered in the ESRC award. |
| Year(s) Of Engagement Activity | 2023 |
| URL | https://www.bbc.co.uk/programmes/m001qr3k |
| Description | Leading a workshop on Naturalistic Experimentation of Child Development |
| Form Of Engagement Activity | A formal working group, expert panel or dialogue |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Postgraduate students |
| Results and Impact | As part of the methodologies developed in this award thus far, PI Ori Ossmy led a workshop on testing children in naturalistic experimentation (based on the work on intelligent toys that was done in this award). The workshop was called NECD and it was an international workshop. The overarching goal of NECD is to bring together researchers of child development who are interested in testing children's natural, real-life behaviour and thereby catalysing discoveries about human development. NECD has been created to educate scientists around the world about various techniques to track the natural behaviour of infants and children and how to visualise and analyse their behaviour. The program included talks by renowned experts in the field (including PI Ori Ossmy) who will give lectures on how to design naturalistic experiments, analyse and visualise real-time naturalistic child behaviour, and build state-of-the-art technology to track child development. The workshop was also an opportunity for attendees to discuss their research and present it in posters and flash talks. NECD specifically aimed to empower young scholars from underrepresented groups and regions by providing them with unique skills to integrate perspectives and methods from engineering and computer science into their developmental work. |
| Year(s) Of Engagement Activity | 2023 |
| URL | https://www.cbcd-necd.co.uk/ |