Modelling the Dynamic Development of Expertise: Integrating Practice, Ability, and Individual Differences

About the Project

Understanding how human expertise develops remains a central question across psychology, education, and cognitive science. Competing theoretical perspectives have emphasised either the dominant role of deliberate practice or the importance of individual differences such as intelligence. While there is broad agreement that both contribute to skill acquisition, it remains unclear how these factors interact dynamically over time to shape the development and retention of complex skills across the lifespan. In addition, other individual characteristics, including personality and motivation, are likely to influence both learning processes and long-term performance.

This project addresses these questions using a dynamical systems approach to modelling skill development. Rather than focusing solely on average trajectories of improvement, the project emphasises regulation and variability within individuals over time. The central aim is to understand how fluctuations in performance arise, whether they are amplified or dampened, and which factors act as regulatory mechanisms governing these dynamics.

The project will develop and apply dynamic modelling approaches, including autoregressive, cross-lagged, and continuous-time models, to examine how skill evolves within individuals. These models provide a flexible framework for capturing both stable and time-varying influences. In particular, the project will distinguish between time-dependent and time-independent predictors. Time-dependent predictors, such as intensive training periods or changes in practice conditions, will be modelled as event-like inputs that perturb the system and whose effects unfold over time. Time-independent predictors, including cognitive abilities, personality traits, and motivation, will be used to explain between-person differences in the underlying system dynamics, such as rates of change and variability.

Empirically, the project will draw on large-scale longitudinal datasets of skill acquisition, including digital trace data from domains such as gaming or structured training environments. These data provide high-resolution measures of performance and practice over time, enabling fine-grained modelling of learning processes.

Methodologically, the project will involve statistical modelling in R and/or Python, with a focus on continuous-time structural equation models and related approaches. The student will develop, compare, and validate competing models of skill development, and examine how different theoretical assumptions (e.g., practice-driven vs. ability-driven accounts) manifest in observed data.

Overall, this project will advance theoretical understanding of expertise by moving beyond static or trajectory-based accounts toward process-oriented models of development. It will provide a unified framework for integrating practice, ability, and other individual differences, and contribute to a more precise understanding of how complex skills are acquired, maintained, and potentially optimised over the lifespan.

Funding Notes

Self funded or externally sponsored students only. Intakes are usually October and March annually. NB The University has some scholarships under competition each year. More details can be found - View Website

References

Vaci, N., Edelsbrunner, P., Stern, E., Neubauer, A., Bilalić, M., & Grabner, R. H. (2019). The joint influence of intelligence and practice on skill development throughout the life span. Proceedings of the National Academy of Sciences, 116(37), 18363-18369.