Jointly fitting models to behavioural and neurophysiological decision-related signals, in principle, enables greater scope for model complexity and richer inferences than can be achieved using behavioural data alone. In recent work, constraining the starting points of evidence-accumulating decision variables to match motor preparation levels reflected in mu/beta EEG activity facilitated the identification of evidence-independent “urgency” contributions to the decision process and drift-rate and non-decision time modulations by speed pressure, which were misestimated by standard models of behaviour alone (Kelly et al, 2021). However, the modelling approach used group-average data, losing individual-level information, and it lacked a formal weighting of behavioural and neural data, potentially biasing inferences. Hierarchical Bayesian modelling provides a way to fit models to each individual while still pooling information across the group, and to weight the fits according to the relative reliability of each information source. Here we developed and applied such a hierarchical modelling approach to the data of Kelly et al, by using model-simulation to determine the joint-probability of single-trial response-times and Mu-beta EEG activity for correct and incorrect responses from a 2-choice motion-discrimination with Easy, Deadline, and Low-coherence conditions. Our results support the presence of an urgency contribution to decision making, and we found evidence for individual differences in decision-making parameters, including the magnitude of urgency. These findings motivate the need to investigate the potential for distinct adaptation strategies involving multiple parameters (eg. correlations between adaptations of different parameters, across participants).