Learning and decision making are interactive processes, yet cognitive modelling of error-driven learning and decision making have largely evolved separately. Recently, evidence accumulation models (EAMs) of decision making and reinforcement learning (RL) models of error-driven learning have been combined into joint RL-EAMs that can in principle address these interactions. However, we show that the most commonly used combination, based on the diffusion decision model (DDM) for binary choice, consistently fails to capture crucial aspects of response times observed during reinforcement learning. We propose a new RL-EAM based on an advantage racing diffusion (ARD) framework for choices among two or more options that not only addresses this problem but captures stimulus difficulty, speed-accuracy trade-off, and stimulus-response-mapping reversal effects. The RL-ARD avoids fundamental limitations imposed by the DDM on addressing effects of absolute values of choices, as well as extensions beyond binary choice, and provides a computationally tractable basis for wider applications.

Working memory (WM)-based decision making depends on a number of cognitive control processes that control the flow of information into and out of WM and ensure that only relevant information is held active in WM’s limited-capacity store. Although necessary for successful decision making, recent work has shown that these control processes impose performance costs on both the speed and accuracy of WM-based decisions. Using the reference-back task as a benchmark measure of WM control, we conducted evidence accumulation modeling to test several competing explanations for six benchmark empirical performance costs. Costs were driven by a combination of processes, running outside of the decision stage (longer non-decision time) and showing the inhibition of the prepotent response (lower drift rates) in trials requiring WM control. Individuals also set more cautious response thresholds when expecting to update WM with new information versus maintain existing information. We discuss the promise of this approach for understanding cognitive control in WM-based decision making.

Joint modelling of behaviour and neural activation poses the potential to provide significant advances in linking brain and behaviour. However, methods of joint modelling have been limited by difficulties in estimation, often due to high dimensionality and simultaneous estimation challenges. In the current article, we propose a method of model estimation which draws on current state-of-the-art Bayesian hierarchical modelling techniques and uses factor analysis as a means of dimensionality reduction to provide further information on which to make inference. The method uses a particle metropolis within Gibbs sampler (PMwG; Gunawan, Hawkins, Tran, Kohn, & Brown, 2020), where the factor structure is estimated within the Gibbs step for the group level. We show the significant dimensionality reduction gained by factor analysis in the Gibbs step of the PMwG, evidence for parameter recovery, a variety of factor loading constraints which can be used for different purposes and research questions, as well as two applications of the method to previously analysed data. This method represents a flexible and usable approach with interpretable outcomes, which relies on data driven analysis as opposed to hypothesis driven methods often used in joint modelling. Although we focus on joint modelling methods, this model based estimation approach could be used for any high dimensional modelling problem. We provide open source code and accompanying tutorial documentation to make the method accessible to any researchers.

No abstract