The development and application of quantitative approaches to investigate spatial processing improvement and cognitive control

This thesis uses quantitative approaches to process behavioural and neural data in order to understand spatial cognition learning and cognitive control. Quantitative measurement was used to clearly identify two distinct strategies for improvement in the mental rotation task, one a departure from mental transformation, the other improvement of mental rotation. Using data from from an experiment on learning in mental rotation, a quantitative model of mental rotation was developed. The model was able to account for the RT distribution and error rates using an LBA decision model and a scale adjusted gamma distribution to account for rotation time. The following two chapters apply a modified version of a previously established signal processing technique to model the change in cued task-switching ERPs as a function of RT. Using this approach we modeled a switch-specific ERP component that increases with RT prior to target onset, providing evidence for switch-specific proactive control. We then used the same approach to investigate how interference following target onset is dealt with, reporting ERPs that suggest reactive control is actively used to resolve both target conflict and cue related processing. The final chapter extends the modeling approach used in the previous two chapters, by making modifications to the algorithm. This new method was evaluated on a simulated dataset, and then applied to neural data from the mental rotation experiment to demonstrate its utility. Although results were encouraging, more testing and development is necessary to optimise this new technique.