Studying Transfer of Learning using a Brain-Inspired Spiking Neural Network in the Context of Learning a New Programming Language

Abstract

Transfer of learning (TL) has been an important research area for scholars, educators, and cognitive psychologists for over a century. However, it is not yet understood why applying existing knowledge and skills in a new context does not always follow expectations, and how to facilitate the activation of prior knowledge to enable TL. This research uses cognitive load theory (CLT) and a neuroscience approach in order to investigate the relationship between cognitive load and prior knowledge in the context of learning a new programming language. According to CLT, reducing cognitive load improves memory performance and may lead to better retention and transfer performance. A number of different frequency-based features of EEG data may be used for measuring cognitive load. This study focuses on analysing spatio-temporal brain data (STBD) gathered experimentally using an EEG device. An SNN based computational architecture, NeuCube, was used to create a brain-like computation model and visualise the neural connectivity and spike activity patterns formed when an individual is learning a new programming language. The results indicate that cognitive load and the associated Theta and Alpha band frequencies can be used as a measure of the TL process and, more specifically, that the neuronal connectivity and spike activity patterns visualised in the NeuCube model can be interpreted with reference to the brain activities associated with the TL process.