Understanding the neural networks in working memory

Submission note: A thesis submitted in total fulfilment of the requirements for the degree of Master of Science (Research) to the School of Psychological Science, Faculty of Science, Technology and Engineering, La Trobe University, Bundoora.

Understanding working memory (WM) as a fundamental cognitive process has showed constant evolution over the last five decades. However, while there is a consensus over the definition of WM as a short-term retention of information, involving decision making and goal-based behaviour, controversy remains on the temporal sequence and location of neural processes of WM. Therefore, this thesis aimed to update a 10 year old neuroanatomical meta-analysis of a common measure of WM, the n-back task, utilizing newly updated specialized software, stricter inclusion criteria and more recent studies. Fifty -seven of 316 papers reviewed fitted the inclusion criteria. Results also revealed a similar fronto-parietal network to that previously described, with greater bilateral parietal involvement and left parietal activation consistently across divisions of task demands, with less hemispheric lateralization dependent on task type, and more sites activated outside the fronto-parietal network. The second study aimed to compare maintenance of visuo-spatial stimuli via neuroimaging of an n-back task utilizing a 1-back design with 3D cube shapes, with maintenance plus manipulation via mental rotation of the shapes. Sixteen healthy adults (9 females; M = 23.94 years, SD = 2.49) performed the task while fMRI scans were acquired. When rotation was not required, right fronto-parietal areas were activated, consistent with previous research. However, when the task involved mental rotation, activation was predominantly in bilateral parietal lobule and left fusiform gyrus. This highlights that WM incorporates more areas as needed by task demands, emphasizing the importance of the parietal contribution to WM. In conclusion, there is a need to move away from traditional functional segregation of cognitive tasks, towards a network inclusion view of cognition.