Using a neurocomputational autobiographical memory model to study memory loss

Autobiographical memory (AM) is a core component of human life and plays an important role in self-identification. Various conceptual models have been proposed to explain its functionalities and describe its dynamics. However, most existing computational AM models do not distinguish AM from other long-term memory. Specifically, during model design, the unique features and the memory encoding, storage, and retrieval procedures of AM were not taken into consideration in prior models. In this chapter, we introduce our neurocomputational AM model, which is consistent with Conway and Pleydell-Pearce’s model in terms of both the network structure and dynamics.We further propose how to apply our parameterized computational model to quantitatively study memory loss in people of different age groups. As such, we provide a suitable tool to evaluate the effect of different memory loss phases in a rapid and quantitative manner, which may be difficult or impossible in experimental studies on human subjects.