Memory-based cognitive framework: A low-level association approach to cognitive architectures

Abstract

At its most fundamental, cognition as displayed by biological agents (such as humans) may be described as being the manipulation and utilisation of memory. A low-level approach to the associative sensory-motor development of cognition is then appropriate, rather than the more common higher-level functional approach. A novel theoretical framework - the memory-based cognitive framework (MBCF) - is proposed based upon these considerations. A computational architecture based on the MBCF is implemented on a mobile robot platform, and experimental results are presented to demonstrate the functionality of the architecture. It is shown that this low-level, bottom-up, approach can produce adaptive behaviours, which may ultimately form the foundation of cognitively flexible agents. © 2011 Springer-Verlag.