Giovanni Pezzulo - "Seeing the world through the lens of a predictive brain: a neuro-computational model"

Recording of the speaker's talk at the Barcelona Brain and Technology Summer School, September 2012. Listen also to an interview with this and other speakers at www.csnetwork.eu/podcast.

"How do we understand the external world and the other agents? We will introduce our ongoing research (using a combined theoretical, empirical and computational approach) on how living organisms (and possibly robots) can understand (1) the external environment, with its affordances, and (2) other people as intentional actors, linking their behavior to (hidden) goals and distal intentions. We assume that all these (epistemic) abilities are intimately associated to the (pragmatic) ability to take intentional actions and to predict their outcomes, and (partially) recruit the same neural and computational mechanisms.

Furthermore, we discuss how these abilities are also important prerequisites fo more advanced cognitive and social skills such as tool use, joint actions, and language understanding. Thus, more generally, the question is: how can living organisms (and possibly robots) develop higher cognition from sensorimotor skills? In keeping with embodied and motor theories of cognition, we propose that the architecture of motor prediction and control of our earlier ancestors was gradually improved to afford cognitive control and executive functions (and in parallel, joint actions and communication in the social domain)."