This paper proposes a brain-inspired cognitive architecture that incorporates approximations to the concepts of consciousness, emotion, and imagination. To emulate the empirically established cognitive efficacy of conscious as opposed to unconscious information processing in the mammalian brain, the architecture adopts a model of information flow from global workspace theory. Cognitive functions such as anticipation and planning are realised through internal simulation of interaction with the environment. Action selection, in both actual and internally simulated interaction with the environment, is mediated by affect. An implementation of the architecture is described which is based on weightless neurons and is used to control a simulated robot. 1

This paper presents a large-scale model of the architecture of the mammalian brain, the core circuit of which carries out inner rehearsal of interaction with the environment to realise a form of cognitively mediated action selection. As it alternates between broadcast to and competition between its component neural assemblies, the core circuit exhibits an episodic dynamics suggestive of cortical processing in discrete frames. The implemented architecture is used to control a simulated robot, and a classic experimental paradigm in which rats performed apparently goal-directed action selection is emulated. 1

(Previously ‘Representation Emergence through Indirect Sensing’) The cognitive significance of representation continues to be the subject of some debate. Some see representation as an integral part of cognition. Others see it as unnecessary or counter-productive, although there is often agreement that representation must be required at the level of reasoning, since by definition this operates on representational tokens. But, using information theory within the framework of Bayesian networks, the paper shows that representation is required not just at the level of reasoning, but also at the level of sensing. The achievement of informational efficiency in a sensory process requires use of representational constructs. Though these are nothing like the disembodied world models of representationalist tradition, they do have all the properties needed for representation at the cognitive level. The indication is then that sensing and reasoning have similar representational requirements. The implications this has for the representation debate are considered.

www.christhornton.eu

In previous papers we presented a robot cognitive architecture organized in three computational areas. The subconceptual area is concerned with the processing of data coming from the sensors. In the linguistic area representation and processing are based on a logic-oriented formalism. The conceptual area is intermediate between the subconceptual and the linguistic areas and it is based on the notion of conceptual spaces. The robot, starting from the 3D information stored in the conceptual area and from the data coming form sensors and processed by the subconceptual area, is able to build a 2D viewer dependent reconstruction of the scene it is perceiving. This 2D model corresponds to what the robot is seeing at any given time. We suggest that the conceptual and the linguistic areas are at the basis of the robot artificial qualia.

The rationality of individual agents is secured for the most part by their make-up or design. Some agents, however – in particular, human beings – rely on the intentional exercise of thinking or reasoning in order to promote their rationality further; this is the activity that is classically exemplified in Rodin’s sculpture of Le Penseur. Do group agents have to rely on reasoning in order to maintain a rational profile? Recent results in the theory of judgment aggregation show that under a range of plausible conditions they do. In a slogan: group agents are made, not born.