In this paper we will argue for a precise interpretation of consciousness —based
on controller mechanics— that renders it not only not epiphenomenal but fully
functional. Even more, this interpretation leads to the conclusion that conscious-
ness necessarily emerges from certain, not excessively complex, circumstances in
the dwelling of cognitive agents.
A characterisation of cognitive control will be needed as a base support of this
argument; and from this initial relatively simple setting, the unavoidable arrow of
evolution will render entities that are not only conscious but also necessarily self-
conscious.
This analysis will provide a stance for the analysis of the phenomenon of con-
sciousness in cognitive agents that is full in-line with fashionable buzzwords like
situatedness and embodiment.
In the case of technical systems, evolutionary pressure also operates in their evo-
lution. Not at the level of individual machines but at the human-mediated level of
product lines and product families (individual machines generally lacking the nec-
essary replicatory capacities for selfish gene evolution). This implies that, sooner
or later, if the initial conditions hold in this context, consciousness will be a neces-
sarly appearing trait of sophisticated machines. This is where we are: identifying
the core mechanics and application constraints for the realisation of consciousness
capabilitites in next generation technical systems. This will imply, necessarily, the
sound characterisation of the expected benefits from making a machine conscious.
2 The Modelling Brain
2.1 The modelling principle
One of the central issues proposed by the research community is the question of
existence of general principles for cognitive systems and of consciousness in particular
Aleksander and Dunmall (2003). These are for example the topics of discussion
formulated by Taylor in a proposal for a special session on ICANN 2007:
• General principles for cognitive systems;
• The pros and cons of embodiment for cognitive systems;
• The desirability or otherwise of guidance from the brain;
• Specific cognitive system designs and their powers;
• Embodied cognitive systems in robot platforms and demonstrations;
• The best future pathways for development of cognitive systems;
ASLab.org / Principles for Consciousness / A-2007-011 v 1.0 Final