Quote  

"The tale of the tuna reminds us that biological systems profit profoundly from local environmental structure.  The environment is not best conceived solely as a problem domain to be negotiated.  It is equally, and crucially, a resource to be factored into the solutions.  This simple observation has, as we have seen, some far-reaching consequences.

"First and foremost, we must recognize the brain for what it is.  Ours are not the brains of disembodied spirits conveniently glued into ambulant, corporeal shells of flesh and blood.  Rather, they are essentially the brains of embodied agents capable of creating and exploiting structure in the world.  Conceived as controllers of embodied action, brains will sometimes devote considerable energy not to the direct, one-stop solution of a problem, but to the control and exploitation of environmental structures.  Such structures, molded by an iterated sequence of brain-world interactions, can alter and transform the original problem until it takes a form that can be managed with the limited resources of pattern-completing, neural-network-style cognition.

"Second, we should therefore beware of mistaking the problem-solving profile of the embodied, socially and environmentally embedded mind for that of the basic brain.  Just because humans can do logic and science, we should not assume that the brain contains a full-blown logic engine or that it encodes scientific theories in ways akin to their standard expression in words and sentences.  Instead, both logic and science rely heavily on the use and manipulation of external media, especially the formalisms of language and logic and the capacities of storage, transmission, and refinement provided by cultural institutions and by the use of spoken and written text.  These resources, I have argued, are best seen as alien but complementary to the brain's style of storage and computation.  The brain need not waste its time replicating such capacities.  Rather, it must learn to interface with the external media in ways that maximally exploit their peculiar virtues.

"Third, we must begin to face up to some rather puzzling (dare I say metaphysical?) questions.  For starters, the nature and the bounds of the intelligent agent look increasingly fuzzy.  Gone is the central executive in the brain--the real boss who organizes and integrates the activities of multiple special-purpose subsystems.  And gone is the neat boundary between the thinker (the bodiless intellectual engine) and the thinker's world.  In place of this comforting image we confront a vision of mind as a grab bag of inner agencies whose computation roles are often best described by including aspects of the local environment (both in complex control loops and in a variety of information transformations and manipulations).  In light of all this, it may for some purposes be wise to consider the intelligent system as a spatio-temporally extended process not limited by the tenuous envelope of skin and skull.  Less dramatically, the traditional divisions among perception, cognition, and action look increasingly unhelpful."