Requirements for a human-like information processing architecture that builds itself by interacting with a rich environment

Aaron Sloman, School of Computer Science The University of Birmingham, England,

Competences of humans that are not usually studied in combination suggest requirements to be met by (a) a human-like robot that develops through interacting with a rich and complex 3-D world and (b) an explanatory theory of how humans (and similar animals) do what they do.

Consideration of a space of niches and designs for different sorts of animals and machines reveals nature/nurture tradeoffs, and indicates hard problems that AI researchers, psychologists and neuroscientists have not addressed, e.g. why a robot or animal that learns through play and exploration needs competences (e.g. perception of and reasoning about affordances) that also seem to underlie human abilities to do mathematics (especially geometry and topology). Perhaps such capabilities are a side-effect of evolutionary processes meeting biological needs in a complex, changing 3-D environment.

The architecture required seems to consist of a complex network of dynamical systems of different sorts that grows itself. E.g.