Will we ever reverse-engineer animal cognition? Alex Kacelnik

Many animals display creative problem-solving abilities beyond those of the most advanced current autonomous robots. This must by necessity be the result of information processing operations in the path from perception to action. Should we be able to identify and emulate the processes that support such performance, our ability to build autonomous robots would enter a new and revolutionary phase, but for the moment we are far from it. Reciprocally, should we be able to build robots capable of performing at least some of the feats displayed by living organisms, we might be able to understand how they (the organisms) do it. Close interaction between comparative cognition and artificial  intelligence is thus a supremely timely multidisciplinary task. As a biologist and comparative cognitivist, I will present examples of problem solving performance that challenge our current capability for algorithmic emulation.