Active shape discrimination with physical reservoir computers

We present the first example of ‘minimally cognitive’ sensorimotor behaviour arising from a body as physical reservoir. By revisiting an experiment introduced by Beer (1996) and replacing the continuous-time recurrent neural network (CTRNN) therein with networks of mass-spring-dampers we demonstrate that bodies may be exploited for more than control and pattern generation and take over some tasks which were previously thought to require a central nervous system.