a Digital domain: The user inputs the actions to perform via a 5-by-5 matrix. This digital matrix is read by a computer and translated into machine code to mechanically actuate the stirrers at the user-defined speed. b Chemical domain: Based on the speed at which the stirrers rotate, the BZ reaction will oscillate between its two states, and oscillating waves will appear in the platform. If the arena does not contain physical boundaries between the cells, the oscillations will not localize and they would spread. If there are physical boundaries between the cells, the oscillations will localize at the specific cells within the arena. c Read-out: Using a camera and image processing, the states of the chemical processor are read by a digital computer. For every cell, the image processing algorithm will classify its state as on or off, depending on if the BZ reaction is oscillating on it. This step will return a 5-by-5 digital matrix representing the state of the system at each cell. d The Input States plot shows the scaling of the number of input states with the number of BZ cells on the experimental platform at different PWM stirring inputs (knxn, where k can be 2, 3 and 4). The Chemical States plot shows scaling of the number of chemical states (defined by the distinct measurable amplitudes of BZ oscillations; see Supplementary Information) with the number of BZ cells on the experimental platform at a different number of measurable oscillation amplitudes ((p + 1)nxn where p describes the different detectable oscillation amplitudes and an additional one for no oscillation). e Evolving microstates: The cells in the arena are weakly connected; therefore, their oscillations will convolve, and be able to perform complex calculations by controlling the stirring speeds into (i) active cells—fast stirring—for inputs, (ii) process cells—slow stirring, (iii) inactive cells—no stirring (see Supplementary Information). f Chemical recurrent state: Because the BZ oscillations have memory, the global state of the medium not only depends on the input, but also on the state of previous iterations.