Morphological computation happens when some computational functionality is realized purely through the physical (and chemical) interaction of system components.
In the realm of artificial cell research, morphological computation is realized at a primitive microscopic level through the physico-chemical interaction of molecules. In particular, the amphiphilic molecules that make up cell membranes provide a computationally rich environment because of their ability to self-assemble into a variety of structures, including micelles and vesicles. On a larger length-scale, these components may aggregate into larger structures in a further self-assembly process.
The programming of these self-assembly processes is highly nontrivial, since the system is complex, in the sense that the structures that emerge from the self-assembly process can typically not be predicted a priori from knowledge of the components.
PACE investigated two forms of morphological computing through self-assembly:
- self assembly and movement of micelles, vesicles, and oil droplets. In this area, evolutionary optimization techniques were used, as described elsewhere in these pages.
- assembly of vesicles into larger-scale structures, also described elsewhere.
A third form of morphological computing was also investigated, through the complementation of artificial cell functionality by technology:
- the microfluidic technology provides its own physical embodiment, which must interact with other molecular and self-assembled structures.
The level of computation achieved was typically quite primitive, typically switching between a small handful of target morphologies. However, the techniques developed will be instrumental in developing more sophisticated programmability of complex and living matter.