Life-like learning in technical artefacts: biochemical vs. neuronal mechanisms

Learning in technical artefacts is based on programming tools, like artificial neural nets. These tools still restrict the individual development of an artefact too much. To allow a proper unrestricted development in changing environments, new principles have to be applied in the construction of technical artefacts. In order to study elementary mechanisms of learning and associated factors, we draw on two biological examples. The study of unicellular biochemical networks can help to detect regularities, which are also present on higher levels of evolution. Imprinting, as a form of individual learning, is based on mechanisms, which show the evolutionary rules of stochasticity, selection and reproducibility on the level of cell tissues. The differences and analogies of biochemical and neural networks show elementary mechanism and principles as well as missing factors for the construction of programming tools in technical artefacts.