LCD-CNRS-Paris 5 and INSERM 9935, Hopital Robert Debré, Paris, France

We have presently no definite idea on how information is coded in neural networks, and we have only a poor knowledge of how the various neurotransmitters and neuromodulators as well as glial cells shape the activity patterns of neural networks. Nevertheless, we can agree, as prudently summarized by Kolb in the beginning of his book on Brain plasticity and behavior (1995), that structural properties of the brain are important in understanding its functions, that behavioural and mental states correspond to brain states, that mental activity, perception, and action arise from the activity pattern of populations of neurones, and that plasticity is a property of the synapse. The pattern of activity of a neuronal network has a “meaning” for other neuronal networks. Presently it is only possible to interpret the activity of a neurone in terms of meaningful aspects of the environment of the brain (that is in terms of stimuli coming from the environment). But it is becoming possible to interpret the activity of a cortical neurone in terms of “meaningful” aspects of its neuronal environment (see, for instance, Oram, Földiak, Perrett, & Sengpiel, 1998; Singer, 1995).