Portal de Periódicos da CAPES

Central pattern generators

2000; Elsevier BV; Volume: 10; Issue: 5 Linguagem: Inglês

10.1016/s0960-9822(00)00367-5

1879-0445

Scott L. Hooper,

Neuroscience and Neuropharmacology Research

Central pattern generators (CPGs) are relatively small, relatively autonomous groups of neurons (neural networks) that produce patterned, rhythmic neural outputs that drive rhythmic behaviours. In addition to generating boring behaviours like walking, CPGs are also responsible for dancing, chewing, swallowing, suckling, copulation and orgasm — all the things that make life worthwhile. Aside from being a source of adolescent humour for electrophysiologists with arrested development, you mean? When first discovered in the 1960s, CPGs were important because they proved that nervous systems can endogenously create output without sensory input; they thus resolved a controversy about whether nervous systems could act only in response to sensory stimulation. Recent work in systems in which all the neurones of a CPG can be identified (such as the pyloric network of the lobster, (Panulirus) interruptus; shown here) has allowed their activity to be explained on the basis of network synaptic connectivity (the network's wiring diagram) and neuronal cellular properties. Actually, no. Work on CPGs has helped to underline the central importance of differing neuronal properties for the function of neural networks. Some CPG neurones are spontaneous oscillators (they repeatedly depolarize and fire bursts of action potentials); others show a delayed excitation in response to being inhibited; others are bistable elements that can be switched between two semi-stable states, one hyperpolarized, the other depolarized and firing. These data suggest that to predict network activity the wiring diagram isn't enough — you also need to know how each neurone transforms the inputs it receives. You might think we have thousands — one for the rumba, one for the tango, one for the waltz, etc. — but that's almost certainly not the case. CPG networks seem to be multifunctional: modulatory input induces them to produce many different outputs, so one network could generate many behaviours. Furthermore, network boundaries are flexible — neurones can switch from one network to another, and networks can be fused into new larger networks that produce patterns different from any of the originals. Well, as it allows each CPG neurone to participate in generating many behaviours, it gives more 'bang' per neurone. If this flexibility exists elsewhere in the nervous system, it would similarly increase the amount of information-processing a given number of neurones could perform. Such multifunctional abilities might, in part, underlie our ability to think and to create new things. Grillner S, et al.: Neural networks that co-ordinate locomotion and body orientation in lamprey.Trends Neurosci 1995, 18:270-279. Marder E, Calabrese RL: Principles of rhythmic motor pattern production.Physiol Rev 1996, 76:687-717. Stein PSG, Grillner S, Selverston AI, Stuart DG: Neurons, Networks and Behavior. Cambridge, USA: MIT Press; 1997.