A NEURAL-NETWORK MODEL OF PHANTOM LIMBS

This paper presents a detailed clinical description of phantom limbs a nd a neuronal network model that provides a comprehensive and parsimon ious explanation of otherwise inexplicable or at least unrelated pheno mena. Simulations of self-organizing feature maps (Kohonen networks) t hat had been trained to recognize input patterns were deprived of part s of their input in order to simulate partial deafferentation. This le ads to reorganization processes that are shown to be driven by input n oise. In patients with an amputated limb, this noise is generated by d orsal root ganglion sensory neurons which are known to fire irregularl y upon laceration. According to this model, the long-standing debate c oncerning non-cortical vs, cortical contributions to the generation of the phenomenon of phantom limbs can be resolved in that it is the per ipherally generated noise that causes cortical reorganization. The mod el can be tested and may have therapeutic implications.