Perceptual correlates of massive cortical reorganization

V.S. Ramachandran, D. Rogers-Ramachandran, M. I. Stewart
Science, 1992 Nov 13, 258(5085):1159-60.

Pons et al. (1) found cortical maps to be capable of an unexpectedly large degree of reorganization. After long-term (12 years) deafferentation of one limb in adult primates, the cortical area of the brain corresponding to the limb became responsive to stimuli applied to the lower face region. This finding extended the previously recognized maximum area of cortical reorganization in adult primates from a mediolateral distance of 1 to 2 mm to about 10 to 14 mm. Because cells that originally received information from the arm can later receive information from the face, we wondered whether stimuli applied to the face would be mislocalized to the arm. To explore this, we have studied (2, 3) localization of touch sensations in two human patients after amputation of one upper limb, and in another patient after amputation of one digit.

We applied light touch or deep pressure to different points on the normal body surface. Stimulation of points even remote from the amputation line evoked precisely localized referred sensations in the phantom limb. We could plot "reference fields," small regions of skin surface that evoked referred sensations in specific parts of the phantom limb (for example, the digits). Our main experimental findings may be summarized as follows:

(i) Points (reference fields) were not randomly distributed. There were two clusters, one on the same side of the face as the phantom limb, and one around the line of amputation. Furthermore, there was precise one-to-one correspondence between these points and those on the phantom limb. (3). (ii) Sensations were referred most often to the hand, especially to the digits with an overrepresentation of the thumb and "pinky." This may reflect the high cortical magnification of these areas. (iii) The referred sensations were modality-specific; for example, a drop of warm water trickling down the face was felt as "warm water trickling down"in the phantom hand. (iv) Reference fields were somatotopically organized. we suggest that this is a direct consequence of the remapping observed by physiologists (1). (v) There was a vivid persistence of short-term "memory" of complex sensations; when we gripped and released the finger adjacent to the amputated finger the patient felt the phantom finger being "gripped," and this sensation persisted for 7 or 8 seconds in the phantom. (v1) Reorganization was relatively rapid. In one patient, our study was carried out 4 weeks after limb amputation rather than 12 years.

That patients "refer" paresthesiae to a "phantom limb is in itself not new. We have attempted to systematically relate such findings to studies of animal physiology (1, 4). For example, we suggest that the reason we found two clusters of reference fields exhibiting topography-one on the face and the other one near the amputation line-is because the hand area in Penfield's homunculus (in the somatosensory cortex) is flanked on one side by the face and on the other side by areas around the line of amputation (for example the upper arm and shoulder). We would therefore expect sensory input from both these regions to "invade" the cortical hand area and provide a basis for referred sensations.

The very existence of phantom limbs might be partially explained by our hypothesis. If tactile and proprioceptive input from surrounding tissue "takes over" the brain areas corresponding to the amputated limb, spontaneous discharges arising from neurons innervating these tissues would be misinterpreted as arising from the missing limb. This hypothesis is different from, although not incompatible with, the idea that phantom limbs result from the persistence of a "neurosignature" in a diffuse neuronal pool (5). Our observation that the changes can occur as early as 4 weeks after amputation is especially interesting since it suggests that the reorganization is a result of the unmasking of "silent" synapses rather than of anatomical changes such as "sprouting." Perhaps, even in normal adults, input from the face projects simultaneously to both face and hand areas in the cortex or thalamus (and input from the hand to both hand and face areas). The unwanted input to the hand area, however, may be subject to tonic inhibition (for example, through an inhibitory interneuron) by the "correct"axon carrying a signal from the hand. When an arm is amputated, this occult input is unmasked through disinhibition. It remains to be seen whether this unmasking is permanent or whether the patients eventually begin to "ignore" the referred sensations.

Whatever the ultimate interpretation may be, however, our findings suggest that the adult mammalian brain has a latent capacity for much more rapid functional reorganization over a much greater area than previously thought, a capacity that could conceivably be exploited for therapeutic purposes.

1. T.P. Pons et al., Science 252, 1857 (1991).
2. V.S. Ramachandran, M. Stewart, D.C. Rogers-Ramachandran, Neuroreport 3, 583 (1992).
3. _____, Soc. Neurosci. Abstr., in press.
4. M.M. Merzenich et al., J. Comp. Neurol. 224, 591 (1984); P. Wall, Philos Trans. R. Soc. London Ser. B 278, 361 (1977).
5. R. Melzack, Sci Am. 266, 120 (April 1992).
6. We thank patients V.Q., W.K., and D.W. for their cooperation; M. Botte and R. Abrams for referring them to us; M. Parsa and O. Gil for assistance; and the Air Force Office of Scientific Research and the Office of Naval Research for support.