Aristotle's illusion in Parkinson's disease: evidence for normal interdigit tactile perception

Abstract

Sensory alterations, a common feature of such movement disorders as Parkinson's disease (PD) and dystonia, could emerge as epiphenomena of basal ganglia dysfunction. Recently, we found a selective reduction of tactile perception (Aristotle's illusion, the illusory doubling sensation of one object when touched with crossed fingers) in the affected hand of patients with focal hand dystonia. This suggests that reduced tactile illusion might be a specific feature of this type of dystonia and could be due to abnormal somatosensory cortical activation. The aim of the current study was to investigate whether Aristotle's illusion is reduced in the affected hand of patients with PD. We tested 15 PD patients, in whom motor symptoms were mainly localised to one side of the body, and 15 healthy controls. Three pairs of fingers were tested in crossed (evoking the illusion) or parallel position (not evoking the illusion). A sphere was placed in the contact point between the two fingers and the blindfolded participants had to say whether they felt one or two stimuli. Stimuli were applied on the affected and less or unaffected side of the PD patients. We found no difference in illusory perception between the PD patients and the controls, nor between the more affected and less/unaffected side, suggesting that Aristotle's illusion is preserved in PD. The retained tactile illusion in PD and its reduction in focal hand dystonia suggest that the basal ganglia, which are dysfunctional in both PD and dystonia, may not be causally involved in this function. Instead, the level of activation between digits in the somatosensory cortex may be more directly involved. Finally, the similar percentage of illusion in the more affected and less or unaffected body sides indicates that the illusory perception is not influenced by the presence or amount of motor symptoms.

Figure 1. Experimental and control conditions.

The experimenter set and maintained the correct position of the subject’s fingers during stimulation. A) Experimental condition: the digits were crossed and a sphere was placed in the contact point between them. This condition is usually associated with Aristotle’s illusion. B) Control condition: the digits were placed parallel and a sphere was placed in the contact point between them. This condition is usually associated with correct perception of an object. C) The additional control conditions for the crossed (left) and parallel (right) finger positions. In the first, the fingers were crossed and one sphere was placed on one finger, near the contact point. In this way it was possible to exclude an association between the crossed position and the “two-stimuli” response. In the second condition, the fingers were parallel and two spheres were simultaneously placed on them. This condition avoided the association between the parallel position and the “one-stimulus” response.

Figure 2. Aristotle’s illusion in patients with Parkinson’s disease (PD) versus healthy controls.

The columns represent the mean percentage of illusion perceived in all the tested digit-pairs of both the affected and the less/unaffected hand. The bars represent the standard error. In the crossed position (A) the patients perceived the illusion like the controls. There were no differences between the two groups also in the parallel finger position (B). Hence, contrary to other aspects of somatosensory perception, Aristotle’s illusion is preserved in PD.

Figure 3. Additional control conditions in PD patients and healthy controls.

The columns represent the mean percentage of correct response averaged across all the digit pairs for the affected and the less/unaffected hand. Control1 refers to the control condition with crossed fingers and one sphere on one fingertip, Control2 refers to the control condition with parallel fingers and two spheres simultaneously placed on the two fingertips. The bars represent the standard error. The two groups showed high and comparable number of correct responses in each additional control condition.