Abstract
A putative tremulous oscillation of the eyes has been proposed to provide the long-sought objective, sensitive and specific diagnostic sign for Parkinson’s disease . Vigorous debate has ensued as to whether this “ocular tremor” is a real phenomenon or an artefact secondary to head motion. Contradictory arguments have centred on the capabilities of magnetic movement measurement technologies, with small numbers of illustrative cases presented to bolster either side. A large, definitive study able to either confirm or reject the original finding on a purely oculographic basis was required. We examined 681 oculomotor recordings from a longitudinal study of 188 Parkinson’s patients and 66 controls. Fourier analysis was used to detect oscilla- tions not only in their raw pupil motion data but also in corneal reflection signals, which can correct for head motion. We replicated the original authors’ findings inasmuch as we detected oscillations in raw pupil position data that often occurred in their reported frequency range of 5.7 ± 3 Hz. They were not, however, present universally in patients and the frequency (if not power) distribution was similar in controls. Crucially, oscillations in that frequency range were abolished when corneal reflection correction was applied to compensate for head-motion. The oscillations in the uncorrected data were strongly related to clinical ratings of somatomotor tremor severity. We found strong evidence that the postulated purely ocular tremor of Parkin- son’s does not exist, other than as an artefact arising from head motion secondary to somatic tremor.