Experimental and clinical data indicate that the cerebellum is involved in the pathophysiology of advanced stages of essential tremor (ET). The aim of this study was to determine whether a dysfunction also affects cerebellar structures involved in eye movement control. Eye movements of 14 patients with ET and 11 age‐matched control subjects were recorded using the scleral search‐coil technique. Vestibular function was assessed by electro‐oculography. Eight ET patients had clinical evidence of intention tremor (ETIT); six had a predominantly postural tremor (ETPT) without intention tremor. ET patients showed two major deficits that may indicate cerebellar dysfunction: (i) an impaired smooth pursuit initiation; and (ii) pathological suppression of the vestibulo‐ocular reflex (VOR) time constant by head tilts (‘otolith dumping’). In the step ramp smooth pursuit paradigm, the initial eye acceleration in the first 60 ms of pursuit generation was significantly reduced in ET patients, particularly in ETIT patients, by ∼ 44% (mean 23.4°/s2) compared with that of control subjects (mean 41.3°/s2). Subsequent steady‐state pursuit velocity and sinusoidal pursuit gain (e.g. 0.4 Hz: 0.90 versus 0.78) were also significantly decreased in ET patients, whereas pursuit latency was unaffected. The intention tremor score correlated with the pursuit deficit, e.g. ETIT patients were significantly more affected than ETPT patients. Gain and time constant (τ) of horizontal VOR were normal, but suppression of the VOR time constant by head tilt (‘otolith dumping’) was pathological in 41% of ET patients, particularly in ETIT patients. Saccades and gaze‐holding function were not impaired. The deficit of pursuit initiation, its correlation with the intensity of intention tremor, and the pathological VOR dumping provide additional evidence of a cerebellar dysfunction in the advanced stage of ET, when intention tremor becomes part of the clinical symptoms, and point to a common pathomechanism. The oculomotor deficits may indicate an impairment of the caudal vermis in ET.
For a development of truly user-centered interfaces we need to take into account not only generic characteristics of human beings but also actual dynamics of attention and intentions of persons involved in an interaction. Modern eyetracking methods are indispensable tools in such a development, as they allow the use of eye movement data for control of output devices, for gaze-contingent image processing and for desambiguation of verbal as well as nonverbal information. The main obstacle on the way to these applications is the so-called “Midas touch problem”: how to differentiate “attentive” saccades with intended goal of communication from the lower level eye movements that are just random or provoked by external stimulation? We report results of our investigations of the problem and present a solution based on a functional classification of fixations correlated with their duration. Several additional solutions are also considered together with the data on the trainability of the human oculomotor system.
It is controversial whether opsoclonus is a cerebellar or brainstem disorder. Two patients whose opsoclonus largely disappeared on eye closure underwent fMRI. A comparison of these two states revealed neither vermal nor brainstem activation but rather a bilateral activation in the deep cerebellar nuclei in excess of what the authors found in healthy subjects. The results support a crucial role of the fastigial nucleus in opsoclonus.