Vergence deficits in patients with cerebellar lesions - Brain

BRAINA JOURNAL OF NEUROLOGY

Vergence deficits in patients with cerebellarlesionsT. Sander,1 A. Sprenger,1 G. Neumann,1 B. Machner,1 S. Gottschalk,2 H. Rambold1 andC. Helmchen1

1 Department of Neurology, University Hospitals Schleswig-Holstein, Campus Lubeck, Ratzeburger Allee 160, Germany

2 Institute of Neuroradiology, University Lubeck, D-23538 Lubeck, Germany

Correspondence to: Thurid Sander, MD,

Department of Neurology,

University Hospitals Schleswig-Holstein,

Campus Lubeck, Ratzeburger Allee 160,

D-23538 Lubeck, Germany

E-mail: [email protected]

The cerebellum is part of the cortico–ponto–cerebellar circuit for conjugate eye movements. Recent animal data suggest an

additional role of the cerebellum for the control of binocular alignment and disconjugate, i.e. vergence eye movements. The

latter is separated into two different components: fast vergence (to step targets) and slow vergence (to ramp and sinusoidal

targets). The aim of this study was to investigate whether circumscribed cerebellar lesions affect these dynamic vergence eye

movements. Disconjugate fast and slow vergence, conjugate smooth pursuit and saccades were binocularly recorded by a scleral

search coil system in 20 patients with acute cerebellar lesions (all ischemic strokes except for one) and 20 age-matched healthy

controls. Patients showed impairment of slow vergence while fast vergence was unaffected. Slow vergence gain to sinusoidal

targets was significantly reduced, both in convergence and divergence direction. Divergence but not convergence velocity to

ramp targets was reduced. Conjugate smooth pursuit eye movements to sinusoidal and to step-ramp targets were impaired.

Patients had saccadic hypometria. All defects were particularly expressed in patients with vermis lesions. In contrast to recent

animal data fast vergence was not impaired in any of our patient subgroups. We conclude that (i) the human cerebellum,

in particular the vermis, is involved in the processing of dynamic vergence eye movements and (ii) cerebellar lesions elicit

dissociable effects on fast and slow vergence.

Keywords: fast vergence; slow vergence; divergence; vermis

Abbreviations: FEF = frontal eye field; FOR = fastigial oculomotor region; IP = posterior interposed nucleus; MST = medial superiortemporal area; NRTP = nucleus reticularis tegmenti pontis; SEF = supplementary eye field; SE = standard error; SPEM = smoothpursuit eye movements.

IntroductionNeurologists usually use eye movements at the bedside as a

sensitive parameter for topodiagnostic localization and pathophy-

siological understanding of involuntary eye movements (Leigh

and Zee, 2006; Ramat et al., 2007). Clinical textbook knowledge

is usually related to conjugate eye movements, i.e. movements

that rotate the eyes simultaneously in the same direction by

about the same amount in frontoparallel planes. Unlike those

conjugate movements, vergence eye movements refer to eye

movements that rotate the eyes simultaneously in the opposite

direction (disconjugate eye movements). Frontal-eyed human

and non-human primates use the vergence system to track

a moving target in sagittal planes or to change the gaze point

in depth. To track a target in 3D space both systems have

to be coordinated. Under natural viewing conditions, shifts of

doi:10.1093/brain/awn306 Brain 2009: 132; 103–115 | 103

Received June 30, 2008. Revised September 22, 2008. Accepted October 21, 2008. Advance Access publication November 26, 2008

� The Author (2008). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved.

For Permissions, please email: [email protected]

Dow

nloaded from https://academ

ic.oup.com/brain/article/132/1/103/290282 by guest on 24 January 2022

the point of fixation usually combine vergence and saccade

movements.

Neurons in the medial superior temporal visual area (MST),

the supplementary eye field (SEF) and the frontal eye field (FEF)

are active during either smooth pursuit or vergence eye move-

ments or their combination (Fukushima et al., 2004; Akao et al.,

2005a, b). MST and the caudal FEF neurons are likely to be

involved in the initiation of vergence eye movements. There

is some evidence that smooth pursuit and vergence eye move-

ments are processed separately in the brainstem (Mays et al.,

1986), and neurons discharging during divergence and con-

vergence are also anatomically separated (Gamlin, 2002). One

downstream pathway from the caudal FEF reaches the nucleus

reticularis tegementi pontis (NRTP) (Suzuki et al., 1999). NRTP

also receives input from superior colliculi and pretectal midbrain

areas and projects to the deep cerebellar nuclei. There is some

evidence from animal experiments that NRTP is involved in the

cortico-ponto-cerebellar circuits for disconjugate eye movements

(vergence) (Gamlin and Clarke, 1995; Gamlin et al., 1996; Gamlin,

2002).

Like conjugate smooth pursuit and saccadic eye movements,

vergence can be distinguished in vergence responses to ramp

(slow vergence) and to step (fast vergence) targets. We have

recently identified impairment of slow vergence in patients with

caudal pontine infarctions involving the NRTP while fast vergence

responses to step targets remained unaffected (Rambold et al.,

2004). In contrast, upper pontine lesions, which spare NRTP,

elicit an additional deficit of fast vergence (Rambold et al.,

2005a). These data provide some evidence that (i) pontine

nuclei are involved in vergence processing and (ii) fast and slow

vergence may be under separate neural control. This is in contrast

to the cortical vergence control which does not seem to show

dissociable activity to slow and fast vergence (Gamlin and Yoon,

2000; Akao et al., 2005b).

For conjugate eye movements (e.g. smooth pursuit) the pontine

nuclei, particularly NRTP, are known to be a crucial precerebellar

relay to the cerebellum, particularly the dorsal vermis (Thielert and

Thier, 1993). Purkinje cells in the vermis discharge during smooth

pursuit and saccadic eye movements in one depth plane (Suzuki

and Keller, 1988; Helmchen and Buttner, 1995; Krauzlis and

Miles, 1998; Barash et al., 1999; Takagi et al., 2000). There

is accumulating evidence that the cerebellum is also involved

in the processing of vergence: (i) cerebellar ablation elicits transi-

ent paralysis of convergence (Westheimer and Blair, 1973);

(ii) patients with degenerative cerebellar disease have impaired

binocular alignment (esotropia during binocular viewing) and dis-

conjugate saccadic dysmetria (Versino et al., 1996); (iii) anatomical

projections of brainstem areas carry vergence signals to the dorsal

vermis and deep cerebellar nuclei (May et al., 1992; Gamlin et al.,

1996) and (iv) vermis is activated in humans during the near

response (Richter et al., 2004). Recently, vergence related neurons

have been recorded in the posterior vermis of monkeys (Nitta

et al., 2008a,b). Accordingly, vermis lesions elicited impairment

of vergence, static alignment (Takagi et al., 2003), and conjugate

smooth pursuit impairment (Nitta et al., 2008a,b). There is some

additional evidence that the deep cerebellar nuclei, i.e. posterior

interposed (IP) (Gamlin et al., 1996; Zhang and Gamlin, 1998;

Gamlin, 1999, 2002) and fastigial nucleus (FOR) (Gamlin and

Zhang, 1996; Zhang and Gamlin, 1996; Gamlin, 1999, 2002)

are involved in vergence.

The aim of our study was to elucidate the role of the cerebellum

in dynamic vergence in patients with acute circumscribed cerebel-

lar lesions. Static misalignments and disconjugate dysmetria

have been studied in cerebellar patients (Versino et al., 1996).

However, no patient study has examined dynamic vergence

yet. We specifically hypothesized that the vermis is implicated

in vergence processing and that vermis lesions impair tracking

in depth. We examined divergence and convergence as well as

slow and fast vergence separately since animal studies and our

previous studies in patients with pontine lesions showed selective

impairment. As fast and slow vergence are age related (Rambold

et al., 2006) we compared patients with age-matched healthy

control subjects.

Methods

SubjectsTwenty patients (mean age: 54 years, range: 26–81 years) participated

in the study. All had an acute cerebellar infarction (53 weeks after

stroke) except for one patient (#13) who had a cavernoma. All sub-

jects gave written informed consent for participation in the study

which was conducted in accordance with the Declaration of Helsinki

and approved by the Ethics Committee of the University of Luebeck.

All patients were clinically examined at the day of recording by two

experienced neuro-ophthalmologists (CH, TS). Symptoms and clinical

signs of patients are listed in Table 1. Patients had no other history

of previous neurological disease affecting the central nervous system.

Most of the patients complained about acute gait unsteadiness and

dizziness. Ataxia of gait and stance was the leading clinical sign.

Neuro-ophthalmological examination revealed cogwheel pursuit in

most of the patients. All of them were able to perform vergence

eye movements to targets slowly moving towards or away from

the patient’s nose. None of them had visible cogwheel slow vergence

eye movements. Fast vergence responses appeared largely unimpaired

in most of the patients. Vision and stereovision (Stereo Optical Co.,

Inc., Chicago, OH, USA) were examined in all patients as well as

the subjective visual vertical. All patients had a visual acuity above

20/25. At the time of eye movement recordings none of the patients

had nystagmus or blurred vision any more; there was no clinical evi-

dence for static misalignment, particularly skew deviation. Binocular

fusion (Bagolini Test) and stereovision were normal (51000 0). All

patients had a high resolution MRI scan of the brain (see below).

Twenty healthy control subjects (mean age: 54 years, range: 26–81

years) without any previous history of neurological symptoms or

diseases participated in the study as controls and were normal on

neurological examination. Neither patients nor healthy subjects were

on medications known to have side-effects on the central nervous

system.

Eye movement recording and stimuliBinocular eye movements were recorded by a scleral search coil system

(Remmel Labs), which has three orthogonal magnetic fields and a

frame size of 180 cm. The two eyes were calibrated using a combined

off-line in vitro and in vivo calibration based on previous studies

104 | Brain 2009: 132; 103–115 T. Sander et al.

Dow



(Rambold et al., 2002b, 2004). Using the inter-ocular distance

(5.8–7.2 cm) the vergence angles of the stimuli were calculated for

each subject individually. The subject’s head was comfortably stabilized

in a natural upright position with a chin rest and the forehead was

kept stationary by a firm head support.

Conjugate eye movements were elicited by a laser stimulus

projected on a flat white tangent screen that was 145 cm from the

subject’s eyes in dark surroundings while vergence stimuli were

produced by a laser stimulus projected onto a horizontal plane 3 cm

below the level of the eyes and aligned in the patient’s midsagittal

plane (Rambold et al., 2002b). Due to the slightly lowered stimulus

platform eye movements were accompanied by small (vertical)

saccades. The laser target (spot diameter: 0.1�, 635 nm) was moved

by mirror galvanometers, precise timing was controlled by a PC.

Experimental paradigmsAs it was our aim to study the effect of cerebellar lesions on vergence

but not saccade–vergence interaction we designed the stimulus to

move either in the midsagittal plane or in the frontoparallel plane in

order to investigate the subsystems separately. All subjects were exam-

ined using the following paradigms: slow and fast vergence, conjugate

smooth pursuit eye movements and horizontal saccades. Due to

the non-linear relation between target distance and target vergence,

we included this non-linearity in the stimulus control in order to elicit

constant vergence angle motion. To prevent anticipatory effects the

duration of fixation before each trial was varied from 1500 to

2500 ms.

Vergence

(i) Slow sinusoidal vergence was elicited by a sinusoidal moving laser

target at 0.12 Hz and 3.6� amplitude (range 4.7–8.3�, peak velocity

1.35�/s). (ii) Slow vergence responses to ramp stimuli of 1.5�/s velo-

city were elicited in divergence and convergence directions (20 repeti-

tions, pseudorandomized); convergence started at 2.8� vergence angle,

divergence at 10.0� vergence angle. (iii) Laser stimuli for divergence

and convergence steps were presented with a vergence angle ampli-

tude of 7� moving between 3.0� vergence angle at distance and 10.0�

vergence angle at nearness.

Conjugate smooth pursuit

Conjugate smooth pursuit eye movements were elicited by a sinusoidal

slowly moving target (0.2 Hz,� 25�amplitude). Smooth pursuit initia-

tion was tested by a step ramp smooth pursuit paradigm (Carl and

Gellman, 1987; Helmchen et al., 2003). Each sequence consisted of

10 foveopetal pursuit step-ramps (3� step, opposite to ramp direction,

20�/s ramp velocity) to either side, presented in a random order.

Conjugate saccadesConjugate horizontal saccades of 10� and 20� amplitude were per-

formed to either side in a pseudorandomized order (four repetitions

Table 1 Clinical data of patients with cerebellar lesions

# Gender Age Symptom Clinical sign

1 f 45 Dizziness, nausea, headache,clumsiness of the left side

Cogwheel smooth pursuit, hemiataxia left

2 f 43 Blurred vision, vertigo, nausea,gait unsteadiness

Downbeat nystagmus, hemiataxia right, gait ataxia,lateropulsion to the right, cogwheel smooth pursuit

3 m 67 Blurred vision, dizziness, nausea, vomiting,dysarthria

Dysarthria, dysphagia, gaze evoked nystagmus, hemiataxia right,cogwheel smooth pursuit, lateropulsion to the right, gait ataxia

4 m 55 Vertigo, nausea, vomiting, gaitunsteadiness,blurred vision

Spontaneous nystagmus to the left, lateropulsion to the left,gait ataxia, cogwheel smooth pursuit

5 m 43 Vertigo, blurred vision, gait unsteadiness,clumsiness of the right side

Hemiataxia right, gait ataxia, lateropulsion to the right

6 m 48 Blurred vision, dizziness, nausea, vomiting Hemiataxia left, gait ataxia

7 m 51 Dizziness, nausea, vomiting Gaze evoked nystagmus, hemiataxia left, lateropulsion to the left,cogwheel smooth pursuit

8 m 63 Dizziness, nausea, gait unsteadiness,clumsiness of the left hand

Gaze evoked nystagmus, dysarthria, hemiataxia left, gait ataxia,lateropulsion to the left

9 m 57 Dizziness, nausea, gait unsteadiness Gait ataxia, lateropulsion to the left

10 m 69 Gait unsteadiness, headache Gaze evoked nystagmus, gait ataxia

11 m 69 Vertigo, nausea, gait unsteadiness,headache

Somnolence, dysarthria, gaze evoked nystagmus, hemiataxia left,lateropulsion to the left

12 f 64 Nausea, vomiting, headache Cogwheel smooth pursuit, gaze evoked nystagmus, hemiataxia left

13 f 51 Dizziness Cogwheel smooth pursuit

14 m 68 Dizziness, clumsiness of the left hand Hemiataxia left, gait ataxia, lateropulsion to the left

15 m 50 Vertigo, nausea, gait unsteadiness,clumsiness of the left side

Gaze evoked nystagmus, dysarthria, hemiataxia left, gait ataxia,lateropulsion to the left

16 f 26 Vertigo, nausea, vomiting, dysarthria,dysphagia

Cogwhell smooth pursuit, gait ataxia

17 m 59 Dizziness, headache Gaze evoked nystagmus, hemiataxia right, lateropulsion to theright, gait ataxia, cogwheel smooth pursuit

18 f 29 Dizziness, nausea, vomiting,gait unsteadiness

Gaze evoked nystagmus, lateropulsion to the right,cogwheel smooth pursuit

19 m 81 Blurred vision, dizziness Cogwheel smooth pursuit

20 m 77 Dizziness, gait unsteadiness Hemiataxia right, gait ataxia, lateropulsion to the right

Vergence and cerebellum Brain 2009: 132; 103–115 | 105

Dow



per direction and each amplitude), starting from gaze straight ahead,

i.e. 10 centrifugally and 10 centripetally.

Data analysisEye movements were recorded unfiltered by a 16-bit AD converter

(NI PCI 6033E) at a sampling rate of 600 Hz. After calibration all

position data were filtered by using a 100 Hz (3-dB value) Gaussian

filter (Matlab, The Math-Works, Natick MA, USA). Horizontal disparity

vergence was calculated as left minus right horizontal eye position,

horizontal version eye position was calculated as cyclopean eye

(average of left and right eye position). Positive values indicate right-

ward, upward, or convergence movement direction and negative

values indicate leftward, downward, or divergence movement direction

of the eyes. Slow eye movements (sinusoidal vergence and version,

ramp stimuli) were analysed using de-saccaded and linearly interpo-

lated eye velocity signals.

Analysis of sinusoidal vergence was performed after applying a

20 sample median filter and a 30 Hz Gaussian filter. Vergence gain

was calculated as the ratio of best fitted vergence velocity to target

velocity.

Slow vergence to ramp stimuli were analysed by using mean eye

velocity values (filtered by 30 sample median filter and 20 Hz Gaussian

filter) for each direction; in order to detect peak acceleration multiple

linear regressions were calculated consecutively for 600 ms beginning

at ramp onset. The slope of the regression with the highest slope and

lowest Euclidian distance was defined as peak acceleration. The inter-

section of the regression line with zero was defined as latency. The

analysis was controlled and adjusted interactively if required. Mean

vergence velocity was quantified over the period of 600–4000 ms.

For the ramp stimuli means of individual responses were derived

from averaged data (control: 13 trials, median, range 10–20; patients:

14 trials, median, range 9–20).

Fast vergence to step target trials were analysed by using filtered

vergence velocity (four sample median filter, 70 Hz Gaussian filter) on

a trial by trial basis similar to the saccade analysis (see below).

Conjugate smooth pursuit eye movements were analysed offline

using an interactive program. Eye position data were low pass filtered

by a 50 Hz Gaussian filter. Pursuit eye velocity was calculated by

differentiating the mean eye position of the eight data points (equiva-

lent 13.3 ms) before and the eight data points (equivalent 13.3 ms)

after that given data point. Sinusoidal smooth pursuit velocity gain

was calculated similar to sinusoidal vergence velocity. Step-ramp

latency, initial acceleration and steady state velocity (interval 400–

800 ms) were analysed as reported previously (Carl and Gellman,

1987; Helmchen et al., 2003). The gain was calculated by the ratio

of eye velocity to target velocity.

Saccades were filtered by a 100 Hz Gaussian filter (–3 dB) and pre-

detected automatically using velocity threshold of 30�/s and subse-

quently searching for peak velocity in a time window of 50 ms.

Begin and end of saccade were detected at the level of 20�/s.

All computerized saccade detections were controlled and adjusted

manually if required.

Mean data are given with standard error unless otherwise stated.

Dependent variables were analysed by analysis of variance (ANOVA),

including a within-subject factor direction (convergence, divergence)

and a between-subject factor group (patient, control). For post-hoc

comparison of two groups or conditions Student’s t test was per-

formed. Statistics was applied using SPSS package (SPSS 15.0.1,

SPSS Inc., Chicago, IL, USA), statistical significance was considered

at P-values 50.05.

SubjectsWe compared mean data of oculomotor parameters between cerebel-

lar patients and controls. Based on our specific a priori hypothesis we

also analysed patients with (n = 11) versus those without (n = 9) vermis

lesions. Due to the low number of patients with floccular lesions (n = 3)

statistical comparison was not possible. There was a large overlap

of patients with vermis and deep cerebellar nuclei lesions (n = 9).

However, seven patients had vermis sparing lesions of the deep cere-

bellar nuclei, which were compared to controls.

MRI scanning and anatomicalreconstructionHigh-resolution MRI was performed with an MRI unit (1.5-T Siemens

Magnetom Symphony; Siemens, Erlangen, Germany) using a contrast-

enhanced T1-weighted spin-echo and a T2-weighted turbospin-echo

sequence with a slice thickness of 3 mm.

Vascular supply assignment was conducted according to pre-

viously validated MR anatomic templates (Tatu et al., 1996). Lesions

were systematically analysed by one experienced neuroradiologist

irrespective of and blinded for the clinical data. They were recon-

structed according to the appropriate axial and sagittal sections

of the MRI atlas of the human cerebellum (Schmahmann et al.,

2000) (Fig. 1). Analysis was focused on specific regions (e.g. flocculus,

uvula/nodulus, deep cerebellar nuclei and posterior vermis, see

Table 2) with respect to regions potentially involved in the control

of vergence.

Results

Lesion dataIschemic strokes were confined to the cerebellum. Sixteen out of

20 patients had an infarction in the territory of the posterior infer-

ior cerebellar artery (PICA; 11 left-sided, 5 right-sided lesions), five

patients suffered from an infarction in the territory of the superior

cerebellar artery (SCA, four right-sided, one left-sided); one (#5)

of these patients had a combined unilateral SCA and PICA infarc-

tion, another patient bilateral infarctions (PICA, SCA, #4, Table 2).

Affected regions of interest with respect to the control of smooth

pursuit and vergence eye movements are listed in Table 2. Eleven

patients had lesions involving the vermis while nine patients

had vermis sparing lesions. Sixteen patients showed lesions of

the deep cerebellar nuclei, nine of them combined with vermis

lesions and seven with vermis sparing lesion of the deep cerebellar

nuclei.

Eye movementsEye movement data will be presented in the sequence of the

following comparisons: (i) patients versus healthy control subjects,

(ii) subgroup analysis of patients with vermis versus vermis sparing

cerebellar lesions, (iii) patients with vermis sparing lesions of the

deep cerebellar nuclei versus healthy controls.


Dow



Comparison of patients versus healthycontrol subjects

Disconjugate eye movements

Slow vergence

In sinusoidal vergence tasks patients showed significantly smaller

slow vergence gain (0.76� 0.03, T(27) = 3.33, P50.01) than

age-matched control subjects (0.89� 0.03) (Fig. 2, Table 3).

There was no difference between the phase shift of patients and

controls (controls = 5.97� 0.94, patients = 8.77� 2.00, T(27) =

�1.12, P = 0.273) during sinusoidal vergence responses. Slow ver-

gence gain was calculated for each direction separately. Analysis

of variance with between-subject factor GROUP and within-

subject factor DIRECTION revealed a main effect of DIRECTION

[F(1,27) = 5.19, P = 0.031] but no interaction of GROUP�

DIRECTION. A post-hoc t test showed no directional gain asym-

metry in patients (gain convergence: 0.77� 0.03, gain divergence:

0.75� 0.03, P = 0.49). For illustration, Fig. 2 shows original

recordings of sinusoidal slow vergence eye movements of one

patient (A, #17) and a control subject (B).

In the ramp paradigm analysis of variance (ANOVA) on mean

vergence velocity revealed no effect of DIRECTION but a

significant interaction of DIRECTION�GROUP [F(1,26) = 4.39,

P50.05]. Post-hoc t tests showed a reduced mean vergence velo-

city in divergence (1.33� 0.05�/s, T(27) = 2.90, P50.01) but not

in convergence direction (1.51� 0.06�/s), when compared with

the control subjects (divergence: 1.55� 0.06�/s, convergence:

1.55� 0.05�/s) (Fig. 3). There was some variability of vergence

velocities (even in healthy controls) but most individual patients

had reduced divergence (but not convergence) velocity compared

to the control subjects. Even when comparing individual velocities

for convergence and divergence there was a consistent reduction

of divergence in most patients.

Analysis of variance for slow vergence latency and acceleration

of DIRECTION (convergence, divergence) and between subject

factor GROUP (patient, control) showed a significant main effect

for DIRECTION (P always50.001) but no interaction. Acceleration

Fig. 1 Topographic distribution of isolated cerebellar infarctions in the territory of either the superior cerebellar artery (SCA, #1–5) or

the posterior inferior cerebellar artery (PICA, #4–12 and #14–20) and the lesion of one cavernoma (#13) according to MRI recon-

struction (Tatu et al., 1996). On the left side section levels of the anatomical reconstruction are shown on axial slices, either at the level

of the upper pons (SCA) or at the level of the pontomedullary junction (PICA). The extent of the lesions is demonstrated in grey. For

the sagittal reconstruction a slightly paramedian slice was chosen. Patient #4 and 5 had combined lesions (#4: SCA right and PICA left,

#5: SCA and PICA left). Note that due to the selected section levels some lesions are not represented in both the axial and the sagittal

slice.


Dow



of divergence (div.) responses to ramp stimuli were smaller

compared to convergence (conv.), both in controls (div.:

10.42� 0.87�/s2; conv.: 13.38� 0.91�/s2) and patients (div.:

9.47� 0.81�/s2; conv.: 14.23� 1.34�/s2). However, there was

no difference between controls and patients in any direction.

There was also no latency difference between controls and

patients, neither in convergence nor in divergence (Table 3).

Fast vergence

Group comparison of vergence responses to step targets (fast

vergence) did not show significant differences between patients

and control subjects with respect to latency, gain, peak velocity,

neither for convergence nor for divergence (Fig. 4, Table 3).

Accompanying small vertical saccades during vergence steps

showed vertical and horizontal disjunctive components. The part

of the vertical disjunctive component on the vertical amplitude

was 3.49� 1.57% and of the horizontal disjunctive component

6.53� 2.17% (Versino et al., 1996).

Conjugate eye movements

Smooth pursuit

Patients showed significantly lower smooth pursuit velocity gain

(0.78� 0.04) than control subjects (0.94� 0.01, T(17) = 3.62,

P50.01) (Fig. 5A) and a significant increased phase shift during

sinusoidal smooth pursuit (patients = 4.05� 0.70, controls =

1.67� 0.46, T(18) =�2.70, P = 0.015). Initial acceleration of

smooth pursuit in the step-ramp paradigm was decreased in

all patients (39.79� 4.03�/s2, T(18) = 4.94, P50.001) when com-

pared with control subjects (79.15� 7.33�/s2) (Fig. 5B). Latency

of smooth pursuit in the step-ramp paradigm was significantly

prolonged in patients (231.95� 12.33 ms) versus controls

(165.77� 4.80 ms, T(18) =�4.61, P50.001) (Fig. 5C). Steady

state velocity pursuit gain was significantly smaller in patients

(0.68� 0.06) than in control subjects [0.86� 0.35, T(18) = 2.35,

P = 0.03].

Saccades

Since amplitude gain did not differ between 10� and 20� target

amplitudes data were pooled. Patients showed hypometria with

a significantly lower saccade gain (0.81� 0.02) than control sub-

jects [0.91� 0.01, T(2,25) = 4.01, P = 0.001] (Fig. 5D).

Comparison of patients with versuspatients without vermis lesion

Slow vergence

In sinusoidal vergence tasks there were significant differences

between patients with and without vermis lesions, when com-

pared with controls [F(2,28) = 6.25 P50.01]. In post-hoc tests,

patients with vermis lesions (0.73� 0.05) but not patients without

vermis lesions (0.79� 0.03) had significantly lower slow vergence

gain when compared to controls (P50.01). Patients without

vermis lesions did not significantly differ from controls (Fig. 6A).

In a subgroup ANOVA of mean vergence velocity in the ramp

task, mean divergence velocity was different between the three

groups [F(2,28) = 6.07, P50.01]. Post-hoc analysis showed

significant differences between patients with vermis lesions

(1.23� 0.09�/s) compared to controls (P50.01) while those with-

out vermis lesions (1.40� 0.04�/s) did not differ from controls

(Fig. 6B). In contrast, there was no difference for mean conver-

gence velocity between these subgroups. Unlike the group of all

patients, latency of divergence response to ramp stimuli was

Table 2 Reconstruction of cerebellar lesion sites

# Age Lesion type SCA PICA AICA Flocculus Uvula/Nodulus

Deepcerebellar nuclei

Tonsil Vermis Hemisphere Cerebellarpeduncle

1 45 Stroke Left � � � � + � � lob IV, V +

2 43 Stroke Right � � � � + � � lob IV, V +

3 67 Stroke Right � � � � + � + lob IV �

4 55 Stroke Right Left � � � + � + Crus I �

5 43 Stroke Right Right � � � + � � Crus I �

6 48 Stroke � Left � � + + + + Crus I, lob VII �

7 51 Stroke � Left � � + + + + Crus I, lob IV-VI �

8 63 Stroke � Left � + � + + + Crus I, II, lob VIII �

9 57 Stroke � Left � � + + + + Crus I, lob VI �

10 69 Stroke � Left � � + + + + Crus I, II �

11 69 Stroke � Left � + � + + + Crus II, Lob VII, VIII +

12 64 Stroke � Left � + � � + + Crus I, II �

13 51 Cavernoma � � � � + + + � lob VIII �

14 68 Stroke � Left � � � + + � Crus I, II, lob VI �

15 50 Stroke � Left � � + � + � VII, VIII �

16 26 Stroke � Left � � � + + � VIII �

17 59 Stroke � Right � � � � + + Crus I, II �

18 29 Stroke � Right � � + + + � Crus I, II, lob VI �

19 81 Stroke � Right � � + + � + Crus I, II �

20 77 Stroke � Right � � � � + � Crus I �

+ = affected, �= unaffected.


Dow



significantly longer in patients with vermis lesions (205.23�

22.3 ms) than in controls [147.36� 9.95 ms, T(17) =�2.72,

P = 0.01] (Fig. 6C). Patients with vermis sparing cerebellar lesions

(152.59� 15.40 ms) did not differ from controls.

Fast vergence

Group comparison of vergence responses to step targets (fast

vergence) did not show significant differences between patients

with and without vermis lesions (Table 3).

Conjugate smooth pursuit

Subgroup ANOVA showed a significant difference in sinusoidal

conjugate smooth pursuit gain between the groups [F(2,18) =

7.37, P50.01]. Patients with vermis lesions (0.74� 0.06) had sig-

nificantly lower smooth pursuit velocity gain than controls

(P50.01) while patients without vermis lesions (0.81� 0.05) did

not differ from controls (0.94� 0.01).

Subgroup analysis (ANOVA) of smooth pursuit in the step-ramp

paradigm revealed that the initial acceleration of both, patients

with vermis [37.42� 3.90�/s2, F(2,19) = 11.92, P = 0.001] and

those with vermis sparing cerebellar lesions (43.95� 9.32�/s2,

P = 0.01) was significantly reduced compared to healthy controls

(79.15� 7.33�/s2). Subgroup ANOVA of smooth pursuit latency

showed a significant difference when both patient subgroups

[F(2,19) = 10.66, P = 0.001] were compared with controls.

Patients with (226.31� 15.6 ms, P50.01) and without

(241� 22.20 ms, P50.01) vermis lesions did not differ in latency.

Conjugate saccades

Saccadic hypometria was found in both patients with vermis

lesions and patients with vermis sparing cerebellar lesions. The

subgroup analysis revealed a difference between controls and

both subgroups of cerebellar patients [F(2,26) = 8.36, P50.01].

Both patient subgroups had a lower saccadic gain than control

subjects [patients with vermis lesion (0.82� 0.02, P = 0.01);

patients with vermis sparing cerebellar lesions (0.81� 0.04,

P50.01)].

Comparison of healthy control subjectsand patients with vermis sparinglesions of the deep cerebellar nuclei

Slow vergence

In sinusoidal vergence tasks mean vergence gain did not differ

between healthy controls and patients with vermis sparing lesions

of the deep cerebellar nuclei [T(17) = 1.72, P = 0.10; controls

0.90� 0.03, patients 0.82� 0.03].

There were no significant differences in the mean vergence

velocity in the ramp task between healthy controls and patients

with vermis sparing lesions of the deep cerebellar nuclei (n = 7),

neither for convergence [T(17) = 0.12, P = 0.86; controls 1.55�

0.05�/s, patients 1.53� 0.07�/s] nor for divergence [T(17) =

1.61, P = 0.13; controls 1.55� 0.06�/s, patients 1.42� 0.04�/s].

Comparison between slow conjugateand disconjugate eye movementsPatients showed a reduced vergence and conjugate smooth

pursuit gain with sinusoidal stimulation, particularly patients with

vermis lesions. However, there was no significant correlation

between both parameters (controls: R = 0.07, P40.05, R2 = 0.00;

patients: R = 0.60, P40.05, R2 = 0.36). There was no significant cor-

relation between the phase shift of sinusoidal vergence and

smooth pursuit (controls: R = 0.15, P = 0.70, R2 = 0.02, patients:

R = 0.56, P = 0.09, R2 = 0.31). Furthermore, there was no signifi-

cant correlation between conjugate smooth pursuit steady state

Fig. 2 Original recordings of (slow) sinusoidal vergence

position and velocity responses of a patient (A) and an age-

matched control subject (B). The upper row shows original

position data (black), the lower row velocity data (original data

in black, velocity fit in blue). The red traces reflect the target

position (0.12 Hz, upper row), respectively, the horizontal

target velocity (lower row). Gain (�SE) of sinusoidal vergence

velocity is plotted for control subjects (black) and patients

(grey); significant differences are indicated by asterisk in C.


Dow



velocity in the step-ramp paradigm and vergence mean velocity

in the ramp paradigm (controls: convergence: R = 0.49, P40.05,

R2 = 0.24; divergence: R = 0.25, P40.05, R2 = 0.06; patients: con-

vergence: R = 0.49, P40.05, R2 = 0.24; divergence: R = 0.42,

P40.05, R2 = 0.17).

DiscussionBased on several lines of evidence we tested the hypothesis that

vergence eye movements are impaired in patients with circum-

scribed cerebellar lesions. This assumption is derived from anato-

mical (May et al., 1992; Gamlin et al., 1996), single unit

recording (Nitta et al., 2008a,b) and lesion studies in animal

experiments (Takagi et al., 2003). Here we show that disparity-

induced vergence to slowly moving targets is impaired in cerebel-

lar patients, particularly in patients with vermis lesions. In contrast,

fast vergence to step targets remained unimpaired. Concomitantly,

patients showed reduced acceleration in the initiation of conjugate

smooth pursuit, diminished smooth pursuit gain and saccadic

hypometria. The main implications of our study are 2-fold:

(i) the human cerebellum is involved in vergence processing and

(ii) the cerebellum, noticeably the vermis, controls slow vergence.

After a brief consideration of methodological aspects of ver-

gence eye movements, we will focus on the implications of iden-

tified vergence abnormalities for the role of the cerebellum in

the control of slow and fast vergence eye movements.

Methodological considerationsThe slighty lowered position of the stimulus platform caused

accompanying small vertical saccades during vergence eye move-

ments particularly during vergence steps. Since (vertical) saccades

enhance horizontal vergence step responses (Enright, 1984; Zee

et al., 1992; Busettini and Mays, 2005), one portion of the ver-

gence step movement is affected by saccadic enhancement.

However, the influence of saccades should be the same for

patients and controls due to the same laboratory setup. The dis-

junctive portion of these accompanying saccades was negligibly

small and has probably no impact on the vergence responses.

Vergence eye movements can be distinguished in vergence to

step (fast or transient vergence) and to ramp targets (slow or

sustained vergence) (Jones, 1980; Schor, 1980; Hung et al.,

1983; Semmlow et al., 1986, 2007; Erkelens et al., 1989b;

Semmlow and Yuan, 2002a,b; Gayed and Alvarez, 2006; Leigh

and Zee, 2006; Straumann, 2007). Both subsystems have different

properties: the fast vergence system is best elicited by stimuli with

large retinal disparity errors; the slow vergence by small disparity

Table 3 Vergence data

Control Patients (all) Vermis Non-vermis

Con Div Con Div Con Div Con DivMean� SE Mean� SE Mean� SE Mean� SE Mean� SE Mean� SE Mean� SE Mean� SE

Fast vergence

Gain 0.89� 0.04 0.86� 0.04 0.84� 0.29 0.81� 0.03 0.81� 0.06 0.78� 0.06 0.86� 0.02 0.83� 0.02Velocity (�/s) 43.70� 3.97 41.41� 2.63 46.89� 3.54 45.25� 2.81 42.33� 3.30 39.46� 4.94 50.08� 5.47 49.31� 2.85Latency (ms) 193.53� 6.04 182.22� 13.62 193.53� 6.62 191.18� 7.52 194.17� 6.04 205.71� 9.66 193.08� 10.75 181.00� 9.96

Slow vergence ramp

Velocity (�/s) 1.55� 0.05 1.55� 0.06 1.51� 0.06 1.33� 0.05 1.48� 0.11 1.23� 0.09 1.54� 0.06 1.40� 0.04Latency (ms) 203.19� 10.96 147.36� 9.95 241.04� 19.27 175.62� 14.29 237.38� 24.20 205.23� 22.3 243.89� 29.89 152.59� 15.40Acceleration (�/s2) 13.38� 0.91 10.42� 0.87 14.23� 1.34 9.47� 0.81 12.94� 2.20 7.97� 1.01 15.23� 1.69 10.64� 1.09

Slow vergence sinus

Gain 0.89� 0.03 0.76� 0.03 0.73� 0.05 0.79�0.03

Fig. 3 Data of vergence to ramp stimuli. Vergence velocity

responses (mean � SE) of one patient with a vermis lesion (red)

and a control subject (black); target velocity is indicated by a

red dashed line; latencies are indicated by coloured arrows in

A. Vergence mean velocity (absolute values = abs., �SE) of

controls (black column) and patients (grey) for convergence

and divergence; significant differences are indicated by aster-

isks in B. Divergence but not convergence is significantly

reduced in patients.


Dow



errors and disparity velocities of less than 3�/s (Erkelens et al.,

1989a). Semmlow and co-workers (1986) suggested stimulus

velocities up to 2�/s to elicit slow vergence, velocities larger

than 4�/s to evoke fast vergence. Accordingly, for slow vergence

we have chosen 1.5�/s and 7�/s disparity angle for fast vergence.

Furthermore these stimulus properties are comparable to our pre-

vious studies in healthy human subjects and patients (Rambold

et al., 2004, 2005a, 2006).

Fig. 4 Fast vergence responses to step targets [vergence position (Verg Pos), vertical eye position (Vert Pos) and peak vergence

velocity (Verg Vel)] in a patient (left) and an age-matched control subject (right) for divergence and convergence. Horizontal target

position is shown in red.


Dow



As the timing of stimulus onset was not predictable it is unlikely

that predictive components influenced our vergence ramp

responses (Kumar et al., 2002).

Recent clinical evidence for selective deficits of vergence track-

ing of a slowly moving target in depth but preserved fast vergence

responses (Rambold et al., 2004) suggested that both types of

vergence have distinctly different neural control and hence differ-

ent anatomical representations. Based on the evidence for pontine

projections to the cerebellum (introduction) we hypothesized that

vergence eye movements to either fast or slowly moving targets

or both may be impaired in cerebellar lesions.

Cerebellum and eye movementsWe specifically examined patients with focal and acute cerebellar

lesions. Unlike previous studies on smooth pursuit in spinocerebel-

lar ataxia or other cerebellar degenerations, lesions of cerebellar

stroke patients can be identified to be confined to the cerebellum.

We focussed our analysis on structures which are known to be

involved in vergence in animal studies and in the cerebellar control

of smooth pursuit eye movements (SPEM), i.e. vermis (Nitta et al.,

2008a), the underlying deep nuclei (Gamlin et al., 1996) and the

flocculus (Tsubuku et al., 2004).

Conjugate eye movements

There are two pathways conveying SPEM signals to the cerebel-

lum: one involves the posterior vermis (lobule VI and VII) and the

underlying deep cerebellar nuclei and the other the flocculus/

paraflocculus (Leigh and Zee, 2006). In our study–except for

three patients–the flocculus was usually not involved. Our cere-

bellar patients showed impaired SPEM, not only in the closed loop

but also in the open loop phase which is in accord with pre-

vious studies on cerebellar degeneration (Moschner et al., 1999)

or infarctions (Straube et al., 1997; Moschner et al., 1999).

Accordingly, latency was prolonged and the initial acceleration

was reduced in the step-ramp paradigm. Smooth pursuit gain

to sinusoidal stimulation was particularly reduced in our patients

with vermis lesions as has been shown elsewhere (Vahedi et al.,

1995). In a non-human animal study (Takagi et al., 2000), vermis

ablation with sparing of the deep cerebellar nuclei in monkeys

caused a decrease in SPEM gain, a decrease in peak acceleration

and a decrease in pursuit velocity, which is in accord with our

patients with vermis lesions. In contrast, vermis was spared in

cerebellar patients examined by Straube and coworkers who

found reduced smooth pursuit initiation using the step-ramp para-

digm (Straube et al., 1997).

Patients also had conjugate saccadic dysmetria with decreased

saccadic gain. As it is well known that experimental and clinical

lesions of the deep cerebellar nuclei cause saccadic hypermetria

(Selhorst et al., 1976; Robinson et al., 1993) the saccadic hypo-

metria in our patients is consistent with vermis impairment (Takagi

et al., 1998). In contrast, saccadic dysmetria is not found in floc-

culus lesions (Rambold et al., 2002a). Therefore both SPEM

and saccade deficits are compatible with lesions of the posterior

vermis.

Vergence

Vergence to slowly moving targets

Vergence to slowly moving targets was impaired in our cerebellar

patients. Gain of slow vergence to sinusoidal moving targets

and slow vergence velocity in the ramp paradigm was significantly

reduced in patients. Therefore our data provide some evidence for

an important role of the cerebellum in the processing of slow

vergence eye movements in humans. This is in accord with pre-

vious animal (Donaldson and Hawthorne, 1979; Gamlin et al.,

1996; Zhang and Gamlin, 1998; Takagi et al., 2003; Nitta et al.,

2008a), human PET (Richter et al., 2000, 2004) and human

lesion studies (Ohtsuka et al., 1993). A functional imaging study

in humans showed activation of the cerebellar hemispheres and

the vermis during the near response and disparity vergence

(Gulyas and Roland, 1994). The slow vergence deficits of our

patients are probably not caused by an impairment of the pursuit

system in 3D space since there was no correlation between both

slow vergence and conjugate smooth pursuit responses.

Interestingly, divergence but not convergence eye movement

responses to ramp targets were impaired. This is in line with an

animal study which showed reduced divergence acceleration,

divergence velocity and delayed divergence responses after

ablation of the dorsal vermis with sparing of the deep cerebellar

nuclei (Takagi et al., 2003). In contrast, in a more recent animal

study convergence (ramp velocity 10�/s) were impaired after cir-

cumscribed vermis inactivation (muscimol) of regions containing

Fig. 5 Horizontal sinusoidal smooth pursuit gain (A), initial

acceleration (B) and latency (C) in step-ramp pursuit and

horizontal saccade gain (D) are shown for controls (black) and

patients (grey); significant differences are indicated by asterisks.


Dow



largely convergence-modulated neurons: a reduction of conver-

gence eye velocity by 20% in the ramp paradigm (Nitta et al.,

2008a). Divergence was less affected (Nitta et al., 2008b). Several

reasons might account for the differential lesion effects on diver-

gence and convergence ramp responses. First, while ablation

affects both cell soma and fibres of passage, muscimol inactivates

neurons only. Secondly, stimulus velocities were different in both

studies. While Takagi and colleagues stimulated small vergence

step targets of 0.5� and 1� (Takagi et al., 2003), Nitta and

colleagues used step targets of 10� amplitude and ramp target

velocities of 10�/s (Nitta et al., 2008a). Finally, convergence-

and divergence-related neurons are anatomically separated

in the cerebellum, not only in the fastigial oculomotor region

(FOR) and posterior interposed nucleus (IP) (Mays et al., 1986;

Gamlin et al., 1996, 2002; Zhang and Gamlin, 1998; Nitta et al.,

2008a) but also in the posterior vermis (Nitta et al., 2008a). Our

selective divergence impairment in the ramp paradigm indicates

separate organization of both, convergence and divergence eye

movements. Divergence-related neurons in the IP (Zhang and

Gamlin, 1998) receive projections from the ventral paraflocculus

(Nagao et al., 1997) and vermis (Leigh and Zee, 2006). Since the

paraflocculus contains divergence-related cells (Tsubuku et al.,

2004) the parafloccular-IP pathway has been suspected to be

involved in the control of divergence eye movements while

the dorsal vermis/FOR pathway may be rather related to conver-

gence (Nitta et al., 2008a). As divergence-related cells were found

sparsely in the vermis when compared to convergence-related

neurons (Nitta et al., 2008a), divergence-related control mechan-

isms may be more susceptible to vermis lesions and may show a

greater vulnerability while convergence may be better compen-

sated in cerebellar lesions. Although these arguments may explain

the divergence deficit they do not account for the additional con-

vergence impairment during sinusoidal vergence.

Esodeviation was found after vermis ablation (Takagi et al.,

2003) but not with vermis inactivation (Nitta et al., 2008a).

Esodeviation was also found in patients with cerebellar degenera-

tion reflecting an increase in convergence tone (Versino et al.,

1996). This might be related to sparing of the fastigial oculomotor

Fig. 6 Vergence to slow vergence stimuli. Gain of sinusoidal vergence velocity is displayed in A for controls (black), patients with

vermis lesions (grey) and patients without vermis lesions (white). Slow vergence mean velocity (B) and latency (C) of the ramp

paradigm are shown for convergence and divergence separately; significant differences are indicated by asterisks.


Dow



region (FOR) since inactivation of these deep cerebellar nuclei also

cause convergence deficits with exodeviation (Scheurer et al.,

2001). Binocular static misalignment was neither reported in

vermis inactivation (Nitta et al., 2008a) nor found in our study.

Vergence to fast moving targets

Step tasks elicit fast vergence eye movements which are coded

by the fast vergence system, in contrast to slow vergence eye

movements which underlie visual feedback. Fast vergence

responses to step targets were unaffected in our patients. This

contradicts animal data with impairment of convergence step

responses after vermis inactivation (Nitta et al., 2008a). After

cerebellar lesions the monkeys showed much slower convergence

velocities (15–20�/s) than prior inactivation (48–87�/s) and had

a prolonged latency of convergence in the vergence step task.

However, divergence responses could not be analysed due to

concomitant saccades. Interestingly, while vermis lesions elicit

impairment of both conjugate eye movement systems, smooth

pursuit and saccades, fast vergence seemed to remain unaffected

in our patients. This is some additional evidence that fast and

slow vergence are under separate neural control. It remains to

be investigated in future studies whether the normal fast vergence

responses in our patients are related to (i) extracerebellar control

mechanisms of fast vergence, (ii) the fact that we missed critical

cerebellar lesion sites in our patient study group, or (iii) more

rapidly adaptive recovery mechanisms in fast vergence.

Eye movements in patients withvermis lesionsBased on animal data we pursued the hypothesis that the cere-

bellar vermis is involved in the processing of vergence eye move-

ments. Fortunately, a large number of our patients showed vermis

involvement which allowed statistical comparison with patients

with vermis sparing cerebellar lesions. This is of interest since sig-

nificant vergence tracking impairment to slowly moving targets

was particularly found in our patients with lesions involving the

vermis. It supports the concept of a crucial role of the cerebellar

vermis in controlling slow vergence eye movements in humans

as assumed by recent animal data (Donaldson and Hawthorne,

1979; Takagi et al., 2003) and imaging studies (Richter et al.,

2000, 2004; Nitta et al., 2008a). While animal lesion studies

of the cerebellum up to now only investigated vergence eye

movements in vermis lesions (Takagi et al., 2003; Nitta et al.,

2008a) our patients’ lesions were larger and not confined to

the vermis. Accordingly, we cannot infer from our data how far

only vermis lesions or other defined cerebellar structures cause

slow vergence deficits. At least two results argue against the

deep cerebellar nuclei to cause the vergence deficits. First, patients

with vermis sparing lesions of the deep cerebellar nuclei did

not show any significant difference in slow vergence compared

to healthy controls. Second, there was no saccadic hypermetria

which is usually found in lesions of the fastigial oculomotor

region (Robinson et al., 1993). Therefore, the deep cerebellar

nuclei alone can probably not account for the slow vergence

deficit in our patients.

In conclusion, this is the first study on a comparative investiga-

tion of fast and slow dynamic vergence in patients with cerebellar

lesions. Our a priori hypothesis of a critical cerebellar role in

vergence was derived from animal lesion and single unit recording

data as well as anatomical ponto-cerebellar projections of ver-

gence modulated neurons. Our patient data implicate the cerebel-

lum, noticeably the vermis, in the control of slow but not fast

vergence. We cannot exclude that lesions of other cerebellar

structures not involved in our patients might evoke fast vergence

deficits. However, we provide additional evidence for a separate

neural control of fast and slow vergence, not only in the pons

(Rambold et al., 2004, 2005b) but also in the cerebellum.

ReferencesAkao T, Kurkin SA, Fukushima J, Fukushima K. Visual and vergence

eye movement-related responses of pursuit neurons in the caudal frontal

eye fields to motion-in-depth stimuli. Exp Brain Res 2005a; 164: 92–108.

Akao T, Mustari MJ, Fukushima J, Kurkin S, Fukushima K. Discharge

characteristics of pursuit neurons in MST during vergence eye move-

ments. J Neurophysiol 2005b; 93: 2415–34.Barash S, Melikyan A, Sivakov A, Zhang M, Glickstein M, Thier P.

Saccadic dysmetria and adaptation after lesions of the cerebellar

cortex. J Neurosci 1999; 19: 10931–9.

Busettini C, Mays LE. Saccade-vergence interaction in macaques. II.

Vergence eenhancement as the product of a local feedback vergence

motor error and a weighted saccadic burst. J Neurophysiol 2005; 94:

2312–30.

Carl JR, Gellman RS. Human smooth pursuit: stimulus-dependent

responses. J Neurophysiol 1987; 57: 1446–63.

Donaldson IM, Hawthorne ME. Coding of visual information by units

in the cat cerebellar vermis. Exp Brain Res 1979; 34: 27–48.

Enright JT. Changes in vergence mediated by saccades. J Physiol 1984;

350: 9–31.

Erkelens CJ, Steinman RM, Collewijn H. Ocular vergence under natural

conditions. II. Gaze shifts between real targets differing in distance

and direction. Proc R Soc Lond B Biol Sci 1989a; 236: 441–65.

Erkelens CJ, Van der Steen J, Steinman RM, Collewijn H. Ocular

vergence under natural conditions. I. Continuous changes of target

distance along the median plane. Proc R Soc Lond B Biol Sci 1989b;

236: 417–40.

Fukushima K, Yamanobe T, Shinmei Y, Fukushima J, Kurkin S. Role of

the frontal eye fields in smooth-gaze tracking. Prog Brain Res 2004;

143: 391–401.

Gamlin PD, Clarke RJ. Single-unit activity in the primate nucleus reticu-

laris tegmenti pontis related to vergence and ocular accommodation.

J Neurophysiol 1995; 73: 2115–9.

Gamlin PD, Yoon K, Zhang H. The role of cerebro-ponto-cerebellar path-

ways in the control of vergence eye movements. Eye 1996; 10: 167–71.

Gamlin PD, Yoon K. An area for vergence eye movement in primate

frontal cortex. Nature 2000; 407: 1003–7.

Gamlin PD, Zhang HY. Effects of muscimol blockade of the posterior

fastigial nucleus on vergence and ocular accomodation in the primate.

Soc Neurosci Abstr 1996; 22: 1034.Gamlin PD. Neural mechanisms for the control of vergence eye move-

ments. Ann NY Acad Sci 2002; 956: 264–72.

Gamlin PD. Subcortical neural circuits for ocular accommodation and

vergence in primates. Ophthalmic Physiol Opt 1999; 19: 81–9.

Gayed BA, Alvarez TL. Quantitative assessment of divergence eye move-

ments to ramp stimuli. Conf Proc IEEE Eng Med Biol Soc 2006; 1:

3954–7.

Gulyas B, Roland PE. Binocular disparity discrimination in human cerebral

cortex: functional anatomy by positron emission tomography. Proc

Natl Acad Sci USA 1994; 91: 1239–43.


Dow



Helmchen C, Buttner U. Saccade-related Purkinje cell activity in theoculomotor vermis during spontaneous eye movements in light and

darkness. Exp Brain Res 1995; 103: 198–208.

Helmchen C, Hagenow A, Miesner J, Sprenger A, Rambold H,

Wenzelburger R, et al. Eye movement abnormalities in essentialtremor may indicate cerebellar dysfunction. Brain 2003; 126: 1319–32.

Hung GK, Semmlow JL, Ciuffreda KJ. Identification of accommodative

vergence contribution to the near response using response variance.

Invest Ophthalmol Vis Sci 1983; 24: 772–7.Jones R. Fusional vergence: sustained and transient components. Am J

Optom Physiol Opt 1980; 57: 640–4.

Krauzlis RJ, Miles FA. Role of the oculomotor vermis in generatingpursuit and saccades: effects of microstimulation. J Neurophysiol

1998; 80: 2046–62.

Kumar AN, Han Y, Garbutt S, Leigh RJ. Properties of anticipatory

vergence responses. Invest Ophthalmol Vis Sci 2002; 43: 2626–32.Leigh RJ, Zee DS. The Neurology of Eye Movements, Vol. 55. New York:

Oxford University Press, 2006.

May PJ, Porter JD, Gamlin PD. Interconnections between the primate

cerebellum and midbrain near-response regions. J Comp Neurol 1992;315: 98–116.

Mays LE, Porter JD, Gamlin PD, Tello CA. Neural control of vergence

eye movements: neurons encoding vergence velocity. J Neurophysiol

1986; 56: 1007–21.Moschner C, Crawford TJ, Heide W, Trillenberg P, Kompf D, Kennard C.

Deficits of smooth pursuit initiation in patients with degenerative

cerebellar lesions. Brain 1999; 122: 2147–58.Nagao S, Kitamura T, Nakamura N, Hiramatsu T, Yamada J. Location of

efferent terminals of the primate flocculus and ventral paraflocculus

revealed by anterograde axonal transport methods. Neurosci Res

1997; 27: 257–69.Nitta T, Akao T, Kurkin S, Fukushima K. Involvement of the cerebellar

dorsal vermis in vergence eye movements in monkeys. Cereb Cortex

2008a; 18: 1042–57.

Nitta T, Akao T, Kurkin S, Fukushima K. Vergence eye movement signalsin the cerebellar dorsal vermis. Prog Brain Res 2008b; 171: 173–6.

Ohtsuka K, Maekawa H, Sawa M. Convergence paralysis after lesions

of the cerebellar peduncles. Ophthalmologica 1993; 206: 143–8.Ramat S, Leigh RJ, Zee DS, Optican LM. What clinical disorders tell

us about the neural control of saccadic eye movements. Brain 2007;

130: 10–35.

Rambold H, Churchland A, Selig Y, Jasmin L, Lisberger SG. Partial abla-tions of the flocculus and ventral paraflocculus in monkeys cause

linked deficits in smooth pursuit eye movements and adaptive mod-

ification of the VOR. J Neurophysiol 2002a; 87: 912–24.

Rambold H, Neumann G, Helmchen C. Vergence deficits in pontinelesions. Neurology 2004; 62: 1850–3.

Rambold H, Neumann G, Sander T, Helmchen C. Age–related changes

of vergence under natural viewing conditions. Neurobiol Aging 2006;

27: 163–72.Rambold H, Neumann G, Sander T, Helmchen C. Pontine lesions may

cause selective deficits of ‘slow’ vergence eye movements. Ann NY

Acad Sci 2005a; 1039: 567–70.Rambold H, Sander T, Neumann G, Helmchen C. Palsy of ‘fast’ and

‘slow’ vergence by pontine lesions. Neurology 2005b; 64: 338–40.

Rambold H, Sprenger A, Helmchen C. Effects of voluntary blinks on

saccades, vergence eye movements, and saccade-vergence interactionsin humans. J Neurophysiol 2002b; 88: 1220–33.

Richter HO, Costello P, Sponheim SR, Lee JT, Pardo JV. Functional

neuroanatomy of the human near/far response to blur cues: eye-

lens accommodation/vergence to point targets varying in depth. EurJ Neurosci 2004; 20: 2722–32.

Richter HO, Lee JT, Pardo JV. Neuroanatomical correlates of the near

response: voluntary modulation of accommodation/vergence in thehuman visual system. Eur J Neurosci 2000; 12: 311–21.

Robinson FR, Straube A, Fuchs AF. Role of the caudal fastigial nucleus in

saccade generation. II. Effects of muscimol inactivation. J Neurophysiol

1993; 70: 1741–58.

Scheurer W, Petz T, Eggert T, Straube A. Static alignment after unilateraland bilateral pharmacological inactivation of the caudal fastigial

nucleus in the monkey. Soc Neurosci Abstr 2001; 27.

Schmahmann JD, Doyon J, Toga A, Petrides M, Evans A. MRI Atlas of

the Human Cerebellum. New York: Academic Press, 2000.Schor C. Fixation of disparity: a steady state error of disparity-induced

vergence. Am J Optom Physiol Opt 1980; 57: 618–31.

Selhorst JB, Stark L, Ochs AL, Hoyt WF. Disorders in cerebellar ocular

motor control. I. Saccadic overshoot dysmetria. An oculographic, con-trol system and clinico-anatomical analysis. Brain 1976; 99: 497–508.

Semmlow JL, Alvarez TL, Pedrono C. Dry dissection of disparity diver-

gence eye movements using independent component analysis.Comput Biol Med 2007; 37: 910–8.

Semmlow JL, Hung GK, Ciuffreda KJ. Quantitative assessment of

disparity vergence components. Invest Ophthalmol Vis Sci 1986; 27:

558–64.Semmlow JL, Yuan W. Adaptive modification of disparity vergence

components: an independent component analysis study. Invest

Ophthalmol Vis Sci 2002a; 43: 2189–95.

Semmlow JL, Yuan W. Components of disparity vergence eye move-ments: application of independent component analysis. IEEE Trans

Biomed Eng 2002b; 49: 805–11.

Straube A, Scheuerer W, Eggert T. Unilateral cerebellar lesions affect

initiation of ipsilateral smooth pursuit eye movements in humans.Ann Neurol 1997; 42: 891–8.

Straumann D. Disconjugate eye movements. Dev Ophthalmol 2007; 40:

90–109.Suzuki DA, Keller EL. The role of the posterior vermis of monkey cere-

bellum in smooth-pursuit eye movement control. II. Target velocity-

related Purkinje cell activity. J Neurophysiol 1988; 59: 19–40.

Suzuki DA, Yamada T, Hoedema R, Yee RD. Smooth-pursuit eye-movement deficits with chemical lesions in macaque nucleus reticularis

tegmenti pontis. J Neurophysiol 1999; 82: 1178–86.

Takagi M, Tamargo R, Zee DS. Effects of lesions of the cerebellar

oculomotor vermis on eye movements in primate: binocular control.Prog Brain Res 2003; 142: 19–33.

Takagi M, Zee DS, Tamargo RJ. Effects of lesions of the oculomotor

vermis on eye movements in primate: saccades. J Neurophysiol 1998;80: 1911–31.

Takagi M, Zee DS, Tamargo RJ. Effects of lesions of the oculomotor

cerebellar vermis on eye movements in primate: smooth pursuit.

J Neurophysiol 2000; 83: 2047–62.Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories

of human brain: brainstem and cerebellum. Neurology 1996; 47:

1125–35.

Thielert CD, Thier P. Patterns of projections from the pontine nuclei andthe nucleus reticularis tegmenti pontis to the posterior vermis in the

rhesus monkey: a study using retrograde tracers. J Comp Neurol 1993;

337: 113–26.

Tsubuku T, Akao T, McCrea R, Kurkin S, Fukushima J, Fukushima K.Purkinje cell activity in the cerebellar floccular region during vergence

eye movements. Soc Neurosci Abstr 2004; 30.

Vahedi K, Rivaud S, Amarenco P, Pierrot-Deseilligny C. Horizontal eyemovement disorders after posterior vermis infarctions. J Neurol

Neurosurg Psychiatry 1995; 58: 91–4.

Versino M, Hurko O, Zee DS. Disorders of binocular control of eye

movements in patients with cerebellar dysfunction. Brain 1996; 119:1933–50.

Westheimer G, Blair SM. Oculomotor defects in cerebellectomized

monkeys. Invest Ophthalmol 1973; 12: 618–21.

Zee DS, Fitzgibbon EJ, Optican LM. Saccade-vergence interactions inhumans. J Neurophysiol 1992; 68: 1624–41.

Zhang H, Gamlin PD. Neurons in the posterior interposed nucleus of

the cerebellum related to vergence and accommodation. I. Steady-state characteristics. J Neurophysiol 1998; 79: 1255–69.

Zhang HY, Gamlin PD. Single-unit activity within the posterior fastigial

nucleus during vergence and accommodation in the alert primate. Soc

Neurosci Abstr 1996; 22: 2034.


Dow



Documents

Vergence deficits in patients with cerebellar lesions - Brain