Psychiatry Research, 38: 173-185

173

Elsevier

Smooth Pursuit Neurological Disorder

and Saccadic Soft Signs

Steven E. Nickoloff, Allen Daniel W. Hommer Received

D. Radant,

Eye Movements and in Obsessive-Compulsive

Robert

Reichler,

October 25, 1990; revised version received February

and

1, 1991; accepted April 9, 1991.

Abstract. Patients with obsessive-compulsive disorder (OCD) demonstrate an increased number of neurological soft signs as well as neuroanatomic abnormalities detected with modern imaging techniques. Quantitative analysis of eye movements has proved fruitful in investigations of other neuropsychiatric disorders with similar findings. Therefore, we studied the smooth pursuit and saccadic eye movements of 8 OCD patients and 12 normal controls using infrared oculography and computerized pattern recognition software. We also measured neurologic soft signs using a standardized rating instrument. Despite having an increased number of neurological soft signs, OCD patients’ performance on a variety of measures of eye movement was not significantly impaired. Neither the severity of obsessions or compulsions nor the number of neurologic soft signs correlated with any of the parameters of eye movement function. We conclude that OCD patients do not have prominent oculomotor dysfunction and that eye movement dysfunction and neurologic soft signs are not inextricably linked. Key Words. Obsessive-compulsive smooth pursuit, saccades.

disorder,

eye movements,

neurologic

soft signs,

disorder (OCD) is characterized by recurrent thoughts, images, ideas, or behavior that the patient considers ego-dystonic and resists. OCD is far more common than has been previously thought, with a lifetime prevalence rate of approximately 3% (Karno et al., 1988; Myers et al., 1984). A number of recent reports have suggested that neurobiologic dysfunction underlies OCD. Neurological soft signs, which are generally defined as nonlocalizing signs of deviant sensory motor performance, are more frequent among patients with OCD than among normal subjects (Behar et al., 1984; Denckla, 1988; Hollander et al., 1990). Other evidence of neurologic dysfunction in OCD includes: high comorbidity and family linkage with other neurologic illnesses (Grimshaw, 1964; Pauls et al., 1986; Swedo et al., 1989a); abnormal electroencephalogram and evoked potentials (Jenike and Brotman, 1984); association with head trauma (McKeon et al., 1984), encephalitis (Schilder, 1938) and birth trauma (Capstick and Seldrup, 1977); high concordance between monozygotic twins (Inouye, 1965); and responsiveness of the disorder to

Obsessive-compulsive

Steven E. Nickoloff, M.D., is Resident; Allen D. Radant, M.D., is Instructor; Robert Reicbler, M.D., is Professor, and Daniel W. Hommer, M.D., is Associate Professor, Department of Psychiatry and Behavioral Sciences, University of Washington, Seattle, and the Seattle Department of Veterans Affairs Medical Center. (Reprint requests to Dr. D.W. Hommer, Geriatric Research Education and Clinical Center 182 B, Seattle VA Medical Center, 1660 S. Columbian Way, Seattle, WA 98108, USA.) 0165-1781/91/$03.50

@ 1991 Elsevier Scientific

Publishers

Ireland

Ltd.

174 specific pharmacologic agents (Jenike et al., 1989a, 19896). The basal ganglia (Rapoport and Wise, 1988; Stahl, 1988) and the frontal cortex (Khanna, 1988; Baxter, 1990) have been proposed as likely sites of dysfunction in OCD. Basal ganglia dysfunction is suggested by localizing data from computed tomography and positron emission tomography (PET) (Luxenberg et al., 1988; Hollander et al., 1990), the association of OCD with Sydenham’s chorea (Rapoport, 1989) and the increased incidence of chorea or tics among patients with OCD (Denckla, 1988). Frontal lobe dysfunction is suggested by magnetic resonance imaging (MRI) and PET studies (Garber et al., 1989; Nordahl et al., 1989; Swedo et al., 19896; Baxter et al., 1990), as well as by the presence of neuropsychological abnormalities similar to those seen in patients with frontal lobe lesions (Flor-Henry et al., 1979; Behar et al., 1984; Rosen et al., 1988). It is difficult to separate the effects of basal ganglia and frontal lobe disease, however, because these two regions are functionally linked (Brown and Marsden, 1990). In addition to the evidence for frontal cortex and striatal dysfunction, there are also suggestions of dysfunction lateralized to the right hemisphere. The evidence includes the presence of left hemibody soft signs (Behar et al., 1984; Denckla, 1988; Hollander et al., 1990), characteristic neuropsychiatric abnormalities (Insel et al., 1983; Diamond, 1988; Rosen et al., 1988; Hollander et al., 1990) and metabolic asymmetry on PET scan (Nordahl et al., 1989). Eye movement dysfunction has been well described in several neuropsychiatric disorders, including schizophrenia (Holzman, 1987; Levin et al., 1988), Parkinson’s disease (DeJong and Jones, 1971; Cipparone et al., 1988), and Huntington’s chorea (Leigh et al., 1983). These disorders are thought to involve dysfunction of the basal ganglia, the frontal cortex, or both. Patients with schizophrenia have an increased number of neurologic soft signs compared with normal controls (Buchanan and Heinrichs, 1988), and patients with basal ganglia disease often demonstrate a variety of abnormalities in sensory motor integration that would be considered soft signs as the term is used in psychiatric research (Cummings, 1986). Siever and Coursey (1985) have reported a significant correlation between smooth pursuit dysfunction, measured using a purely qualitative rating approach, and neurological soft signs. This concordance suggests that eye movement dysfunction and neurologic soft signs may be manifestations of similar types of brain dysfunction. There has been only one report of eye movements in OCD, and this study examined only smooth pursuit in a qualitative manner and did not compare OCD patients with normal controls (Siever et al., 1986). We used high resolution infrared oculography and computerized pattern recognition software to compare the eye movements of OCD patients with those of normal controls. We also used standardized rating instruments to assess neurological soft signs and psychiatric symptoms. The purpose of these investigations was twofold: (1) Since eye movement studies can provide quantifiable and lateralizing information about sensorimotor integration, we wished to characterize any possible eye movement dysfunction among patients with OCD. (2) Since eye movement dysfunction and neurological soft signs appear to be asoociated in schizophrenia and perhaps in other neuropsychiatric disorders, we wished to determine if these two phenomena were also linked in OCD. We used two eye movement tasks: one designed to elicit smooth pursuit tracking

175 and another designed to elicit visually guided as well as predictive saccades. We examined smooth pursuit since it has been extensively studied in schizophrenia. We examined saccades because studies in both human and nonhuman primates (Hikosaka, 1990) indicate that predictive saccades are particularly impaired by lesions of the basal ganglia or frontal cortex. In addition, recent preliminary reports suggest that patients with Huntington’s disease (Tian et al., 1990) and schizophrenia (Hommer et al., 1990) are impaired in their ability to generate predictive saccades while their visually guided saccades remain relatively intact. Methods Subjects. We studied eight patients who met DSM-III-R criteria (American Psychiatric Association, 1987) for OCD. The patients (5 male and 3 female) ranged in age from 21 to 45 (mean age = 36.5 years, SD = 7.8). None of the patients had focal neurologic disorder or major medical illness. Six were receiving fluoxetine, one was receiving clomipramine, and one was medication free. Twelve normal control subjects (6 male and 6 female) ranged in age from 24 to 49 (mean age q 38.2 years, SD = 9.9). All control subjects were free of significant psychiatric, neurologic, or medical illness and took no medications on a regular basis. Both control subjects and patients were required to abstain from alcohol and all medication (except fluoxetine or clomipramine among OCD patients) for 24 hours preceding the eye movement recording. Neuropsychiatric Measures. Each subject was administered an examination for neurologic soft signs by one of the authors, using the Neurologic Evaluation Scale (Buchanan and Heinrichs, 1988). The examination includes a variety of tasks, which assess fine motor coordination, involuntary movements, sensory function, and visuospatial skills. Hemispheric dominance was determined by asking subjects to demonstrate which eye, arm, or leg they would use to perform a variety oftasks (Buchanan and Heinrichs, 1988). Previous studies have shown high intrarater and interrater reliability for evaluation of most neurologic soft signs (Rosen et al., 1988). Severity of obsessive and compulsive symptoms among OCD patients was measured by the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS; Goodman et al., 1989) and severity of overall psychiatric symptoms by the Brief Psychiatric Rating Scale (BPRS; Overall and Gorham, 1962). Eye Movement Procedure. The subject was seated 43 cm in front of a video monitor on which a small target was displayed against a black background in an otherwise dark room. The subject’s head was stabilized with a bite bar and head rest. Horizontal eye movements were recorded using an infrared photoelectric limbus detection eye tracking device (Eye-trac model 2 10, Applied Sciences Laboratories, Waltham, MA), which is accurate to within 0.25” of visual angle and has a time constant of 4 ms. The analog output of the device was sampled at 1000 Hz using a 1Zbit analog-to-digital converter. Data were collected only from the eye for which the most rapid and accurate calibration could be obtained. Eye Movement Tasks. The smooth pursuit task consisted of the following: the target moved back and forth horizontally over 30” with a constant velocity of 11” jsec and a I.4set fixation period between ramps (a “trapezoidal pattern”) (Fig. IA). This was followed by a l- to 2-min period allowing the subject to rest and the equipment to be recalibrated. Next, a sacaddic task was performed, consisting of the following: For the first 10 set, the target stepped back and forth from IO” right of the center of the screen to 10” left at pseudorandom time intervals between 0.85 and 2.0 sec. Next, the target’s behavior became completely regular, stepping from the left to the right position at a fixed interval of 0.85 set for a total of 26 sec. Finally the target motion became irregular again for the final 20 sec. The initial pseudorandom portion of the task was designed to elicit visually guided saccades, as the timing of the target’s movement was impossi-

176 Fig. 1. Eye movement activity in a patient with obsessive-compulsive (OCD) c-. . \ * .

disorder

10 Seconds CATCH-UP SACCADES

•I

1 Second

SACCADES 1 Second

1 Second Panel A shows the pattern of target (dashed line) and eye movement (solid line, of a typlcal OCD patient with OCD during the smooth pursuit task. Panel B shows 2 catch-up saccadesof 1.4” and 0.9”, respectively. Panel C shows anticipatory saccades of 1.4” and 2.6”, respectively. Note the transient slowing of eye velocity following the anticipatory saccades. Panel D shows a square wave jerk consisting of 2saccades in opposite directions of 3.1° and 1.7”, respectively, separated by an interval of smooth pursuit 310 ms in duration.

ble for the subject to predict. The regular portion was designed to elicit internally guided predictive saccades, as the regular behavior of the target encouraged subjects to anticipate the target motion. The final pseudorandom segment was designed to elicit visually guided saccades and measure any tendency of subjects to perseverate by continuing to attempt to make saccades that anticipated the target motion despite the fact that accurate prediction was now impossible. Each task lasted approximately 56 sec. The instructions for both tasks were the same: the subjects were told simply to keep their eyes on the target and follow it as rapidly and as accurately as possible. They were not given any information about regularities in target movement nor were they told to attempt to predict its motion. Data Analysis. The eye movement data were analyzed using a computerized pattern recognition program. Eye movements were divided into discrete segments, and then each segment was classified as either saccade or smooth pursuit. Artifactual segments caused by eye blinks showed a distinct morphology (two high velocity intervals separated by < 1.5 times the duration of the initial interval or > 3 changes in the sign of the acceleration during the interval) and were removed from the analysis by the pattern recognition software. Saccades were identified on the basis of peak velocity (> 35” /set) and initial acceleration (> 4,000” isec2). Segments not meeting velocity and acceleration criteria for saccades or artifacts were considered smooth pursuit or fixation. The intervals of smooth pursuit were then divided by an iterative process until no segment differed from perfect linearity by more than 0.175” / 100 ms of segment

177 duration or a total of 0.6” irrespective of the duration of the segment. This procedure produced individual smooth pursuit segments with a mean duration of 355 ms (SD q 100). The pursuit velocity during each of these intervals was calculated by determining the linear slope of the segment using its end points. Finally, the pursuit velocity for each subject over the total duration of target motion was calculated by averaging the velocity of each pursuit interval adjusted for its duration. During the smooth pursuit task, eye movements during fixation or within 250 ms of a change in target motion were discarded from the analysis, as these movements may not represent normal pursuit (Abel and Ziegler, 1988). Smooth pursuit gain was defined as mean eyevelocity divided by target velocity. Saccades that interrupt smooth pursuit may represent several distinct subtypes; we developed algorithms which assigned each saccade to one of the following categories (Abel and Ziegler, 1988): Saccades in the direction of target motion that ended behind the target position were considered catch-up saccades. Saccades that began behind the target and ended ahead of the target were also considered catch-up if they ended less than half as far from the target as they began. Catch-up saccades bring gaze that has been lagging behind the target closer to the target (Fig. 1B). Saccades in the direction of target motion and beginning when the gaze is ahead of the target were considered anticipatory as were saccades that started behind and ended ahead of the target if their final position was at least twice as far from the target as their initial position. Anticipatory saccades take the gaze ahead of the target in the direction of its motion (Fig. 1C). Square wave jerks (SWJs) were defined as two saccades in opposite directions, separated by 50 to 400 ms of smooth pursuit with a pursuit gain > 0.5 (Fig. ID). These criteria separate SWJs from the other types of intrusive saccadic events. The few saccades that did not meet criteria for either anticipatory, catch-up, or SWJ were classified as other. Total numbers of saccades were converted to standardized units of frequencyimin of actual smooth target motion. The reliability of the saccadic recognition program was assessed by visually inspecting each record and comparing the judgment of a human observer with that of the computer with regard to saccade subtype. Fewer than 1 in 50 saccades were classified differently. In almost ail of these cases, the analysis program excluded events as artifacts that the human observer was willing to consider mildly anomalous saccades. For purposes of consistency, the results of the computer analysis were used. Two aspects of saccadic function were examined: saccadic metrics and the ability to learn to predict target motion. Saccadic metrics consisted of saccadic amplitude, peak velocity, and latency from target appearance to beginning of the saccade. We measured these variables for visually guided saccades during the initial and final pseudorandom segments oftarget motion as well as for predictive saccades during the middle, predictable segment. Visually guided saccades were defined as saccades made toward a visible target. Predictive saccades were defined as saccades made toward the location at which no target was currently visible but at which a target soon was likely to appear. Saccadic learning was evaluated in two ways: We measured the latency of the primary tracking saccade for each change in target position, irrespective of whether it was visually guided or predictive. The latencies were averaged over blocks of five target steps, to generate learning curves for saccadic latency as a function of target behavior (Fig. 3). The reduction in saccadic latency as a function of the pattern of target movement (learning curve data) was analyzed using a two-way analysis of variance with repeated measures on the target steps factor. We also counted the number of saccades that interrupted fixation in unsuccessful attempts to anticipate target motion after the target had again become unpredictable during the final pseudorandom segment. This provided a measure of the subject’s tendency to perseverate. Variables were compared between the OCD patients and the normals, using two-tailed t tests. In addition, the data for each variable were divided into means for leftward and rightward target motion, and compared for significant laterality differences using two-tailed paired r test. Finally, the relationship between subject characteristics, neurologic soft signs, and rating instrument scores with each of the parameters of eye movement function was examined using Pearson product-moment correlations. All data are expressed as mean f SD.

178 Fig. 2. Visually

guided saccades

1 Second

1 Second Panel A illustrates visually guided and predictive saccades that occur early during the segment of predictable target motion. Dashed line indicates target motron and solid line indicates eye movement. This tracing is from one of the obsessive-compulsive disorder patients. Note that eye movements follow target movements by approximately 200 ms. Panel B is taken from later during the predictable segment of target motion of the same subject. It demonstrates how saccadic latency decreases as the subject learns to predict the regular behaviorof the target. Note that the first 2 saccades begin at virtually the same time as the target moves.

Fig. 3. Decreasing

saccadic

latency with task learning PSEUDO-RANDOM ‘PREDICTABLE

200 3 1 5 z

loo-

E! 4

o-,

I 0

, 10

20

30

40

TARGET STEPS Saccadic latency of both obsessive-compulsivedisorder (OCD) patientsand control subjectsdecreased as they learned to anticipate the regular movement of the target. There was no difference in the saccadic learning of the 2 groups. Note the rapid return to baseline latencyaseach groupswitched to visuallyguided saccades when the target’s behavior became unpredictable.

179 Results Neuropsychiatric Measures. The OCD and normal control groups did not significantly differ in age, sex, or cerebral dominance (Table 1). The severity scores for the OCD patients on the Y-BOCS for obsessions was 11.1 k6.7, and for compulsions 10.5 f 4.5. The total BPRS score of the OCD patients was significantly higher than that of the controls (Table 1). OCD patients also had significantly more neurologic soft signs than normal controls (Table 1). Table 1. Subject characteristics

and neuropsychiatric Normals

Mean

Measure Number

12

Age iyr)

38.2

Sex:

SD

Hemispheric

dominance:

L/R

9.9

1 o/2

NES

1.8

Y-BOCS Obsessions Compulsions

21.9

BPRS

Mean

SD

Significance

7.8

NS

0

6/6

M/F

measures

OCD patients

1.9

5.6

36.5 513

NS (x*1

612

NS ($1

8.9

4.6

p < 0.001

11.1 10.5

6.7 4.5

-

31.6

a.8

p c 0.01

Note. OCD = obsessive-compulsive disorder. NES = Neurological Evaluation Obsessive-Compulsive Scale. BPRS = Brief Psychiatric Rating Scale.

Scale. Y-BOCS

= Yale-Brown

Smooth Pursuit Task. There were no significant differences between normals and OCD patients in smooth pursuit gain, overall frequency of intrusive saccades, and frequency of each saccade subtype (Table 2). When each of the above variables was divided into rightward and leftward eye and target motion and analyzed separately, there were no significant laterality differences either among the controls or OCD patients (data not included in table). Table 2. Smooth pursuit eye movements Normals Measure Smooth

pursuit

gain

Total saccades Frequency (min-1) Amplitude to) Catch-up

OCD patients

Mean

SD

Mean

SD

Sianificance

0.942

0.056

0.960

0.082

NS

1.74

33 0.58

64.4 1.50

35 0.66

NS NS

(min-1) (“I

31.8 1.51

11 0.34

21.5 1.44

11 0.14

p < 0.06 NS

saccades (min-1) (“)

28.9 2.06

13 1.1

20.1 1.35

21 1.2

NS NS

8.33 2.34

12 1.7

3.37 2.27

5.3 1.2

NS NS

80.9

saccades

Frequency Amplitude Anticipatory Frequency Amplitude

Square wave jerks Frequency (min-1) Amplitude to ) Note. OCD = obsessive-compulswe

disorder.

180 Saccadic Metrics. Table 3 displays the quantitative results for saccade metrics. During the initial pseudorandom segment of target motion, both OCD patients and control subjects made accurate visually guided saccades, and there were no differences between their peak saccadic velocities. However, OCD patients made visually guided saccades with a significantly shorter latency than controls did during the initial pseudorandom segment. During the final pseudorandom segment, there were no significant differences between the saccade metrics of OCD patients and control subjects. OCD patients made significantly larger amplitude and more accurate predictive saccades than controls during the middle predictable segment of target movement. As would be expected on the basis of their larger amplitude, there was a trend for these saccades to have a higher peak velocity among OCD patients. There were no significant differences in the latency, accuracy, or peak velocity of leftward or rightward saccades during any of the conditions of target motion among either control or OCD groups (data not included in table). Table

3. Saccadic

metrics Normals Mean

Measure

OCD patients SD

Mean

SD

Significance

Visually guided saccades Initial

pseudorandom

Amplitude

Peak velocity Latency

segment 20.3

co) t”/sec

f ms)

Final pseudorandom Amplitude Latency

20.1

1.8

NS

59

516

44

NS

212

17

186

20

t = 3.20, p < 0.01

segment

IO I

Peak velocity

1.4

519

18.9 i”/seci

i ms 1

1.7

19.7

1.1

NS

478

44

502

51

NS

212

23

193

25

NS

Predictive saccades Middle

predictable

Amplitude

1’

Peak velocitv

segment

)

16.6 i”/sec)

Note. OCD = obsessive-compuhve

2.7

421

80

19.1 481

1.3 49

t= 2.47, ~~0.05 t = 1.90, p < 0.10

disorder

Saccadic Learning. For the combined groups of OCD patients and controls, there was a highly significant reduction in saccadic latency as a function of predictable target motion. Both groups learned to anticipate regular changes in target position (multivariate analysis of variance, main effect for target step factor: F = 26. I; df = 9, 162;~ < 0.0001). There was no significant interaction, however, between thediagnostic group factor and the target step factor as would be expected if one group learned to anticipate better than the other (F 0.4; df = 9, 162; NS). Fig. 3 reveals that not only did both groups learn to anticipate target motion when it was adaptive to do so, but also neither group perseverated in this response once the target returned to pseudorandom behavior. This is confirmed by the observation that there was no significant difference between OCD and normal subjects in the number of q

181 saccadic attempts to anticipate target motion during the final pseudorandom segment when accurate anticipation was impossible (OCD patients: 2.6 + 2.4 saccades; controls: 2.5 + 2.6 saccades). Correlations. There were no significant correlations between subject characteristics, neurologic soft signs, BPRS scores, or Y-BOCS score and any of the parameters of eye movement function. Power Calculations. Since we found little significant difference in eye movements between OCD patients and controls, we examined the possibility that the failure to find differences might be due to a type II statistical error. We performed power calculations to determine if our sample was large enough to detect differences between OCD patients’ and normals’ smooth pursuit gain, if the OCD patients had a level of impairment comparable to that reported among schizophrenic patients (gain 0.86) (Levin et al., 1988). We found that there was less than a 10% chance of failing to detect this degree of impairment (Bartko et al., 1988). q

Discussion This study examines the relationship between eye movement performance and neurologic soft signs in a group of adult OCD patients. Our principal finding is that OCD patients performed at least as well as normal controls on a variety of tests of oculomotor function, despite showing a significantly increased number of neurologic soft signs. In no instance were any of the parameters of eye movement performance significantly impaired among OCD patients when compared with the eye movements of normal controls. There were no significant differences in laterality when rightward eye movements were compared with leftward movements, either between groups or within groups. The increased neurologic soft signs scores we found among OCD patients are consistent with previously reported results in larger samples (Denckla, 1988; Hollander et al., 1990). We found no correlations between the number of soft signs and any of the parameters of eye movement performance. These results indicate soft neurologic signs and eye movement dysfunction are not inextricably linked and probably reflect different underlying brain dysfunctions, at least in OCD. Only two of the eye movement parameters we measured differed significantly between OCD patients and normal controls. In both of these parameters (saccadic latency during the initial pseudorandom segment and amplitude of predictive saccades during the predictable segment of target motion), the OCD patients performed better than the normal controls. That two out of the many parameters compared are significantly different may represent a chance finding. However, on nearly all the measures of eye movement function, OCD patients had a superior mean score compared with that of the normal controls, although these differences did not reach statistical significance. One possible explanation for this slightly superior saccadic performance is that fluoxetine itself improves oculomotor function. Unfortunately, there have been no examinations of the effects of fluoxetine on saccades. Other possible explanations include: (1) whatever cerebral dysfunction produces OCD also leads to a slight improvement in oculomotor function, (2) changes in brain function

182 secondary to the symptoms of OCD (e.g., increased arousal) improve eye movements. If future studies confirm the presence of slightly superior oculomotor performance among patients with OCD, then further work could determine the effects of fluoxetine on eye movements and investigate the relationship of eye movements to other measures of cerebral function. The utility of eye movements analysis for the understanding of brain function in neuropsychiatric disease ultimately depends on our knowledge of the functional neuroanatomy of saccades and smooth pursuit. The control of smooth pursuit eye movements is known to involve at least three cortical and several subcortical regions (Tusa and Ungerleider, 1988). The widely dispersed nature of these circuits makes smooth pursuit performance of limited value for the precise localization of brain dysfunction. On the other hand, since so many brain regions are involved in the control of smooth pursuit, the excellence with which OCD patients execute these movements makes it extremely unlikely that they suffer from any diffuse or global cerebral dysfunction. The functional neuroanatomy of saccadic eye movements is more localized than that of smooth pursuit, particularly for saccades based on anticipation and learning. Alexander et al. (1986) have described an oculomotor control circuit linking prefrontal cortex, basal ganglia, and thalamus. This circuit seems to be particularly important in the control of nonvisually guided saccades based on memory or learning (Hikosaka, 1990). This basal ganglia-thalamocortical oculomotor loop is one of a family of parallel circuits described by Alexander et al. (1986). Each of these circuits serves specific regions of frontal or limbic cortices. This family of basal ganglia-thalamocortical circuits has been suggested as sites of dysfunction in both OCD (Rapoport and Wise, 1988) and schizophrenia (Robbins, 1990). However, while the cortical portions of these circuits appear to hyperactive in OCD (Swedo et al., 19896; Baxter et al., 1990), they appear to be hypoactive in schizophrenia (Buchsbaum, 1990). These differences in cortical metabolism at rest correspond to the differences in oculomotor performance when OCD and schizophrenic patients are compared to normals. Further research will be required to determine if the decreased cerebral metabolism observed in schizophrenic patients is related to impaired eye movements or the increased cortical metabolism observed in OCD patients is related to their slightly superior performance. Finally, we attempted to assess hemispheric asymmetry by comparing rightward with leftward eye movements. Previous reports have associated lateralized lesions in neurological patients with dysfunction of eye movements in the ipsilateral direction (Sharpe et al., 1979; Thurston et al., 1988). These lesions have involved relatively large areas of cortex. Our findings of no lateralized differences in the eye movements of OCD patients provide no support for the idea that there is lateralized cerebral dysfunction in OCD. Our results do not exclude the possibility of lateralization of dysfunction in OCD, however, since the region(s) of dysfunction could be small and distinct from structures involved in the control of eye movements. In addition to those discussed above, there are several other constraints on our conclusions: (1) Most of the OCD patients were receiving fluoxetine, whereas the normals were all free of medication. It is possible that there may be state-dependent oculomotor impairment in OCD that improves with treatment or, as already men-

183 tioned, that fluoxetine itself affects eye movements. (2) Our sample size is small, and despite the results of a power calculation suggesting that we could detect decreased smooth pursuit gain if it had been present, other aspects of eye movement may have been subject to type II errors. (3) The range of eye movement tasks we used was limited, and the tasks were relatively simple. More complex tasks, such as antisaccade and remembered-target tasks, place a greater demand on higher cortical functions, and may elicit subtle deficits in eye movement performance among patients with OCD. Acknowledgment. This work Research Service.

was supported

in part

by a grant

from

the VA Medical

References Abel, L. and Ziegler, A. Smooth pursuit eye movements in schizophrenics-What constitutes quantitative assessment? Biological Psychiatry, 24141-761, 1988. Alexander, G.E.; DeLong, M.R.; and Strick, P.L. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annual Review of Neurosciences, 9:357381, 1986. American Psychiatric Association. DSM-III-R: Diagnostic and Statistical Manual of Mental Disorders. 3rd ed., revised. Washington, DC: Author, 1987. Bartko, J.J.; Pulver, A.E.; and Carpenter, W.T., Jr. The power of analysis: Statistical perspectives, Part 2. Psychiatry Research, 23:301-309, 1988. Baxter, L.R. Brain imaging as a tool in establishing a theory of brain pathology in obsessive compulsive disorder. Journal of Clinical Psychiatry, 5 l(2, Suppl.):22-25, 1990. Baxter, L.R.; Schwartz, J.M.; Guze, B.H.; Bergman, K.; and Szuba, M.P. PET imaging in obsessive compulsive disorder with and without depression. Journal of Clinical Psychiatry, 51(4, Suppl.):61-70, 1990. Behar, D.; Rapoport, J.L.; Berg, C.; Denckla, M.B.; Mann, L.; and Cox, C. Computerized tomography and neuropsychological test measures in adolescents with obsessive-compulsive disorder. American Journal of Psychiatry, 141:363-369, 1984. Brown, R.G., and Marsden, CD. Cognitive function in Parkinson’s disease: From description to theory. Trends in Neuroscience, 13:21-28, 1990. Buchanan, R.W., and Heinrichs, D.W. The Neurological Evaluation Scale (NES): A structured instrument for the assessment of neurological signs in schizophrenia. Psychiatry Research, 27~335-350, 1988. Buchsbaum, M.S. The frontal lobes, basal ganglia, and temporal lobes as sites for schizophrenia. Schizophrenia Bulletin, 16:379-390, 1990. Capstick, N., and Seldrup, U. Obsessional states: A study in the relationship between abnormalities occurring at birth and subsequent development of obsessional symptoms. Acta Psychiatrica Scandinavica. 561427-439, 1977. Cipparone, L.; Ginanneschi, A.; Degl’Innocenti, F.; and Porzio, P. Electra-oculographic routine examination in Parkinson’s disease. Acta Neurologica Scandinavica, 77:6-I 1, 1988. Cummings, J.L. Subcortical dementia: Neuropsychology, neuropsychiatry and pathophysiology. British Journal of Psychiatry, 49:682-697, 1986. DeJong, J.D., and Jones, G.M. Akinesia, hypokinesia, and bradykinesia in the oculomotor system of patients with Parkinson’s disease. Experimental Neurology, 32:56-68, 197 I. Denckla, M.B. Neurological examination. In: Rapoport, J.L., ed. Obsessive Compulsive Disorder in Children and Adolescents. Washington, DC: American Psychiatric Press, 1988. Diamond, B.M. Neuropsychology of obsessive compulsive disorders. Presented at the 43rd Annual Meeting of the Society of Biological Psychiatry, 1988. Flor-Henry, P.; Yeudall, L.T.; and Kiles, Z.J. Neuropsychological and power spectral EEG investigations of the obsessive compulsive syndrome. Biological Psychiatry, 14: I I9- 130, 1979.

184 Garber, H.J.; Ananth, J.V.; Chiu, L.C.; Griswold, V.J.; and Oldendorf, W.H. Nuclear magnetic resonance study of obsessive-compulsive disorder. American Journal of Psychiary, 146:1001-1005, 1989. Goodman, W.K.; Price, L.H.; Rasmussen, S.A.; Mazure, C.; Fleischmann, R.L.; Hill, C.L.; Heninger, G.R.; and Charney, D.S. The Yale-Brown obsessive compulsive scale: I. Development, use, and reliability. Archives of General Psychiatry, 46: 1006-1011, 1989. Grimshaw, L. Obsessional disorder and neurological illness. Journal of Neurology, Neurosurgery and Psychiatry, 21~229-23 I, 1964. Hikosaka, 0. Role of basal ganglia in saccades. Review of Neurology, 145:580-586, 1990. Hollander, E.; Schiffman, E.; Cohen, B.; Rivera-Stein, M.A.; Rosen, W.; and Gorman, J.W. Signs of central nervous system dysfunction in obsessive-compulsive disorder. Archives of General Psychiatry, 41~27-32, 1990. Holzman, P.S. Recent studies of psychophysiology in schizophrenia. Schizophrenia Bulletin, 13:49-75, 1987. Hommer, D.W.; Radant A.D.; and Nickoloff S.E. Saccadic eye movements in schizophrenia: Evidence for impaired sensorimotor gating of internal representations. Society of Neuroscience Abstracts, 16(Part 2): 1348, 1990. Inouye, E. Similar and dissimilar manifestation of obsessive-compulsive neurosis in monozygotic twins. American Journal of Psychiatry, 121:1171-1175, 1965. Insel, T.; Donnelly, E.; Lalakea, M.; Alterman, I.S.; and Murphy, D.L. Neurological and neuropsychological studies of patients with obsessive-compulsive disorder. Biological Psychiatry, 18:741-751, 1983. Jenike, M.A.; Baer, L.; Summergrad, P.; Weilburg, J.B.; Holland, A.; and Seymour, R. Obsessive-compulsive disorder: A double-blind, placebo-controlled trial of clomipramine in 27 patients. American Journal of Psychiatry, 146: 1328-1330, 1989a. Jenike, M.A., and Brotman, A.W. The EEG in obsessive compulsive disorder. Journal of Clinical Psyciahtry, 45: 122-124, 1984. Jenike, M.A.; Buttolph, L.; Baer, L.; Ricciardi, J.; and Holland, A. Open trial of fluoxetine in obsessive-compulsive disorder. American Journal of Psychiarry, 146:909-910, 1989b. Karno, M.; Golding, J.M.; Sorenson, S.B.; and Burnam, M.A. The epidemiology of obsessive-compulsive disorder in five US communities. Archives of General Psychiatry, 45:1094-1099, 1988. Khanna, S. Obsessive-compulsive disorder: Is there a frontal lobe dysfunction? Biological Psychiatry, 24:602-613, 1988. Leigh, R.J.; Newman, S.A.; Folstein, S.E.; Lasker, A.G.; and Jensen, B.A. Abnormal oculomotor control in Huntington’s disease. Neurology, 33:1268-1275, 1983. Levin, S.; Luebke, A.; Zee, D.S.; Hain, T.C.; Robinson, D.A.; and Holzman, P.S. Smooth pursuit eye movements in schizophrenia: Quantitative measurement with the search coil technique. Journal of Psychiatric Research, 22: 195-206, 1988. Luxenberg, J.S.; Swedo, S.E.; Flame& M.F.; Friedland, R.P.; Rapoport, J.L.; and Rapoport, S.1. Neuroanatomical abnormalities in obsessive-compulsive disorder. American Journal of Psychiatry, 145:1089-1093, 1988. McKeon, J.; McGuffin, P.; and Robinson, P. Obsessive-compulsive neurosis following head injury: A report of 4 cases. British Journal of Psychiatry, 144: 190-192, 1984. Myers, J.K.; Weissman, M.M.; Tischler, G.L.; Holzer, L.E.; and Leaf, P.J. Six-month prevalence of psychiatric disorders in three communities. Archives of General Psychiatry, 41:959-967, 1984. Nordahl, T.E.; Benkelfat, C.; Semple, W.E.; Gross, M.; King, A.C.; and Cohen, R.M. Cerebral glucose metabolic rates in obsessive-compulsive disorder. Neuropsychopharmacology, 2:23-28, 1989. Overall, J.E., and Gorham, D.R. Brief Psychiatric Rating Scale. Psychological Reports, 10:799-8 12, 1962.

185 Pauls, D.L.; Towbin, K.E.; Leckman, J.F.; Zahner, G.E.; and Cohen, D.J. Gilles de la Tourette’s and obsessive-compulsive disorder. Archives of General Psychiatry, 43: 1180- 1183, 1986. Rapoport, J.L. Obsessive Compulsive Disorder in Children and Adolescents. Washington, DC: American Psychiatric Press, 1989. Rapoport, J.L., and Wise, S.P. Obsessive-compulsive disorder: Evidence for basal ganglia dysfunction. Psychopharmacology Bulletin, 24:38O-384, 1988. Robbins, T.W. The case for frontostriatal dysfunction in schizophrenia. Schizophrenia Bulletin, 16:403-412, 1990. Rosen, W.G.; Hollander, E.; Stannick, V.; and Liebowitz, M.R. Test performance variables in obsessive-compulsive disorder. Journal of Clinical and Experimental Neuropsychology, 10:73, 1988. Schilder, P. The organic background of obsessions and compulsions. American Journal of Psychiatry, 94: 1397, 1938. Sharpe, J.A.; Lo, A.W.; and Rabinovitch, H.E. Control of the saccadic and smooth pursuit systems after cerebral hemidecortication. Bruin, 102:387-403, 1979. Siever, L.J., and Coursey, R.D. Biological markers for schizophrenia and the biological high risk approach. Journal of Nervous and Mental Disease, 1734-16, 1985. Siever, L.J.; Insel, T.R.; Hamilton, J.; Nurnberger, J.; Alterman, I.; and Murphy, D.L. Eye-tracking, attention and amphetamine challenge. Journal of Psychiatric Research, 21: 129135, 1986. Stahl, S.M. Basal ganglia neuropharmacology and obsessive-compulsive disorder: The obsessive-compulsive disorder hypothesis of basal ganglia dysfunction. Psychopharmacology Bulletin, 24:370-374, 1988. Swedo, SE.; Rapoport, J.L.; Cheslow, B.S.; Leonard, H.L.; and Ayoub, E.M. High prevalence of obsessive-compulsive symptoms in patients with Sydenham’s chorea. American Journal of Psychiatry, 1461246249, 1989a. Swedo, SE.; Schapiro, M.B.; Grady, C.L.; Cheslow, D.L.; Leonard, H.L.; Kumar, A.; Friedland, R.; Rapoport, S.I.; and Rapoport, J.L. Cerebral glucose metabolism in childhoodonset obsessive-compulsive disorder. Archives of General Psychiatry, 46:518-523, 19896. Thurston, S.E.; Leigh, R.J.; Crawford, T.; Thompson, A.; and Kennard, C. Two distinct deficits of visual tracking caused by unilateral lesions of cerebral cortex in humans. Annals of Neurology, 23:266-273, 1988. Tian, J.R.; Zee, D.S.; Folstein, S.E..; and Lasker, A.G. Predictive saccades in Huntington’s Disease (HD). Society of Neuroscience Abstracts, 16(Part 2): 1083, 1990. Tusa, R.J., and Ungerleider, L.G. Fiber pathways of cortical areas mediating smooth pursuit eye movements in monkeys. Annals of Neurology, 23:174-183, 1988.