Abstract

The significance of cannabinoid signaling for human cognition and motor control is still poorly understood. Here, we have investigated acute behavioral effects of oral δ-9-tetrahydrocannabinol (THC) with oculomotor paradigms in 12 healthy human subjects. Compared to baseline testing: (i) THC increased latencies of reflexive visually guided saccades, while their accuracy was not affected; (ii) latencies of memory-guided saccades were unaffected, but THC modulated accuracy of these eye movements by increasing average gain and gain variability; (iii) frequency of anticipated memoryguided saccades and antisaccade errors was increased; (iv) the saccade amplitude/peak velocity relationships were not affected. These results show that THC acts on selected aspects of saccade control, namely spatial attention shifts, fine tuning of volitional saccades, spatial working memory and inhibition of inappropriate saccades. The pattern of effects suggests modulation of neuronal activity in substantia nigra pars reticulata and/or dorsolateral prefrontal cortex and sparing of the eye fields and the final motor pathway for saccades. Behaviorally, our findings reflect the distribution of CB-1 cannabinoid receptors in the human neocortex, basal ganglia and brainstem and provide evidence for participation of the cannabinoidergic system in high level control of saccades and associated cognitive functions. Saccadic eye movements may provide an objective measure of motor and cognitive effects of cannabinoids.

Introduction

The recent discovery of cannabinoid receptors and their endogenous ligands has prompted intense research on the pharmacological and physiological properties of the cannabinoidergic system (Felder and Glass, 1998; Ameri, 1999; Iversen, 2000). In the central nervous system, cannabinoidergic effects are mediated by type 1 cannabinoid receptors (CB-1 receptors) (Devane et al., 1988; Matsuda et al., 1990; Huestis et al., 2001). In the human brain, these receptors show high concentrations in substantia nigra pars reticulata, globus pallidus, cerebellar cortex, hippocampal formation, cingulate cortex, mediodorsal thalamus and dorsolateral prefrontal cortex (Herkenham et al., 1990; Glass et al., 1997). Evidence accumulates that dysfunction of endogenous cannabinoid signaling in some of these regions is involved in the pathogenesis of schizophrenia (Andreasson et al., 1987; Leweke et al., 1999; Dean et al., 2001) and Huntington’s disease (Glass et al., 1993, 2000; Richfield and Herkenham, 1994). It has further been speculated that this distribution pattern may also explain cognitive and psychedelic effects of exogenous cannabinoids (Herkenham et al., 1990; Glass et al., 1997). While there is indeed ample evidence that cannabinoids modulate hippocampus-dependent memory processes in humans (Sullivan, 2000), previous studies on cannabin-oidergic effects on vision, attention, short-term memory and motor control have been less conclusive (Solowij, 1998; Iversen, 2000). At present it is unclear whether such effects relate to modulation of neuronal activity in extrahippocampal brain regions with high densities of CB-1 receptors. Functional brain imaging in cannabis-intoxicated humans has not fully resolved this issue, since it has proven difficult to disentangle local blood flow and metabolic changes that are related to CB-1 receptor activation from secondary changes caused by cognitive and psychedelic effects of cannabinoids, e.g. hallucinations, imagery or mood changes (Volkow et al., 1996; Mathew et al., 1997; O’Leary et al., 2000).

In the present study we have investigated the behavioral significance of the cerebral CB-1 receptor distribution by studying acute effects of the main psychoactive ingredient of Cannabis sativa, δ-9-tetrahydrocannabinol (THC), on performance of healthy humans in three well-established saccade paradigms. These eye movements are controlled by an extensive network that includes parietal and frontal cortices, basal ganglia, brainstem and cerebellum and involve elementary cognitive functions such as attention, short-term memory and inhibition of reflexive behavior (Pierrot-Deseilligny et al., 1995; Leigh and Zee, 1999). Dysfunction in any of these regions yields distinct and localization-specific patterns of saccade disturbances (Pierrot-Deseilligny et al., 1995; Leigh and Zee, 1999). It is therefore possible to infer from changes in saccade parameters, derived from a single and relatively short eye movement recording session, the functional status of the diverse neuronal substrates of the network. This is a decisive advantage when examining acute effects of THC, since both subjective intoxication and plasma levels of THC and its metabolites change rapidly after drug intake (Iversen, 2000). Here we have examined whether (i) the cannabinoidergic system is involved in control of saccadic eye movements and associated cognitive functions and whether (ii) possible cannabis-induced changes in saccade parameters relate to the known distribution of CB-1 receptors in the human brain.

Materials and Methods

Subjects

Subjects were 12 volunteers (eight women and four men) with a mean age of 27.2 years (range 24–31 years) recruited from the staff and students of the medical faculty of the Humboldt University Berlin. None of the subjects had a history of neurological or psychiatric disorders. All subjects had some prior experience with cannabis, but did not use cannabis regularly. None of the subjects used cannabis during a period of 4 weeks preceding participation in the study. Prior to inclusion in the study, pregnancy and drug abuse was ruled out by means of standard urine tests. Written informed consent was obtained from all subjects before participation in the study, which was approved by the local ethics committee and governmental institutions (Bundes-Opiumstelle, Germany) and conducted in conformity with the Helsinki Declaration.

Study Protocol

Oculomotor testing was performed on two subsequent days at 10 a.m. On both days, food was not allowed from 0 a.m. to 12 a.m., but subjects were allowed to drink water at will. No nicotine and alcohol was allowed 1 day before and during the study. On the first day, subjects were tested with oculomotor paradigms during a recording session of 25–30 min duration, including two breaks of several minutes to avoid fatigue (control condition). On the second day, subjects received an oral dose of 10 mg THC (Marinol; Roxane Laboratories, USA) at 8 a.m. Oculomotor testing was started 2 h after THC intake (THC condition). Subjects quantified subjective effects of THC by means of an analog intoxication rating scale (0 = no intoxication, 10 = maximum intoxication) 0, 1, 2, 3, 4, 6 and 9 h after drug intake. In parallel, repeated blood samples were taken for determination of plasma levels of THC and its two main metabolites, i.e. 11-hydroxy-δ-9-tetrahydrocannabinol (11-OH-THC) and carboxy-tetrahydrocannabinol (THC-COOH). 11-OH-THC is a psychoactive metabolite with a potency comparable to that of THC and binds with high affinity to the CB-1 receptor (Ameri, 1999; Iversen, 2000). THC-COOH is the most abundant inactive THC metabolite. Its long persistence in blood and urine provides a sensitive measure of previous cannabis use, even several days after a single drug exposure (Ameri, 1999; Iversen, 2000). During the entire study, subjects were supervised by an experienced physician.

Eye Movement Recordings

Eye movements were recorded by horizontal infrared oculography of the right eye (Eyetracker; AMTech, Weinheim, Germany). Data were sampled at a frequency of 200 Hz. The system had a spatial resolution of 0.3° and a horizontal linear range of >20° bilaterally. Subjects were seated in complete darkness to avoid an external spatial reference frame. The subject’s head was fixed to the recording system by means of a bite bar with individual dental impressions. Visual cues were presented at a distance of 120 cm with a horizontal array of red light-emitting diodes (LEDs). LEDs were 5 cd/m2 in luminance. Calibration trials with two lateral targets at 15° eccentricity were performed regularly during recording sessions.

Paradigms

In the visually guided saccade task (Fig. 1A), subjects were instructed to fixate on a central fixation point. Then, the central fixation point was switched off and a visual cue was presented in a pseudorandom position at either 10, 12.5, 15, 17.5 or 20° eccentricity, right or left of the central fixation point. Subjects were instructed to move their eyes directly and as accurately as possible to the cue as soon as it appeared. After 1000 ms, the cue was switched off and the central fixation point was re-illuminated. After an inter-trial interval of 2500–3500 ms the next trial began. A total of 30 trials was performed.

In the memory-guided saccade task (Fig. 1B), subjects were instructed to fixate on a central fixation point. Then, a visual cue was presented for 100 ms in one of the positions used in the visually guided saccade task, while subjects continued fixating. After a delay of 5000 ms the central fixation point was switched off and subjects moved their eyes directly and as accurately as possible to the remembered cue position. After 3000 ms, the central fixation point was re-illuminated and after an inter-trial interval of 5000 ms the next trial began. A total of 30 trials was performed.

In the antisaccade task (Fig. 1C), subjects were instructed to fixate on a central fixation point. Then the central fixation point was switched off and after 200 ms (‘gap’) a visual cue was presented at 20° eccentricity right or left of the central fixation point. Subjects were instructed to move their eyes in the direction opposite to the cue as soon as it appeared. Subjects were given no instructions for saccade accuracy. After 1000 ms, the cue was switched off and the central fixation point was re-illuminated. After an inter-trial interval of 2500–3500 ms the next trial began. A total of 30 trials was performed.

Data Analysis

Oculomotor data were analyzed off-line using EYEMAP software (AMTech, Weinheim, Germany). In the visually guided saccade task and antisaccade task, the first saccade after cue onset was chosen for analysis (Pierrot-Deseilligny et al., 1991a). In the memory-guided saccade task, the first saccade after central fixation point offset was analyzed (Pierrot-Deseilligny et al., 1991b; Ploner et al., 1999). Additional saccades were generally small and frequently occurred several hundreds of milliseconds after the first saccades. Their frequency was similar in both conditions (control 32.8 ± 8.5% of trials; THC 34.9 ± 8.7%; P = 0.88, Wilcoxon signed ranks test) and they were thus not further analyzed. In the visually guided saccade task and in the memory-guided saccade task, latencies, peak velocities and horizontal amplitudes were measured. In the antisaccade task, the percentage of misdirected saccades, i.e. saccades directed towards the cue, was calculated. In addition, we quantified the frequency of memory-guided saccade anticipations, i.e. premature saccades performed during the memory delay of the memory-guided saccade task.

None of the oculomotor variables showed significant right–left differences. Hence, for statistical analysis, rightward and leftward saccades were pooled in each subject. Medians were used to describe a subject’s average saccade latency. For analysis of velocities of visually guided saccades and memory-guided saccades, a subject’s saccade peak velocities were plotted against saccade amplitudes. For each subject, regression analysis was then performed assuming an exponential relationship between both variables described by the formula: 

\[saccade\ peak\ velocity\ {=}\ \mathit{V}_{max}\ {\times}\ (1\ {-}\ e^{{-}amplitude/\mathit{C}})\]
where Vmax is the asymptotic peak velocity and C is a constant (Leigh and Zee, 1999). In the visually guided saccade task and memory-guided saccade task, saccade accuracy was expressed as gain, i.e. the ratio saccade amplitude/target eccentricity. Thus, a gain of 1 indicates a precise saccade, a gain 1 hypermetria and a gain <1 hypometria. In both tasks, medians were used to describe a subject’s average saccade gain (average hyper- or hypometria of saccade endpoints, i.e. systematic error of saccades) and interquartile ranges to describe a subject’s saccade gain variability (scatter of saccade endpoints, i.e. variable error of saccades) (Ploner et al., 1998, 1999). In the memory-guided saccade task, trials with premature saccades were excluded from analysis of saccade accuracy. Two-tailed Wilcoxon signed ranks tests were used for statistical comparison of oculomotor data between the control and THC conditions. Moreover, in the THC condition, Friedman’s analysis of variance and two-tailed Wilcoxon signed ranks tests were used for analysis of subjective intoxication and plasma levels of THC, 11-OH-THC and THC-COOH. Spearman’s correlation coefficients were used for correlation analysis.

Results

Subjective Intoxication and THC Plasma Levels

All subjects reported clear subjective intoxication (‘high’) in the THC condition, but remained fully cooperative during oculomotor testing. No serious unpleasant side-effects were reported. For subjective intoxication and plasma levels of THC, 11-OH-THC and THC-COOH, a peaked relationship to time after drug intake was found (Fig. 2). For all variables, analysis of variance revealed significant differences between different time points after drug intake (d.f. = 6, χ2 ≥ 41.8, P ≤ 0.0001). Two hours after drug intake, i.e. at the beginning of eye movement recordings, all variables were significantly different from zero (P ≤ 0.008 for all comparisons). No correlation was observed between individual subjective intoxication and individual plasma levels of THC, 11-OH-THC and THC-COOH 2 h after drug intake (P ≥ 0.16 for all correlation analyses).

Oculomotor Results

In the THC condition, latencies of visually guided saccades were slightly but significantly longer than in the control condition (control mean 200 ± 7.2 ms; THC mean 214 ± 6.4 ms; P = 0.008). For saccade velocities, regression analysis revealed very similar asymptotic peak velocities in both conditions (control mean Vmax 462 ± 17.5°/s; THC mean Vmax 469 ± 23.1°/s; P = 0.88). Likewise, amplitude constants did not significantly differ between control and THC conditions (control mean 7.38 ± 0.45; THC mean 8.03 ± 0.45; P = 0.64). From these values and inspection of the cumulative peak velocity/amplitude plots (Fig. 3) it is evident that no slowing of visually guided saccades occurred in the THC condition. Likewise, accuracy of visually guided saccades was not significantly affected by oral THC, as both average gain and gain variability did not statistically differ between the control and THC conditions (P = 0.2 and P = 0.062, respectively; Figs 4 and 5).

Unlike for visually guided saccades, we found no latency prolongation for memory-guided saccades. If at all, latencies of memory-guided saccades in the THC condition tended to be shorter than in the control condition (control mean 311 ± 15.3 ms; THC mean 301 ± 13.3 ms). However, this difference was not statistically significant (P = 0.31). As for visually guided saccades, regression analysis revealed almost identical asymptotic peak velocities (control mean Vmax 339 ± 19.0°/s; THC mean Vmax 330 ± 16.7°/s; P = 0.48) and amplitude constants in the control and THC conditions (control mean 7.14 ± 0.88; THC mean 7.07 ± 0.74; P = 0.58). In addition, fitted lines of cumulative peak velocity/amplitude plots for both conditions were largely congruent (Fig. 3), suggesting that oral THC induced no significant slowing of memory-guided saccades. However, in contrast to visually guided saccades, we found significant differences for saccade accuracy (Figs 4 and 5). Compared with the control condition, average gain of memory-guided saccades increased significantly in the THC condition (P = 0.041; Fig. 5A). A subject’s memory-guided saccades in the THC condition were thus significantly less hypometric than in the control condition. At the same time, gain variability increased significantly in the THC condition (P = 0.019; Fig. 5B). In other words, memory-guided saccades in the THC condition exhibited an increase in variable targeting errors compared to the control condition.

In the THC condition, subjects showed a significant increase in anticipatory saccades during the memory delay of the memory-guided saccade task (P = 0.011; Fig. 6). Obviously, subjects had difficulties in withholding the premature execution of a prepared saccade while fixating on the central fixation point during the memory delay. A similar deficit was found in the antisaccade task, where subjects showed a significant increase in antisaccade errors (P = 0.008; Fig. 7). It is evident that subjects had more difficulties in suppressing erroneous reflexive saccades to visual cues in the THC condition compared to the control condition. However, there was no significant correlation between frequency of antisaccade errors and frequency of anticipatory saccades in the memory-guided saccade paradigm (P = 0.29).

We calculated the individual differences between the THC and control conditions for all significant results (latency of visually guided saccades, average gain and gain variability of memory-guided saccades, memory-guided saccade anticipations and antisaccade errors). None of these variables correlated significantly with individual subjective intoxication or individual plasma levels of THC and 11-OH-THC 2 h after drug intake, i.e. at the beginning of eye movement recordings (P ≥ 0.1 for all correlation analyses).

Discussion

In the present study we have examined acute cerebral effects of oral THC in humans with oculomotor paradigms. THC did not affect reflexive and volitional saccades equally, but significantly modulated several distinct parameters of saccadic eye movements while sparing others, thus suggesting modulation of selected neuronal substrates of the network controlling saccades. These results provide evidence that the cannabinoid-ergic system is involved in control of saccadic eye movements. In the following we will discuss how our findings relate to previous studies of oculomotor control and to the distribution of CB-1 receptors in the human brain.

In the only previous study on cannabis effects on saccadic eye movements, Baloh et al. (1979) observed no acute effects of smoked THC (100 μg/kg body wt) on latencies, peak velocities and accuracy of visually guided saccades in regular cannabis users. While our findings are otherwise in agreement with this study, we found a significant effect of 10 mg oral THC on latencies of visually guided saccades. Since saccadic latencies result from a series of visual, attentional and premotor processes, latency prolongation may arise in any of these processing steps (Becker, 1989; Leigh and Zee, 1999). However, the lack of latency prolongation of memory-guided saccades shows that this effect cannot be due to delays in visual processing or to impaired disengagement of fixation. Indeed, the significant increase in anticipated memory-guided saccades during the memory delay rather suggests that fixation is more easily disengaged from the central fixation point in the THC condition. Latency prolongation of visually guided saccades may thus be due to slowing of attention shifts to the cue location or delayed programming of saccades. Unchanged latencies of memory-guided saccades are compatible with this hypothesis, since in this paradigm attention is already shifted to the remembered cue location and an eye movement is programmed before the central fixation point is turned off and the memory-guided saccade is executed (Bruce and Goldberg, 1985). Functional imaging studies have shown that both spatial attention shifts and programming of saccadic eye movements are controlled by overlapping anatomical networks that include cortical premotor regions for saccades, like the frontal eye field (FEF) for volitional saccades, the intraparietal sulcus for reflexive saccades, regions of prefrontal and parietal association cortex and the basal ganglia (Corbetta et al., 1998; Gitelman et al., 1999; Perry and Zeki, 2000). However, dysfunction of the FEF and intraparietal sulcus has been shown to yield increased latencies of memory-guided saccades (Pierrot-Deseilligny et al., 1991a,b; Rivaud et al., 1994; Gaymard et al., 1999), changes in the relationship between saccade amplitudes and peak velocities (Deng et al., 1987; Funahashi et al., 1993; Dias and Segraves, 1999) and increased hypometria of saccades (Pierrot-Deseilligny et al., 1991b; Rivaud et al., 1994; Gaymard et al., 1999; Ploner et al., 1999). Since we did not observe any of these effects, an action of oral THC on cortical premotor regions of the saccadic system appears unlikely. Increased latencies of visually guided saccades in our study may therefore rather reflect THC effects on association cortices or subcortical regions involved in spatial attention shifts. This is corroborated by the fact that even doses of up to 22 mg smoked THC do not affect smooth pursuit eye movements (Flom et al., 1977; Baloh et al., 1979), which are invariably impaired with FEF dysfunction (Rivaud et al., 1994). The lack of cortical premotor effects in our study is in agreement with the observation that CB-1 receptor density in motor and premotor cortices is considerably lower than in association cortex (Glass et al., 1997).

The selective changes in systematic and variable errors of memory-guided saccades suggest that they arise from modulation of parts of the saccadic system that are mainly devoted to volitional saccades. Since we found no evidence for THC effects on the FEF, the dorsolateral prefrontal cortex (DLPFC) and the basal ganglia are likely candidate regions (Pierrot-Deseilligny et al., 1995; Leigh and Zee, 1999; Hikosaka et al., 2000). While even large lesions of the DLPFC yield no significant systematic errors of memory-guided saccades (Funahashi et al., 1993; Ploner et al., 1999), hypometria of these eye movements in the presence of unimpaired visually guided saccades is a consistent feature of patients with basal ganglia pathology, e.g. in Parkinson’s disease (Lueck et al., 1990; Vidailhet et al., 1994; Vermersch et al., 1996, 1999). Much like in our THC-intoxicated subjects, stimulation of the subthalamic nucleus (STN) in these patients selectively increases average gain of memory-guided saccades, probably by modulating neuronal activity in the substantia nigra pars reticulata (SNpr) (Rivaud-Péchoux et al., 2000). This major output nucleus of the basal ganglia projects onto the superior colliculus (Leigh and Zee, 1999; Hikosaka et al., 2000) and shows a much higher CB-1 receptor density than any other region of the basal ganglia implicated in saccades (Herkenham et al., 1990; Glass et al., 1997). In the SNpr, CB-1 receptors are mainly present on axon terminals of neurons in striatum and STN, which in turn receive inputs from FEF and DLPFC (Breivogel and Childers, 1998; Ameri, 1999; Leigh and Zee, 1999; Hikosaka et al., 2000). These data and modulation of neuronal activity in the SNpr by cannabinoids and CB-1 receptor antagonists suggest an important SNpr-mediated role of the cannabinoidergic system in the fine tuning of motor control (Breivogel and Childers, 1998; Ameri, 1999). Our findings directly support this hypothesis by showing a possibly SNpr-mediated selective increase in average gain of memory-guided saccades. In addition, our results suggest that this modulating influence of cannabinoids mainly affects volitional movements.

Whether the SNpr also accounts for variable errors of memory-guided saccades is unclear, since this error type has to our knowledge not been investigated with basal ganglia pathology. It is therefore possible that increased variable errors of memory-guided saccades indicate loss of SNpr-mediated control of volitional eye movements, which may become noisy and less sharply tuned. However, behavioral and lesion studies have shown that systematic and variable errors of memory-guided saccades may also occur independently, with systematic errors arising in FEF or downstream areas of the saccadic system and variable errors reflecting spatial working memory processes in the DLPFC (White et al., 1994; Ploner et al., 1998, 1999; Wang, 2001). Within the human neocortex, the DLPFC contains the highest density of CB-1 receptors (Glass et al., 1997). Lesions of this region mainly increase variable errors of memory-guided saccades, while latencies and velocities are not affected (Funahashi et al., 1993; Ploner et al., 1999). It therefore appears possible that a THC-induced increase in variable errors of memory-guided saccades indicates additional impairment of spatial working memory functions of the DLPFC. The coexistence of spatial working memory deficits and CB-1 receptor anomalies in DLPFC of schizophrenics (Park and Holzman, 1992; Dean et al., 2001), THC-induced spatial working memory deficits in rats (Jentsch et al., 1997) and THC-induced modulation of prefrontal neurotransmitter systems (Auclair et al., 2000; Ferraro et al., 2001; Pistis et al., 2001) lend further support to this hypothesis and may point to a role of the cannabinoidergic system in the regulation of spatial working memory.

THC-induced increases in frequency of memory-guided saccade anticipations and antisaccade error rates were a prominent finding in our subjects and indicate an impairment both in withholding the premature execution of a programmed volitional saccade and in suppression of reflexive visually guided saccades. These inhibitory functions are mediated by the DLPFC and the SNpr by virtue of their inhibitory projections on the superior colliculus (Pierrot-Deseilligny et al., 1995; Everling and Fischer, 1998; Hikosaka et al., 2000). Dysfunction in either structure may lead to increased frequency of memory-guided saccade anticipations and increased antisaccade error rates (Hikosaka and Wurtz, 1985b; Pierrot-Deseilligny et al., 1991a,b; Everling and Fischer 1998). Moreover, as in our subjects, evidence from human patients suggests a partial independence of both deficits (Walker et al., 1998; Broerse et al., 2001). It is thus not possible to unequivocally infer from these parameters on dysfunction of either the DLPFC or the SNpr. Unimpaired smooth pursuit eye movements in THC-intoxicated subjects (Flom et al., 1977; Baloh et al., 1979) are compatible with both possibilities, as neither the DLPFC nor the SNpr are implicated in control of these eye movements (Leigh and Zee, 1999). Since both regions contain very high CB-1 receptor densities (Herkenham et al., 1990; Glass et al., 1997), an additive effect likewise appears possible. However, it is obvious from our findings that the cannabinoidergic system is involved in inhibitory control of inappropriate saccades to visual and remembered targets.

Unchanged amplitude/peak velocity relationships of visually guided and memory-guided saccades and normal accuracy of visually guided saccades argue against significant THC effects on the final common motor pathway of the saccadic system in midbrain, pons and cerebellum. Dysfunction of the brainstem reticular formation, the superior colliculi or the dorsal vermis of the cerebellum is usually accompanied by slowing of saccades (Hikosaka and Wurtz, 1985a; Takagi et al., 1998; Leigh and Zee, 1999). Moreover, dysfunction of the latter two structures or the cerebellar caudal fastigial nuclei yields impaired accuracy of both reflexive and volitional saccades (Hikosaka and Wurtz, 1985a; Robinson et al., 1993; Kanayama et al., 1994; Vahedi et al., 1995; Takagi et al., 1998). Our results thus complement the very low CB-1 receptor density in superior colliculi, brainstem reticular formation, pontine nuclei and deep cerebellar nuclei (Herkenham et al., 1990; Glass et al., 1997). Our findings, as well as the lack of any THC-induced impairment of smooth pursuit eye movements (Flom et al., 1977; Baloh et al., 1979), further suggest that cannabinoids do not significantly modulate the dorsal vermis of the cerebellum (Vahedi et al., 1995; Takagi et al., 2000). This is an unexpected result, as the cerebellar cortex generally contains a high density of CB-1 receptors (Herkenham et al., 1990; Glass et al., 1997). Since convincing behavioral evidence for cerebellar effects of THC in humans is not available, the absence of significant cerebellar oculomotor effects is difficult to interpret, but may point to regional differences in CB-1 receptor density in the cerebellar cortex.

In conclusion, the results from this study strongly suggest participation of the cannabinoidergic system in high level control of saccadic eye movements, in particular in spatial attention shifts, fine tuning of volitional saccades, spatial working memory and inhibition of inappropriate saccades. These findings can be explained by modulation of neuronal activity in SNpr and/or DLPFC and are consistent with the high density of CB-1 receptors in these regions. To the best of our knowledge, this is the closest brain–behavior relationship reported so far for the extrahippocampal components of the primate cannabinoidergic system. Saccadic eye movements may thus provide an oculomotor model that allows for further exploration of CB-1 receptor-mediated behavioral effects of cannabinoids in humans and non-human primates.

Figure 1.

Tasks used in the study with infrared oculographic recording traces. R, right; L, left; Stim, stimulus; Eye, eye movement recording. (A) Visually guided saccade task. (B) Memory-guided saccade task. (C) Antisaccade task. Asterisk indicates antisaccade error.

Figure 1.

Tasks used in the study with infrared oculographic recording traces. R, right; L, left; Stim, stimulus; Eye, eye movement recording. (A) Visually guided saccade task. (B) Memory-guided saccade task. (C) Antisaccade task. Asterisk indicates antisaccade error.

Figure 2.

Time course of subjective intoxication and plasma levels of THC and its metabolites. Dotted line, time of drug intake; gray bar, duration of eye movement recordings. (A) Subjective intoxication as quantified with an analog intoxication rating scale. (B) Plasma level of THC. (C) Plasma level of 11-OH-THC, the psychoactive metabolite of THC. (D) Plasma level of THC-COOH, the main inactive metabolite of THC. Data are presented as group means ± SE.

Figure 2.

Time course of subjective intoxication and plasma levels of THC and its metabolites. Dotted line, time of drug intake; gray bar, duration of eye movement recordings. (A) Subjective intoxication as quantified with an analog intoxication rating scale. (B) Plasma level of THC. (C) Plasma level of 11-OH-THC, the psychoactive metabolite of THC. (D) Plasma level of THC-COOH, the main inactive metabolite of THC. Data are presented as group means ± SE.

Figure 3.

Relationship between saccade amplitude and saccade peak velocity for visually guided saccades and memory-guided saccades in the control condition and THC condition. Pooled data from 12 subjects. Dots represent individual saccades. Fitted lines were calculated assuming an exponential relationship between both variables that progressively saturates towards asymptotic maximum peak velocities. (A) Visually guided saccades, control condition. (B) Visually guided saccades, THC condition. (C) Memory-guided saccades, control condition. (D) Memory-guided saccades, THC condition. Note that for both types of saccades the slope of the fitted lines is very similar in the control and THC conditions. THC causes no downward shift of the fitted lines, i.e. slowing of visually guided saccades or memory-guided saccades.

Figure 3.

Relationship between saccade amplitude and saccade peak velocity for visually guided saccades and memory-guided saccades in the control condition and THC condition. Pooled data from 12 subjects. Dots represent individual saccades. Fitted lines were calculated assuming an exponential relationship between both variables that progressively saturates towards asymptotic maximum peak velocities. (A) Visually guided saccades, control condition. (B) Visually guided saccades, THC condition. (C) Memory-guided saccades, control condition. (D) Memory-guided saccades, THC condition. Note that for both types of saccades the slope of the fitted lines is very similar in the control and THC conditions. THC causes no downward shift of the fitted lines, i.e. slowing of visually guided saccades or memory-guided saccades.

Figure 4.

Targeting errors of rightward visually guided saccades and memory-guided saccades expressed as gain in the control condition (white background) and THC condition (gray background). Example results from a single subject. Dots represent individual saccades. A gain of 1 indicates a precise saccade, a gain >1 hypermetria and a gain <1 hypometria. Note very similar targeting errors of visually guided saccades in both conditions. Note an increase in average gain (systematic error) and gain variability (variable error) of memory-guided saccades in the THC condition compared to the control condition.

Figure 4.

Targeting errors of rightward visually guided saccades and memory-guided saccades expressed as gain in the control condition (white background) and THC condition (gray background). Example results from a single subject. Dots represent individual saccades. A gain of 1 indicates a precise saccade, a gain >1 hypermetria and a gain <1 hypometria. Note very similar targeting errors of visually guided saccades in both conditions. Note an increase in average gain (systematic error) and gain variability (variable error) of memory-guided saccades in the THC condition compared to the control condition.

Figure 5.

Targeting errors of visually guided saccades and memory-guided saccades in the control condition (white bars) and THC condition (gray bars). Group results from 12 subjects. (A) Systematic error. Note that there is no significant difference between average gain of visually guided saccades, while there is a significant increase in average gain of memory-guided saccades in the THC condition compared to the control condition. (B) Variable error. Note that there is no significant difference between gain variability of visually guided saccades, while there is a significant increase in gain variability of memory-guided saccades in the THC condition compared to the control condition. Bars represent group means ± SE. *P < 0.05.

Figure 5.

Targeting errors of visually guided saccades and memory-guided saccades in the control condition (white bars) and THC condition (gray bars). Group results from 12 subjects. (A) Systematic error. Note that there is no significant difference between average gain of visually guided saccades, while there is a significant increase in average gain of memory-guided saccades in the THC condition compared to the control condition. (B) Variable error. Note that there is no significant difference between gain variability of visually guided saccades, while there is a significant increase in gain variability of memory-guided saccades in the THC condition compared to the control condition. Bars represent group means ± SE. *P < 0.05.

Figure 6.

Frequency of memory-guided saccade anticipations in the control condition (white bar) and THC condition (gray bar). Group results from 12 subjects. Bars represent group means ± SE. *P < 0.05.

Figure 6.

Frequency of memory-guided saccade anticipations in the control condition (white bar) and THC condition (gray bar). Group results from 12 subjects. Bars represent group means ± SE. *P < 0.05.

Figure 7.

Frequency of antisaccade errors in the control condition (white bar) and THC condition (gray bar). Group results from 12 subjects. Bars represent group means ± SE. *P < 0.05.

Figure 7.

Frequency of antisaccade errors in the control condition (white bar) and THC condition (gray bar). Group results from 12 subjects. Bars represent group means ± SE. *P < 0.05.

We are grateful to Markus Ploner for helpful comments on the manuscript. Special thanks are due to Anke Dirks for assistance during manuscript preparation. This work was supported by research grants from the Forschungsförderung der Charité to A.M.S. and C.J.P. and by grants from the Deutsche Forschungsgemeinschaft to C.J.P., F.O. and S.D. (Pl 248/2-1, GRK 423).

References

Ameri A (
1999
) The effects of cannabinoids on the brain.
Prog Neurobiol
 
59
:
315
–348.
Andreasson S, Allebeck P, Engstrom A, Rydberg U (
1987
) Cannabis and schizophrenia: a longtitudinal study of Swedish conscripts.
Lancet
 
2
:
1483
–1485.
Auclair N, Otani S, Soubrie P, Crepel F (
2000
) Cannabinoids modulate synaptic strength and plasticity at glutamatergic synapses of rat prefrontal cortex pyramidal neurons.
J Neurophysiol
 
83
:
3287
–3293.
Baloh RW, Sharma S, Moskowitz H, Griffith R (
1979
) Effects of alcohol and marijuana on eye movements.
Aviat Space Environ Med
 
50
:
18
–23.
Becker W (1989) Metrics. In: The neurobiology of saccadic eye movements (Wurtz R, Goldberg ME, eds), pp. 13–67. Amsterdam: Elsevier.
Breivogel CS, Childers SR (
1998
) The functional neuroanatomy of brain cannabinoid receptors.
Neurobiol Dis
 
5
:
417
–431.
Broerse A, Holthausen EAE, van den Bosch RJ, den Boer JA (
2001
) Does frontal normality exist in schizophrenia? A saccadic eye movement study.
Psychiat Res
 
103
:
167
–178.
Bruce CJ, Goldberg ME (
1985
) Primate frontal eye fields. I. Single neurons discharging before saccades.
J Neurophysiol
 
53
:
603
–635.
Corbetta M, Akbudak E, Conturo TE, Snyder AZ, Ollinger JM, Drury HA, Linenweber MR, Petersen SE, Raichle ME, Van Essen DC, Shulman GL (
1998
) A common network of functional areas for attention and eye movements.
Neuron
 
21
:
761
–773.
Dean B, Sundram S, Bradbury R, Scarr E, Copolov D (
2001
) Studies on [3H]CP-55940 binding in the human central nervous system: regional specific changes in density of cannabinoid-1 receptors associated with schizophrenia and cannabis use.
Neuroscience
 
103
:
9
–15.
Deng SY, Goldberg ME, Segraves ME, Ungerleider LG, Mishkin M (1987) The effect of unilateral ablation of the frontal eye fields on saccadic performance in the monkey. In: Adaptive processes in visual oculomotor systems (Keller EL, Zee DS, eds), pp. 201–208. Oxford: Pergamon Press.
Devane WA, Dysarz FA, Johnson MR, Melvin LS, Howlett AC (
1988
) Determination and characterization of a cannabinoid receptor in rat brain.
Mol Pharmacol
 
34
:
605
–613.
Dias EC, Segraves MA (
1999
) Muscimol-induced inactivation of monkey frontal eye field: effects on visually and memory-guided saccades.
J Neurophysiol
 
81
:
2191
–2214.
Everling S, Fischer B (
1998
) The antisaccade: a review of basic research and clinical studies.
Neuropsychologia
 
36
:
885
–899.
Felder CC, Glass M (
1998
) Cannabinoid receptors and their endogenous agonists.
Annu Rev Pharmacol Toxicol
 
38
:
179
–200.
Ferraro L, Tomasini MC, Gessa GL, Bebe BW, Tanganelli S, Antonelli T (
2001
) The cannabinoid receptor agonist WIN 55.212-2 regulates glutamate transmission in rat prefrontal cortex: an in vivo and in vitro study.
Cereb Cortex
 
11
:
728
–733.
Flom MC, Brown B, Adams AJ, Jones RT (
1977
) Alcohol and marijuana effects on ocular tracking.
Am J Optom Physiol Optics
 
53
:
764
–773.
Funahashi S, Bruce CJ, Goldman-Rakic PS (
1993
) Dorsolateral prefrontal lesions and oculomotor delayed response performance: evidence for mnemonic ‘scotomas’.
J Neurosci
 
13
:
1479
–1497.
Gaymard B, Ploner CJ, Rivaud-Péchoux S, Pierrot-Deseilligny C (
1999
) The frontal eye field is involved in spatial short-term memory but not in reflexive saccade inhibition.
Exp Brain Res
 
129
:
288
–301.
Gitelman DR, Nobre AC, Parrish TB, La Bar KS, Kim Y-H, Meyer JR, Mesulam MM (
1999
) A large-scale distributed network for covert spatial attention.
Brain
 
122
:
1093
–1106.
Glass M, Faull RL, Dragunow M (
1993
) Loss of cannabinoid receptors in the substantia nigra in Huntington’s disease.
Neuroscience
 
56
:
523
–527.
Glass M, Dragunow M, Faull RLM (
1997
) Cannabinoid receptors in the human brain: a detailed anatomical and quantitative autoradiographic study in the fetal, neonatal and adult human brain.
Neuroscience
 
77
:
299
–318.
Glass M, Dragunow M, Faull RL (
2000
) The pattern of neurodegeneration in Huntington’s disease: a comparative study of cannabinoid, dopamine, adenosine and GABA(A) receptor alterations in the human basal ganglia in Huntington’s disease.
Neuroscience
 
97
:
505
–519
Herkenham M, Lynn AB, Little MD, Johnson MR, Melvin LS, De Costa BR, Rice KC (
1990
) Cannabinoid receptor localization in brain.
Proc Natl Acad Sci USA
 
87
:
1932
–1936.
Hikosaka O, Wurtz RH (
1985
) Modification of saccadic eye movements by GABA-related substances. I. Effect of muscimol and bicuculline in monkey superior colliculus.
J Neurophysiol
 
53
:
266
–291.
Hikosaka O, Wurtz RH (
1985
) Modification of saccadic eye movements by GABA-related substances. II. Effect of muscimol in monkey substantia nigra pars reticulata.
J Neurophysiol
 
53
:
292
–308.
Hikosaka O, Takikawa Y, Kawagoe R (
2000
) Role of the basal ganglia in the control of purposive saccadic eye movements.
Physiol Rev
 
80
:
953
–978.
Huestis MA, Gorelick DA, Heishman SJ, Preston KL, Nelson RA, Moolchan ET, Frank RA (
2001
) Blockade of effects of smoked marijuana by the CB-1-selective cannabinoid receptor antagonist SR141716.
Arch Gen Psychiat
 
58
:
322
–328.
Iversen LL (2000) The science of marijuana. Oxford: Oxford University Press.
Jentsch JD, Andrusiak E, Tran A, Bowers MB, Roth RH (
1997
) δ9-Tetrahydrocannabinol increases prefrontal catecholaminergic utilization and impairs spatial working memory in the rat: blockade of dopaminergic effects with HA 966.
Neuropsychopharmacology
 
16
:
426
–432.
Kanayama R, Bronstein AM, Shallo-Hoffmann J, Rudge P, Husain M (
1994
) Visually and memory guided saccades in a case of cerebellar saccadic dysmetria.
J Neurol Neurosurg Psychiat
 
57
:
1081
–1084.
Leigh RJ, Zee DS (1999) The neurology of eye movements. Oxford: Oxford University Press.
Leweke FM, Giuffrida A, Wurster U, Emrich HM, Piomelli D (
1999
) Elevated endogenous cannabinoids in schizophrenia.
Neuroreport
 
10
:
1665
–1669.
Lueck CJ, Tanyeri S, Crawford TJ, Henderson L, Kennard C (
1990
) Antisaccades and remembered saccades in Parkinson’s disease.
J Neurol Neurosurg Psychiat
 
53
:
284
–288
Mathew RJ, Wilson WH, Coleman RE, Turkington TG, De Grado TR (
1997
) Marijuana intoxication and brain activation in marijuana smokers.
Life Sci
 
23
:
2075
–2089.
Matsuda LA, Lolait SJ, Bownstein MJ, Young AC, Bonner TI (
1990
) Structure of a cannabinoid receptor and functional expression of the cloned cDNA.
Nature
 
346
:
561
–564.
O’Leary DS, Block RI, Flaum M, Schultz SK, Boles Ponto LL, Watkins GL, Hurtig RR, Andreasen NC, Hichwa RD (
2000
) Acute marijuana effects on rCBF and cognition: a PET study.
Neuroreport
 
11
:
3835
–3841.
Park S, Holzman PS (
1992
) Schizophrenics show spatial working memory deficits.
Arch Gen Psychiat
 
55
:
105
–116.
Perry RJ, Zeki S (
2000
) The neurology of saccades and covert shifts of attention. An event-related fMRI study.
Brain
 
123
:
2273
–2288.
Pierrot-Deseilligny C, Rivaud S, Gaymard B, Agid Y (
1991
) Cortical control of reflexive visually guided saccades.
Brain
 
114
:
1473
–1485.
Pierrot-Deseilligny C, Rivaud S, Gaymard B, Agid Y (
1991
) Cortical control of memory-guided saccades in man.
Exp Brain Res
 
83
:
607
–617.
Pierrot-Deseilligny C, Rivaud S, Gaymard B, Müri RM, Vermersch A-I (
1995
) Cortical control of saccades.
Ann Neurol
 
37
:
557
–567.
Pistis M, Porcu G, Melis M, Diana M, Gessa GL (
2001
) Effects of cannabinoids on prefrontal neuronal responses to ventral tegmental area stimulation.
Eur J Neurosci
 
14
:
96
–102.
Ploner CJ, Gaymard B, Rivaud S, Agid Y, Pierrot-Deseilligny C (
1998
) Temporal limits of spatial working memory in humans.
Eur J Neurosci
 
10
:
794
–797.
Ploner CJ, Rivaud-Péchoux S, Gaymard BM, Agid Y, Pierrot-Deseilligny C (
1999
) Errors of memory-guided saccades in humans with lesions of the frontal eye field and the dorsolateral prefrontal cortex.
J Neurophysiol
 
82
:
1086
–1090.
Richfield EK, Herkenham M (
1994
) Selective vulnerability in Huntington’s disease: preferential loss of cannabinoid receptors in lateral globus pallidus.
Ann Neurol
 
36
:
577
–584.
Rivaud S, Müri RM, Gaymard B, Vermersch A-I, Pierrot-Deseilligny C (
1994
) Eye movement disorders after frontal eye field lesions in humans.
Exp Brain Res
 
102
:
110
–120.
Rivaud-Péchoux S, Vermersch A-I, Gaymard B, Ploner CJ, Bejjani BP, Damier P, Demeret, S, Agid Y, Pierrot-Deseilligny C (
2000
) Improvement of memory-guided saccades in parkinsonian patients by high frequency subthalamic nucleus stimulation.
J Neurol Neurosurg Psychiat
 
68
:
381
–384.
Robinson FR, Straube A, Fuchs AF (
1993
) Role of the caudal fastigial nucleus in saccade generation. II. Effects of muscimol inactivation.
J Neurophysiol
 
70
:
1741
–1758.
Solowij N (1998) Cannabis and cognitive functioning. Cambridge: Cambridge University Press.
Sullivan JM (
2000
) Cellular and molecular mechanisms underlying learning and memory impairments by cannabinoids.
Learn Mem
 
7
:
132
–139.
Takagi M, Zee DS, Tamargo RJ (
1998
) Effects of lesions of the oculomotor vermis on eye movements in primate: saccades.
J Neurophysiol
 
80
:
1911
–1931.
Takagi M, Zee DS, Tamargo RJ (
2000
) Effects of lesions of the oculomotor vermis on eye movements in primate: smooth pursuit.
J Neurophysiol
 
83
:
2047
–2062.
Vahedi K, Rivaud S, Amarenco P, Pierrot-Deseilligny C (
1995
) Horizontal eye movement disorders after posterior vermis infarctions.
J Neurol Neurosurg Psychiat
 
58
:
91
–94.
Vermersch A-I, Müri RM, Rivaud S, Vidailhet M, Gaymard B, Agid Y, Pierrot-Deseilligny C (
1996
) Saccade disturbances after bilateral lentiform nucleus lesions in humans.
J Neurol Neurosurg Psychiat
 
60
:
179
–184.
Vermersch A-I, Gaymard BM, Rivaud-Péchoux S, Ploner CJ, Agid Y, Pierrot-Deseilligny C (
1999
) Memory-guided saccade deficit after caudate nucleus lesion.
J Neurol Neurosurg Psychiat
 
66
:
524
–527.
Vidailhet M, Rivaud S, Gouider-Khouja N, Pillon B, Bonnet A-M, Gaymard B, Agid Y, Pierrot-Deseilligny C (
1994
) Eye movements in parkin-sonian syndromes.
Ann Neurol
 
35
:
420
–426.
Volkow ND, Gillespie H, Mullani N, Tancredi L, Grant C, Valentine A, Hollister L (
1996
) Brain glucose metabolism in chronic marijuana users at baseline and during marijuana intoxication.
Psychiat Res
 
67
:
29
–38.
Walker R, Husain M, Hodgson TL, Harrison J, Kennard C (
1998
) Saccadic eye movement and working memory deficits following damage to the human prefrontal cortex.
Neuropsychologia
 
36
:
1141
–1159.
Wang X-J (
2001
) Synaptic reverberation underlying mnemonic persistent activity.
Trends Neurosci
 
24
:
455
–463.
White JM, Sparks DL, Stanford TR (
1994
) Saccades to remembered target locations: an analysis of systematic and variable errors.
Vision Res
 
34
:
79
–92.