Read at the 124th annual meeting of the American Psychiatric Association, Washington, D.C., May 3 -7, 1971.

This work was supported in part by Public Health Service grants MH-I3358 and M H-I9172 from the National Institute of Mental Health and by a contract from the New York State Narcotic Addiction Control Commission.

In the present experiment, tests of short-term memory, reaction time, and time estimation were employed. While the latter two have been studied previously, no experiment has compared the differential effects of marijuana on performance in both the visual and auditory modalities, as was done in the present situation. In addition, heart rate and blood sugar were measured. (The latter has been used in the past as an index of hunger [31.) Finally, the relation of cortical activity, as reflected by electroencephalogram (EEG), to behavioral change was studied.

 

Method

 

Subjects. Ten first-year medical students served as paid subjects. Their mean age was 22 years, six months (standard deviation six months). The subjects were selected from a larger sample on the basis of an interview and EEG and had similar histories of marijuana use. Each subject participated in six sessions: two placebo, two "high" dose sessions, and two "low" dose sessions, the order of which was randomized.

 

Procedure. Each subject was brought to the EEG laboratory and made comfortable on a couch in a supine position. Electrodes were applied for EEG recording and the subject was given appropriate instructions. The protocol used is shown in table 1. Double-blind procedures were employed, and minimal subject-experimenter interaction occurred.

 

Marijuana. The marijuana, supplied by the National Institute of Mental Health, had a 1.5 percent delta-9-tetrahydrocant1abinol (THC) content. Two doses were used as follows:

Oregano was used as filler material and as placebo because it is similar to marijuana in appearance, odor, and taste.

 

Each subject smoked the first cigarette at his own rate but was paced as to the length of draw and maintenance of the inhalation. There were no significant differences in number of puffs per cigarette for each subject or mean smoking time among sessions.

Tasks. The three tasks were presented in prearranged random order. For short-term memory, the subject was presented with a consonant trigram, e.g., DKF, and required to recall that trigram either immediately or after filled retention intervals of 0, 6, 12, or 18 seconds. One list of 16 trigrams was used at each session, four trigrams at each retention interval. Trigrams were presented on a preprogrammed auditory tape.

 

Simple reaction time in both the visual and auditory modalities was measured by requiring the subject to respond as rapidly as possible to either a single auditory stimulus or a single visual stimulus. There were 24 auditory and 24 visual trials counterbalanced for practice and fatigue effects; the preparatory interval was varied.

 

For time estimation, the method of reproduction in which the subject has to reproduce the duration of a prior stimulus was used. As in reaction time, time estimation was measured in both the visual and auditory modalities by the presentation of either a single visual or a single auditory stimulus. Three stimulus durations were employed: one second, two seconds, and five seconds. There were 24 auditory and 24 visual trials, and each stimulus duration occurred eight times (randomly) in each modality.

 

The EEG was recorded from the right occipital-vertex derivation, using a Grass polygraph. The magnetic tape records were analyzed by period analysis, using an IBM 1800 digital computer, 320 samples per second in 20-second epochs.

 

Results

 

in the high-dose conditions and the effect was greater with longer delays in reporting. the trigrams. The data were analyzed by a three-way analysis of variance, which indicated that the effects of sessions (Fs.45 = 3.89), retention intervals (F3.27 = 41.20), and the interaction between sessions and retention intervals (F15,135 = 2.01) were significant beyond the .05 level (figure 1).

 

Reaction time. As with short-term memory, low-dose marijuana did not affect auditory or visual reaction time, but reaction time in both modalities was longer (Fs.39 = 4.08; p < .05) with higher doses (figure 2). Auditory reaction time was shorter than visual (F1.8 = 16.28; p < .01), but this finding is not specific to marijuana.

 

Moment-to-moment fluctuation in reaction time did occur but with equal frequency in each group, including placebo.

 

Time estimation. Estimation of time was unaffected by either dose of marijuana; typically, however, smokers report that the perception of time is changed under the influence of marijuana. That such was not the case here may be related to the method of measurement used—that of reproduction; this method has been found not to be sensitive to changes in time estimation (6). ,

 

Blood sugar. Blood sugar levels were not significantly affected by marijuana; however, they did increase significantly over time (F,9 = 8.19; p < .05). When questioned at the end of each session, one-half of the subjects indicated increased hunger, the remaining subjects indicated no increase. As blood sugar was measured once before smoking and once at the end of the experiment, it is possible that changes may have occurred between these measurements that were unobserved.

 

 

 

Heart rate. This measure was clearly altered by marijuana. Heart rate was measured (simultaneously with EEG) before smoking, immediately after smoking, and 20 minutes, 40 minutes, and 60 minutes after smoking. In the placebo condition, heart rate decreased consistently from presmoking levels. Heart rate under the high-dose conditions increased immediately after smoking and then gradually decreased, reaching presmoking levels approximately 40 minutes after smoking (figure 3). Heart rate after low doses also increased immediately after smoking and then decreased, reaching presmoking levels within 20 minutes; it then followed the course of placebo.

An analysis of variance indicated a drug effect on heart rate (F,„, = I 1.19; p < .001) reflecting differences between low and high dose, and between high-dose and placebo conditions. There was also a significant difference in heart rate depending on the time it was measured (F4,6 = 35.07; p < .001). In this instance, heart rate was different at each point from every other point except for pre-smoking versus 20 minutes after smoking, and 40 minutes after smoking versus 60 minutes after smoking. The interaction of drug and time of heart rate.measurement was also significant (F16,,,,, = 12.16; p < .001), representing changes in heart rate in both high and low conditions immediately after smoking.

 

EEG. The EEG results are presented in summary here and more fully elsewhere (7). The EEG effects were rapid in onset and changed at the time of the first postsmoking record (immediately after smoking). The principal changes were an increase in percent time alpha (8-13.5 Hz), and decreases in percent time theta (4-7.5 Hz) and beta (18.5-24.5 Hz) activities. Using regression analysis, the differences between high dose and both low dose and placebo effects are seen principally in the differences jn intercepts.

 

Discussion

 

These data support and extend the reports by Rodin and associates (4) and Well and Zinberg (3). On the tasks that were adversely affected by marijuana (short-term memory and reaction time), behavioral changes were observed only at the high dose. At the low dose, performance was indistinguishable from placebo. It is interesting to note, however, that even though overt behavior is unaffected by low doses of marijuana, heart rate, EEG, and possibly other physiological measures not observed in this experiment are altered with low doses.

 

It is difficult to extend the definition of "high" and "low" from the laboratory to the social situations in which subjects smoke marijuana. However, even knowing the amount of marijuana consumed in social situations is not adequate, as its chemical constituents, and therefore its potency, may vary from a similar amount of marijuana used on another occasion. In fact, even in the laboratory there is a need to determine and carefully monitor the content of active constituents in marijuana and its extracts throughout any given study. Ongoing analyses in our laboratory and the report by Rodin and associates (4) indicate that the potency of marijuana, measured by its constituents such as THC, deteriorates rapidly. The content of delta-9-THC is used as the standard for defining potency of cannabis preparations. This approximation may be invalid since constituents other than delta-9THC may be the "active" principal constituent, and indeed some recent studies have been focused on other derivatives (8).

 

The direct relationship between dose and effect with cannabis suggests that this compound follows established principles of clinical pharmacology; studies that fail to identify the type of material, dose and amount of chemical constitutents, and assay methods and standards ought to be ignored.

 

REFERENCES

 

1. Clark LD, Nakashima EN: Experimental studies of marihuana. Amer J Psychiat 125:379-384, 1968

2. Melges FT, Tinklenberg JP, Hollister LE, et al: Marihuana and temporal disintegration. Science 168:1118-1120,1970

3. Weil AT, Zinberg NE, Nelsen JM: Clinical and psychological effects of marihuana in man. Science 162:1234-1242,1968

4. Rodin EA, Domino EF, Porzak JP: The marijuana-induced "social high" JAMA 213:1300 1302,1970

5. Hollister LE: Marijuana in man: three years later. Science 172:21-29,1971

6. Carlson VR, Feinberg 1: Individual variations in time and judgment and the concept of an internal clock. J Exp Psycho] 77:631-640,1968

7. Volavka J, Dornbush R, Feldstein S, et al: Marijuana, EEG, and behavior. Ann NY Acad Sci (in press)

8. Christensen HD, Freudenthan RI, Gidlgy JT, et al: Activity of Y- and .1"-tetrahydrocannabinol and related compounds in the mouse. Science 172: 165-167,1971