The use of cannabis in Egypt: A behavioural study

Sections

Introduction
Procedure
Results and discussion
Relation to use of other agents acting on the CNS
Commencement and continuation of cannabis use
Motivation to start, to interrupt and to resume regular taking of cannabis
Effects
Criminal behaviour
B. OBJECTIVE TESTS
Conclusions
Summary
Acknowledgements
Bibliography

Details

Author: M. I. SOUEIF
Pages: 17 to 28
Creation Date: 1971/01/01

The use of cannabis in Egypt: A behavioural study *

Ph.D. M. I. SOUEIF Cairo University, Cairo, Arab Republic of Egypt

Introduction

This is the fourth of a series of reports on the study of hashish consumption in Egypt conducted at the National Centre for Social and Criminological Research in Cairo since November 1957. The first report concentrated on problems of method including construction, standardization and estimation of reliabilities and validities of two interviewing schedules intended for administration to hashish users and controls (C.I.H.C.E., 1960). The second publication reported on a pilot survey carried out in the city of Cairo on 204 hashish takers and 115 controls (C.I.H.C.E., 1964). The third report disclosed the results of another pilot investigation conducted on 49 experimentals and 40 controls, dwellers of semiurban and rural parts of Egypt. These findings were presented together with various comparisons between urban (Cairo) and rural users (Soueif, 1967).

The present paper is intended to report briefly on the results of the administration of the interviewing schedules and a number of objective psychological tests to 850 hashish takers and 839 non-takers. A full report is now in preparation.

Procedure

Subjects

The experimental group included the whole male population convicted exclusively for hashish use and detained in Egyptian prisons during the period from June 1967 to March 1968 who admitted that they had been using the drug at least once per month throughout the year preceding imprisonment. Ages ranged from 15 to slightly over 50 years with an average of 39 years (±10.5). 460 subjects were kept in prisons situated in big cities and intended for urban offenders, while 390 subjects were villagers detained in rural prisons. 72% of the group were married, 21% bachelors, 6% divorced and about 1% widowed. 60% were illiterate and the rest distributed among various levels of education with only 6 subjects attaining the level of high school certificate. None were university graduates. Approximately 25% of the group were skilled labourers and the rest were distributed among all sorts of unskilled jobs. As to income, 65% could earn E£15 or more monthly.

* The work reported in this paper has been conducted by the Committee for the Investigation of Hashish Consumption in Egypt (C.I.H.C.E.) under the auspices of the National Centre for Social and Criminological Research. The author is Chairman of the Committee, responsible for drawing the plan of the work, designing the analysis of the data and writing the report. Members of the Committee are: Dr. S. A. Zaki, M.D., A.M. El-Sayed, M.A. (Secretary of the Committee), Z. A. Darweesh, B.A. and M. A. Hannourah, B.A

Controls were defined as those subjects who never took hashish or any other narcotic substance as specified in Act 182 passed in June 1960 by the UAR Government as to the prohibition and/or regulation of the use of narcotic substance (Soueif 1967). Thus our controls, consisting of 839 subjects were selected according to this criterion of no drug abuse from among the convicts detained in the same prisons where experimentals were detained. ** Ages ranged from 15 to slightly more than 50 years with a mean of 33.1 years (±9.75). Some 454 subjects were selected from urban prisons and 385 from rural ones. Of the group, 58.4% were married, 36.9% bachelors, 3.6% divorced and about 1.1% widowed. 54.8% were illiterate; the rest were distributed among various levels of educational achievement, with 44 subjects attaining the level of high school certificate and 9 university graduates. About 26% of the group could be classified as skilled labourers while the rest were mostly unskilled. Regarding income, 50% earned E£15 or more per month.

It will be readily seen that (in so far as measures of central tendency go) the control group is fairly well matched to experimentals regarding age, urban or rural backgrounds, and percentage of skilled labourers. The experimentals have a better stand as to income and marital status while the opposite is true considering literacy.

To reach controls we had to depend on key men (Inkeles and Bauer, 1959) and on an unstructured preliminary interview.

Interviewing and testing

The standardized experimental and control interviewing schedules were individually administered, to users and on users respectively, by interviewers who had the minimum qualification of a bachelor's degree in Psychology, and who had received an intensive course of training (expressly designed for this investigation) which took them 3 months to complete (Soueif, 1967; Maccoby & Maccoby, 1954).

The two schedules were standardized with respect o sequence of topics, sequence of items and to wording. The experimental interview included 258 questions with a wide range of complexity. The points of investigaion covered 18 main topics: subject's personal estimation of the relative indulgence in cannabis use in different sectors of Egyptian society; cannabis use as practised by the interviewee; general appearance; perception and thought processes; temperament and mood change; social milieu of the interviewee, sexual behaviour, self image, interviewee's opinion about legislation prohibiting cannabis use and traffic, cannabis effect on appetite for food, drink and general energy output, cumulative effects of taking the drug for a long time, family background, history of the habit in the family, pattern of relationships within the interviewee's conjugal family, work conditions, recreational activities, structure and dynamics of the" hashish party" and finally personal data.

The control interview was composed of 147 items, again with a wide range of complexity. This schedule was only a replica of the experimental schedule except for the items dealing with first-hand experience of the drug, which were excluded. Re-take reliabilities of all the items were established for experimentals and controls separately over a period ranging from 7-15 days. Minimum estimate of reliability for any item to be endorsed in the interview was fixed at 66% agreement, a product moment correlation of 0.50 % and a coefficient of contingency or phi 0.40 between take and retake Soueif, 1967). Median estimates of reliability obtained were as follows:

  1. The experimental schedule: 85 % agreement. 0.74 as coefficient of correlation. (N=45)

  2. The control schedule: 85% agreement. 0.90 as a coefficient of correlation. (N=45)

Twelve objective tests, generating 16 test variables, were also administered to all subjects to examine a number of hypotheses suggested by our pilot studies. In addition to considerations of reliability and validity the choice of tests was influenced by two main factors:

  1. They should be widely known among psychologists in Egypt and abroad, and should be as culture free as possible, so that they would serve to make possible future comparisons across researchers and across cultures.

  2. Brevity and simplicity of administration were also taken into account to allow the completion of all the interviewing and testing required within one session.

The tests consisted of the following:

  1. Tool Matching: This test was designed to assess form perception. Subjects are required to rapidly compare pictures of tools alike except for differences in shading. Time allowed was 4 minutes.

  2. H Marking, was made to assess the ability for aiming. Subjects are instructed to draw quickly one line through each letter H. Time allowed was 30 seconds.

  3. Speed, was made to assess motor speed. Subjects are requested to put three dots anywhere inside each rectangle (1.5 X 2.5 cm) filling as many rectangles as they can in 30 seconds.

  4. Mark Making: This test was constructed to measure aiming and motor speed. Subjects are asked to make marks like these ____ in each square (8 X 8 mm), filling as many squares as they can in 60 seconds (Anastasi, 1968).

    Tool Matching, H Marking, Speed and Mark Making are all parts of the well known General Aptitude Test Battery (GATB). L. Cronbach holds that the GATB is designed with an efficiency that has never been exceeded (Cronbach, 1960).

  5. Trail Making (Part A): This test comprises numbers enclosed in small circles which are scattered randomly over the test sheet. Subjects are required to draw lines connecting numbers 1 through 25 in proper sequence and speed is emphasized in the instructions (Reitan, 1955).

  6. Initial Reaction Time: Four cards, I, IV, V and VI of the Rorschach test of inkblots were utilized for this purpose. The score was timed in seconds elapsing between exposing subject to each card and the start of verbal response.

  7. Digits Forward and Backward: This is part of the Wechsler scale for intelligence. It is a test for immediate memory. D. Wechsler maintains that, "it has been used for a long time by psychiatrists as a test for retentiveness and by psychologists for all sorts of psychological studies" (Wechsler, 1944).

  8. Bender Gestalt: This test consists of nine simple designs and involves reproduction by drawing. The test as used in this work was modified in the way described by Shapiro, Field and Post (1957). Each subject was first asked to copy all the designs and immediately after completing this job was requested to draw them from memory. The scoring system devised by Miss Lofving was adopted (Shapiro et al., 1957). This system procured a five point rating scale of excellence for deriving a score for copying and a score for recall.

  9. Time Estimation: A period of 3 minutes was a criterion, during which subject was told to have a rest out not to leave the testing situation. The tester did not engage subject in conversation. A stopwatch (hidden from subject) is used and by the end of the period subject is asked to estimate the length of time passing since he was told to have a rest. The rest period was always in interval between two testing periods (Soueif et al., 1964).

  10. Length Estimation: Three sticks, made of plain wood, were shown to subject one at a time. The tester would hold the stick horizontally at about 20 cm above the table, and 30 cm from the testee, and would then say, "Look at this piece of wood. What do you think is its length in centimetres? Mind you, one centimetre is exactly this length", showing subject the exact length of one centimetre as marked on the pen the tester is using. This last demonstration is only made on showing the first stick. The lengths of the sticks were 34, 25 and [?]cm (Soueif et al., 1964).

All interviewing and testing took place within the prisons where subjects were detained. Testees were seated comfortably all through the interviewing and testing and the tester was always left alone with subject in the testing room. Tests were administered in random order, and always at the end of the interview.

Results and discussion

A. THE INTERVIEWING SCHEDULES

Patterns of drug use

Smoking was found to be the prevalent route of administration (89.4%) either by josah, a smoking pipe, 61.7%) or cigarettes (10.6%) or a combination of both (17.1 %). Very few habitués (1.3 %) take the drug orally only, and the rest (9.3%) combine smoking with some method of oral administration.

Smoking as a preferred way of drug use was found to be related to the number of times per month the drug was taken. Thus 87% of heavy users (those who take the drug more than 30 times) v. 93% of moderates those who take it 30 times or less) invariably smoke the drug. The difference between the two percentages is statistically significant beyond 0.01 (c.r.=3). Nevertheless, smoking was independent of the urban-rural background of the drug-taker. Smokers among our rural sample (91%) exceeded smokers among urbans (87.7%) but not significantly so.

39% of our experimentals took hashish at definite hours of the day. More heavy takers (42%) than moderate users (29.3%) expressed this tendency. The disparity between the two percentages is highly significant (c.r. =3.61).

FIGURE I

Incompatibility between times of the day when cannabis takers actually took the drug and times when they preferred to take it

Full size image: 12 kB, FIGURE I

Almost all habitués prefer to take the drug in the evening (94.2%) and they do take it at this time of the day (97.2%). Regarding other times of the day, however, we found a certain degree of incompatibility between what they preferred and what they actually did. This is clearly shown in figure I. That this discrepancy is related to the degree of drug dependence is attested to by the fact that it is more pronounced in heavy users than in moderate users (figure II). Inspection of the two parts of figure II discloses another factor which again is related to the degree of attachment to the drug. Many heavy users reported taking the drug in the morning (41.2%), at midday (31.1%), and in the afternoon (52.1%). Moderate takers deviated much less from their expressed preference.

The big majority (82.5%) takes hashish in groups of 4-6 members, on an average, and only 17.5% take it on their own. Heavy users do not differ markedly from moderate ones in this respect. City-dwellers have, however, a significantly higher percentage of lonetakers (23.9%) than users from rural areas (10%) (c.r. =5.6).

Volume of drug consumption and some correlates

By volume is meant the amount of drug consumed at a time, and the number of times a month the drug is taken. 67.3% of our group used to take an amount of cannabis ranging between 0.54 and 1.08 gm at a time, with 22.4% taking less than 0.54 gm and 10.3% taking more than 1.08 gm. There is a relationship, though not a significant one, between dose at a time and frequency of consumption in a month. Thus 31.2% of moderate users, but only 18.2% of heavy users, tend to take less than 0.54 gm at a time, while 13.2% of heavy users as compared to only 4.9% of moderate users take more than 1.08 gm at a time (figure III). It has also been observed that only 18.1% of city dwellers take less than 0.54 gm at a time as compared to 27% of takers from rural areas (figure IV).

FIGURE IIa

Incompatibility between times of the day when moderate takers actually took the drug and times when they preferred to take it

Full size image: 12 kB, FIGURE IIa

FIGURE IIb

Incompatibility between times of the day when heavy takers actually took the drug and times when they preferred to take it

Full size image: 12 kB, FIGURE IIb

70% of convicts from town and 66% of convicts from rural areas take hashish more than 30 times a month. The discrepancy between the two percentages has no statistical significance. Looking more closely into the data it was found that 86% of heavy takers use the drug more than 60 times a month with very slight differences between urban and rural inhabitants. It will be recalled that the average frequency of drug-taking by the non-prison Cairo sample was reported elsewhere to be 12 times a month, with the upper 10% exceeding 50 and the lower 10% taking it no more than 5 times. At the same time the corresponding figures for the non-prison rural inhabitants were 8 times, with 2% taking more than 50 times and 36% less than 5 times (Soueif, 1967). Obviously, therefore, the prison group of takers should be considered as markedly different from the non-prison group.

We found a significant relationship between frequency of hashish use and age at onset of the habit; the earlier the onset the higher the frequency (figure V). When we divided our group into those who started taking the drug regularly before the age of 22 (N=488) and those who began at 22 or after (N=354) the relationship became very striking. Whereas 76% of the starters before 22 were heavy takers, only 56% of the beginners after 22 belonged to the same category. The discrepancy between the two percentages is highly significant (c.r.=6.1).

FIGURE III

Quantities of hashish (per gram) consumed per session by heavy takers and moderates

Full size image: 11 kB, FIGURE III

FIGURE IV

Quantities of hashish (per gram) consumed per session by urban and rural users

Full size image: 11 kB, FIGURE IV

FIGURE V

The relationship between age at onset of regular drug use and frequency of drug taking per month

Full size image: 7 kB, FIGURE V

Frequency of drug use was also found to be related to divorce. 6.2% of heavy takers compared with 2.2% of moderates reported that they were divorced. The difference is statistically significant (c.r. =3).

Heavy takers did not differ markedly from moderates as regards the situation (i.e. the external circumstances) which occasioned the start of drug use. When asked about their conscious motives for starting hashish use, decidedly more heavy takers than moderates mentioned imitation (62.5% v. 53.4% respectively), ryingt to behave as "real" men (78.4% v. 58.9%) and seeking euphoria (86.3% v. 79.1%). On the other hand significantly more moderates (35.6%) than heavy takers (23%) reported to have made one or several attempts at interrupting hashish use (c.r.=3.73).

More heavy takers than moderates gave neurotic responses to three items selected from Taylor's scale of manifest anxiety (Taylor, 1953). Thus more heavy takers (25.4%) than moderates (16%) stated that they slept badly, lay awake in bed long before falling asleep (38.7% v. 29.8%) and that they wake up very early in the morning without being able to go back to sleep (34.5% v. 25.2%). In all the three cases the discrepancies between both groups are significant (c.r.'s=3.26, 2.59 and 2.82 respectively).

Relation to use of other agents acting on the CNS

Opium

6% of our group admitted that they used to take opium. In the urban part of the group the corresponding percentage is 34.5%. This exceeds what was previously reported about the Cairo sample (26%) (Soueif, 1967). More heavy takers (34.3%) than moderates (25.7%) tend to take opium. The difference is statistically significant at a reasonable level of confidence (c.r.=2.16). When we divided the group into urbans and rurals the same trend emerged again within each subgroup though it lost its statistical significance within the latter. Opium taking was also found to be slightly more prevalent among urban habitués (34.5%) than among the rural group (28%). Moreover, we were able to establish a positive relationship between duration of hashish use and opium taking. This can be readily seen in figure VI.

Alcohol

More hashish takers (22.7%) than controls (9.1%) (regardless of urban or rural background) were found to be alcohol drinkers. The difference is highly significant (c.r.= 7.7). As in the case of opium, more, alcohol drinkers were found among heavy hashish users (27%) than among moderate users (13.5%) and the difference was again highly significant (c.r.=4.82). Duration, however, proved irrelevant in this respect.

Coffee, tea, and tobacco

Hashish takers tended to take coffee (the thick Turkish brew) to a greater degree than controls did, but the difference was not statistically significant. Also, urban hashish users tended to take coffee more than country dwellers, but again the difference did not reach any acceptable level of significance. Many more heavy users (67.5%) than moderate users (49.1%), however, took coffee and the difference was highly significant (c.r.=5.14).

FIGURE VI

The relationship between opium taking and duration of hashish consumption

Full size image: 9 kB, FIGURE VI

Tea use differentiated at a high level of confidence between experimentals (96.5%) and controls (79.3%) (c.r. = 11.2). Urban users did not differ from rural habitués on this item, nor did heavy users differ from moderate users.

Tobacco smoking was much more prevalent among experimentals (91.6%) than among controls (32.4%). But, like tea use, tobacco smoking showed no difference between urban (92.6%) and rural users (90%), nor between heavy takers (92.2%) and moderate users (90.8%).

Noteworthy is the fact that the majority of our experimentals who reported taking coffee, tea and smoking tobacco admitted that they did so before starting to take cannabis.

The conclusion can, therefore, be drawn that cannabis users tend to seek agents acting on the CNS more than non-users do. Moreover, heavy cannabis users, compared to moderate users, showed a greater tendency to indulge in alcohol and coffee use (and tobacco smoking too though this item was not very marked). It is our impression that in accounting for opium taking, as presented by our cannabis consumers, the " need for CNS effects" should be taken into consideration. Methodologically, the least that can be said about this way of putting the problem of taking other drugs in addition to cannabis is that it helps to bring in the concept of " individual differences" that should play an important role as a basic component in any theorizing about drug taking behaviour.

Commencement and continuation of cannabis use

42% of our group began to take cannabis before the age of 20. Out of this subgroup 32.8% had their first experience with the drug before completing 16 years of age (early beginners). On the other hand 17.6% of the total group did not start the habit before the age of 28 years (late beginners). These findings differ markedly from corresponding findings in the Cairo sample (Soueif, 1967).

A greater number of early than late beginners used hashish in the morning (53.4% v. 15%) at noon (44.8% v. 12.2%) and in the afternoon (56% v. 36.7%). The disparities in all these cases are statistically significant. More early beginners (61.2%) than late ones (24.7%) reported to have had some member (or members) of the family using cannabis. Again more early beginners (36.4%) than late ones (17.6%) with one parent reported dead, stated that the remaining parent was re-married. The discrepancy is significant (c.r. =2.95). Early beginners reported more incidence of separation and/or divorce between parents (15%) than did late commencers (7.8%). The discrepancy is just short of the 0.05 level of significance (c.r.= 1.78).

Thus early commencement of cannabis use is associated with exposure to the influence of a drug consuming example within the family circle, combined with the effect of the disappearance of father or mother and the intrusion of a step-parent.

Motivation to start, to interrupt and to resume regular taking of cannabis

Asked about their conscious motives for taking hashish for the first time subjects stated numerous motives ranking in the following order: conformity to a group of personal friends (87.8%), seeking euphoria (86.6%), behaving like "real men" (72.5%) imitating others (59.5%), curiosity (57.5%), forgetting about one's personal problems (42.8%), alleviating a mood of depression (40.5%), as medicine for some body ailments (27.3%) and lastly for the enhancement of sexual enjoyment (23.8%). By and large this way of ranking conscious reasons for taking hashish for the first time, was found to prevail among our experimentals irrespective of urban or rural background, heavy taking or moderation and early or late commencement of the habit. Moreover, it does not differ much from what was reported about the Cairo sample (Soueif, 1967). It is worth investigating whether this conclusion would again emerge when studying comparable groups in societies other than the Egyptian.

27.3% of the group stated that they had interrupted cannabis taking once or twice. Out of the early starters only 17.1% did so compared with 31.7% of the late ones. The difference is significant at 0.01 (c.r.=2.82).

When asked about their conscious motives for the resumption of drug taking, our consumers stated three main reasons in the following order: the impact of the occasion, to be able to bear troubles and having leisure time to spare. Almost the same pattern emerged when we compared early with late beginners, and again when comparing heavy takers with moderate takers.

When asked whether they still wished to stop taking the drug 78.5% of the total group answered in the affirmative. Looking closely into our data, however, we found a marked difference between early (60%) and late beginners (93.2%). We also found a significant difference between heavy takers (75.1%) and moderates (85.6%) (c.r.=3.78). The reasons behind this wish, as stated by our experimentals, were as follows (in their order of frequency) irrespective of heavy taking v. moderation, and/or early v. late onset of the habit: financial reasons, fear of the law, health reasons, and lastly worry about social status.

Effects

We do not have any test data describing the immediate effects of the drug. Rather, we depend, for this purpose, on subjective reports procured by our experimentals about their past experience with the drug.

Euphoria (83.1%), acquiescence (71.9%), hesitancy (57.4%), suggestibility (53.5%), carefreeness (58.7%) and a tendency towards gregariousness (50.1%) were all reported by our experimentals as changes of the mood immediately effected by drug taking. It should be noted that these percentages left the rest of the group reporting either no change in mood, or a change towards the opposite poles of the moods described by the percentages of their fellows referred to above.

Various percentages of our experimentals reported different patterns of distortion in perception of time, distances, object sizes, colours and sounds. 53% of the group said they used to perceive time as passing quicker than usual whereas only 13% perceived it passing slower. This is contrary to the subjective reports we got from the Cairo experimentals, but is in agreement with what was reported about Upper Egyptian hashish takers (Soueif, 1967).

  1. 9% of the group, when under the drug effect, saw distances getting longer than usual, but 52.5% saw them neither longer nor shorter. Again, this finding is in conflict with what was reported previously (C.I. H.C.E., 1964).

We are not in a position to explain this disparity between the present and the previous findings. Suffice it to say, here, that the distortion of time perception as described by our present group proved to be rather independent of urban or rural background and of frequency of drug taking per month. But to perceive distances as longer than they usually seem was significantly more frequent in heavy takers (41.6%) compared with moderates (31.4%) (c.r.=3.02).

For a number of hashish users there is an inclination to perceive sizes of various objects as larger (18.6%), colours brighter (34.4%) and voices louder (19.9%) and more articulate than usual (34.8%).

Thought processes were also reported to reflect acute effects of hashish consumption. Fluency of ideas is reported by 68% of the group to have been accelerated. Parallel to ideational fluency, changing topics of conversation at a high speed was reported by 63.9%. Accuracy of thinking was reported by 88.3% to be unaffected under immediate drug effect, and 77.4% stated that they often tried to solve some of their personal problems when under hashish influence, and that the solutions they worked out were usually quite adequate. The majority (88.4%) also maintained that they could recall these solutions later, and that they even used to try to put them into action. As to memory 48.3% claimed that it was unaffected, 38.7% that it was improved and 13% that it was impaired.

Criminal behaviour

Asked whether they thought hashish takers tended to criminal actions more than non-takers, 6% of our experimentals answered in the affirmative. Moreover, out of this small minority 11 subjects maintained that hashish takers had a criminal tendency but never actualized it. The idea, however, entertained by non-takers concerning this point is totally different. 56% of our controls held that hashish takers had criminal tendencies and that they did commit criminal acts.

To test the validity of this notion we examined the actual criminal records of 553 experimentals, detained in the three largest prisons (Turah, Quanater and Marg) and of 458 controls from the same prisons. To guarantee the highest possible authenticity we did not use the local files kept in the prisons, but rather those kept at the Central Record Office in the Ministry of Interior. In comparing our two groups we took into account all criminal offences other than those having to do with narcotics (using and/or selling). 5.7% of hashish takers v. 13.5% of controls were found to have had criminal records previous to their arrest. The discrepancy between these percentages is highly significant (c.r.= 2.86). We, also, found that controls tended to exceed experimentals regarding the average number of crimes committed by each of those having criminal records (5.3 v. 4.5 crimes respectively).

B. OBJECTIVE TESTS

On the whole, objective test data reveal consistently significant differences between hashish users and controls. This can be readily seen in tables I and II. Controls from urban districts and from Lower Egypt obtained significantly better scores than comparable cannabis takers on most of the tests, e.g. speed and accuracy of psychomotor performance and memory span for digits and designs. On Length Estimation (discrepancy, irrespective of direction of errors) urban controls fared significantly better than comparable experimentals. But on Time Estimation urban non-takers erred significantly more than takers. In Lower Egyptians practically no difference came out between the two groups of users and non-users on Length Estimation. On Time Estimation (discrepancy), however, Lower Egyptian controls again erred significantly more than experimentals.

The results concerning Time Estimation seem to go against popular expectation, and indeed against predictions based on our interviews. To be sure, our interviewees were asked about the immediate (not the chronic) effect of the drug on time perception. Nevertheless one would rather expect subjects who are frequently exposed to the experience of some distortion of time perception to err on tasks involving time estimation, more than those who are not. It is tempting, here, to raise a question about the behavioural validity of the popular notion and of the subjective statements of cannabis takers in this respect. Subjects, under the immediate effect of various doses of smoked cannabis should be objectively tested to estimate time periods of various lengths. Though the results of such testing may shed some light on our query, still there may always be other possibilities to be investigated. One such possibility could be that chronic and immediate effects do not go in the same direction. In other words we might be facing, here, something like a paradoxical effect of the drug when it is taken very frequently for considerably long periods.

TABLE I

Means, standard deviations, medians and quartile deviations obtained by urban cannabis takers and controls on objective tests

 

Cannabis takers

Controls

 

Test variables

M

SD

Mdn

Q

M

SD

Mdn

Q

t

between

d.f.

Tool Matching
11.89 4.97 11.61 3.43 12.61 5.33 12.52 4.08 2.09a
means
891
H Marking
13.46 4.97 13.11 3.42 15.23 4.85 14.85 3.52 5.10 c
means
801
Speed
19.43 5.90 19.55 4.10 21.02 7.09 20.55 4.67 3.45 c
means
802
Mark Making
14.53 6.23 14.43 3.91 17.04 7.68 17.00 4.96 5.06 c
means
795
Trail Making
96.99 52.8 83.50 28.2 100.6 65.5 83.47 34.2 0.75
means
599
I.R.T.
9.18 5.11 7.95 3.11 7.99 4.76 7.03 2.67 3.63 c
means
904
Length Est. (+)
7.01 9.83 3.76 2.47 7.44 13.1 3.53 1.77 0.24
medians
894
Length Est. (-)
14.02 12.6 9.56 10.1 12.89 12.4 8.04 9.26 1.46
medians
900
L. Est. (discr.)
19.40 12.4 17.28 9.17 18.48 15.1 14.27 10.2 2.61 b
medians
893
Time Est.
5.08 3.1 4.39 1.36 5.69 3.77 4.60 1.38 2.65 b
means
899
T. Est. (discr.)
2.89 2.65
2. 10
0.80 3.56 3.31 2.36 1.43 3.17 b
means
809
Dig. Forward
4.55 0.93 4.46 0.65 4.54 1.07 4.44 0.71 0.15
means
901
Dig. Backward
2.75 1.15 2.94 0.47 2.94 1.12 3.01 0.46 2.52 a
means
902
Dig. F + B
7.31 1.75 7.33 0.86 7.47 1.83 6.54 1.02 5.28 c
medians
900
B. Gestalt (Copy)
14.37 2.99 14.21 2.94 14.91 3.61 14.61 2.57 2.40 a
means
863
B. Gestalt Recall
6.97 3.87 6.56 2.67 8.14 4.01 7.73 2.84 4.33 c
means
849

a Statistically significant beyond 0.05

b Statistically significant beyond 0.01.

b Statistically significant beyond 0.001.

Length Est. (discrepancy) and Time Est. (discrepancy) were computed irrespective of direction of error.

TABLE II

Means, standard deviations, medians and quartile deviations obtained by rural cannabis takers and controls (from Lower Egypt)

 

Cannabis takers

Controls

 

Test variables

M

SD

Mdn

Q

M

SD

Mdn

Q

t

between

d.f.

Tool Matching
10.68 4.44 10.39 3.27 12.64 5.38 12.00 3.63 2.34 a
means
555
H Marking
13.28 4.73 12.94 3.38 15.71 4.59 14.76 3.83 2.64 b
means
501
Speed
18.54 6.23 18.80 3.84 21.80 6.66 20.96 4.60 2.74 b
means
465
Mark Making
14.70 6.71 14.46 4.73 16.51 7.53 16.46 5.25 1.39
means
476
Trail Making
102.33 50.8 92.86 30.3 86.96 46.7 76.04 25.2 0.14
means
329
I.R.T.
8.03 5.40 6.68 2.72 7.31 4.04 6.24 2.43 0.89
means
554
Length Est. (+)
6.33 8.09 3.86 1.93 5.85 7.52 3.58 1.79 0.12
means
553
Length Est. (-)
11.70 11.90 7.14 9.58 13.07 12.85 7.92 9.52 0.27
means
570
L. Est. (discr.)
17.73 12.1 15.18 8.82 17.76 12.87 14.7 9.52 0.01
means
560
Time Est.
5.55 3.47 4.84 1.27 5.80 3.53 4.77 1.63 0.42
means
556
T. Est. (discr.)
2.86 1.96 2.34 0.67 3.40 3.14 2.33 1.37 2.33 a
means
498
Dig. Forward
4.34 0.99 4.22 0.68 4.48 1.07 4.35 0.70 1.61
means
570
Dig. Backward
2.60 1.66 2.87 0.55 2.78 1.31 2.76 0.88 1.43
means
561
Dig. F + B
6.87 1.79 7.01 1.07 7.26 2.00 7.21 1.27 2.42 a
means
558
B. Gestalt (Copy)
13.76 2.97 13.71 1.85 14.48 3.30 14.33 2.36 2.70 b
means
552
B. Gestalt (Recall)
6.54 3.25 6.48 2.40 7.71 4.02 7.20 3.14 3.75 c
means
543

a Statistically significant beyond 0.05

b Statistically significant beyond 0.01.

b Statistically significant beyond 0.001.

Length Est. (discrepancy) and Time Est. (discrepancy) were computed irrespective of direction of error.

Table III is not all that clear. On 13 out of 16 test variables there were no significant differences between takers and non-takers. These results might be due to the fact that the Upper Egyptian experimentals and controls were not equated on education nor on skilled labour. On both variables experimentals had a better stand than controls. Thus 62.3% of the former were illiterate and 29.6% knew how to read and write (semi literate), in contrast to 72.5% illiterate and 16.3% semi-literate among the latter. We also found that 25.6% of experimentals were unskilled labourers, while 72.5% of the controls were to be classified under the same category.

TABLE III

Means, standard deviations, medians and quartile deviations obtained by rural cannabis takers and controls (from Upper Egypt)

 

Cannabis takers

Controls

 

Test variables

M

SD

Mdn

Q

M

SD

Mdn

Q

t

between

d.f.

Tool Matching
9.41 3.63 9.18 2.66 9.14 3.48 8.77 1.99 0.52
means
186
H Marking
12.52 4.63 12.07 3.67 12.23 4.38 12.38 2.97 0.44
means
188
Speed
18.65 6.26 18.04 4.04 16.77 5.67 16.06 3.07 2.18a
means
190
Mark Making
13.57 6.05 13.14 4.16 13.04 5.54 12.14 4.48 0.93
medians
179
Trail Making
105.69 38.7 99.06 27.54 100.2 46.5 85.82 27.2 1.11
medians
82
I.R.T.
7.43 3.52 6.76 2.06 7.47 3.82 6.80 2.32 0.08
means
194
Length Est. (+)
8.53 10.5 4.31 4.83 4.25 3.87 3.35 1.68 3.71c
means
181
Length Est. (-)
13.32 14.4 4.41 10.45 18.94 14.7 14.78 11.9 2.67b
means
189
L. Est. (discr.)
20.54 12.9 19.00 10.82 22.08 13.6 19.17 10.3 0.72
means
183
Time Est.
5.44 3.53 4.25 1.50 5.63 3.65 4.43 1.79 0.36
means
185
T. Est. (discr.)
3.16 2.24 2.26 0.92 3.47 2.31 2.48 1.84 0.78
means
130
Dig. Forward
4.29 0.81 4.27 0.67 4.09 0.82 4.03 0.70 1.72
means
194
Dig. Backward
2.27 1.19 2.57 0.71 2.42 1.17 2.69 0.70 0.89
means
194
Dig. F + B
6.56 1.83 7.05 1.35 6.51 1.81 6.69 1.10 0.19
means
194
B. Gestalt (Copy)
13.58 3.09 13.77 2.19 13.13 3.53 12.7 2.27 1.75
medians
188
B. Gestalt (Recall)
6.23 3.14 6.00 1.96 5.64 3.11 5.04 1.85 1.68
medians
184

a Statistically significant beyond 0.05

b Statistically significant beyond 0.01.

b Statistically significant beyond 0.001.

Length Est. (discrepancy) and Time Est. (discrepancy) were computed irrespective of direction of error.

Comparisons were also made between groups of experimentals and controls equated for level of education.

The results are disclosed in tables IV, V and VI. It is interesting to note that the higher the level of education of the users and comparable non-users the larger the discrepancy between their respective test scores. Thus the disparities are minimal in the illiterate groups (table VI), maximal in those attaining the level of high school education (table IV) and somewhere in between among semi-literates (table V). Of particular importance also is the fact that the test scores were far more affected by illiteracy than by cannabis consumption. Cannabis takers among the high school subjects obtained better scores than did illiterate and semi-literate controls on most of our test variables.

TABLE IV

Means, standard deviations, medians and quartile deviations obtained by cannabis takers and controls who attained the level of high school education

 

Cannabis takers

Controls

 

Test variables

M

SD

Mdn

Q

M

SD

Mdn

Q

t

between

d.f.

Tool Matching
13.91 5.38 13.93 3.07 16.33 5.56 17.00 3.94 3.15b
means
247
H Marking
15.40 4.67 16.00 3.53 19.78 4.86 19.71 3.32 6.33c
means
231
Speed
25.32 4.96 25.67 3.92 27.34 6.28 27.50 4.50 2.50a
means
227
Mark Making
20.50 5.70 19.62 3.36 23.15 7.22 23.32 4.90 2.88b
means
226
Trail Making
81.27 28.3 75.32 22.3 68.12 29.4 60.08 21.0 3.04b
means
227
I.R.T.
7.56 3.96 6.42 2.86 6.66 2.70 6.13 2.26 1.70
means
244
Length Est. (+)
3.98 3.60 3.25 1.63 5.35 7.44 3.43 1.71 2.07a
means
240
Length Est. (-)
11.86 3.15 7.91 9.03 8.28 7.75 5.59 4.86 6.88c
means
247
L. Est. (discr.)
15.89 12.2 11.88 9.34 12.58 9.24 9.02 5.57 2.04a
means
244
Time Est
4.58 2.25 4.44 1.32 5.00 3.12 4.20 1.35 0.53
medians
243
T. Est. (discr.)
2.50 1.89 1.95 0.75 3.00 2.86 2.10 0.83 1.48
means
209
Dig. Forward
4.78 0.77 4.81 0.57 5.36 1.04 5.22 0.67 4.46c
means
245
Dig. Backward
3.25 1.00 3.20 0.55 3.58 1.11 3.63 0.67 2.27a
means
245
Dig. F + B
8.04 1.54 8.00 0.86 8.93 1.73 8.96 1.13 3.92c
means
243
B. Gestalt (Copy)
16.38 3.02 16.83 2.18 17.74 3.04 17.61 2.24 3.09b
means
242
B. Gestalt (Recall)
9.06 4.35 8.88 3.63 10.93 3.79 11.21 2.42 3.06b
means
244

a Statistically significant beyond 0.05

b Statistically significant beyond 0.01.

b Statistically significant beyond 0.001.

Length Est. (discrepancy) and Time Est. (discrepancy) were computed irrespective of direction of error.

TABLE V

Means, standard deviations, medians and quartile deviations obtained by semi-literate cannabis takers and controls

 

Cannabis takers

Controls

 

Test variables

M

SD

Mdn

Q

M

SD

Mdn

Q

t

between

d.f.

Tool Matching
11.97 5.12 11.66 3.36 12.98 4.40 13.78 2.69 3.78c
medians
438
H Marking
14.65 4.90 14.23 3.04 15.83 4.14 15.36 3.02 2.60b
means
392
Speed
21.60 5.56 21.31 3.54 22.60 5.26 22.50 3.58 1.80
means
381
Mark Making
17.37 5.86 17.73 3.18 17.33 5.58 18.83 5.16 1.51
medians
384
Trail Making
94.72 49.8 84.18 29.2 85.74 38.4 78.34 24.2 2.06a
means
408
I.R.T
8.43 5.11 7.10 2.89 8.00 5.14 6.45 1.82 0.87
means
437
Length Est. (+)
7.09 8.48 3.87 3.01 5.91 8.28 3.47 1.73 1.45
means
402
Length Est. ( - )
11.84 11.6 6.50 9.42 12.06 11.0 8.60 9.19 1.57
medians
442
L. Est. (discr.)
17.19 9.96 15.50 8.79 17.16 11.3 14.50 8.60 0.77
medians
431
Time Est.
4.74 2.86 4.17 1.37 5.41 3.34 4.53 1.40 2.21a
means
436
T. Est. (discr.)
2.86 2.14 2.25 0.75 3.21 2.96 2.23 0.96 1.40
means
395
Dig. Forward
4.78 1.00 4.76 0.63 4.71 0.95 4.74 0.65 0.73
means
442
Dig. Backward
3.00 1.01 3.05 0.44 3.07 0.91 3.10 0.51 0.78
means
441
Dig. F + B
7.80 1.75 7.78 0.97 7.82 1.54 7.77 0.81 0.12
means
441
B. Gestalt (Copy)
15.36 2.80 15.18 1.89 15.43 3.30 15.50 2.67 0.23
means
435
B. Gestalt (Recall)
8.07 3.43 8.05 2.42 8.60 3.59 8.35 2.59 1.57
means
429

a Statistically significant beyond 0.05

b Statistically significant beyond 0.01.

b Statistically significant beyond 0.001.

Length Est. (discrepancy) and Time Est. (discrepancy) were computed irrespective of direction of error.

TABLE VI

Means, standard deviations, medians and quartile deviations obtained by illiterate cannabis takers and controls

 

Cannabis takers

Controls

 

Test variables

M

SD

Mdn

Q

M

SD

Mdn

Q

t

between

d.f.

Tool Matching
10.30 4.20 10.07 3.06 10.25 4.44 9.68 3.26 0.02
means
937
H Marking
12.25 4.58 11.94 3.09 12.70 4.14 12.5 2.98 1.53
means
860
Speed
16.80 5.32 16.70 3.51 17.01 5.14 17.2 3.98 2.19a
means
847
Mark Making
12.18 5.54 12.15 3.53 12.64 5.92 12.7 3.94 0.60
means
836
Trail Making
109.47 54.6 93.31 29.8 133.24 71.0 115.2 45.8 3.39c
means
365
I.R.T
8.64 4.80 7.58 2.73 8.00 4.36 7.10 2.66 2.25a
means
963
Length Est. (+)
7.12 9.60 3.89 3.00 7.76 13.1 3.65 1.83 0.25
medians
951
Length Est. (-)
14.30 13.55 9.28 10.5 16.26 14.5 11.0 11.5 1.72
medians
965
L. Est. (discr.)
20.30 12.66 17.5 9.53 22.33 16.1 19.1 10.8 2.21a
medians
956
Time Est.
5.57 3.30 4.73 1.61 6.10 4.00 4.84 2.10 0.15
means
951
T. Est. (discr.)
2.50 1.89 2.09 0.71 4.08 4.12 2.52 2.64 1.43
medians
821
Dig. Forward
4.18 0.87 4.13 0.58 4.00 0.80 3.97 0.47 3.33e
means
970
Dig. Backward
2.40 1.21 2.74 0.67 2.43 1.25 2.72 0.62 1.15
means
967
Dig. F + B
6.59 1.76 6.85 1.04 6.55 1.83 6.68 0.87 0.35
means
964
B. Gestalt (Copy)
13.03 2.78 13.06 1.73 12.92 2.79 13.0 3.08 0.60
means
917
B. Gestalt (Recall)
5.80 3.14 5.37 2.14 6.03 3.05 5.81 2.05 0.46
means
886

a Statistically significant beyond 0.05

b Statistically significant beyond 0.01.

b Statistically significant beyond 0.001.

Length Est. (discrepancy) and Time Est. (discrepancy) were computed irrespective of direction of error.

The positive association between education and excellence of performance on our objective tests can be accounted for in a number of ways. One possible explanation is that some learning transfers from the school to performance on the tests. If this is so, then our results denote that less transfer takes place in cannabis takers than in comparable non-takers. Equating transfer to learning (since after all transfer is learning how to learn) (Woodworth and Schlosberg, 1954; Mowrer, 1960) a testable hypothesis may be formulated as follows:

The effect of learning on speed and accuracy of performing simple psychomotor tasks and on rote memory goes at a slower rate in cannabis takers compared with controls. In other words, cannabis users are slow learners compared to non-users.

Stated in this way, the idea seems plausible as it ties up neatly with a number of reported findings on the effect of depressants as to increasing the rate of building up reactive inhibition (Trouton and Eysenck, 1961), with observations cited by various writers to the effect that hashish takers seem to lack the ability to learn (viz. to generalize) from experience, and with a whole host of other findings and hunches relating to personality and to motivation.

This hypothesis, however, should not be taken as stating a causal relationship between drug consumption and slow learning. What we have been presenting shows only the limits of function deficit that could be causally related to drug effect, if by some other design of experiment any such relationship could be established.

Conclusions

The work reported in this paper provides two sets of reasonably reliable answers to a number of questions concerning hashish consumption in Egypt; one set of answers was elicited by standardized interviews, and the other was triggered by objective tests for the assessment of well defined psychological functions.

Of particular importance among our factual findings and inferences are the following:

  1. That cannabis takers crave for agents acting on the CNS more than non-takers do;

  2. That both early beginners of the habit of cannabis taking and heavy users display more socio-psycho-pathology than late beginners and moderates respectively;

  3. That cannabis users have a lower criminal record than controls; and

  4. That cannabis takers in general are slow learners compared with non-takers.

The full report, under preparation, will, we hope, give the answers to scores of questions but there will always remain other aspects of cannabis use that would need further investigation. Elucidation of the immediate effects of various doses of cannabis (of known Δ 9-THC content) by various routes of administration on objective test performance is one such aspect. Another aspect would be a comparative study of immediate and long term effects of cannabis and other mood modifying drugs on a number of operationally defined psychological functions. A third aspect would be a comparative study of the responses given by other samples of cannabis takers in other societies representing different cultural frameworks.

For a well-integrated scheme for the development of objective knowledge about the effects of cannabis on man, comparative studies (across societies and across drugs) are of utmost value, since they provide the only valid basis for sound generalization. The point that should always be promoted, as fundamental to the achievement of this objective, is the adoption of a systematic scientific approach utilizing standardized tools, control data and adequate techniques of statistical analysis.

Summary

  1. This paper reports on the results of interviewing and testing 850 Egyptian cannabis takers and 839 controls, all being male prison inmates.

  2. Two standardized interviewing schedules, one for experimentals and one for controls were administered to the respective groups, together with 16 objective test variables, speed and accuracy of psychomotor performance, initial reaction time, immediate memory, visuo-motor co-ordination and time and length estimation.

  3. Smoking was found to be the prevalent method of cannabis use. The majority of hashish takers use the drug in the evening. Most of the habitués use hashish in groups, yet more urban than rural dwellers are lone users.

  4. Frequency of drug taking per month and of attempts at interruption of hashish use was found to be related to age at onset of the habit; the earlier the onset the higher the frequency of drug taking and the fewer the attempts at interruption.

  5. Comparison of heavy takers with moderates revealed the following: incidence of divorce was more frequent among the former than among the latter; fewer attempts at interrupting hashish use were reported by heavy takers; more heavy takers than moderates take opium, alcohol and coffee; more heavy takers than moderates stick to smoking as the preferred way for hashish use; more heavy takers than moderates take cannabis in the day time as well as in the evening.

  6. A positive relationship could be established between duration of hashish use and opium taking.

  7. Early onset of regular cannabis consumption was found to be associated with exposure to influence of a drug-taking person as an example within the family circle, combined with the effect of the disappearance of father or mother, and the intrusion of a step-parent.

  8. Regarding frequency of criminal offences, other than those dealing with pushing and/or taking narcotics, hashish takers were found to have a significantly lower record than controls.

  9. The differences between hashish takers and controls on objective tests were positively correlated to the level of education of the comparable groups; the higher their educational achievement the larger the discrepancies.

  10. Controls scored significantly better than hashish takers on most of the objective tests. Only on Time Estimation did experimentals earn consistently better scores than controls.

Acknowledgements

Thanks are due to Dr. A.M. Khalifah, Director of the National Centre for Social and Criminological Research, for his untiring support without which this work would have been impossible; to the prison authorities who showed an admirable degree of flexibility to meet the requirements of this research, and to Mr. M. El-Sayed for his aid in data processing and computation.

Bibliography

Anastasi, A., Psychological testing, London: Macmillan, 3rd ed., 1968.

Committee for the Investigation of Hashish Consumption in Egypt, Hashish consumption in Egypt; research in progress; Report I: The interviewing schedule; construction, relibility and validity. Cairo: Publications of the National Centre for Social and Criminological Research, 1960 (in Arabic).

Committee for the Investigation of Hashish Consumption in Egypt, Hashish consumption in Egypt; research in progress; Report II: Hashish users in Cairo City: A pilot study, Cairo: Publications of the National Centre for Social and Criminological Research, 1964 (in Arabic).

Cronbach, L. J. Essentials of psychological testing, New York: Harper, 2nd ed., 1960.

Inkeles, A. and Bauer, R. A. The Soviet citizen, Cambridge: Harvard University press, 1959.

Maccoby, E. E. and Maccoby, N. The interview: A tool of social science, Handbook of social psychology, G. Lindzey ed., Cambridge: Addison-Wesley, 1957, 449-487.

Mowrer, O. H. Learning theory and behavior, New York: J. Wiley, 1960.

Reitan, R. M. The relation of the Trail Making Test to organic brain damage, J. consult. Psychol., 1955, 19, 393-394.

Shapiro, M. B., Field, J. and Post, F. An enquiry into the determinants of differentiation between elderly organic and nonorganic psychiatric patients on the Bender Gestalt Test, J. ment. Sci., 1957, 103, 364-374.

Soueif, M. I. Hashish consumption in Egypt, with special reference to psychosocial problems, Bulletin on Narcotics, 1967, 19/2, 1-12.

Soueif, M. I., Abd-el-Nabi, S. and Helny, A. Objective assessment of psychiatric changes produced by reserpine in Egyptian schizophrenics, Acta Psychol., 1964, 22, 85-99.

Taylor, J. A. A personality scale of manifest anxiety, J. abn. soc. Psychol., 1953, 48, 283-290.

Trouron, D. and Eysenck, H. J. The effects of drugs on behaviour, Handbook of abnormal behaviour, New York: Basic Books, 1961.

Wechsler, D. The measurement of adult intelligence, Baltimore: Williams and Wilkins, 3rd ed. 1944.

Woodworth, R. and Schlosberg, H. Experimental psychology London: Methuen, 3rd ed., 1954.