INFLUENCE OF MAGNESIUM PEMOLINE ON LEARNING TO READ

• Document Type: SPECIALCOLLECTION
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): 00146159
• Release Decision: RIPPUB
• Original Classification: U
• Document Page Count: 14
• Document Creation Date: December 17, 2024
• Document Release Date: January 15, 1983
• Publication Date: April 9, 1969

ANNEX
SAINT FRANCIS MEMORIAL HOSPITAL
900 Hyde Street, San FrancisCo
CURRENT PROBLEMS IN RESEARCH �
Influence of Magnesium Pemoline on Learning to Read
Progress Report
This paper is a preliminary report on an experiment
designed to test the influence of magnesium pemoline on a
complex learning task, learning to read.
Learning may be defined as the modification of behavior
by experience, or stated more simply, the acquisition of '
skill or knowledge. Memory is the capacity to recall past
thoughts, ideas and mental images. Sometimes the definition
of memory is extended to include the capacity to perform
previously learned skills. For practical purposes, the
words learning and memory describe similar or identical
things. Learning is a process; memory is a capacity or a
storage bank.
For the past five decades it has been accepted gen-
erally that 4!"1P process of learning must be a chemical or
zt physical and chemical phenomenon. � However, very littic
was known about its details. Quite recently, a mass of
research has converged on the problem, as illustrated by
one bibliography of 571 papers (1).
The vast majority of the reports in this area deal
with experiments on animal subjects and in most instances
the learning tasks are extremely simple, such as learning
T-mazes and learning a conditioned avoidance response in.
a jump-out apparatus. Where human subjects have been used,
learning tasks have been limited to problems such as those
using a discrimination-reaction apparatus, or by reproduction
of a design or picture, exposed and then removed from sight.
Many investigators exhibit an understandable tendency to
interpolate data from experiments of this kind to practical
problems of education, mental retardation or senile memory
deficits.
It is now feasible to test the interpolations from
simple learning tasks to a complex, time-extended learning
prc'blem, specifically, learning to read. A new system of
instr,.:ction, Conversational Reading, provides a means for
accelerated reading instruction(2). Persons who are literate
but are not necessarily trained teachers perform the-
teacning role. Reading skills can improve up to several
-grade levels durin an 8-12 weeks teaching pe _d, The
system of instruction is well adapted to prison teaching
and was used in a prison, the California Medical
Facility at Vacaville, for the present study:
There is not complete unanimity of opinion regarding
the action of magnesium pemoline. Plotnikoff reported
that the drug enhanced the acquisition and retent4ion of
a conditioned avoidance response to electric shock in
rats, in contrast to methamphetamine, which did not enhance
this response (3). Beach and Kimble, using a similar appa-
ratus, found that rats injected with magnesium pemoline
had an increased spontaneous activity, that they tended
to jump more Quickly at a conditioned stimulus, but they
did not exhibit "enhancement by magnesium pemoline" of
learning and memory (4).
Concurrently, observations were being made on the
effects of magnesium pemoline on human subjects. Cameron
administered the drug to a group of patients with senile
brain changes and found an increase in alertness and a
reported improvement in the ability of subjects to repro-
duce geometric drawings (5). Ronald Smith, using refined
psychological methods for measuring short-term memory,
found no facilitation of learning, memory or performance
in normal adult men (6). Cameron criticized Smith's
conclusions, stating that Smith tested his subjects 3 hours
after drug administration, whereas Cameron felt that the
drug achieved statistically significant "improvement"
only after one month of adm'llie"..r='-4-n (7). rmc,erm sub-
mitted a table in this paper which showed an increase of
"Mean I.Q." from 73.5 to 82.2 over a month. Also, Cameron
implied that "brain-damaged humans" might respond better
to magnesium pemoline than normal subjects.
The literature regarding magnesium pemoline which
has been cited may be summarized as sometimes open to �
criticism of experimental method, sometimes contaminated
by anecdotal material, and generally contradictory. One
of the most interesting controversial points in the lit-
erature was the question of whether magnesium pemoline
acted to stimulate RNA polymerase activity. Glasky and
Simon reporting in the affirmative (8), and Morris, et.al.
defending the negative (9).
Experimental Procedures
The present study was undertaken at the California
Medical Facility at Vacaville, a state prison*. Volunteer
Supported by a research grant from Abbott Laboratoriea,
North Chicago, Ill. Grateful acknowledgement is made to the
Department of Corrections of California, to Lester J. Pope,M,D.
Superintendent, C.M.F., and to Ralph Urbino, Research Direc-
tor, Solano Institute for Medical and Psychiat-ric Research. �
-2-
�
.40ti=earsasau
�
subjects were selected from the prison population according
to the following criteria: 1-Q. 85 or above afr2 2 ,31:i. yea
below the level of reading skill which would be expected
from schooling and I.Q. From a pool of 50 subjects, 20
pairs of men were selected, so that each member of a pair
was as close as possible to his opposite number in, I.Q.,
schooling, measured reading skill, race and cultural back-
ground. Through a system of random numbers, the men in
each pair were split to form the experimental and control
groups. Thus, in the beginning, experimental and control
groups were made as comparable as possible. Later losses
of men, principally through transfers out of the institu-
tion, but in some cases because of abnormal initial lab-
oratory findings such as elevated SGOT, resulted in-some
replacements which were not paired as accurately as the
original group. Members of.the experimental group received
a 25 mg. tablet of magnesium pemoline each morning; members
of the control group received a placebo, :Throughout the
experiment there was no instance wherein any individual,
subject, inmate teacher, or investigator broke the code.
All subjects were led to believe that they were taking the
drug; no subject ever questioned this. There were no .
illnesses attributable to-the drug, and no complaints of
adverse reactions.
The principal teaching activity was carried out
between 6 P.M. and9 P.M.evenings. Individual instruction
was supplemented by language laboratory tapes and by coor-
dinated assigned reading. Enthusiasm for the program was
great. One 17 year-old, deemed unable to sign a waiver for
liability immunity because of his age, carried his petition
to remain in the study to such an administrative level
that he was allowed to remain in the teaching program,
without medication or placebo. He is not included in the
statistics.
Most subjects completed the entire 60 lessons of .
.the Basic Program of Conversational Reading, Approximately
12 weeks. They were tested prior to the experiment, at .
the 40th Lesson (8 weeks), and after the 60th Lesson. A
few subjects were transferred out of the institution before
completing the 60 Lessons, and for these men, test scores
run only to Lesson 40. �
Measurement of reading skills deserves some discussion.
A cardinal rule, often disregarded, is that a method for
teaching a skill such as �reading must be measured by an
instrument or by instruments extrinsic to the method being'
studied. Otherwise, if the measurement is intrinsic to the
method, such as a vocabulary test made up of words taught,
spuriously high improvement scores are found. In the
- 3 -
v�ferrs.v.sloarionrasasg..--uunviadv,,zie4 4it55,,,JSK
present study, two quite different commercially-available
tests were used, the Stanford Achievement Test for Reading
and the Gilmore Oral Reading Test. The former has a word-
meaning or vocabulary section and a paragraph-meaning, or
comprehension section. The Gilmore is a test wherein the
subject reads selected and standardized material aldua and
is questioned on content. It is scored according to
vocabulary and comprehension. All subjects were tested
with both of these instruments before medication or instruc-
tion began, at the 40th Lesson, and at the end of the experi-
ment after completion of the study. A few subjects were
transferred from the institution after the 40th Lesson
and had no testing after the 60th Lesson.
Results
The results of this experiment are expressed in
reading test scores, or measures which are designed to
indicate the grade level.of a subject, measure his improve-
ment with.training, and in this experiment determine if
magnesium pemoline has a measurable effect on the learning
process. Measuring instruments are two commercially-avail-
able tests, one of which (Stanford Achievement Test) is
r14,-on4-=.4 c4lcon4.- r=aer14ng =1p411co ^A-11=r ((i4lm--.
Oral) is based on oral reading, followed by questioning to
determine comprehension. Alternate forms of the tests
are used to avoid practise effects.
The actual scores of the tests are expressed in
grade levels. Thus if a subject scored 4.0 before training
and 5.5 at the end of 60 lessons, it would be concluded
that he increased in reading skill, according to the test,
by 1 1/2 years.
Experimental and control groups in this -experiment
were compared with regard to both tests and at testing
after the 40th Lesson and after the 60th Lesson. Both
groups improved, but there was a consistent tendency for
the control group to improve more than the experimental
group. Although the average differences sometimes appeared
to be appreciable, simple statistical measures of signifi-
cance of difference failed to show that any single differ-
ence was significant. It was our opinion that the array
of differences favoring the control group could not be
manipulated statistically as a set of independent variables,
since all were part of a single experiment.
4
"-\
It is possible that there are more appropriate
ideas regarding statistical interpretation. Therefore,
we have decided to confine ourselves to presentation of
raw data and means in this preliminary report.
Table I indicates the grade level reading scores
of 22 experimental subjects on the Stanford Achievement
Reading Test, and the Gilmore Oral Reading Test before
training or medication, after forty lqssons (8-10weeks)
and after sixty lessons (12 or more weeks). Table II
is similar to Table I, except that control group data are
presented. Table III presents the means of the Stanford
and Gilmore tests for experimental subjects. Table IV
presents the means of the Stanford and Gilmore tests for
control subjects.
Finally, Table V presents the mean grade level
gains in reading for the control and the experimental
groups, after forty lessons and after sixty lessons, on
the Stanford, the Gilmore and the means of the two inde-
pendent tests. This table represents the average gain in
grade level years. The average gains range from .61 years
to 1.77 years for the learning period. Comparison of the
control group scores with the magnesium pemoline experi-
mental group scores indicates an 11 out of 12 superiority
of control group gains over experimental group gains.
One question of experimental design was thought
to deserve consideration. It was stated earlier that the
original experimental and control groups were set up with
subjects in each group paired for I.Q., tested reading
level and other pertinent variables. Later, with drop-
outs and transfers, it was necessary to introduce new
subjects in one or the other groups who did not have oppo-
site numbers. To check the possibility that these changes
may have introduced new factors, a table was made which
included only subjects who were among the original pairs.
Table VI presents the means of the Stanford and Gilmore
tests for paired individuals only. Members of each pair
are opposite one another. It will be noted that controls
improved on the average by 1.32 years, while experimental .
subjects improved .67 years, at the 40th Lesson. Similar
differences are seen at the 60th Lesson level although there
were 4 drop-outs among the control group. It is thus apparent
that the observed but not statistically significant differ-
ences between experimental and control group exists when
the cases are limited to those originally paired.
- 5 -
Discussion
One interpretation of the data presented in this
paper is that no evidence is adduced to support the hypo-
thesis that magnesium pemoline, administered in a daily
dose of 25 mg. over a period of many weeks, facilitates
learning in a complex, long-range learning situation,
specifically a reading training program. The conclusion
suggests itself that generalization regarding the functions
of "learning" and "memory" from earlier experiments may
have been prematuke. The possibility suggests itself that
the animals in Plotnikoff's experiment and ths' human sub-
jects in Cameron's experiment may have performed as they
did because they were stimulated or made more alert, and
not because their learning was reinforced. A controlled -
human experiment reported by Gelf and et pl., demonstrates
the stimulant effects of magnesium pemoline of fatigued
subjects (10).
Close examination of the data reported here leads
to another interpretation. At Lesson 40 and again at
Lesson 60, both the Stanford and the Gilmore measures of
reading proficiency consistently show the control group
to be leading the experimental group in reading improve-
ment. Preliminary calculations not reported here indicated
that no single comparison of control and experimental
groups was statistically significant. No final conclusion
can be made that the control group subjects in this
experiment were better learners than those given magnesium
pemoline. However, the consistency of the data could lead
to the speculative hypothesis that magnesium pemoline could
have a deleterious effect on learning and memory.
.A modern view of learning is that it can be divided
into at least two phases, an early, largely electrochemical
or reverberation circuit phase, and a later consolidation
phase which depends on the synthesis of specific neuronal
nucleoproteins (11). Conceivably, a drug .could-have a
favorable effect on the first phase and a deleterious
effect on the second phase. .
The data presented in this paper which are at most
suggestive that magnesium pemoline may have an adverse
effect on learning could be related to the findings of
Burns et al. (12). Subjects were required to learn a
complex discrimination-reaction problem. Magnesium
pemoline, as well as amphetamine were reported to have a
possible deleterious effect on leazning, although there
"was an insufficient number of cases to afford statistically
significant results. The Burns experiment would certainly
be an example of first-phase memory, while the experiment
reported here, dealing with long-term accuisition of
reading-skills, is an example of second-phase memory.
One criticism of this study would be that the dose
of magnesium pemoline was insufficient. It is possible
that administration of larger amounts of the drug would
clarify some of the problems which have been raised, and
at the same time afford an opportunity to search for side-
actions of magnesium pemoline.
Summary
Prisoner volunteers, interested in improving educa-
tional deficiencies in reading, were given an intensive pro-
gram in reading training over, a period of 10-12 weeks and
concurrently given a daily dose of 25 mg. of magnesium
pemoline. Control subjects, equally motivated and simi-
larly selected, received the same training and placebo
medication. The tested reading skills of both groups of
subjects improved markedly. By test, control group subjects
improved consistently more than experimental group subjects
but the differences in improvement did not reach levels of
statistical significance.
James A. Hamilton, Ph.D., M.D.
Farel D. Footman, B.A.
Number 6
April 9, 1969
References Cited
1. BOgoch, S.: The Biochemistry of Memory (Oxford Univ.
Press, London, 1968), p. 219-243.
2. Hamilton, J.A., Brimley, G.M., Footman, F.D., Schauf,E.T.,
and Petraske, A.R.: Current Problems in Research: Con-
versational Reading. St. Francis Memorial Hospital,
San Francisco, 1967.
3. Plotnikoff, N.: Magnesium Pemoline: Enhancement of
Learning and Memory of a Conditioned Avoidance Response.
Science,151, p. 703 (1966).
4. Beach, G. and Kimble, D.P.: Activity and Responsivity
in Rats after Magnesium Pemoline Injections. Sciencei155,
p. 698 .(1967).
5. Cameron, D.E.: Evolving Concepts of Memory. Mss., 1966.
This manuscript was Dr. Cameron's address at the meeting
of the Society for Biological Psychiatry as reported in
Time, June 24, 1966.
6. Smith, R.G.: Magnesium Pemoline: Lack of Facilitation
in Human Learning Memory and Performance Tests. Science,
155, P. 603 (1967).
7. Cameron, D.E.: Magnesium Pemoline and Human Performance.
Science,157, P. 958 (1967).
8. Glasky, A.J., and Simon, L.N.: Magnesium Pemoline:
Enhancement of Brain RNA Polymerases. Science,151,
P. 702 (1966).
9. Morris, N. R., Aghajanian, G.K., and Bloom, F.E.:
Magnesium Pemoline: Failure to Affect in vivo Synthesis.
Science, 155, P. 1125 (1967).
10. Gelfand, S., Clark, L.D., Herbert, E.W., Gelfand, D.M.,
and Holmes, E.D.: Magnesium Pemoline: Stimulant Effects
on Performance of Fatigued Subjects. din. Pharm. and
Therapeutics, 9, p. 56 (1969).
11. Flexner, Dissection of Memory in Mice with
Antibiotics. American Scientist, 56, p. 52 (1968).
12. Burns, J.T., House, R.F., Fensch, F.C., and Miller, J.G.:
Effect of Magnesium Pemoline and Dextroamohetamine on
Human Learning. Science, 155, p. 849 (1967).
Subject
Table I
Grade Level Reading Scores
Magnesium Pemoline
Stanford Reading Achievemeni-
r 13
16
as 17
n 21
Base
Score
4.5
2.5
4.3
5.9
23
4.2
za 25
5.7
6
5.5
a 28
5.9
38
5.4
e 45
5.1
id 47
4.1
49
5.2
50
2.1
el 51
5.9
Ir-.4
,.......
..,r ...1. 1
an 59
2.3
ngs 60
4.9
� 48
5.1
52
3.5
ns 54
4.8
.on JC
2.9
f YC
3.9
Sum
988
Mean
4.49
N
22
Improvement
After
Lesson
Forty
After
Lesson
Sixty
6.7
7.5
3.5
3.9
5.3
4.3
6.4
7.1
5.1
5.1
7.2
7.5
7.1
7.3
4.9
5.5
4.7
5.2
4.9
5.1
4.2
4.6
5.4
6.7
3.0
3.3
5.1
5.2
5.3
5.5
3.6
3.9
4.9
5.3
6.5
4.3
5.6
6.3
6.5
3.4
4.5
4.3
5.1
1121
1147
5.10
5.46
22
21
.61
.97
Gilmore Reading Test
After After
Base
Score
Lesson Lesson
Forty Sixty
5.1
5.2
5.5
4.5
4.4
4.4
4.1
5.2
5.5
5.2
6.6
7.1
4.2
4.7
5.2
3.2
5.4
7.4
5.3
6.6
6.5
4.5
4.8
4.9
4.3
5.0
6.1
4.4
5.1
5.6
4.5
5.5
5.3
4.1
5.4
6.2
3.1
3.3
3.6
5.0
5.8
5.G
5.2
6.0
5.6
2.8
4.0
4.5
3.6
5.5
6-1
3.8
5.8
3.6
4.9
5.1
4.1
4.7
5.2
1.0
1.5
3.0
3.8
3.9
4.2
894
1093
1126
4.06
4.97
5.36
22
22
21
.91 1.30
F
1
Table .11
Grade Level Reading Scores
Control Group
Subject
12
: 14
.15
rd 18
rt 22
S 30
e 31
33 .
35
i 36.
s 37
.n 46
. 58
41
son
65
24
27
132
t41
t57
!SP
Stanford Reading Achievement
After After
Base Lesson Lesson
Score Forty Sixty
63
,
� Sum
Mean
� Improvement
Gilmore Oral Reading Tesi
After After
Lesson Lesson
Forty Sixty
Base
Score
�
5.2
5.1
6.1
5.9
5.0
3.7
3.5
5.0
3.2
5.5
5.7
8.0
6.9
4.2
5.9
5.1
6.4
5.9
4.9
5.8
4.3
6.8
9.0
4.7
6.2
4.9
4.7
5.2
5.4
4.9
4.0
7.8
9.5
4.5
5.9
1.9
4.9
6.1
2.2
4.6
5.3
5.2
6.8
4.2
5,5
4.2
5.8
6.1
4.3
5.3
6.3
7.0
4.3
5.3
5.2
6.3
6.1
3.4
6.4
4.1
5.5
7.4
4.4
5.8
4.4
4.2
3.2
4.4
4.5
4.9
7.2
6.1
4.8
6.1
5.4
6.4
6.7
4.6
6.4
4.6
7.3
4.7
6.6
4.5
4.2
4.3
4.4
6.0
5.9
5.8
7.1
4.7
5.8
4.7
5.2
3.2
5.8
3.4
5.2
4.6
4.7
5.3
4.4
4.5
1014
1278
1084
967
1220
4.61
5.81
6.38
4.40
5.55
22
22
17
22
22
_1.20
1.77
1.15
5.5
5.5
6.9
5.8
5.8
5.3
6.8
4.5
5.4
5.5
5.8
5.5
6.0
4.1
6.5
6.8
5.0
967
5.69
17
1.29
_
p.
Table III
Grade Level Reading Scores
Magnesium Pemoline
Subject Mean of Stanford and Gilmore Tests
Number After After
Base Lesson Lesson
Score Forty Sixty
13
4.7
5.9
6.5
16
3.5
3.9
4.2
17
4.2
5.2
4.8
21
5.6
6.5
7.3
23
4.2
4.8
5.2
25
4.5
6.3
7.6
26
5.4
6.8
6.9
28
5.2
4.9
5.2
38
4.9
4.9
5.7
45
4.7
5.0
5.4
47
4.3
4.8
5.0
49
4.6
5.4
6.5
50
2.6
3.2
3.4
51
5.5
5.5
5.4
53
5.2
5.7
5.5
59
2.6
3.8
4.2
60
4.2
5.2
5.7
48
4.5
5.9
6.2
52
3.5
4.6
5.3
54
4.4
5.5
5.8
JC
1.9
2.5
3.8
YC
3.9
4./
4.7
Sum
941
1104
1202
Mean
4.28
5.02
5.46
N
22
22
22
Improvement
.74
1.18
Table IV
Grade Level Reading Scores
._ _Control Group
Subject 'Mean of Stanford and Gilmore Tests
Number After After
Base Lesson Lesson.
Score Forty Sixty
12
5.5
5.1
5.8
14
3.4
4.5
5.3
15
4.9
6.9
6.9
18
5.0
6.1
5.9
22
4.5
6.5
7.5
30
5.1
4.9
5.3
31
4.3
6.9
8.1
33
2.1
4.8
5.3
;
35
4.7
5.3
6.1
36
_ 4.2
5.5
5.8
37
4.7
5.8
6.5
46
4.3
6.4
5.8
58
4.2
5.6
6.7
61
4.5
4.3
3.7
63
4.9
6.6
6.3
65
5.0
6.4
6.8
24
4.7
7.0
27
4.4
4.3
32
5.9
6.5
41
4.7
5.5
57
3.3
5.5
--SP
-4.6
5.2
Sum
988
1250
1030
Mean
4.49
5.68
6.02
22
22
17
Improvement
1.19
1.53
���
. , "Ii";X:t � rk..c
Table V
Mean Grade Level Gains in Reading
Control Group
Magnesium Pemoline
Forty
Lessons
Sixty
Lessons
Forty
Lessons
Sixty
Lessons
�
� Stanford
1.20
1.77
.61
..97
Achievement
Gilmore
1.15
1.29
.91
1.30
Oral
;
Mean of
--Stanford and
1.19
1.53
.74
1.18
Gilmore
Table Vi
Comparison of Subjects Originally Paired
Mean of Stanford and Gilmore Tests
Control
Group
After
Lesson
Sixty
Subject
Magnesium Pemoline
Subject
Base
Score
After
Lesson
Forty
Base
Score
After
Lesson
Forty
"%fter
Lesson
Sixty
24 4.7
7.0.
13
4.7
5.9
6.5
12 5.5
5.1
5.8
51
5.5
5.5
5.4
46 4.3
6.4
5.8
59
2.6
3.8
4.2
27 4.4
4.3
---
16
3.5
3.9
4.1
18 5.0
6.1
5.9
26
5.4
6.8
6.9
14 3.4
4.5
5.3
45
4.7
5.0
5.4
41 4.7
5.5
---
17
4.2
5.2
4.8
30 5.1
4.9
5.3
28
5.2
4.9
5.2
31 4.3
6.9
8.1
53
5.2
5.7
5.9
22 4.5
6.5
7.5
60
4.2
5.2
5.7
57 3.3
5.5
---
47
4.3
4.8
5.0
58 4.2
5.6
6.7
25
4.5
6.3 ;
7.6
37 4.7
5.8
6.5
38
4.9
4.9
5.7
63 4.9
6.6
6.3
21
5.6
6.5
7.3
15 4.9
6.9
6.9
50
4,6
3.2
1.A
65 5.0
6.4
6.8
49
4.6
5.4
6.5
Sum 729
940
�
769
717
830
896
Means 4.56
5.88
6.41
4.48
5.18
5.60
-.
N 16
16
12
16
16
16
Improvement
1.32
1.85
.70
1.12