TRUTH DRUG (T.D.) WITH ATTACHMENTS

• Document Type: SPECIALCOLLECTION
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): 00144773
• Release Decision: RIPPUB
• Original Classification: U
• Document Page Count: 32
• Document Creation Date: November 22, 2024
• Document Release Date: January 15, 1983
• Publication Date: April 5, 1946

1
5 April 1946
fiElORAUDUfl FOR THE FILE:
SUBJE.C.T: Truth Drug (T.D.)
The following informati2hIpassed out by --
CY /) .. � � 40 ,r 7_,,i: ;el! in discussions vi.th,,,,....-.,..,..,,
/47 -: -- � . . . � � . -.- ..
on 30 and 31 January 1946. This mem(randum should be read
C.- in conjunction with the notes prepared by on this
subject, and with the report of the 055 board which conducted
linical tests, copies of Waich are attached.
A. THE DRUG: �
1. The Drug is a liquid acetate preparation of cannibas
indicia (marijuana or hashish). It is prepared only by
chemists of the, .4-and
is produced as an oil Which has no taste, odor or colcr.
The preparation is an involved process, done principally
in the Laboratory. There is no reason
to believe that any other nation or group is familiar with
the preparation and use of this particular drug. ,This arhg
is almost impossible to analyze, and, for this reason, there
would be little.danr:er of eompromise if samples were to be
obtained by any unfriendly agency. This oil is not miscible
in any liquid except alcohol or similar product, an' l then only
partially. Accordingly, it cannot nornally be used in a drink
as it retains its globulartorm, and being henvier tAart water
will sink to the bottom. if vigorously shaken in a liquid, it
will separate into tiny droplets and remain in a state of
visible suspension hntil the nroplets giadually sink. The
is now experimenting with a tinctuie which
will allow proper proportions of the drug to be insoTted into a
drink.
B. PPETARAFIOU AID alT7I;lInfr'ATION
1. At the pre:lent Lime the oil is prepared by thr,
in ampules of 3 c.c. Inasmuch as the maximum dose
is .03 c.c. the present ampules are too large for our use, and
it should be suggested that ampules of .5 c.c. capacity be pre�
pared. The normal minimum dosage is.01 c. c. In any case, it
ii
fikt
- 2 -
s
has been found that the Drug must be inE:ested in order to be
effective.. This can be done either by eatin(7, as in candy
or food, or by inhalation, by means of a cigarette. It has
been found that the cigarette method generally gives a quicker
effect. In either case the drug is inserted into the- food
or tabacco by means of a hypodermic syringe. In the .,case of a
cigarette, the needle should be thrust into the cicarette along
its axis for about one inch. Slowly withdrawing the needle,
the plunger should be slowly depressed so that the drug is
depositod evenly along this portion of the cigarette. Care
must be taken to avoid the drug 'seeping thmugh and staining
the paper. This procedure should be followed rith both ends of
the cigarette in order to iisure that the minimum dosage is
absorbed. It is obvious that the smokinr, habits of an in-
dividual will largely determine the amount of the drug which
he inhales. The operator should have observed these habits
in order that an adequate, but not excessive, dosage be insured.
In other cases, -the drug can be injected into any type of food,
such as mashed potatoes, butter, saladdressing, or in such things
as candy. By tills method a closer' check Can be maintainnci on
the actual amount of the drug which is -inresLed.
C. ::17.) In:Er:TS
1. The eperator er..n 1o11; for a reaction from Ura.ciraroLte -
method in about:15 minutoes and from the food met:led in about
ono 'half hour :Al 1.6 atinutes. These are not, alwayS dependable �- -
rules. 'rho Lime of reaction will differ with each person, and
it depends considerably on the physical conditiOn of the subjec:;�,,
the ameunt of fo�-41 or liquor he has already cci.sumed, the length
of Lime since his last meal, etc. The operator...must not bocume
J.:vatic:it if the effect is slov in appearing. Caro must be
ta';:en not to admin7.nter unnecessar:7 amounts _of the dr.ii? because,
alt:louch the c1ru ill not ihj�....re t.he Abjeet, -,.1.11.zprobably
cause him to lose consciousness in the sae mahher s ..ohe who is.
completely drunk, and no 'further q testionin;-, he possible.
An adminis:a�ul,ion of Irora .01 c.c. LO .03 C .0 run:Iran-1
have an effect la:; tinf,, t ons huLr. !lowevor, I t
been neces:;ary for the administration to 1.;a1e ;dace o-rer 2 e.)r
3 hours. Therefore, it is e.ssential operaLion of 4.hin
kind not be attempt:xi uhless there, is reasonablo cerrtaintv that
there will be no interruption over a period of several hours.
4 .,
- -
2. Tho effects of tho drug are similar in many ways
to the effects produced by alcohol. The brain reacts as
when intoxicated, but no delusions or halucinations are
prodaced -as is the cane in the une of dracs such az ncapolamine.
The drug appears to ram: all inhibitions and to deaden the
areas of thlbrain which sovorn an individual's discretion and
caltien. It also eccentuaten the senses and ma;:os mahlfent arc!,
strong characteristics of Lie individual. Sexual inhibitions
are lowered, and the sere of humor tn accentuated to the point
where any statemeat or situation can become extremely ninny
to Cle subject. On the other hand a person's unpleanant
charact'cristics may al!,n be 1,eightened. It may bo stated
that, generally r7ca1.ing, the reaction will be one of ereat
loqvatity and hilarity.
3. In all cases a prereqaisite for t:p effoedve use of
this druc is the creation of a friendly, .s7mpathetio, informal
basis for a discusnicn, which should extend over a period of
several hours. It nest not be used on a hostile subject, but
may nY'etimus be 1:sed on a nubject who is sespous, but not
hostile. It may prove desirable to spend considerable time
Ouring previoui`meetinn and discussions bofore hrinLIne about
proper envirenment for the use of the drno. Some cases of
its uses follow:
�
1
. a. Az oxa:Tles of I. forooping, the story was 'told
cf a visit to a Pri:Pner of War Camp in Virginia where only
bop-notch German military prisoners mare kept for interreoations..
Line Geman submarine -commander, who was censidercd oltstanding
in his field, ha: been the sele survivor of his submarine'a
sinking and IL was thought he had valeable information con-
cerning: (1) the depth taldlich the submarines cnuld ge with
safety, and (2) the morale of Germui submarine crews. Both
of these items wore of oxLremo interest to our Navy at the
time. Tiany attempts hld bef:n made to olytain. this iuforAution
from the commander withollb success as he na:;-extrenalyary.
At the tLue of the in question the commander was in-
vited over be the officers' club for a few social drinks,
which was more or less customary. After a few drinks and �
some ecnversation, it coild ho seen that while ho would talk
freely on mo!;t all matters,�he was well aware of the fact that
an attaapt mijht be made to obain informalpn from him. The
cpkertunity was found to rive him a ri;71rette which had been
loaded and after appromimntely half an hour had passed frau the
first administration, one of the company, in a round-about and
innocent manner, lead up to the question of maximum depth of the submarine.
.-7
�
The commander, although at this time definitely under the
effects of the drug, was still %very and indicated he-would
not give information on this point. Some time later after
two more cigarettes had been consumed, the talk was led around
to morale and he freely conversed about the general lowering
of morale of German submarine personnel. However, no infor-
mation was over obtained concerning maximum depth.
b. Another cano is that of a notorious dope peddler
and racketeer in New York City who was well-known to the
operator from his days of law enforcement work. The dope
peddler was telephoned and asked to come to see the
operator who described himself as being connected with
secret government work and who was in a position to give
the dope peddler an opportunity to help his country in the
War. Despite the character of the peddler, he evidently
harbored some feelings of patriotism and presented himself
when it was explained that it was thought his connections
in Sicily could be used for espionage purposes. Conversation
along this line was carried on for some time and no reference
Was made to any illegal activities of the dope peddler.
Although.the:,dope peddler did not drink liqubr, he was per-
suaded to take lust one liqueur to keep the operator -
n
company'. Also during the time, three loaded cigarettes
had been given to him at different times. The effect was �
cuite rapid and about 20 minutes after the third cigarette,
the peddler told freely of the many illegal activities in
which he had been and was then engaged concerning the "-
smuggling of narcotics. He then also nsmed many of his
associates and officials who were also involved in till's-
activity, although he was well aware that the person to
whom he was talking was, and might be in the future,
connected with law enfercoment activity.
A little later the peddler complained of feeling
very light-headed, as though he had had too much liquor.
He discarded this explanation, however, becaus:e he had
had only one drink. He did state that he had, been having.
a little trouble with his digestion and he had not eaten
for approximately 24 hours. The operator to cover up
this situation convinced him that the one liqueur on an
empty stomach and his fasting had made him sick and susEep-
tible to the alooholic fumes. Ho accepted this explanation.
c. Another case was one which illustvated tho rule
that a friendly basis must first be established and that
suspicion must be allayed: This case involved the
questioning of approximately 30 army officers in the
space of one day. All of the officers were suspected of
being Communist and the interview here being held to tr.; to
;7.
establish the suspicion. It should be noted that beca4se
of the speed with which the interrogations were carried out,
no mutuality of feeling could be first established, and
also because most of the men were in fact Communists and
were suspicious, their fears concerning the questioning were
not quieted. Nevertheless, the drug was administered in,
cigarettes and five of the thirty, even under the adverse
conditions, gave full information concerning their communist
affiliations.
4. It must be remembered that when the effects of the
drug wear off, it is probable that the subject will be
completely aware ct the fact that he talked freely and
was .possibly indiscreet. For this reason, it is desirable
where ever poeviblo to associate the use of the drug with
liquor, in order that this effect will not arouse suspicion.
Nevertheless, it ehould alsb be remembered that the subject.
subsequently may be expected to feel remorse, embarrassment
or alarm over things he knows he has said. Extreme reactions
in such ovent_could be: a) flight; b) suicide, or c)
personal danger for the operator. It is possible that
reactions of-,this type might outweight the benefits which
might be gained by the use of the drug.
D. COVER
1. Innemuch as the drug, syringe, alcohol for' cleaning,
are all contained in a kit, the posoescor of such a kit rust
have a logival reason 2or its presence In his effects. The
person in whose.care the drug and the equipment is exit:Hiatt:id -
should be covered by a story, preferably backed by documentary
medical evidence, which indicates that he is receiving insulin
shots, self administered, allergy shots, or something
similar. He should be prepared to demonstrate his cover
story if the occasion demands.
E. MISCELLANEOW
1. Tests hnvo shown that the dru,7 mny remain potent in
a cigarette for a couple of weeks end in one case it was
found that considerable reactixl Was caueed L7 o cigarette
in which the drur had been introduced approximstely six
months before. A.= a general rule, the dr17; would keep even
better in food unless the food itself were to spoil.
2. Possession of the drug is contrary to Federal law,
-7�
C
and a special license, obtainable from the Bureau of
Narcotics is necessary to make legal its possession.
This fact may be ueed to rrevent the retention of this
drug by persons not in the employ of the Government and
who should have no access to the drug.
3. The syringe must be graduated in .01 c.c. This is
very important in order to measure the doosge accurately.
A recommended model is that called the tuberculin syringe,
which is so graduated and which has a capacity of .25 cc.
After each use the syringe should be washed out with
alcohol to prevent gumming. The ampules containing the
drug are now packed in 3 cc containers. This is too much,
and a .5 cc or smaller ampule should be developed to our
request. The entire set, includinz cover equipment should
De orepered in a small medical kit.
shorld contacted by, .as a source
of the drug and the equipment.
4. Regarding knowled=wof this drug, it s e9ms that a
small number 'of people in and "are aware of its
existence. The drug defies all but the most expert and-
searching annlysis,qnd for practical purposes can be
considered beyond analysis.
�
I.
///11 2 Y
�
In the Spring of 1042 '.7ieneral Donovan assembled a
group to inquire into the possibilities of dei;e1oping�a
drug which would have the characteristic-of causing a
subject'to whom it was administered to tell the truth about
matters where it would be contrary to the interests of
subject. Prominent psychiatrists, biochemists, ote.,
collaborated under the direction of, of
Research and Development Department of 033 in an attempt to
produce such aNdrug.. It was alleged that the Russians had
such a substance and that the German:. also enployed a drug
in connection with the interrogation of prisoners of war.,
�
This group made extensive research on all possible drugs
including scopalamine, barbituates, peyote,' etc., and found
none of these substances to be satisfactory for various reasons.
One defect was the production of halucinations, particularly
manifested in the use of scopalamine. The cormkttLee finally
decided that the most practical substance available,was.a
deritive of cannabis Indicia (mvrijuana or hashish)
.chemists developed an accetate of this substance which
represonta the active inzredients and this Is the "TD" now
available. Most of the technical experiment was first carried
on by who was
then on the,Planninc Staff. Attempts were made to inject
the substance into a room in a spray, but this was unsatis-
factory because the carryinr scent was irritatinc to throat
and eyes. The cit:arette method was then extensively used
with some success, except that it was found dosage could not
be ronfrolled. Later nnd myself obtained
the issistance of the and made extensive experiments with
a croup of personnel on the various for-1s of sOminiStra-
tion, the results of which seemed to indicate internnl
administration -war the most satisfactory.
NumerousTield experiments vnre carried on, by the writer
' in connection with the project under personal
direction with some success. Some SUCCO3S was lind qn projects
undertaken on behalf of,the -locally. It was
� �
found that the Eretest success was achieved when an opportu-
nity was afforded to "set the stare" in euch a fashion as to
lull the suspicions of the subject that he mi:ht be the object
of interrocation. It was used in z:onnectIon wkth formal PaN
interror:ations with little or no results. It inks used in
connection with informal POW intorrorntionS where come detree
of "opine technique could be (Applied, with Food results.
.7.7,01111MT,r7TP77
TT. I:PORT
IVISTIlATIOA 07 U:T. OF TD IU INTERRJ0ATION
41th the technical nociatance of.
� lb
1
-
�c�-�
Thu object of this investiLation ,:9:n to
�
discover
a drug or drugs that would induce the betrayal of secret
information.
On April 1, 1D45, thc, ,roup at
boE;an work on the assigned problem of .
ascrrtnc whether tetrahydrocannabinal aceLate, herelifter
known an TD, is effective in oxtractinr, information during
interrocation of individuals who aro reluctnnt to yield it.
soon nfter this 1nverti7ation wn7 Initiated, it ,dis-
rovfrPd that ti,r(ffectivenwrr: at :,he 1ri. Je:lend:: to e lar0
extent on the persontility and moLivation of- te ubjecl beinc
interro7nted. The inventiat;ion was Lherefore lirondelu-,r1 to
include a study of the drur in rcla'c,Ion to prA-ron?lity nnd
motivation.
=OD
Oeneral Plan
Tho rloneral plan of tic 0:T01.1r:out Taw ds folloqs:
Personnel of the Offle of Stratecic Services who had Informa-
tion or a confidential or secret nature, were to be interro-
gated before and fter the adminictration of TD, to determine
how much of this information they could be induced to reveal.
If more information could be obtained while the individual
-2-
was under the influence of TD than at other tines, TD could
be considered an effective acont In interrocation. In-order
to determine the rulation of personality to the effectiveness
of Ti), the personality oC each subject ws7 .exenZned b moans
of psycholor:ical tests r,nta psychiatric interview.
Subjects.
Forty subjects were used in thls experlrent, of whom
twenty-four were enlivted 110 sizteen offier7. of the
Office of CtrateFic Zervices. They had been members of this
orcanization or periods of t:Ale vary117, from one 11')n1;h to
several years, and bad performed duties of various decrees
\
of ipoit;c andic:pon:libility.
All of the onlIrted exce7t tv:o were 7nnt without
special instructions. They were t;ohl only thet tl.oy would be
psychol,cical terqin7.- Two enlisted nen and all of tl:e
officers had been riven strict instructiJas bcfor, sul;mittinc
to L:le c:odoriment. This "brcifinc" consisted of Instructions
1) that on attempt would bo made tod,cco7er what Colt:l('n they
had performed and 2) that they were not to cii1e this .
informntion. The secret natnr or theIr irlronnnti3n ViR3
strersed.
Ratinr- of subjects' persJnalities
A series of psychol),Lical retinc, instruments was
administered to each subject. The results of these, tot-,ether
with a psychiatric interview wore used to determine the
individual's structure. structure. The following instruments were
used:
1. norschach.Test
2. Cornell Index
Je Cornell -.7ord form
4. Coodonouch Drawina-a-man- test
5. Self Ratinz toot
6. Work Interest Blank
7. P-S Experience Blank
B. Psychiatric interview
Intiniro.:ation
Ladh individual was subjected to nn interroshtion
in andtempt to Cisco/or what duties he had performed. Intorro-
cation was carried .out in an informal atmosphere before,
Curtnn and efler tht administration of TD �
Intorrocation wnr recorded by an of n 151c &phone
LlectricorC liccordirK7-Roprodu-in 7adhrne. .The
subject was not aware that his conversation ries beinc recorded.
These recordin7i; formed a nermanent record of the poychiatric
Illterview and of the interro;ntion of thn Entjectn, by means
ol which it was 'possible to ascertain 1.:Lc rate of speedh and
�
loquacity durina any peridd.
Rate of speech ant: loquacity were a;certglned bi
counting the number of words uttered by the -subject in two
minutes out of every- five durina the interrozetion and by
neacurins the lenath of time he talked in any two minute
plriod. By plotting those findinas it was possible to
to present graphically the effect of TD on rate and loquacity.
�
-4-
, -
?MIT II
Psychogalvanometer
About one-half the subjects were examined by.
means of the psychoralvanometer to explore the possibilities
of usin: th(i tenimiclue for uncoverinL; infor:ation no.t .easily
elicited. The tcchnicito employed was os:;ontially that made
ur.,,e of, In "lie cletc,ction" and will be described Lel�D�..v.
nmouilt of TI)
In on otterpl; t C.rytr.rmine t.:Le nr-,71,,nt of
verious ciunntItier from .01 to .00 cc by inhalation and
.01 to .02 cc by mouth, were administei�ed to the. forty
cub:ccLo. I. vnrj C vnred 1;11n1; ihhalntion nriounts
frort .01 to cc pr:ekl or no c-,ffoct
amounts of .00 cc were toxic in rao:zt oi. individualc to
v:horitliey ven ; .01 cc by mouth was discovered�Lo be
ef ective in four of five ividual, while a s developed
such a severe toxic reaction after .02 cc by mouth that this
�-1
quantity was not given again. The optinal amount by Inhalation
war f�-nind to be .04 cc in one ei;.arette, an .01 cc when
administered In rood. The s;nounts 1'ri'rr o rep recent the
drw-, 1�efore dilutio..1; in term); of t',.e volume
r '.r!lrienicdvo the opLinly.1 wo1:1.3 .(-; twice t.:11oLo
mentioned.
-1., .4,-, .�1.
�
These quantities are average; with a few indi-
viduals they may cause toxic reations, while with others
they will be inadequete in :Iroducinf�the meaximum desired
effe7:t loquacIty. There is no method ovnilale at pre: exit
to determine which individuals will become toxic with the
recommended quantities and which will require more than that
recommended to obtain the optimal effFet. It is, tereforc,
rr tirt the rr:erage arount be j�iven In all insances
the first time the druG lo !,drInistered.
None of the subjects used in this investi.iation was
ewer� that he wls under the influence or cliemical nc,ent
excr-)t Ui t:le nc� ihotnce in v1;11.C.) no:lerrItely severe toxic
reactIpns developed.
Latent- Period
When rlycn
bz� inhalntion, the effect ot Ti) beertrie
anrarent after a latent: perloC or und ;;-,:enty
When uy- mouth the latent; rc.Q.-ioil was between
one-1).:11 r.' aly1 one !lour.
Duration of actiun
When Given by inhalation the :aximum cfL'6ct or TD
wr.:.; manifest in twenty Lo thIrty minuc:; Fr-i that time
on it becomes less effect:Lye and Its net.I.on_alr-ost, entirely
disappeared in two to three hours.
When administered by mouth the maximum effect is
reached in one and one-half to two hours. This effect lasted
w
7
1 .
:4
I.
-6-
several hours and then gradually disappeared over the
course of six to ten hours. (see table I)
Effect of Repented Administration
Mien TDwes rriven a secend time by.inhalation
\
within a half to one hour, the effect was opproximately
equivalent to that when dollble the amount was given initially.
(Administration by mouth wn: never repeated on the same
day/ When TD was 7lven either -by inhalatioa or by r...)Lth
twenty-four hours after thu firot adninisLI�ation tha s'econd,
was more toxic then the orirInnl even thoncji both may have
been of the sae size. It would seem fror these findin:s
that the first edmihistretionnmaitizes the individual:Ito
the second and thnt n7enL sheuld not; be riven a second �
� ,
time within twenty-four hours.
Toxic Effects'of TD -
While in most subjects the effect of TD wao mainly
to increase feolinzs of: well bcin:, self-confidence, amiabi-
lity end loquacity, In some instancez retions resulted.
Thus, of rorty subjects studied five hid Iilfl toxic and
three moderately toxic rocctions.
The mild reactions were ell in 'Persons who had
teceived .05 or .04 cc of TD in one ciGarot1.0 and vier()
generally manifested by slurrinG of speech and sometimes by
sensations of warmth, rosltessness or irritability.
6.� ���
������0
� - - "ST, ".� '
-7-
II
��
Of the moderately tx:ic reactions one took the
form of an acute anxiety state characterized by paniC.and
war; the result of .03 cc TD in one ciGarette. It, ia interestia
to note that whqe nnder the influence oC TD this subject.
divulced all his secret information in a rual:*Of talk.
This subject's greatly ineresed loquacity while in a state
of near panic, was in contrazt to the decreased loquacity
url117 found in those .thone anxiety rade them cautious or.
unfriendly.
The second subject in this group had been :iven
.03 cc of TD In n cigarette on the ftrst day of his stay
and a7nin on-tbe rocond 6ay. On the f...econd day he deve.1)ped
such symptoms az dizzinezrz, unstrointevtinal distress,
palpitation, intense lar,situde, heacachej "sense of Impnding.
. -
disaster", difficulty In concentration and serious memory -
defect for recent events. His loquaci - %::no greatly decreased.'
The third sul)ject or tms sroup W115 the only person
to ec was administered in foo,.;.- His reaction was
one of etrome and 1nca aci1;&In eiiicty, kluO.ng which he
was irritable, threateninc, complained of 41-rzinesz and. a'
feeling of "bcing. two difff.reat poplc".--Decau.rn of the
nature of hi:: symptoms no attempt was mde nt interro:ation.
Although he talked a sreat decl during the eppode ho gave
no military information.
41. .������ �
-3-
1
.1'
4
a
4
1
Effects of TD in terms of loquacili
The effect of TD on loquacity was to 'names� the-
.
output of woDds In a given period of tine. This war accom-
plished not so much by the subject's talking faster but 14
his tnkine more Or the time, that Is witll fet7er or vlibrtor
pauses, and with losu prompting. As shown in Tnble II for
eiGht "normal" sub.iocts the average rates of speech before and�
after TD were 3.5 and 3.7 words per second, respectively.
This represents on indicnificnnt chance. Thu nvern:e per-
centrue or tine talked (number of seconds the subject talked,
divided by the 120 secondn of the two minute period studied).
however, rose�from forty percent to fifty-sin per cent, an
increase of forty per cent. Mc, inr!rerPe In pf,recntn!...,7! of .
tire talked renced.fr.m. nineteen percent to eic.hty-four per
cont. In other words, nubjects talhed nnither faster nor -
�
slower after TD) but they talked more 'of the tire bedene
they paused less often or for shorter periods, or Deeded to
be stimulated less often by comments from the interrocntor.
As shown in Table III In three other. subjects there
wan either no increase or slicht decreese tla loquaity.follow-
Inc the administration of PD. It is cicnifIcant thnt while
the eicht subjects described qbovo were juC, cc ti little
or no porconality clinturbances on the is of psychiatric
interview and psychol)gical tests, these throo.were judged
to have significant degrees of personality disturbances as
�������
umm�
114�11�
" � �
, , ,-2-1.�����,..7.:4�0��� - e�-� � � �-�--g�-�;�� � - � ������
� _ _
����
-9-
manifested in subject 11 by an irlieture personality, and
in subjects L and R by botli rebelliousnoss and
In brief, eight of eleven subject:7 studied hnd
sicnificant Inc -care in loguncity, one .1o2-2, no increase and
two talked less.' The (:A7ect 2C 'fl) t'hen, incry7c
loquac1t7, exce7t In tlIone :ndivjdoalr dis-
turbances maniferted n3 in-aturfty, rebell:o=nrr or
is rown 'c,he rffnct rern%r(7) n,111in-
In tnble IV
istranc of 7) on 1.v,scit-. 1)c 7rrn
instnnce tlInt while the effect of the firs1;
of TD ww; to incrcase loquacity the seno:Id [locrea:ed loquacity.
Fowevcr, on the day' afUer secopd ndninstveton lor.uacity
nc.nin IncreascO.
I': would 1-.11nn ercn Froln thenc fThni
nnount of TD stinulate-loquneity in "normnl" jriviil
lri Vr i: lrIC i'tintr deprcrn locrItcity.
In f1J7uree I:nnd It arc rhown two typist:1 inrtances
of t.he cffecte of TD ozi loquacitr
The subject whose 12quanity is :;'i: in i'ltre I
was given TO by routh (cub,Inct J). I will be reeft
dny 1:efore TD the nvera7,e percentnro of ti.r1c tnll-nd In nnch
two minute sn-irle wns thl.rty-soven nnrnent,_nnl 91,it nftrr
TD was sixty-eicht percent, an eighty-four percent incronse.
,
-10-
. -
Thic increase resulted benau,e the subjectts averaGe level
of talk after TD was at ti-e level of output that represented
his maximum before TD. Thue, the highest pereentage of tote
talkPd in any two minute period before .TD wac ei:hty-Cour,
percent and afterTD was eiuhty-sevon
before TD the subject was e::tremely variable in his output.;
the pr.ecenteGe of. time tc117ed in each two mlnute period
varied from seven to eight-i-four. Ti) he cclIsistontly
talked at the hich level, the secnnde Lallrcd ia each two
minute period vnryinz, from fifty Co el.:-four. Thiu
is shown in another way by the standard deviation of twenty-.
three before.p and ten after; and by the coeffitIent of
variability o: fifty-one before arvl felve aftAr. Mcce
rtatintical data 1n6icate tl,nt tefo"e Tr. the :.111ijact ::ometimes
talked a Erf.st deal and sometimes very little, while .f4.)r _
rome time arter the admintotratIon of TI) the prc6ictabl7
talked a great deal.
In figure Iris rhn the loquacity of n cubject
(3ubject 3) who was given T1) by iriiialation. It will bee, eon
that in this rubject there was Lin lncreaor .2f.cy 76rcent
in loquacity - from an nverare of '61.irty:e!.7111; percent of
the t,wo minute nanple onoken before 7) ;: r
ni.:71.1t percent !After TD. ThIn increase wn:: reachecl in
twenty minutes. Folllwinr: this increase there was a Gradual
decrease In loquacity to thirty percent of the two minute
sample, which was renched eighty minutes after the administra-
� �
..-
�..
. �
farm?
tion of TD.
Effects of TD on self-confidonce and rnpport .
�
rrom the point of view of InterroLotion an important
effect of TD w' t} induction .o,' o pf self-confi-
'
donee and security. Durinc the period that the subject was
under the influence of TD they actod is if thsy were in
corrIond of the 11-,Iletion, vnr] If 1:11cy hod the obtlity to
dfrcurn forbidden tanic.s. revcclin import:7nt 7-rterIal.
77i: -a: confir.!,cd by I:he co,imcnts. of the cl;ject in re:ponse
to t.:le (1,stton, "Did yol: feel diffPreutly on any day that
yo l7 w^re beret" ktypiecl rc:Iponsc wa, "On --nnnin 1.he day--
.
I felt very solf-confizlent nn r:F if I could 7.1cle yo:1 do,
what I wanted." The day mentionc(.: as in all cc c' t:10 (2.Ty
on which Ti) ;16 been :iven. This effcct is siZiilar,:to tlmt
prodi,ced by mbderato am.)unts of alcoholic bo.,;era;:es. It wa;
durin:: the period In wcii the nvlbjcct :1:1;7 :;yr:1; solf-co:Iridenco.
t1'.at e revenled signiCIont in:orr7at1ou.
The followinn nre e7amplen of one manner in which
inforrabion Virk3 revealed. 11.1)ject 7%, n Quarivennasfer. Corps
officer, while under the Influence of Ti) nai3 .rt t. a .tirio when he
wo: ny'!eoLltn:ly unwIttlnr.-ly i:r.r.7artef1 the tnforma-
.
tion that he hod otfendorl the Office of SLrateric Services
School In Wyonlin. Since the school had been in operation
only at a time when the subject claimed that ho was abroad
the disclosure of his attendance at this school constituted
a break in his cover story.
fott7f
Et" �
17�
J�k
;.; ,�.s� �
'
Subject B, a medical officer, tried to convince
the interrocator that he worked with Chinese cuerillas.to
teslh them medical practices. However, while under tho
influence of TD ho told a story of how he kept the venereal
disease rate low anon c his troops. He was especiall explicit
in what he did with troops. fresh from America. It was
obvious from this information, therefore, that this offierts
duty consisted of work wit Anericsn troops ruv". not with
Chinese cuerilln's. :Thorny nfter tellinr- this stor: the
subject apparently realized thnt he had revenled information
and attempted n cover story. It is si7nificant th t without.
TD he did not.). discuss rtn mi1ur military or medical exper-
ience that ni:.,ht have revealed his duties.
The relation of personality disturbaLices to interro-
cntion with and without TD
Mir
Each 3ubjoct wns rated by the investicators on
sixty-four itons of anannistic rind personality data. The
criterion used to judo the offectivenemi of TD was whether
or not nore infornation had teen obi ne with than without
TD. This criterion was urrd -eardless of,w1..othcr_the
interrocetor }led correctly Interpreted the information ;_;iven.
Tjbulii C an)unt of nilitnry Informa-tion
obtained from ".lorr.lal" inrlividunls and thw1-6 with neuroses
revealed that information was more readily obtained from
subjects who were sociable, cheerful and had outcoinc
personalities. It was more difficult to obtain information
from subjects whose anxieties made them cautious and un-
friendly. The reason for this is probably that those without
perconality disturbances cnn feel enolth self-confidence
to be willinG to risk discussinc sicnificant topics, while
those who are anxious nre'unwillinc to take this risk.
It should be otatcd here, however, thnt this investi-
crtion was carried out only on a small :roup of IlbjecLs in
lower echelons. ;illether persoas of 7,reater
would be affected in the same way is not known.
Part II. The Ilse of tlie psych2i7alvanometer an a technique
for tbe detecion 01: concealed infornaetion.
rothod.
The Sommer-Fordham avnrttus connintinr: of a.
Pnthometer and an Esterline-Anus continuous recorder vas
used.
The sn'%ject was eomfortatly ::eated with hlo back
to L:-,e apparatus. Zinc electrodes (2 7: 3 cri) were atached
to tbe pair; of the band and brachial.surface of 1.;;.(1 forearm.
These were hcpt in place be moans of elc!astic:gands, Lind
Coed contact further Insured by facin:: the electrodes with
electrode jelly.
The resistance of 1:be subject and-o'6her haze
rt.:tic:lin:3 were no4-ed, and the nature of the experiments
�
explained to him.
1
-14-
�
In each trial the Galvanometric reins were
taken to the nearest microampere deflection and ranked As
recards macnitude of Galvanic response.
As In all experiments with the ps:reoGalvtinomothr.
ss.
it was assumed that those stimuli elicI'Linc the :rtest �
galvanic response were most sicnificant*affectively to the
subject. In addition, two other criteria were utth7ed:
a) the consi:!ten.:.o of the rubjecv,'r verbal Tesponsc, snd
L) the attenuatinc; effect. or reTetitio:.. Dr ctimull In �
successive trials. Ar In "lie detection" =periments it
was assumed that those stimuli to which the subject cave
false respoliss were those which showed vulatlyely
dear-car(' in calvanomrtric resction in S'.1CCO.17C trials.
Followinc each e.-npriment on.tho.
recording tape and PatIpmeter were transcribed to record
� �N�
sheets, to-ether with the rubject's verbal response.
Subjects were 2tudind under twp e:,.perimcntal
conditions. These are subsequently dezi-,_;ned az the "number
e:Teriment" and %he "InforrIfItion exnjriment"..;
A. Fiumbcr ::xperiment
In thir r7periment, which wnr irrunlly carried out
firrt, the sub:Jr-et was esl:ed t.D keep In ;iind ona ilC 11 a
sheet: of pttper a number fron one Lo Len. 1.111s number was
not made known to the operator. The subject was told that
the operator would try to discover the number. The subject
,
MOM..
4
�
-16-
was instructed to answer "yes" or "no" when he was asked
if he were thinking of a particular number, not necessarily
truthfully but with the intention of confusing the operator.
The operator spoke each number in turn, repeating them for
\
several trials. As each number was called the response on
the psychogalvanometer and the spoken response of the subject
were noted. The number consistently evoking the greatest
response from the apparatus was usually found to be the one
chosen by the subject.
These experiments were carried out at least once
and more often two or throe times, both before and after
the administation of TD.
Each subject had brought with him a list bf five
places where he had worked and five jobs ho ma have done.
One of each of these was true, the others false. It was
the object of this experiment to determine the correct area
and activity by meant' of an interpretation of the response
of the subject to questining, as indicated on the psychoga1-
vanometer.
The subject was prepared as for the- number experi-
ment. The operator then asked twenty questions concerning
the five areas and the five activities. The vPrbai response
and the action of the psychogalvanometer were noted.
Results
16,
A. Number experiments
The procedure described above was carried out on
ten subjects both before and after the administration of
z . �
the drug.
1
II
In two individuals in whom it was possible to draw
the correct conclusion on the basis of galvanometric road-
ings in 100% of the trials before the administration of the
drug, this was possible in only 114'ty percent of the trials
after the administration of TD.
In one individual whose numters were ascertained
in 1005 of the trials before the administration of TD it
was not 7lo5sib1e to draw the correct inference on thc basis
-
of psychogalvanometrie readings aftor the administeation
_
In three individual: in whom the oorrect number
could be postulated on:the basis of psychogluanonctric
readings in fifty percent of the trial: before the adminis-
tration of TD, this was possible in i00% of te trials
.� -
after the administration of TD.
In one individual in whom correct irCeronces based
on the readings of the psychognlvrnoneter wore never possi-
ble before the administration of TD, they were possible in
100% of the trials after the administration of TD.
-17-
Of two individuals whose numbers were correctly
ascertained in 100% of the trails before the administration
of TD this accuracy was maintained after tho administration
of the druc.
In one individual it was not possible to draw the
correct conclusion on the basis of psychocalvanometric
readincs either before or after tl'e riministration of TD.
Before the administration of TD it was possible
0 ascertain on the basis of psychpcalvanometric readincs,
the correct number in at least fifty percent of the triala
in eicht of the ten subjects.
After t aclminIstration of TD it was possible to
ascertain on the basis of ps7chocslvr.nometric readin7,s, the.
correct number in at least fifty percent of the trIelyin
eight of the ten subjects.
The avern7e percentace of scluracy for esch croup
of eicht subjects in whom it 'vas possible to drivl the correct
conclusicn of the basis of galvanometric readinr,s in at
least half of the trials, was found to be eiclit-one pprcent
in the eizht tested before the administration of TD arid
eichtysoven percent in the ciht tested after the administra-
tion of TD.
In all, 14 subjects were tested beforo the adminis-
tration of TD. In fifty percent of the subjects the correct
inference was always drawn on the basis of psychogalvanometric
-1E3-
readings; in eighty-six percent in half of 'the trials; and
in fourteen percent never.
In all, thirteen subjects were tested one to three
hours after the administration of .03 - .06 cc of TD by in-
halation or .01 cc by mouth. In sixty-one percent of' the
subjects the correct conclusion was always drawn on the basis
of psychogalvanometric readings; in seventy-seven percent in
half of the trials; nnd in twenty-three percent never.
The following is a summary of the fity-one number
experiments performed on seventeen'subject, when the presence
or absence of.the drug is ignored: In sixty-six percent it .
wa., possible-to draw the correct conclusion on the basis of
psychogalvanometric readings. In thirty-four this was not
possible.
This, while there is a better than chance possibir
lity of ascertaining numbers by means of response on Vire
psychogalvanometer, it will be seen that the use of TD is
of no value in increasing the usefulness-of this technique.
It will benoted that in two-thirds of: the trials on seventeen
subjects, It was possible by means of the,psychogalaanometer
to ascertain the numl.er the subject had chosen and was trying
to hide. It will also be seen that TD has no influence"on
determining the frequency with 'which It was -possible to
ascertain which number the subject had chosen, and th:t
there was no correlation between the frequency with which
the number of a single individual was ascertained before and
-
.4
1
after TD. There was more often a chanGe, either an increase
or a decrease (70%) than a consistency (302) in the frequency
with which It was imusible to ascortain nn indIvidualls
number after TD..
B. Information experiments
Seventeen subjecta were examined by moans of the
psychoGalvanometer.
Of two indivic.ivals tested both before and after
the a6ministration of TD correct concluaions about the .area
and activity wore drawn on the basic of psychoGalvanometrie
readinGc in orie. In the other this was not possible.
Of teven tested before or at le.aut twenty-four.
hours r.fter the adninIstrntion of TD, corrf:c0 Inferenf;o:
were drawn al;out the area and activit: )n ts.le liasin, or the
Galvanometric'readincs, in four. In two, either the arca
1.�
or actl.vity was'ascertaincd ruir3 in one tic ripp,ns o!' %he
psychoonlvanometer was:entirely mislen:lInG.
Of the eIGht subjects tested Jne to four hours
after the administration of .03-.06 c:u of TD j ihhalation
.or .01 co by mouth all ill:formation was accertdined.on the
basic of Galvanometric ...'otalnoo In fcrlr. � In the other four
either the area or activity was ascertined.
DIZCUSSION
Al]. skillful exsminers know and make use of the
fact that interrogation in facilitated if the subject can
be mado to ovoraome Ms suspicions and look on the examiner
in a friendly fashion. Rapport results in loquacity and
loquacity usually results'in the betrayal of Guarded
information.
Since loquacity is affected by the attitude of
tic individual to,7erd the person to le,cm b is talkinr., any
drugs used in interrozation sould be used in conjunction
with te tried and effective methods of conducting such
inquiries. �t ts nhvious, for instance, that .the efficacy
of u drug n be Oestroyed if 417c suct tecores
to and sv::pIc1s (If the nxs!..inf.,r end therefor silent..
Thus, one subject hnd talking volubly for almost two
hours. At the end of that time he wan ',T1c1 that c was
one topic IA, ind bee ll avoidin: f:nd If-at 'ie it op:)ear
that nis topic was imr,crtant. Almcat Lr.,-cdiately the
subject stopped talkinG and it became e7ceedlia-ay difficult
for sn hour thereafter to in-:lace him to disclis eytn the
most innocuoue topiss.
Thn chemical su'sstance r.'03", widely used In feci-
_
litatinG interroGstion Is alcohol. Like alcohol, Tr. 13
effective in fscilitating interrogation in that it encourages
aimiability and self-confidenco, thus establishinG rapport,
and directl7stimuletes loquacity. Another advantace of
TD is the fact that beton:7e of its tantelesannsa, and
effectivenew in pmalldoees, it can be adniretstered withdut
\
the knowledze of the subject. Not on17 has:alcohol inme
of the disadvantaLes of TD marked variabilit7 in indivi-
dual s'llspoptibility and thn developr,Int of "drunkenness"
It M.ay also enzonder s.J::pioiounneas tn thn bidividtial being
plind with Orink.
th tho subjpottr hostility and suspiciousneos
lulled, he can be induced to disc,les indirectly information
he wishes t,-.1**114,:te. .Aa In n pn7lbotherapeutic .intorvinw, or
when unler tLe inflncnIce or a'cohol, rlibjelt nnlor t'Ae
InClanon of Tn e:tarko on (7: oonvc7ca1;'on 1i i 1.o .Continued
even rhen guarded mterial ir approached, as long as he is
not aware of this arproach. Thus, the major's1 sl7ou1d be
uncovered :11oNly Tad indir',:tly. bccors conscious
clanzer,nnd the hie ,-,1.11,3 or-
nncret informa-
tion is threatened, the subject 1),.�co,;nr oner4i. 'or thtr
renroa dirct lonstisno yhov1,1 bn avoided, -;rnze2t.whe-esn in
thn judomcnt thn Intr.rro7ator thn: arn appropriato.
Interrontion 17 for.illtstel w17en ths vlbject has
foclinzs of resentment e:;out 'nis Amy capt7rienen since these
feelinissnake him less sensitive to innter warnings that he
should not discuss secret information and make him more
�
IS
-22-
amenable to having the conversation channelized in desired
directions. This is especially true if the subject welcomes
an opportunity for mental catharsis.
TD, like alcohol, should be administered only after
some measure of rapport has been established.. 'It ismore
effective when it is administered during an interrogation
that is carried out in a relaxed and informal social situation
rather than in a formal military or civilian interrogation.
In the experiments reported here the establishment
of rapport with all subjects was facilitated by the fact
that the examiners were not the enemy, end by the subject's �
knowledge that betrpyins information to the expellmenters
was not as dangerous as betraying it to the enemy. .This
indicates that the druc would lsse some of its ,effectiveness
if the subject were being interrogated by an enemy acett, as
long as the subject remained hostile and auspicious.
The inferences from these observations must, un-
fortunately, be limited since the work had to be discontinued
before suitable comparison of the effect of TD could be made
with those of caffeine, benzedrine, scopolamine and the -
barbiturates.
I�� 4. � � -.����� � � ��
V..
-23
sutr:A RE
4
1
�
1
�
1. Experiments were performed to ascertain
whether TD facilitated interrogation.
2. TD was found to increase loquacity. This
increase averaged forty percent with a range of from
nineteen to eighty-four perent.
3. In addition, TD improves rapport, and increases
self-confidence and feelings of security in the subject, so
that he believes he can enrage in 'conversation without
betraying confidential or secret information. TD induces
the subject to relax his cuard and to betray inPormation,
usIlally Indirectly.
4. In this limits study anxio..1s, cautious and
obsessive persons were found to reveal information lose
often than those Without nuch personality,disturbancea:
In a few subject: with personality disturbances manifested
by ',maturity, hostility and rebelliousness it was found
that loquacity decreased after the administration of TD.
5. TD can bo compared in nation arl.a effectiveness
in interrogation to alcoholic beverages. It has in added
advantage in that it is admininstrable without the knowledge
of the subject rince it is testeless nn f.1 effective in cmall
amounts.
. �,
6. While there is some evidence to indicate that
the psychogalvanometer can be useful in uncovering information,
TD does not increase this usefulness,