J. Pgchiat.
MENTAL DISORDER
"The monogenic theory of
genet. Basel, 8, 50-56.
g. "Zum Ersatz des Weinberg
`crfahrens'; zugleich ciii Beitra
Erhuichkeit des Erkrankungsalte.
enie." Z. gcs. .,Venral. Psjchiai.,
By EDGAR MILLER
of E.G. T.
Although most theories of E.C.T. assume a
beneficial therapeutic effect, the evidence for
this from well-controlled studies is somewhat
equivocal 6. However, at least two controlled
rrtture has accumulated on tills form of studies
, 68 have indicated that depressed
treatment. Yet, despite this vast patients treated by E.C.T. have a better clinical
chiati
-rature and the passage of over 30 years of outcome than those not so treated. Campbell
- ::rimcntal opportunity, no predominant or
12 draws attention to the idea that E.C.T.
:vincing rationale for the use of electro may be effective in speeding tip recovery that is
::vulsive therapy E.C.T. has emerged; under way rather than by actually causing
r,luna's 53 original theory of the incorn- recovery. There is evidence in favour of this
24.
;:ibility of schizophrenia and epilepsy having and Slater's 69 reassessment of Karagulla's
-ra long discredited. E.C.T. remains, therefore,
4 data is also consistent with this idea.
empirical form of treatment.
A large number of studies agree that recovery
The immediate effect of E.C.T. is most following E.C.T. is better for affective disorders,
tiecable at a physiological level, but it is especially depression, than for
other disorders.
* `1ally prescribed for its alleged psychological Similarly, some recent research 13,
, 66 has
* n-ct on affective symptoms; and hence, in confirmed the common clinical impression
that
ing to explain the effects of E.C.T., both recovery rates are
better for eudogenous than
siologicai and psychological theories will be for exogenous depressions.
However, studies
interest. However, it is the purpose of the comparing the response to treatment of different
* `;ent paper to review only psychological
diagnostic groups have generally not controlled
!qries of E.C.T.
for differences in untreated remission rates.
theoretical approaches towards explaining
It seems reasonable to conclude that, com
effects of electroconvulsive shock E.C.S. on
pared with most other psychiatric treatments,
- mats will
also be considered. As the work on the therapeutic efficacy of E.C.T. is reasonably
with animals has been more empirical well established. Even if Riddell's
65 con
vs corresponding work with humans, it has
clusion that an unequivocal proof is lacking is
io more consistent findings and a sounder
accepted, there is at least a strong indication
the inception of the use of artificially
convulsions as a therapeutic agent in
Meduna and the modification of this
of treatment by the use of electric
rents by Cerletti and Bini in ig8, a vast
Since
seed
5 by
-thod
. "CheF Krankheitshild und K:
I schizophrcncn Geseliwistern."
103,
113, 301-351
Psychological Theories of E.C.T.: A Review
965. "A study of pairs of sibs:
w mental disorder." Ph.D. t
,ondon.
963. "Manic-depressive heredie;
:hizophrenia." J3rit. J. Psychics:.,
,,
1967,
230-277.
i The Therapeutic Efficacy
:.s.
re/i Worker, M.ILC. I'syi
ondon, S.E.5; now Lecture'
ml Taiwan University .Ffoi
*s for theorizing. Hence, theories drawn
:n animal research may be able to give useful
* !
for explaining the effects of E.C.T. on
:sans.
-
ondon, 1965
that E.C.T. is more effective that no treatment
at all, especially in eases of endogdnous de
pression.
ii Memory Disturbances
Almost all authorities who have considered
this problem agree that E.C.T. causes a degree
of memory loss for events preceding the treat
ment. Experimental verification of this has been
provided to, , and Cronholm and his
associates have made an attempt to explore the
parameters of this effect ig, 20, 2!..
:t
Xeeds to be Evplained?
considering theories in detail it is
ary to be clear about what phenomena are
`-e explained. Therefore, the main and the
established phenomena will be briefly
cnted below.
:fore
*
*
-
301
302
PSYCHOLOGICAL THEORIES OF E.C.T.: A REVIEW
iii confusion
A large number of authors, especially those
advocating the intensive use of E.C.T., report
that E.C.T. produces confusion, the extent of
which seems to depend on the number and
frequency of the treatments 10, 30, 43, 72.
Paradoxically, besides itself producing con
fusion, E.C.T. is used effectively in certain
conditions, such as severe manic excitement, in
which confusion is present as part of the clinical
picture. In these latter conditions E.C.T.
appears to clear up conflision rather than cause
it, but this may well be because the E.C.T. acts
upon the cause of the original confusion thus
reducing it to a greater extent than any added
confusion due to the E.C.T.
1
iv Other Psychological .EJfects
Investigations of the effects of E.C.T. on a
large number of psychological tests, on motor
speed and on perception, have also been carried
out but the results are oftel1 contradictory and
do not lead to clear cut conclusions. The
important efkcts of E.C.T. appear to be the
first three given above.
A point also worth noting here is the fact that
in all forms of convulsive therapy it is the
convulsion itself which is necessary for the
effects to occur and not some other incidental
feature of the treatment situation. This has been
demonstrated for both drug induced convulsions
7, iG and E.C.T. 22.
In the actual discussion of the psychological
theories that have been put forward to explain
the effects of E.C.T. it seems most convenient to
split the theories into two fairly well defined
groups:
a Those influenced by theories of a psycho
analytic nature.
b Others.
a Psychoanalytic Thcories
A large number of theories have been put
forward from within a psychoanalytic frame
work. These were often initially applied to drug
induced convulsions, but most such theorists
made little distinction between thug induced
convulsions and E.C.T. , 35. As it is the
convulsion itself which seems to be the dir
agent in all forms of convulsive therapy
reasonable to apply theories originally dc;i:'
to account for the effects of drug-iu.,
convulsions to E.C.T. also.
Among this group of theorists, all agree, c;:
explicitly or implicitly, that E.C.T. li
effects by assisting the process of tepressiou,,
is hence opposed to psychoanalysis, which
at recovering repressed material yo. All'!
goes further than any other theorist with
concept of repression in that he uses it to arc
not only for the beneficial effects of E.G.]'.
for the amnesia and other negative sequeL'
treatment. 1-Ic claims that these effects are
what he would predict on the basis of so
repression having taken place.
The various psychoanalytic theories diIk
to the hypotheses put forward concerning
mechanisms which either cause or accozup..
repression. The three most common s::
hypotheses are discussed below and have 1
used in varying permutations and combinat
by the different theorists.
p
`Flu,
:?
fc
act
i-cti'
r
.,t
S
* lad
`Pla
lug
SVet
The Regression Hypothesis:
A number of authors 2, 3, 31, 58, 62 l;.
regarded the treatment as producing regrc:'
of behaviour to infantile, or even pre-n.Y
levels. Usually the regression is thought of
being psychologically induced by the strc
involved in the treatment situation, tho'
Power 62 feels that the regression is physic..
induced by the convulsion. Power argue's t
the tonic and clonic phases of the eonvu!'
are similar to movements seen in the foetal
of both man and sheep, and that these therc
represent a return to a foetal level of ncr'
functioning. Others, holding that the regres:
is psychologically induced, relate post-CS
vulsive behaviour, such as apparent sUC!
movements or faecal smearing, to the Frcut'
stages of psychosexual development 31, 58
assert that the oedipal conflict is reactivated
73.
1
As with most hypotheses within this C.1
there has been little attempt at exper1m'.
verification. Cameron i i claims that caV
observation of patients following E.C.T. d
elyt
Pt
W1
tI te
* tre
[1
Cnc
.efl
`"`Cr
:lcc
Pflti
`hO
re'
Cur
dru
Ills
1111''
lit
r
_
__
C'
-`-------
BY EDGAR MILLER
ch seems to be the effe. .rcveal infantile speech or behaviour in the
that this has been described, by child
Df convulsive therapy
theories originally desh thologists such as Piaget or Gesell. One basic
effects of drug-indt ilculty with this hypothesis is that it is
also.
:iicult to test convincingly, as it is always.
of theorists3 all agree, Ci .ible for those so predisposed to interpret
:itly, that E.C.T. ha clam features of the clumsy, semi-voluntary
.e process of repression, -haviour of a semi-conscious and confused
psychoanalysis, which .ient as "infantile".
;ed material 70. Abse In conclusion, this hypothesis is definitely
ny other theorist with :i-provcn. Even purely as a description of
in that lie uses it to acc *t-convulsive behaviour regression is not very
ficial effects of E,C.T. :ivincing it.
other negative sequelc
s that these effects are
lict on the basis of st: Fear Hypothesis
The central tenet of this hypothesis is that it is
cen place.
.oanalytic theories difk fear induced by the treatment, rather than
ut forward concerning * actual efl'ects of the convulsions, which is the
itlier cause or accorn :ctive agent. This hypothesis has received
ree most common t :isiderable support i, 2, 3, 27, 31 and was
ssed below and have 1' ct strongly advocated when drugs were used
iutations and combina: induce convulsions. As Good si strongly
:pliasized, a large number of patients underrists.
f ing leptazol Cardiazol, Metrazol therapy
hwed unmistakable signs of fear and it is' also
tcly that many patients undergoing -E.C.T. are
without fear. However, to jump from this
ors 2, 3, 31, 58, 62
tnt as producing regre' the argument that fear is the effective element
`antile, or even prc-r. treatment is not supported by the expcri
:egression is thought atal evidence.
y induced by the sir `:ook 17 compared ratings of fear in 275
:atment situation, tli ients undergoing convulsion therapy with
the regression is physi sments of clinical improvement, and con
rulsion. Power argues !rd that, if anything, the trend was in the
phases of the convr `rse direction to that predicted by this
aents seen in the foeta' :-`athesis. Cook performed no statistical analy
p, and that these ther' but the writer, using Cook's published data,
a a foetal level of ncr been able to show that the reported trend in
holding that the regrc reverse direction was significant at the per
.nduced, -relate pOst" 1. level. Two other investigations 7, iS
leptazol therapy have used the technique
;uch as apparent su
smearing, to the Frcr romparing patients treated in the normal
ii development 31, 5: and with the injection of a similar amount
d conflict is reactivatet. -tug too slowly for a convulsion to occur.
in latter
procedure is reported to create as
:i fear as the standard procedure. Neither
aotheses within this
investigations produced results support
attempt at experirt
n x i claims that cc: yhe fear hypothesis.
more recent experiment 23 compared
nts following E.C.T.
`
-
-
-
-
303
degree of fear as measured by the T.A.T. and
other measures with rated improvement in
patients treated by E.C.T. and a control group
who were anaesthetized hut had no convulsion
induced. Again no relationship was found
between fear and clinical improvement, though
the measures of fear used were rather poor and
it seems that ratings of fear and improvement
were not done independently.
The experimental evidence does not, there
fore, support this hypothesis, and one experi
ment even shows a significant trend in the
opposite direction 17. To argue, as does Abse
i, that such experimental results do not
disprove tl1is hypothesis, as the fear may be at
an unconscious level, is not permissible. Ad
vocates of this hypothesis, such as Abse himself,
base their original arguments on the fact that
patients do show overt signs of fear.
The Punishment L&Po1heszs
A third main hypothesis occurring in psycho
analytic theories is to postulate that the patient
regards the treatment as a form of punishment
2, 3,46, 6, 8. Korson t6, p.4' states that:
"The individual delivers himself into the hands
of a strict, but in the end forgiving, parent figure,
who will mete out punishment justly and allow
atonement and delivery from evil. Acceptance
of punishment allows the patient to assuage his
conscience, fear and anxiety becoming un
necessary once retribution has taken place."
This hypothesis assumes guilt to be a central
feature of illnesses treated effectively by E.C.T.,
and this is in agreement with the general
clinical impression that en dogenous depressivcs,
who often show strong guilt feelings, respond
most favourably to E.C.T. Other apects of the
hypothesis are less satisfactory. It assumes that
the patient identifies the doctor with his parent.
This, in turn, implies a regression to childhood,
at least to a stage at which the child is dependent
upon parental sanctions and discipline.
There has been no direct test of this hypo
thesis, although Lockwood gj has tested a
derivation of it based upon Rosenzwcig's 67
classification of responses to frustration. Lock
wood argued that intropunitive subjects should
show the best response to E.C.T., but failed to
PSYCHOLOGiCAL THEORIES OF E.C.T.: A REVIEW
3°4
confirm this in two separate experiments. porary, in both human and animal brains
Flowever, this is not a crucial test, as the E.C.T. 4., 50.
The evidence is consistent with the possi
plinishment hypothesis could be valid without
Lockwood's extcnsion of it being valid also. The of a small degree of diffuse brain damage hr.
punishment hypothesis remains, therefore, non- caused by E.C.T. Unfortunately this still l÷..
the structural theory unproven, as it is necc.
proven.
to demonstrate that the damage involved I,
main agent of behavioural change and
Summary of Psychoanalytic Theories
merely an incidental feature of the treatniez.
The general criticism can be made of nearly
all psychoanalytic theories that the' rely
heavily on factors in the treatment situation Theories InvoIcing A innesia
other than the convulsion itself e.g. the patients'
These predominantly centre around
fear of treatment, thus ignoring the consistent
suggestion that the treatment-induced amlic
finding that the convulsion is the effective agent might
be responsible for the beneficial eli'
in treatment. In addition they rely on a back
ii, 40, 13 59, 70. It is often pointed out ft
ground theory which is in itself unproven.
amnesia is usually greatest for experiences wli
Psychoanalysis seems, therefore, not to be a
come immediately before treatment, and
fruitful starting point for an explanation of the
psychotic episodes, being usually recent,
effects of E.C.T.
more likely to be affected by amnesia than
more normal experiences from the subjcv
b .Won-Psychoanalyiic Theories
distant past. Janis io also feels that t
Somatic Theories
amnesic action of E.C.T. becomes a new leari
There are a large number of somatic theories. defence mechanism enabling the subject
All have in common the obvious inference that similarly banish stressful experiences occurt
if some fairly permanent behavioural change subsequent to treatment.
occurs in the patient as a result of treatment,
The amnesic effect of E.CLT. is the appare
such a change must be correlated with some rationale for treatment by "rearessive" E.G..
somatic change within the nervous system. A 30, `13 or "depatterninc" as it is called
large number of suggestions have been made as Cameron x i. The general principle
to the change that occurs 26, but these regressive E.C.T. is the intensive use of E.G.]
generally lie outside the scope of this review.
at rates of one or more treatments per day in:
Psychological research has been attracted the patient is in a totally amnesic state, a!
towards one particular type of somatic theory confused and often doubly incontinent. As t
which assumes that E.C.T. has its effects by patient recovers, he learns, or is taught, ii'
damaging nerve cells 29, 73. One group of and better adjusted patterns of behavior'
experimenters 71 found that their subjects, Several authors have enthusiastically report
after three E.C.T.s, showed a definite change in the results of this form of treatment on chro:
Rorsehach responses tQwards the pattern of patients, but there has been no attempt .81
responses produced by patients with diffuse controlled study. There is only one report of t
brain damage. Others have found that short- use of regressive E.C.T. that has come to d
term response to treatment is correlated with conclusion that it is of no value fl.
positive results on the amylobarbitone test for
A variation of this theory is the suggest!
brain damage ip.
that there is a differential loss of mate'
The worth of such findings, and of findings relating to the patients' psychopathology i:
from similar experiments, is stronglyjeopardized 40. Cameron ii suggests that as the amneY
by the general unreliability of such indices of following intensive treatment recedes, tlit'
brain damage. A more direct approach is further events are recalled to the extent tIL'
obviously to be preferred, and there is some they are compatible with the emerging fram'
evidence of vascular changes, probably tem work of the patients' behaviour. If the patteni'
-
-
BY EDGAR MILLER
an and animal brains af:.
crging into a normal pattern of behaviour
events related to previous abnormal
javiour will be incompatible and hence not
-nsistent with the possibil
diffuse brain damage bel:
.:allcd. Janis 39 interviewed patients before
nfortunately this still lea'.
.d after EXIT, and found that, as compared
unproven, as it is necessa: ;ih controls, the treated patients showed
1uesia which was more apparent both for
the damage involved is
avioural change and r -cent material and for anxiety provoking
feature of the treatment. rerial. However, an adequate test of the
vpotliesis of differential action of the amnesic
.-Thct of E.C.T. is almost impossible because of
norm
:-e difficulty of controlling for the degree of
intly centre around t -irning of both normal and pathologically
treatment-induced amneE dated material.
e for the beneficial ciTe., Direct experimental investigations of the role
It is often pointed out Li `wed by amnesia in the therapeutic effects of
eatest for experiences whi: .C.T. are not available. Information of some
before treatment, and `kvance comes from work with unilateral
being usually recent, a47, 52, where the general finding is
fected by amnesia than t? `:at unilateral E.C.T. can be as therapeutically
:icflCes fi-om the subjec :rective as bilateral E. c.'r. but gives significantly
s memory impairment. This suggests that the
4.0 also feels that 1'
:37. becomes a new lean :erapeutic effect of E.C.T. does not depend on
enabling the subject inesia, hut it is a possibility that even the
ssful experiences occun; ilaterally treated cases had some degree of
: muesia which was adequate for therapeutic
ient.
:t of E.C.T. is the appare 9cacy. No untreated controls were used to
ent by "regressive" E.G. -t for this.
The evidence from unilateral E.C.T. points
:erning" as it is called
he general principle `cay from, but does not definitely disprove,
the intensive use of E.G codes of E.C.T. based on amnesia. A direct
re treatments per clay liE perimental test of this theory would be useful.
totally amnesic state, a
doubly incontinent. As t
Theories
learns, or is taught,
d patterns of behavia' Other theoretical approaches 35, 57 have
re enthusiastically repor !pha,5ized the "shock" aspect of shock
rim of treatment on chrc' `:atment. The treatment is regarded as giving
has been no attempt ac patient some sort of psychological jolt to
crc is only one report of i -lug him face to face with reality. Foulds 28
C.T. that has come to cd a similar idea in postulating that the effect
E.C.T. was to break up painful thoughts.
of no value 75.
suggest
`ulds'
the
is
own experiment gave confirmatory
is theory
fferential loss of mate' zults, but others 68 were unable to replicate
ents' psychopathology :` findings.
uggests that as the amn Hetherington 36 felt that depression is
treatmear recedes, tNrked by motor retardation but also by
recalled to the extent t !r.activity of thinking. E.C.T. helps the
with the emerging fra: `pressive by abolishing motor retardation and
`behaviour. if the patie'ducing psychic retardation. His own cx-
305
perimental results are consistent with this view,
but this study has been criticized on the grounds
that the experimental and control groups were
not properly matched 12.
.
,
i
-
-
- -
Research wit/i Animals
The possibility presents itself that the results
of work on animals might help towards im
proving the rather dismal theoretical picture of
EXIT. given above. Accordingly, the major
trends in theorizing about the effects of E.C.S.
on animals will be outlined.
Before doing this, certain differences between
human and animal studies must be noted.
Firstly, theorizing in animal research has been
mainly concerned with the phenomenon of
retroactive amnesia R.A. and obviously not
with therapeutic change. Secondly, E.C.S. is not.
given with an anaesthetic, whereas E.C.T. is.
Although the convulsion is the effective thera
peutic agent in E.C.T. it is possible that the
anaesthetic has a minor effect as it has been
shown, on animals, that anaesthetics have an
R.A. effect of their own i 6i. There seems no
good reason to suppose that an electrically
induced convulsion in an animal is a different
phenomenon from such a convulsion induced in
a human and, therefore, providing the relevant
procedural differences are borne in mind, it is
reasonable to use theoretical ideas derived from
animal research to suggest theoretical
approaches for work with humans.
Neural consolidation Theory
This is the most prominent theory in E.C.S.
research with animals, and often other theories
in this field were inspired by alleged inade
quacies in this theory. Briefly, the theory
assumes that for any memory to become
established it is necessary for the original
memory traces, which are only temporary lit
form, to be transferred to a more permanent
form, i.e. consolidated. An E.C.S. occurring
within the time period that consolidation is
taking place will break up the process of
consolidauon, causing R.A. for material not
already consolidated.
Early experiments 25, 74 showed retention
to be a negatively accelerated function of the
306
PSYCHOLOGICAL TIIEORtES OF E.C.T.
interval between learning and E.C.S. and were
generally accepted as demonstrating the validity
of this theory. More recently these early
experiments have been criticized, as they used
several learning trials, thus giving the early
trials time to consolidate before the later trials
came along 6i. More satisfactory experiments
using one learning trial followed quickly by a
i have, however, upheld the
single E.CLS.
consolidation theory. Other experiments using
the same initial design but going on to give
further learning trials and E.C.S.s have con
cluded that other factors come into operation
when several E.C.S.s are given is, 38.
Chevalier ti, again using a one learning trial
and single E.C.S. design, has shown that the
R.A. effect remains undiminished over 30 days.
Lewis and Maher 48 have brought together
a number of results inexplicable by the con
solidation theory. One experiment 5 has
shown that a series of E.C.S.s given a few days
prior to learning have a proactive effert. Others
have shown that a series of E.C.S.s given a
few days after learning, and hence long after
what would generally be considered as the
consolidation period, can also disrupt retention.
Brady 8 also found that when a learned
response was obliterated by a series of E.C.S.s
given some time after learning, there was some
degree of recovery of the response go days after
the last E.C.S. As the breaking up of consolida
tion should give a permanent loss, this last result
is also inexplicable by the consolidation theory.
The consolidation theory has been universally
upheld by experiments using the one learning
trial followed by a single E.C.S. paradigm.
However, other experiments give results un
accountable for by this theory. It is of im
portance from the point of view of subsequent
discussion to note that in the latter group of
experiments a series of E.C.S.s has always been
used; no one has yet demonstrated proactive or
retroactive effects after a long delay from a
single E.C.S.
Goiiflict Theory
This is analogous to the fear theory of the
psychoanalysts for E.C.T. It suggests that E.C.S.
has its effect by being an aversive or fear-
A REVIEW
provoking stimulus. Coons and Miller z:;
an experiment replicating sonic features
earlier experiment by Duncan 25 obi::
results which they interpreted as showing
fear was induced by E.C.S. Unfortunately,
experiment failed to control for the eff:t
E.C.S. alone, and so the effect of E.C.S.
confbsed with other variables gy.
Experiments using one aversive learning u..
quickly followed by a single E.C.S., in a di
test of this theory, have shown that the anw
effect of E.C.S. is stronger than any indt:
fear
si. However, other experimen:
have found that after a series of E.C.S.s Ii
been given apparently aversive effects of E.C,
do appear 15, 25.
As the conflict theory cannot account for L
results of experiments using the one Icarni:
trial and single E.C.S. paradigm. it cannot
used to displace the consolidation then:
Experiments giving several E.C.S.s have fin::
effects attributable to fear but which could
explained in other ways. It would seem reas:.
able to suppose that undergoing a series
E.C.S.s is an "unpleasant" experience for
rat; the problem is whether any fear occurri:.
is of significance in explaining the effects
E.C.S.
Gompethzg Response Theory
This theory, originally proposed by Adat:
and Lewis
to account for an apparer$
deficiency in the consolidation theory, assun:
that some aspect of the response to the E.CJ'
becomes conditioned to stimuli in the surround
ings. The most recent statement of the theory:
by Lewis and Maher 48, who suggest that ti
coma following the seizure is due to "protecti''
inhibition" and that components of tfi
inhibition become conditioned to surroundii
stimuli in the experimental apparatus.
If this theory is correct, it would be predict'
that E.C.S. given in the same location a:
learning took place would disrupt learning nntr
more than when given in a dissimilar situatlofl
The original proposers 6 tested this anobtained confirmatory results, but in a sini1l
experiment Quartermain ci al. 64 found tliJ
the location in which E.C.S. was given V3
T.: A REVIEW
BY EDGAR MILLER
Miller 4vant1 as Would be predicted by the
g stimulus. Coons and
featureolidation theory. However, the two cx
some
iment replicating
25 olnentS did differ in the timing and number
Duncan
xperiment by
shpwiiC.S.5, with Quartermain ci a!. using a
as
interpreted
hich they
E.C.S. nd short learning E.C.S. interval,
induced by E.C.S. iinrortunat4
and Lewis using several B. C.S.s and
eMams
the
for
nt failed to control
learning
E.C.S. interval.
E.Cger
of
effect
the
lone, and so
it seems that the experiment 64
tin,
variables
37.
with other
the single E.C.S. and short learning E.C.S.
iments using one aversive learni,
supports the consolidation theory.
zial
in
Eollowed by a single E.C.S.,
longer
learning E.C.S. interval and a
a
ti
the
that
shown
us theory, have
E.C.S.s
is used it appears that other
of
L
any
than
stronger
E.C.S. is
into
which could be attributed
come
play
51. However, other expeflfs
E.C.Speting
as suggested by Adam
responses,
of
md that after a series
or
explained
in other ways.
of,ewis,
effects
aversive
en apparently
r i5, 25.
conflict theory cannot accountLFOClOT Explanation of E.C.S.
I
,f experiments using the one
main conclusion that seems to follow
cai
it
paradigm,
I single E.C.S.
above review of theories of E.C.S. is
rhe
consolidation
the
displace
two-factor
explanation seems to be
hava
ents giving several E.C.S.s
Experiments
using one learning trial
ced.
uributable to fear but which
quickly
by
a single E.C.S. have
seemrd
would
d ih other ways. It
9ently given evidence in support of the
a
undergoing
suppose that
consolidation theory. There longer
is an "unpleasant" experienc
oclg
E.C.S. intervals and several E.C.S.s
problem is whether any fear
used, results inexplicable by the
etbeen
the
nificance in explaining
dation theory have been dbtained. Hence
js that an E.C.S. given within the period
;required for the consolidation of memory
frill disrupt consolidation, but the massing
Response Theory
S.s results in other eiiècts. The remaining
by
proposed
heory, originally
0'n is to explain these other effects.
an
for
account
to
vis
theory, possible that the additional effects due to
y in the consolidation
E.C.S.s are the result of competing
e aspect of the response to th
or fear, but another explanation is
sues
the
in
stimuli
conditioned to
The giving of several E.C.T.s to humans
r..
the
of
statement
recent
most
tn to result in confusion, and it is
and Maher 48, who suggest
"pce
likely that giving several E.C.S.s
to
due
lowing the seizure is
tsult
in an analogous state of "confusion"
components
and that
would explain the inability
Confusion
surrat.
to
conditioned
i become
ials
to
lcarn
following massed E.C.S.s,
the experimental apparatus.
p
the
failure
to
perform learned responses
be
would
theory is correct, it
given a comparatively
locmveral
E.C.S.s
same
IS. given in the
after
learning.
In the latter case it
learniitle
disrupt
would
took place
5C expected that when the confusion had
dissimilar
a
in
given
n when
tested tC to disperse the learned response would
lmnal proposers 6
2r, as was found by Brady 8. Pearlman
confirmatory results, but in
using drug-induced convulsions,
foi6',
64.
al.
ci
nt Quartermain
gi'O proposed a similar two-factor ex
was
E.C.S.
which
:ion in
307
planation of E.C.S. The loss of learned responses
was accounted for in terms of disruption of
consolidation with a short learning-convulsion
interval and in terms of confusion with a longer
learning-convulsion interval.
The difficulties in postulating a state of
confusion in animals receiving massed E.C:S.s
are twofold. Firstly, such a concept as clouding
of consciousness, which is part of the general
psychiatric description of confusion, is almost
meaningless when applied to animals. However,
it would be possible to define confusion in
animals operationally in terms of other features
of confusion, such as spatial disorientation or
possibly impairment of attention. Secondly, it
is not yet possible, on the evidence available, to
completely eliminate the conflict or competing
response theories as explanations of the effects
of massed E.C.S.s, but experiments could be
designed to test these theories against the
hypothesis of confusion.
Diseussiox
In 1948 Gordon was able to write a paper
entitled "Fifty shock therapy theories" 32.
Since then, theories have continued to multiply,
and it is obvious that the present review has had
to be content with a sample of the principal
psychological theories. It is also obvious that
none of the theories discussed so far is near to
being considered adequate. Doubtless the
steady proliferation of theoretical speculation is
related to the lack of success of earlier attempts.
We are thus left with two further problems
for discussion. Firstly, there is the problem of
why the theories put fonvard have been so
lacking in success. Secondly, whether it is
possible to suggest ways for a more successful
theoretical approach.
"
11 7y Flare Theoretical Attempts Failed?
The reasons for failure appear, basically, to
be threefold. Firstly, knowledge of the actiology
of illnesses treated by E.C.T. is minimal. If it
were possible to reliably implicate a process or
processes, of whatever nature, as being the cause
of endogenous depression, then this would give
an excellent lead to investigations and theorybuilding with E.C.T. Coiwersely, of course, an
308
PSYCHOLOGICAL THEORIES OF E.C.T.
explanation of the effects of E.C.T. would help
investigations into the nature and causes of
menial illnesses, particularly depression.
Secondly, and more seriously, the work on the
effects of E.C.T. has revealed' little well-estab
lished information, even when the vast literature
on the subject has been carefully combed 63.
U is known fairly definitely that E.C.T. is
beneficial in cases of endogenous depression,
that it has an effect on the recall of past events
and that it causes confusion. Systematic
exploration of these effects has generally not
been attempted; a notable exception being the
work of Cronholm and his associates.
A third reason for the lack of success in
theorizing is that in many instances the theories
put forward have been. highly speculative, based
on clinical impressions only, and not tied down
to experimentally established phenomena. Such
theories, besides often being difficult or im
possible to test, have no more validity than the
impressions or assumptiolis on which they are
based and seem merely to confuse the issue.
Suggestions for Belier Lines of Approach
A much-neglected source of inspiration for
workers concerned `with E.CLT. is the work on
the effects of E.C.S. in animals. This is clue
presumably to the fact that most work on E.C.T.
is done by psychiatrists, whilst that on E.C.S.
with animals is done mainly by experimental
psychologists. The animal work could present a
useful model for E.C.T. research, as it is much
better executed as far as the experiments are
concerned, the relevant parameters are well ex
plored at least for the R.A. effect and the
theorizing is more closely tied to experimental
findings.
The main suggestion coming from the above
review of animal research is that of a two-factor
theory of 13.0.5. in terms of the interruption of
memory traces and also a confusion effect
produced when several E.C.S.s are administered
within a short period of time. That this latter
confusion effect may be particularly important
in explaining the effects of E.C.T. on human
psychiatric patients is also suggested by the
frequent mention of confusion in clinical reports
of E.C.T. It is possible, for example, that
E.cJ,'s effectiveness may be partially spurious,
A REVIEW
as a series of E.0.T.s may mask the pat;.
illness because of the induced state of conf,
in much the same sort of way as a se
E.C.S.s given to a rat have been found to;
a learned response.
1..
Unfortunately, little
perimental research has been done on
E.C.T. conflision. That which has been :
. 72 has done little more than slv,
existence experimentally.
Experiments using unilateral E.0.T.
are also of theoretical interest. These s,
that memory disturhances and confusion fbi:
ing E.C.T. may be less if the electrode
placed unilaterally as opposed to the ii
bilateral placements. Further experinit;:
exploration of the use of unilateral RC'
could contribute greatly to the determinatio:.
the role played by memory disturbanc
confusion, and possibly other factors, in
,iiT.
litt.'
therapeutic effects of EXIT.
CoNcLusioNs
The main, unavoidable conclusion arhi:
out of this review is that we have come scBa'
little way towards explaining the effects
E.C.T. Explanations that have been put forscr
tend to have been speculative and unconviuci:.:
The outstanding fact that emerges is that
explanation of the phenomena can only be
adequate as the data on which it is based.
the experimental data on the effects of E.C1
are so poor, the biggest contribution of tb
C'
t. C.
etical significance in this field must come
more careful and detailed exploration of ii:
effects of E.C.T. arid the parameters on wbit
they depend. Without this, further theoretic
development will be gravely hindered, if fl
impossible.
It is further suggested that help in i
solution of the problem may be obtained ft
utilizing the results of experimentation
animals as such research has certain yntagr
over that using human patients. %Vork
unilateral 13. C.T. also presents intercstlt-'
possibilities.
ACKNOWLEDGMENTS
The writer would like to acknowledge the hclpfut adv;'
of Mr. C. 13. Blakcrnore of the Institute of Psychiatry a
Mr. A. R. Dabbs of Oakwood Hospital in the prcparat1
of ttiis paper.
`I-
-.1
L
I
LE'IEW
BY EDGAR MILLER
O.T.s may mask the patiej
RnxutneEs
the induced state of confus. 1tm, D. W. 1940. "The psychology of convulsion
ne sort. of way as a scrie5
therapy." 3. zncni. Sri., 86,
1942. "Rationale of convulsion therapy." Bri!.
i rat have been fotmd to
1
J med. Psyc]toL, ig, 26227o.
ise. tlnfoi tunately, little
,
1944 "Theory of the rationale of convulsion
rch has been done on Pj
therapy." ibid., 20, 35-50.
That which has been d. ADT,J. P., and ES5MAN, W. B. mg6i. "Ether induced
retrograde amnesia for one trial learning in
ae little more than show
mice." Science, 133, 1477-1478.
ent-tIl
ADAMS, H. E., and LEWIS, D. J. Ig6aa. "Electra
mg unilateral E.C.T.
convulsive shock, retrograde amnesia and
etical interest. These sugt:
competing responses." J. camp. PJysiol. Psycho!.,
irbances and confusion foll
55, 299-301.
be less if the electrodes . - - i96th. "Retrograde amnesia and com
peting responses." Ibid.,
302-305.
ly as opposed to the
IILAVR0cK, M. F., Low, A. A., and SACII5, M. 1939.
cuts. Further experimet
"Influence of fear, pharmacologic action and
ie use of unilateral E.Ci
convulsion in Metrazol therapy." Arch. .Vn,rol.
and Psychiat., 42, 233-236.
teatly to the determinatiot!
ov memory disturban BRADY, J. V. 1952. "The effect of electroconvulsive
shock on a conditioned emotional response: the
DsStbly other lhctors, rn
significance of the i,iten-aE between the emotional
of E.C1.T.
conditioning and the electroconvulsive shock."
3'. camp. Physiot P.cyclwl., 45' 9-.-13.
- lli'cr, H. F., and CELLAR, L. `95.1. "The
JONCLUSIONS
effect of clectroeonvulsive shock on a conditioned
tvoidable conclusion ant1
emotional response as a function of the temporal
is that we have come v
distribution of the treatment." Ibid., 4, 54-7.
s explaining the efFcctsi. llncNoEv1Axs, j. C. ig. The PJfl'ct of Repeated
Electroshock on Learning in Deftrcssion. Heidelberg.
is that have been put fonv
pcculative and unconvinci; CASIER0N, D. IL. igfio.a "The production of differ
ential amnesia as
factor in the treatment of
act that emerges is that
schizophrenia." Camp. Psychial., I, 26-34.
phenomena can only bLi CAMPBELL, D. igGo. "The psychological effects of
cerebral electroshock." In: handbook of ;lbnorino!
ia on which it is based.
Psychohi,gr. ed. Eysenek. London.
ata On the elThcts of
atENEY, r'i. w. P., 1&rrit, ?vt., and GAnsstn:, K. F.
ggest contribution of thc1'
ig6. "The diagnosis of depressive syndromes
*
a this field must come ft.
and the prediction of E.C.T. responses." 11th'. 3.
Psychiat., III, 659-674.
detailed exploration of;
td the parameters on wi:1 CFEEvALIEIt, J. A. 1965. "Permanence of amnesia
after a single post trial electroconvulsive seizure."
out this, further theoret
ramp. Pip'skl. Psycho!., 59, 125-127.
3.
e graveLy hindered, if Cuoaovast, S. L., and SCIIILLEIt,
P. H. sg6. "Shortterm retrograde amnesia in rats." ibid..
73-78.
ggested that help in
`1cm may be obtained Ii' CohEn, L. H. 1939. "The therapeutic significance of
fear in the Metrazol treatment of schizophrenia."
ts of experimentation
Amer. 3. Psychic!., 95, 1349-1357.
arch has certain advantaL Coon, L. C. 1940. "Has fear any therapeutic
significance in convulsion therapy?" 3. mcix!. Sri.,
uman patients. Work
86, 484-490.
also presents xnteres:
Coons, IL. IL., and MILLER, N. IL. 1960. "Conflict
versus consolidation of memory traces to explain
`retrograde amnesia' produced by E.C.S." 3.
3WLEDGMENT5
camfi. Physio!. Psycho!., 53, 52.1.-53t.
o acknowledge the helpful ac
CRonnoLil, B., and BL0SIQUIST, C. mgg. "Memory
sf the Institute of Psychiatry
disturhanee after eleetroconvulsive therapy. 2.
vood Hospital in the prcpara'
Conditions one week after a series of treatments."
Aria psychic:. Scand., 34, 18-25.
-
309
and M0LAIWEa, L. mg. "Memory disturbances after eleetroconvulsive therapy. z. Con
ditsons 6 hours after electroshock treatment."
Ibid., 32, 280-306.
21. - - 1964.' Memory disturbances after eleetro
convulsive therapy. 5. Conditions one month
after a series of treatments." Ibid., 40, 212-216.
22. - and Orrosson, J. 0. igGo. "Experimental
studies of the therapeutic action of electro.
convulsive therapy in endogenous depression."
ibid., g, Suppl. `45, 69-97.
23. CiwnoTon, IL., BIULL, N. Q., EsDuson, S., and
CELLER, IL. 1963. "The role of fear in electra
convulsive treatment." 3. nero. men!. Dis., i6,
20.
-
29-33.
24, mInER, C. IL., CULLINAN, C. and KovnsemLn 13.
`952. "Results of treatment of schizophrenia in
a state hospital: changing trends since the advent
of electroshock therapy." A.M.A. Arch. Neural.
Psyrhiat., 67, So-OS.
2. DuncAn, C. P. igq. "The retroactive effect of
electroshock on learning. 3. romp. PlysroL
Psycho!., 42, 32-44.
26. FI.nsimxc.J. C. 1956. "An inquiry into the nseehan
isus of action of electric shock treatments."
3. BC/V. went. Dis., 124, 440-450.
27. FLE5CEIER, J. zg4. "Time discharging function of
the convulsive seizure." Ibid., gfi, 274-285.
28. FouLos, C. A. r952. "Temperamental differences in
maze performance: Part I LThe effect ofdistraction
and of cleetmeonvolsive therapy on psychomotor
retardation. Bri!. 3. Psycho!., 43, 33-41.
29. FREEMAN, W., and XV.vns,J. W. 1944. Psyrhosurgr.
Springfield.
30. GLUECK, B. G., Rriss, H., and BERNARn, L. IL. 1957.
"Regressive shock therapy." Psychia!. Quart, s,
117-136.
31. Gooo, R. 1940. "Some
? s'ations on the Psych?.
logical aspects of cardmazol therapy. ,7. ment. Scm.,
86, 491-50!.
32. Connox, M. `ot8- "Fifty shock therapy theories."
Miii!. Surgery, 103, 397-401.
33 H.euti, 13. 0. 1949. T/zr Organizrx!ion of Behaviour: .1
Neuropsycho!ogica! Theory. New York.
34. FIin.nmtunx, C., and LiEnnlur, IL. zgp. "Biopsies on
the brain following artificially produced eon
vulsions." Arch. .iVcuro!. Psychiot., 46, 5fS-552.
35. HEMPHILL, K. IL. 1940. "Studies in certain patho
physiological and psychological phenomena in
epilepsy." 3. men!. Sci., 86, 799-818.
g6. HETHERLNGTON, R. 1956. "The effects of E.C.T. on
the efficiency and retentivity of depressed
patients." Bri!. 3. uxed. Psycho!., 29, 258-269.
37. Hunspavir, W. J., and CERBRANDT, L. K. tg6.
"Electroconvulsive shock: conflict, competition,
and ncuroanatomical functions." Psycho!. Bull.,
63, 377-383.
38. - McGAIJCII, J. L., and Timosisoic, C. XV. 1964.
"Aversive and amnesic effects of electrocon
vulsive shock." 3. romp. Physio! Psycho!., 57, 6 1-64.
:310
PSYCHOLOGICAL THEORIES OF E.C.T.: A REVIEW
39.jANI5, I. L. igoa. "Psychologic effects of electric
convulsive therapies I. Post-treatment amnes
ias."J. nero. men!. Dis., in, 359-382.
40.' jANts I L. sg5ob. "Psychologic effects of clectric
convulsive treatments III. Changes in affective
disturbances." Ibid., III, 469-489.
4!; Kanrc, R L., Fzsn, M., and Vnxsmia,
A. 1956.
"Relation of arnobarbitol test to clinical improve
ment in electroshock.' Arch. .iVeurol. Psychiat., 76,
23-29.
.42.: KARAGULLA, S. rgo. "Evaluation of electric
convulsive therapy as compared with conserva
tive methods of treatment in depressive states."
men!. Sri., gG, io6o-iogi.
43. KENNEDy, C. j. C., and A!ccuEL, ID. 1948. "Re
gressive electric-shock in schizophrenics refractory
to other shock therapies." Psychiat. Quart., 22,
3 17-320.
KILOH, L. G., CHILD, j. P., and L,vrxaa, C. i6o.
- 44.
* "Endogcnous depression treated with Iproniazid:
- a follow-up study." 3. men!. Sri., joG, 1425-1428.
45: KING, it. A. zg6. "Consolidation of tlsc neural trace
itt memory: investigation with one trial avoidance
conditioning and E.C.S." 3. romp. Physic!.
Psycho!., 59, 283-284.
* 46. Koitsox, S. M. tg.tg. "The successful treatment of
an obsessive compulsive neurosis with narco
synthesis followed by daily electroshocks."
j. nero. ozei!. Dis., 109, 37-41.
and Fnosv, I.
- 47. LANcAsTER, N. P., SrnsxEnv, A. R.,
r g8. "Unilateral electroconvulsive tlserapy."
`3. men!. Sri., 104, 221-227.
* 48. Lawis, ID. J., and Mxnrtit, B. .4. tg6. "Neural
consolidation and electroconvulsivc shock."
Psycho?. Rev., 72, 225-240.
49. Locrcworrn, W. 1950. "Some relation between
response to frustration punishment and outcome
of electric convulsive therapy: an experimental
* study of psychiatric theory." Gomp. Psychcl.
-monogr., 20, 121-186.
:50. - M.know, L. ig6. "Brain changes in electroshock
* therapy." Amer. 3. Psychiat., "3, 337-347.
51.MAU5EN, M. C., and McGxuort, j. L. igfli. "The
* effect of E.C.S. on one-trial avoidance learning."
j. romp. Physic!. Psycho!., 54, 522-523.
-52. MARTIN, Vt L., FORD, H. F., MCDONALD, B. C., and
`TowLitn, M. L. xg6. "Clinicnl evaluation of
unilateral E.S.T." Amrr. 3. Pychia1., 121,
:1087-1090.
-53. MEDUNA, L. V. xg. "Tcrsuchc übcr die biologisclle
*
Beeinflussung des Ablaufes der Schizophrenic;
Kampfer und Cardiazolkrampfe." .. ges. .Weurol.
:Ps.yrhia!., 252, 235-262.
54.. MENDEL5, J. zg6. "Elcctroconvulsive therapy and
depression. II. Significance of endogenous and
rreactive syndromes." Dii!. 3. Psychic!, 11,,
682-686.
55. MICnAEL, S. T. `gM. "Impairment of mental
function during electric convulsive therapy."
A.M.A. Arch. .Afeuro?. Psychic!., 71, 362-368.
a
6. MILLET, J. A. P., and Morn, B. P. Iq*
certain psychological aspects of eli-cit
therapy." Psyrhosom. Med., 6, 226-236.
57. Morto.r, j. j. B. tg.jo. "Shock as a prepar4'.
readjustment." 3. Psycho?., 10, 3'3-32i
8. Mossa, E. P. `946. "Electroshock and
structure." 3. nero. men!. Dis., zo., 296'
g. Mvnnsox, A. 1943. "Borderline cases tr.
electric shock." Amer. 3. Psychiot., Too, Go. PAcELLA, B. L., B*urniut, S. B., and lCA,.I.',-,J
ig1n "Variations in eleetrocncepht:
associated ivith electric shuck therapy ` I
with mental disorders." Arch. .iVrurol. !`
Ghirago, 47, 367-384.
6t. Paxn,.u.tx, C. A., S5IARPLESS, S. K., andjtstvie
1961. "Retrograde amnesia produr,
anaesthetic and convulsant agents." j.
Phj'siol. Psycho?., 54, 109-112.
62. Powan, T. ID. `g-is. "Psychosonsatic regre.
therapeutic epilepsy." Psychosom. Mccl., ,
290.
63. Pnoxno, N. H., Srrrnnnv, R., and BERG, K. i".
"Twenty years of shuck therapy in An,-:
1937-1957; an annotated bibliography."
Psycho?. Monogr., 62, 233-329.
64. QUARTERMA1N, ID., PAULINO, R. M., and Mi:.:
N. B. 1965. "A brief temporal gradi'-.'
retrograde amnesia independent of situat.
change." Scienre, 149, 1116-11 i8.
6. RIDDELL, S. A. 1963. "The therapeutic cflirac.
E.C.T." Arch. gen. Psychic!., 8, 42-52.
66. Rosn, J. T. 1963. "Reactive and endog':
depressiotis-rcsponsc to E.C.T."Bri!.J. Pt..
109, 213-217.
67. RosEszsvExo, S. 1938. "The experimental mr:L
mcnt of types of reaction to frustration." Explorations iii Personality. ed. Murray.
yorl
68. Srw'iao, M. B., CAMPOELL, ID., HAsnus, A.. Dasvsnnnny, j. P. rg8. "Effects of I-.'
upon psychomotor speed and the `diorra
effect' in depressed psychiatric patients." J. Sri., zo., 681-695.
6g. SLATEn, E. T. 0. rgi. "Evaluation of C1
convulsion therapy as compared with C-:
vative methods of treatment in depressive 5"
*
Ibid., g7, 567-56g.
B. 1946. "Shock therapy: psyri:
theory and research." Psycho?. lieU., 43, 21''71. STENnACK, A., VIITAMXK1, R. 0., and KUEKONi
5957. "Personality changes in electrornnv::
treatment." .4r!ez psychic!. neural. ScanS-.
3'I5-359.
72. SnetIEnSKILL, J., Ssasw-i, W., and MExL5, D
1952. "An evaluation of postelecit'
confusion with the Reiter apparatus." .-knc'
Psychic!., ioU, 835-838.
73. TANNER, H. `95°. "Physiological and psyclI0h1'
factors in electroshock as criteria of therapy
nero. men!. Dis., III, 232-238.
*
70. STAINDRO0K,
BY EDGAR MILLER
B. P. 1945. `1
cts of clcctrosl,
226-236.
as a preparation
ThoMSON, R., and DEAN, W. 1955. "A further study
of the retroactive effect of E.C.S." 3. camp.
P/rysiol. Psychol., 48, 488-491.
311
75. WElL, P. L. 1950. "`Regressive' electroplexy in
schizophrenia." 3. urea!. Sri., gG, 514-520.
,
rock and persona:
296-302.
lie cases treated
cc/riot., I00, 355-3
and JCALTNOWSK;
1ectroencephalogr
k therapy of patie
rch. JV?r,rol. I'Syc/,[
*, 104,
C., arndJMtvuc, M
aesia produced
l
.t agents." .7. w
romatic regression.
1105001. A/ct!., 7,
md BERG, K. 19
therapy in Amcd.
bibliography." Cr
9.
t. NI., and MILl:
emporal gradient!
:ndcnt of situatic:t
herapeutic euIicac
8, 42-52.
ive and endogen
:.T." Bril. 3. Psych.,
xpcrirncntal mean.
to frustration." 1
ed. Murray. N
D., HARRIS, A., 9
"Effects of E.C
and the `disnraci
tric patients." 3. m!
Evaluation of dcc
mpared with con1
.t in depressive stat'
dgar
therapy. psc SOC
Miller, B.Sc., clinical Psychologist, Oakwood Hospital, A faidstone, Ecu!; Present address:
Unit for Research on A.Icdical .4jrJ'lications of Psychology LTnirersity of Gambridge, 5 Salisbury Villas,
Station Road, Cambridge
.,and Kunnox&
;es in electroconvul
L neural. Scant!.. Ii'cceived
and Mtu..s, 13.
of postclectrosb
apparatus." Anrrr
icaI and psycbo1og
:riteria of therapy.
8.
29
iiarc/j, sgCG