Prog. Neuro-Psychopharmacol, b Biol. Psychiat. 1987. Vol. 11. pp. 179-184 0278-5846187 $0.00 + .

50
Printed in Great Britain. All rights reserved. Copyright 0 1987 Pergamon lournals Ltd.

MEMORY EFFECTS OF RESTRICTED ENVIRONMENTAL

STIMULATION THERAPY (REST) AND POSSIBLE
APPLICATIONS TO ECT

PETER SUEDFELD, CARMEN E. RAMIREZ, RONALD A. REMICK,

and JONATHAN A. E. FLEMING

Departments of Psychology and Psychiatry, University of British Columbia

Vancouver, British Columbia, Canada

(Final form, December 1986)

Abstract

Suedfeld, Peter, Carmen E. Ramirez, Ronald A. Remick, and Jonathan A. E. Fleming: Memory
Effects of Restricted Environmental Stimulation Therapy (Rest) and Possible Applications to
ECT. Prog. Neuro-Psychopharmacol. & Biol. Psychiat. 1987, 11: 179-184.

Restricted environmental stimulation (REST) has been shown to facilitate learning and
memory in both human and animal experimental subjects. This paper reports early data
from a test of the usefulness of REST in reducing post-ECT amnesia in depressive
patients. Two such patients were placed in a quiet, dimly illuminated room for 2-4 hrs.
after recovering from each ECT administration in a series of treatments; three others,
following standard practice, were returned to their normal hospital rooms. Measures of
memory (verbal, numerical, nonverbal, life event, and self-rating) were given prior to
the first ECT treatment; after the first post-recovery session; after the last
post-recovery session; and one week after the last ECT administration. The major
difference found was that the REST group showed an improvement in self-rated memory
functioning from the first to the last ECT administration that was 15 times as great as
that reported by the control group. This finding is interesting because of the major
role played by self-reported memory disturbances in the scientific, clinical, and
popular evaluation of ECT. The sample size is being increased, as it must be for any
reliable conclusions to be drawn from this study.

Keywords: amnesia, ECT, memory, restricted environmental stimulation, REST.

Abbreviations: electroconvulsive treatment (ECT), restricted environmental stimulation

therapy (or technique) (REST).

Introduction

Most of the research literature on the memory effects of ECT has been devoted to
two primary issues. One of these is to establish what these effects actually are. and
the other is to find methods that would prevent or at least minimize adverse effects.
These endeavours have been proceeding for several decades, as the possibility that ECT
may cause serious and long-lasting cognitive deficits first arose and then became a
widely publicized issue. As any review of popular or professional writings indicates,
the problem is still attracting considerable attention.

The answer to the first question. the nature of ECT effects, depends on many
factors. Among these are the kind of memory and memory disruption being tested:
long-term, short-term. iconic. verbal, nonverbal, anterograde. retrograde. etc. The
kind of test used is also important. since a wide range of objective, self-report, and
observational measures do not necessarily yield identical results. Last, such variables
. .
as the tunmg, number and frequency of tests, the use of alternate but equivalent test
forms, the inclusion of appropriate control conditions, etc., also affect the
conclusions. To summarize the data. there may be no definitive answer to the general

179
180 P.Suedfeld
etal.

question. The results are quite mixed, although we can agree that there is certainly
short-term disturbance of memory. The data on anterograde amnesia, or interferencewith
new learning. are less consistent; and long-term memory deficits appear to be primarily
evident in self-report rather than objective indices. However, the literature suffers
from practice effects. confounded designs, inconsistent choice and inappropriate
comparison of tests. the use of subject groups of uncertain or mimed diagnostic
categories, and other shortcomings of rigour.

As to the minimization of memory loss, most researchers have addressed themselves

to aspects of ECT administration. The literature discusses and compares variations in
waveform. power, and electrode placement as the most frequently identified critical
variables. In general, the argument has been that unilateral shock to the nondominant
hemisphere results in less disturbance, at least to verbal memory, than bilateral shock:
that suprathreshold intensities are more likely to cause amnesia than threshold levels;
and that there may not be much difference between the effects of sinusoidal and pulse
waves if intensities are more or less equated. However, here again the data are mimed
and there have not been enough studies with designs that disaggregate the variables.
Other factors related to the treatment that may be manipulated to reduce amnesic effects
may be oxygenation of the patient, snaesthesia. muscle relaxation, and the number and
frequency of shocks (Malitz & Sackeim, 19861.

One problem with this approach is that clinicians cannot allow the memory
consequences of a particular combination of factors to dictate its use. For example.
some types of shock delivery may be technically more difficult than others, as has been
claimed for unilateral as compared to bilateral electrode placements. Some may have
adverse effects on other processes, even if they disrupt memory less. Perhaps most
importantly, procedures that cause minimal memory disruption msy not be as effective in
clinical treatment as some more disruptive techniques.

The procedure that we are now testing t% minimize memory loss in this context does
not manipulate the parameters of ECT itself. It therefore avoids the potential problems
described above. Instead, it involves the use of an environmentalmodification after
each ECT administration,an environmental modification that has been shown to affect
retention and amnesia in ways that make it a plausible technique for ameliorating
disruption.

The modification is generally referred to as REST, or Restricted Environmental

Stimulation Therapy. Ita major characteristicis the reduction of ambient stimulation in
the subject's immediate global emironment, and a concomitant but lesser reduction in
response latitude. Although several specific techniques have been tested (Zubek, 19691,
the most frequently used methods today are chamber and tank REST. Respectively, these
involve the participant lying on a bed in a darkened, quiet room or floating in a warm,
dense saline solution within a dark and quiet enclosure (Suedfeld, 1980).

Both of these variants of the technique have been used CliniCallY to treat a wide
variety of psychological and psychophysiologicalproblems. Among the former are health-
endangering habit patterns, such as smoking, overeating and alcohol abuse; among the
latter are stress-relateddysfunctions including tension headache, insomnia, essential
hypertension, and gastrointestinalsymptoms. Stimulus reduction has also been used in
conjunction with more traditional psychotherapeuticand behavioural interventions. in
the conditions mentioned previously as well as in the treatment of various psychotic
states, neurotic symptoms and adjustment problems (Fine & Turner, 1985; Suedfeld &
Kristeller. 1982). Incidentally, one factor shared by ECT and REST is the vast amount
of misinformation and distorted criticism of the techniques by mass media and special
interest groups; another is their well-demonstratedclinical efficacy (Suedfeld. 1980).

The experimental literature on the effects of REST has addressed the issue of
memory. A number of studies has shown quite reliably that REST can assist recall, both
for essentially artificial material such as nonsense syllables and digits and for more
realistic material such as connected verbal passages (Suedfeld, 1969). One study, in
fact, indicated that subjects undergoing 24 hours of chamber REST remembered more of a
long passage after the experiment than they had immediately after hearing the passage
 Environmental and memory loss after
restriction ECT 181

prior to REST, where8s control subjects showed the expected forgetting curve during an
intervening %-hour period during which they engaged in normal activities, The
difference was not due to initial learning, which W8S about the same for both groups;
nor to rehearsal, since the two groups reported equivalent and low rates of thinking
about the material between test administrations (Grissom. 1966).

The fact that REST has been shown to assist recall by reducing forgetting in human
experiment81 subjects does not necessarily imply that the technique will counteract
the effects of memory-disruptingevents. Here. however. there are analogues from animal
research that seem encouraging. In three studies, experimenters administered
electroconvulsiveshock sufficient to induce extensive retrograde amnesia to groups of
rats. The responses learned previous to the shock included appetitive as well as
avoidance and escape responses. In each study, one group of animals was put immediately
after shock into a dark. isolated cage; other groups were put back into their home cages
or into other illuminated and/or group housing. In 811 three studies, what we might
Call the REST group showed little or no amnesia. in significant contrast to the other
treatment groups. The duration of RgST in these c8ses varied from one hour to three
days (Calhoun, Prewett. Peters & Adams. 1975: Hinderliter, Smith & Misanin. 1976;
Peters, Calhoun & Adams. 1973).

To sum up, we have two starting points for the hypothesis that post-ECT REST may
reduce amnesia: the positive effects of REST on recall in human experimental subjects.
and its action in eliminating post-shock amnesia in animals. Since REST is pleasant and
relaxing for human participants, has no known adverse side- or after-effects, and is
easy and economic81 to administer. there appear to be no counterindicationsfor testing
the hypothesis.

One problem of which potential users or investigatorsof this technique must be

aware is the response of some patients' rights advocates. We encountered this issue
when our planned study became public. The objections seemed to be based on two
misconceptions. One ~8s that BCT would be administered as part of an experiment; the
other ~88 that REST is a torture technique that would be imposed upon unwilling and
helpless patients. who were already being tortured by EXT. Most of these attacks were
abandoned after the project was cleared by the ethical review boards of the University.
the Faculty of Medicine, and the Hospital. We also explained in many letters and
telephone calls that the participants would be depressed patients who were
to receive RCT as part of their prescribed therapy and who volunteered for REST, and that
REST itself has been repeatedly and consistently shown to be a mild and frequently
enjoyable experience for psychiatric patients as well as for other subject groups.

Methods

The final design of the study was as follows. To be eligible for inclusion,
patients had to be diagnosed as depressive; functioning at a sufficiently good level to
give informed consent and to respond to orientation interviews and the memory tests; and
must not have received RCT within the 12 months prior to participating in the study.
Patients who met these criteria and who were echedufed for ECT at Shaughnessy Hospital
(later extended to the IJBCHealth Sciences Centre Psychiatric Hospital) were interviewed
by a member of the research team. If they agreed to participate, they were randomly
assigned to either the REST or the control group. ECT was administered as sinusoidal
current (Medcraft machine) to the right hemisphere only, for 0.5 to 1.0 sec. 8t 110-170
volts. Standard anaesthesia (methohexital,succinylcholinefwas administered prior to
ECT. Treatments were given every third day when possible, end most patients received
8-12 treatments in the series.

The environmental manipulation occurred after the patient had undergone ECT and
then recovered from the snaesthetic in the usual recovery room. REST patients were
taken to 8 small room. from which all furniture except the bed had been removed. The
room was dimly illuminated, and w8s loc8ted so thst ambient noise w8s quite low. This
Somewhat modified REST condition had previously been used with autistic children
(Suedfeld & Schwartz. 1983), and was felt to be less anxiety-arousingthan complete
derkness. An intercom was instslled, and a monitor just outside the room kept constant
watch over the patient. Requests for something to drink, to go to the toilet, etc..
were responded to by the monitor. as were any requests to come out of the room. Since
182 P. Suedfeldet al.

there was no a priori basis for a particular session length, we decided that two hours
would count as a completed session for purposes of data analysis. and that four hours
would be the maximum that a patient would be permitted to remain in REST. All of the
four REST-group patients run so far have completed at least two hours in each scheduled
session. Patients assigned to the control condition spent the equivalent time in their
own hospital room. The same schedule was followed after each ECT administration in the
series, regardless of whether memory tests were given on that day.

Testing sessions were scheduled for one or two days previous to the first
administration of ECT; immediately after the first REST or control session, which in
turn came immediately after recovery from the first ECT administration of the series;
immediately after the equivalent period following the last ECT of the series; and one
week after completion of ECT. Each testing session was preceded by an interview and
conversation to relax the patient and establish rapport; nevertheless, some patients
declined to be tested on some occasions. The acceptability of such a decision was
always communicated to the patient to prevent any feelings of anxiety or guilt.

The tests included both long- and short-term memory items, both objective and
self-evaluation formats, and verbal as well as nonverbal materials (Ray-Davis. digit
span, word learning, delayed recall. memory for life events, Squire self-rating scale of
memory function). Both retrograde and anterograde amnesia were tested. and equivalent
forms of the test were used when available.

Results

At this time, only preliminary results are available. from the first five patients
who participated in the experiment at Shaughnessy Hospital. Two of these were in the
REST and three in the control condition. We must emphasize the small sample size, and
therefore the tentativeness of any comparisons. The projected total size is
approximately twenty (ten in each group). Some additional subjects have already been
run but data analysis on their performance is incomplete; others are scheduled for
future participation.

The results so far do show some patterns. To begin with, one REST subject found
the scheduled four-hour sessions in that environment to be quite tolerable. The other
responded well during the first two hours, and then requested to discontinue the
sessions. Two other patients. run more recently (for whom other data are not
yet available), also remained at least two hours in each REST session without any problem.
Thus, fears that REST would be very disruptive or stressful for post-ECT depressive
patients were unfounded, at least for these individuals.

Looking at the results of the objective memory tests, we found that both the REST and
control subjects tended to show declines on most tests from the pre-ECT administration to
the testing immediately following the first shock. However, the one-week followup essent-
ially demonstrated recovery to pre-treatment levels. There was considerable individual
variation, which may have overwhelmed any group differences in such a small sample; in any
case, the objective tests showed no consistent differences between the REST and the control
groups.

However, one striking difference was found. This was on self-rated problems of
memory. One of the tests (Squire. Wetzel & Slater, 1979) requested the patient to rate
his or her evaluation of current memory problems or lack of them, compared to the period
just prior to hospitalization. This was done on a +4 to -4 scale, positive numbers
indicating improvement and negative ones indicating disruption. Again there were large
individual differences, both in the magnitude and the pattern of ratings across
administrations. But there was a consistent pattern, in the comparison between the
ratings given after the first ECT session and those obtained after the last treatment:
REST subjects averaged an improvement of 47 scale points. while controls'averaged under
3 points.

Conclusion

One fact that has been established so far is that our patients do not find REST to
be particulary aversive or difficult to tolerate. Objective tests of memory have shown
 Environmental restriction and memory lossafter
ECT 183

" encouraging recovery to pre-ECT levels, regardless of experimental condition; whether

the environmental variable will remain nonsignificantwith a larger subject sample is
not yet known.

Our most interesting finding at this point is that REST-group patients evaluate
their own memory as showing much greater improvement from the first to the last post-ECT
period than controls. This datum, strongly favouring the REST intervention, is
compatible with clinical practice in prescribing quiet and bedrest for patients
following ECT administrations. The finding is particularly significant because
self-reported disturbances show the most consistent memory loss in the general ECT
literature, seem to be most distressing to patients, and have given rise to the most
hostile public attacks on the use of ECT as a therapeutic tool. The planned additional
subjects will help to answer the question of whether the difference is reliable and
long-lasting.

Acknowledgements
This research was made possible by funding from the British Columbia Health Care
Research Foundation. The contributions of Rodney Day, Alistair Wallbaum and Glenn K.Y.
Wong to the running of the study. and the cooperation of the staff and patients of the
Shaughnessy Hospital and the UBC Health Sciences Centre Hospitals, are gratefully
acknowledged.

References
Calhoun, K.S., Prewett. M.J.. Peters, R.D.. & Adams, H.E. (1975). Factors in the
modification by isolation of electroconvulsiveshock-producedretrograde amnesia
in the rat. Journal of Comparative and Physiological Psychology, e. 373-377.

Fine, T.H. & Turner, J.W. Jr. (Eds.). (1985). Proceedings of the First International
Conference on REST and Self-Regulation. Toledo, OH: Medical College of Ohio.

Grissom. R.J. (1966). Facilitation of memory by experiential restriction

after learning. American Journal of Psychology,2. 613-617.

Hinderliter. C.F.. Smith. S.L. & Misanin. J.R. (1976). A reduction of

ECS-produced amnesia through post-ECS sensory isolation. Bulletin of the
Psychonomic Society. 1, 542-544.

Malitz. S. & Sackeim. H.A. (Eds.) (1986). Electroconvulsivetherapy: Clinical

and basic research issues. New York: Annals of the New York Academy of Sciences,
Vol. 462.

Peters, R.D.. Calhoun, K.S.. & Adams, H.E. (1973). Modification by

environmental conditions of retrograde amnesia produced by ECS. Physiology and
Behavior. 11. 889-892.

Squire, L.R.. Wetzel. C.D.. & Slater. P.C. (1979). Memory complaint after
electroconvulsivetherapy: Assessment with a new self-rating instrument.
Biological Psychiatry, 14, 791-801.

Suedfeld. P. (1980). Restricted environmental stimulation: Research and

clinical applications. New York: Wiley.

Suedfeld. P. h Kristeller. J.L. (1982). Stimulus reduction as a technique in

health psychology. Health Psychology, 1. 337-357.

Suedfeld. P. & Schwartz, G. (1983). Restricted environmental stimulation

therapy (REST) as a treatment for autistic children. Journal of Developmental and
Behavioral Pediatrics, 4. 196-201.

Zubek. J.P. (Ed.). (1969). Sensory deprivation: Fifteen years of research.

New York: Appleton-Century-Crofts.
184 P. Suedfeld et al.

Inquiries and reprints requests should be addressed to:

Dr. Peter Suedfeld

Department of Psychology
Univ. of British Columbia
Vancouver, B.C.
Canada
V6T lW5