1. Brhav. Thhrr. & Erp. Psychmr. Vol. 10. No. 3. pp ?I l-217.

Prmred in Great Bntain.

1989
0

InxF-79lQ89
no0 + 0.00
1990 Pergamon
Press plc.

EYE MOVEMENT DESENSITIZATION: A NEW TREATMENT FOR

POST-TRAUMATIC STRESS DISORDER
FRANCINE
Mental

Research

Institute,

SHAPIRO
Inc., Palo Alto, California

Summary -The
use of saccadic eye movements
for treating post-traumatic
stress disorder is
described.
The procedure
involves eliciting from clients sequences
of large-magnitude,
rhythmic saccadic eye movements while holding in mind the most salient aspect of a traumatic
memory.
This results in (1) a lasting reduction
of anxiety,
(2) changes in the cognitive
assessment
of the memory,
and (3) cessation of flashbacks,
intrusive thoughts,
and sleep
disturbances.
The procedure
can be extremely effective in only one session, as indicated by a
previous controlled study and a case history presented here. It does not require a hierarchical
approach,
as in desensitization,
or the elicitation of disturbingly
high levels of anxiety over a
prolonged
period of time, as in flooding. Some speculations
are offered concerning
the basis
for the effectiveness
of the procedure.

A post-traumatic stress disorder (PTSD; DSMIII, 1980) is characterized by anxiety attacks,

sleep disturbances, flashbacks, and intrusive
thoughts - related to a traumatic event, as
well as by a variety of irrational beliefs (DeFazio, Rustin, & Diamond, 1975). Two populations susceptible to this disorder are sexual
assault/molestation
victims and war veterans
(Burgess & Holmstrom, 1985; Figley, 1978), in
both of which traumatic memories seem to be
central to the manifestations (Keane, Zimering, & Caddell, 1985).
It is widely believed in the behaviorally
oriented therapeutic community that successful
treatment of PTSD requires some form of
exposure to the traumatic cues to overcome
avoidance behavior and to allow for desensitization (cf. Fairbank & Brown, 1987; Fairbank
& Keane, 1982; Fairbank & Nicholson, 1987).
Requests
U.S.A.

for reprints

should

be addressed

to Francine

Some have maintained that cognitive reassessment, in terms of redefining and finding
meaning in the event and alleviating inappropriate self-blame, is also a relevant aspect of
treatment (Janoff-Bulman, 1985). However, it
is not apparent that either systematic desensitization (SD) or flooding is particularly effective; and some members of the therapeutic
community have voiced concern regarding the
exposure of stress victims to the high levels of
prolonged anxiety used in flooding (Fairbank
& Brown, 1987).
Following encouraging clinical experiences
with the eye movement procedure on approximately 70 clients and volunteers, a systematic
study was made of 22 rape/molestation
and
Vietnam veterans (Shapiro, 1989). In this
study, the subjects (ages 11-53 years: 2 = 37
years) reported traumatic memories that had

Shapiro,

Ph.D.,

14850 Oka

Road

#I2

Los Gatos.

CA 95030.

Edirors nofe - The technique

described
in this article is out of the usual run. The results that are claimed in posttraumatic stress disorder are of great magnitude and rapidly achieved. It is very much to be hoped that the findings will be
independently
replicated.
I myself have been encouraged
by observations
with respect to three areas of continuing
disturbance
in a case of post-traumatic
stress disorder that I had largely overcome
by other methods.
One source of
disturbance
was the image of Mrs. K., a hostile insurance agent who was involved in the litigation following the causative
accident. When asked to imagine Mrs. K., the patients anxiety level went up to 100 SUDS. This was not diminished by 20
saccadic movements.
However.
30 more rapid movements
two minutes later brought
the SUD level down to 0.
Thereafter,
the patient was consistently
able to think of Mrs. K. calmly.

211

212

FRANCINE

persisted for l-47 years (_I?= 23 years). They

had received previous therapeutic treatment
for 2 months-25 years (a = 6 years). Their
presenting
complaints
included
intrusive
thoughts, flashbacks, sleep disturbances, low
self-esteem,
and relationship
problems. A
pivotal aspect was the memory of one or more
traumatic incidents. The dependent variables,
measured at the initial session and one and
three months later, were: (1) anxiety level, (2)
validity of a positive self-statement/assessment
of the traumatic incident, and (3) presenting
complaints.
The results of the study indicated that a
single session of the eye movement desensitization (EMD) procedure was sufficient to
desensitize
completely
subjects traumatic
memories and dramatically alter their cognitive
self-assessments. That is, while subjects in the
placebo group revealed little or no change in
their high anxiety level and low perceived
validity of their desired cognitions, the EMD
treatment group experienced a marked decline
in anxiety, as seen in a pre-post shift of 7.45 to
.13 on a O-10 Subjective Units of Disturbance
(SUD) scale (Wolpe, 1982) and a concomitant
increase in rated validity of cognitions. This
treatment effect was maintained virtually unchanged when assessed three months later and
was accompanied by behavioral shifts which
included the alleviation of the subjects primary presenting complaints.
Subsequent to this study, a number of new
clients were treated for PTSD symptomatology
during single desensitization sessions or multiple therapeutic sessions. One of these cases
is described here. First, however, a detailed
description of the EMD procedure is presented. Finally, potential underlying mechanisms are proposed.

General Procedure
The effect of saccadic eye movements was
discovered accidentally by the author upon
noticing in herself that recurring, disturbing

SHAPIRO

thoughts were suddenly disappearing and not

returning.
Careful self-examination
ascertained that the apparent cause was that the
authors eyes were involuntarily moving in a
multi-saccadic manner when the disturbing
thoughts arose. The thoughts disappeared
completely and, if deliberately retrieved, were
no longer upsetting. The author then made
conscious use of these movements with a
variety of volunteers and clients to explore
systematically their therapeutic possibilities.
The present EMD procedure evolved from the
observations
garnered during hundreds of
treatment sessions.
The traumatic memory is treated by requiring that the client maintain in awareness one or
more of the following: (1) an image of the
memory; (2) the negative self-statement or
assessment of the trauma; and (3) the physical
anxiety response. Simultaneously, the therapist induces multi-saccadic eye movements by
asking the client to follow the repeated side-toside movement
of the therapists finger.
Although the optimal condition occurs when
all three representations
are held simultaneously in the clients consciousness,
the
presence of any one of them can be sufficient
to achieve full desensitization.

Measurements
The anxiety level associated with the
traumatic memories is assessed by means of the
11-point (0 = no anxiety; 10 = highest anxiety
possible) SUD scale (Wolpe, 1982). This correlates with objective physiological indicators
of stress (Thyer, Papsdorf, Davis. & Vallecorsa, 1984), and is customarily used to monitor anxiety during the SD procedure.
Since irrational beliefs are a part of the
PTSD syndrome and cognitive therapy aims
to restructure
them (DeFazio,
Rustin, &
Diamond, 1975; Keane et al., 1985), shifts in
clients self-assessment of the traumatic incident or their own participation in the event are
also monitored. The measure, which was de-

Eye

Movement

vised by the author, is referred to as the

Validity of Cognition scale (Shapiro, 1989). It
is a seven-point (1 = completely untrue; 7 =
completely true) semantic differential scale
that aims to provide a rapid assessment of the
relevant beliefs.
Step-by-Step Procedure
Clients are first asked to focus on the
memory from which they wish relief and then
to isolate a single picture representative of the
entire memory (preferably the most traumatic
point in the incident). It is unnecessary for
them to describe or discuss the memory or
picture in detail; mere awareness of the image
will allow desensitization to proceed.
In order to assess the belief statement about
the incident, clients are asked What words
about yourself or the incident best go with the
picture? Most express such beliefs as I am
helpless, I should have done something, or
I have no control. For clients who have
difficulty generating an assessment statement,
the therapist may suggest some alternatives
after asking them to describe their feelings
about the past incident. Only those belief
statements that are recognized by clients as
clearly applicable to them and the incident
should be used and, where possible, in verbatim form.
Clients are then directed to concentrate on
the traumatic picture and the words of the
belief statement, to (1) assign a SUD level to
them and (2) identify the physical location of
the anxiety sensations. They are next asked
how they would prefer to feel and to supply a
new belief statement reflecting the desired
feeling (e.g., I have control, I am worthy,
I did the best I could). They are then to
judge by means of the seven-point Validity of
Cognition scale how true the new statement
feels to them (i.e., their gut level response).
Next, the folllowing instructions are given:
What we will be doing is often a physiology
check. I
need to know from you exactly what is going on. with as

Desensitization

213

clear feedback
as possible.
Sometimes
things will
change and sometimes they wont. I may ask you if the
picture changes - sometimes it will and sometimes it
wont. Ill ask you how you feel from .O to 10 sometimes it will change and sometimes it wont. I may
ask if something else comes up-sometimes
it will and
sometimes
it wont. There are no supposed
tos in
this process. So just give as accurate feedback as you
can as to what is happening,
without judging whether it
should be happening or not. Just let whatever happens.
happen.

In part, these instructions are designed to

reduce performance anxiety, confusion, and
the effects of potential demand characteristics.
They are particularly important in light of the
fact that clients will often find difficulty initially
in accepting the changes that are occurring and
will make such statements as: This is too

easy, . This cannot be happening; I must be

blocking (i.e., the emotions are not as
strong); I must be doing something wrong the picture is changing/Im having difficulty
bringing the picture up/I cant see it clearly.
During the EMD procedure,
the picture,
anxiety level, and cognitive statement do indeed undergo rapid alteration.
Therefore,
clients must be reassured that the process is
proceeding normally and that they should
Just let whatever happens, happen and report without judging whether it should be
happening or not.
Clients are then instructed to (1) visualize
the traumatic scene, (2) rehearse the negative
statement (e.g., I am helpless), (3) concentrate on the physical sensations of the anxiety,
and (4) visually track the therapists index
finger. The finger is moved rapidly and rhythmically back and forth across the line of vision
from the extreme right to extreme left at a 1214 inch distance
from the clients face, two
back-and-forth
movements
per second.
The
distance traveled by the hand on each sweep is
at least 12 inches. Very rarely, clients may
respond better to a diagonal movement
across
the midline of the face from their lower right to
to contralateral
upper
left (i.e., chin-level
brow-level).
The back-and-forth
movement
of
the therapists
finger is repeated
12-24 times,
each such grouping being defined as one set.

214

FRANCINE SHAPIRO

For the occasional client who is unable to

track the moving finger or finds this aversive,
the therapist can use a two-handed approach of
positioning each index finger on opposite sides
of the clients visual field at eye level and
alternatively lifting them. The client is instructed to move the eyes from one finger to
the other as each is raised.
After each set of saccades, clients are asked
to: Blank it (the picture) out and take a deep
breath. They are then instructed to bring up
the picture and words again, to get in touch
with the feeling generated, and to give a SUDS
rating from 0 to 10. If the SUDS level does
not decrease after two sets of saccades, the
client is asked, Did the picture change?, or
What do you get now/Does anything else
come up? If a new memory has been revealed, it is desensitized before returning to
the old picture.
Any new memories reported are usually
associated with the trauma in some way.
Frequently, they are (1) other examples of the
same type of occurrence (e.g., another incident
of molestation),
(2) another event that is
linked by the commonly shared belief statement (e.g., Im not worthy, exemplified by a
molestation incident or a failure at school), or
(3) other occurrences involving the same person (e.g., beating or molestation by father). If
the new memory is emotionally loaded, it
should become the new focal point and should
be desensitized before continuing with the
original memory. Upon returning to the latter,
it will frequently be discovered that the SUDS
level has dropped considerably.
Clients are asked periodically, with respect
to the picture, cognition, and memory, What
do you get now? Their answers are used as
evidence of change, since they often reveal
new insights, perceptions,
or alterations of
the picture (e.g., I didnt do anything wrong;
The picture seems further away). If an
answer reveals that a new associated limiting belief has arisen (e.g., I did something wrong,
along with I have no control), this belief is
included with the picture during the next set.

If the new cognition is a positive one, clients

are directed to Think of that, along with the
picture, and then the therapist induces a new
set of saccades. The instruction, Think of
that (with respect to whatever new insights or
observations have emerged during the set)
often causes clients to continue generating new
insights and cognitions that approximate more
closely the desired self-statement. Only when
the insights are no longer constructive (i.e., no
longer add to an understanding or emotional
adjustment to the situation) are clients directed
to return to the original picture.
If a reduction in SUDS level fails to occur
after two sets of eye movements, it is extremely important to search for a mismatch
of picture, cognition, or emotions. That is, if a
new picture or memory has been revealed, it
may not fit the cognition already being used.
Likewise, if the emotion has changed, the
cognitive component may be incongruent. For
instance, the cognition, It was shameful may
have applied to the feeling of guilt which was
the initial emotion associated with the trauma,
but not with the feeling of sadness which
replaced it. In order to continue the desensitization process, the cognition is dropped completely. The same is true if the picture becomes
altered so that it is no longer congruent with
the cognition.
Parenthetically, it should be noted that the
traumatic picture may change to a more neutral one (e.g., the rapist disappears from the
bedroom or a leering face changes to a smiling
one), making it difficult or impossible for the
client to retrieve the original image. In this
case, it is possible to resume the desensitization
process (assuming that it has ceased) by discontinuing the cognition or replacing it with the
desired self-statement. If the original picture or
facsimile can be retrieved, it is preferable to
continue desensitization with it in mind rather
than with the altered version. If the picture
disappears completely, clients are instead instructed to, Think of the incident.
If the SUDS level remains unchanged, but
clients can identify a body location (e.g.,

Eye Movement

tightness in the stomach) for the feeling/

emotion, they are instructed to concentrate
only on the body sensation while new sets of
saccades are generated. They then usually
report a change of body sensation and the
SUDS level decreases. When the body sensations of anxiety cease changing or disappear
entirely (as is most often the case), clients are
instructed to return to the picture of the
trauma and the standard EMD procedure is
resumed.
When the SUDS level reaches 0 or 1
(generally after 3-15 sets of saccades), clients
beliefs in the validity of the desired cognitions
are tested by asking, How do you feel abut
the statement [desired cognition]
from 1 (completely untrue) to 7 (completely true): Regardless of their assigned
Validity of Cognition rating (i.e., even if it has
already reached 7). clients are asked to
visualize the original picture, along with the
desired cognition, and another set of saccades
is generated. Then the question is repeated and
a new measure taken. If self-rated validity of
the statement consistently increases, this process is repeated. If another memory and/or
cognition seems to be interfering, the entire
procedure is repeated on the new material.
The latter situation is exemplified by a
Vietnam veteran who was attempting to accept
as valid the cognition, I can be comfortably in
control (Shapiro, 1989). When asked to respond to the validity of the statement after the
original image had been desensitized, he said,
I am not worthy to be comfortably in control. This cognition of lack of worth was
related to a different trauma which needed to
be desensitized, and then still another trauma,
having to do with failure, was revealed.
When these two additional traumas were desensitized, the client assigned a rating of 7 to
the words: I can be comfortably in control.
When no other trauma or competing cognition
is elicited and clients indicate that the positive
cognition is installed (usually after one-tothree additional sets at a validity level of 6 or
7) the EMD procedure is terminated.

Desensitization

21.5

Follow-up sessions have consistently demonstrated that the picture and cognition remain
altered. Most often, the emotional level of
0--1 SUDS is maintained, although occasionally a new emotion arises (e.g., anger,
instead of the earlier anxiety). It appears that
the predominant emotion will be desensitized
during the first session, allowing other previously masked emotions to surface. Very
often this changing of emotions occurs during
the initial treatment session, at which time they
are all desensitized. If the emotion surfaces
later, however, the EMD procedure can be
used to desensitize it at that time.

Case Study
A 63-year-old women had been raped 15
months previously. Her presenting complaints
were daily intrusive thoughts, flashbacks (including those consistently invoked by seeing
the guard dog that she had purchased after the
rape), inability to be alone, nightmares, and a
self-described feeling of being spacey (i.e.,
extremely forgetful of things that had come
naturally before, such as fastening seatbelts,
remembering wallet and checkbook, etc.). Her
summation statement was, I have lost control
of my life and the details of my life.
The therapist treated three memories of the
rape during a single 50-minute session: (1)
seeing the masked rapist appear from around
the corner, holding a gun; (2) oral copulation;
and (3) vaginal penetration.
The opening
cognition for all three was, Im overwhelmed; the desired cognition was, Its
over. The desensitization proceeded in standard form, with the client holding in awareness
the picture, cognition, and anxiety level. The
initial memory was of the appearance of the
masked rapist. The client provided an opening
SUDS level of 10 which was reduced to 0
(complete desensitization)
in four sets, at
which time she was no longer able to maintain
clearly the original picture and was asked

216

FRANCINE

therefore to Just think of the rape and the

words, Its over.
At that point the two other memories of the
rape were treated. Her representative reaction
during four sets of saccades was one of going
from pure disgust (at oral rape; SUDS =
7) to Its diffused, I have the power and
control (SUDS = 3) to I have the feeling
that he wasnt even there (SUDS = .O). At
the end of the session the client gave a Validity
of Cognition level of 7 to the statement, Its
over.
At a three-month
follow-up session the
client reported that all of the presenting complaints had ceased immediately after the initial
session and that she had remained undisturbed, focused, and peaceful during the
intervening months. When she was asked to
retrieve the picture at the follow-up, she was
able to bring up only vague blurs and reported
a SUDS level of 0. The rated validity of the
cognition, Its over remained at 7 and she
remarked spontaneously, I have control over
my life now.

Discussion
It is apparent that the EMD procedure is
extremely effective in desensitizing traumatic
memories characteristic of PTSD and eliminating attendant complaints. The basis for the
effectiveness is, however, unclear in that the
technique was not derived from a theoretical
position and there is, at this time, insufficient
empirical evidence to justify conclusions. This
in no way detracts from the usefulness of the
procedure,
of course. Nevertheless,
some
speculations about its underlying mechanisms
may be of interest.
One hypothesis comes from Pavlovs (1927)
suggestion that traumatic incidents upset the
excitatory/inhibitory
balance in the brain, causing a pathological change in the neural elements. It may be proposed that such a change
would maintain the incident (i.e., the picture
and negative self-assessment) in its original

SHAPIRO

anxiety-producing
form. Thus, the pathological change of neural elements would block the
usual progression of information-processing
to
resolution and the incident would be maintained in active memory and triggered as
intrusive thoughts, flashbacks, and nightmares
(cf. Horowitz & Becker, 1972).
Perhaps, then, when saccadic movements
are induced simultaneously with the cognition
and image associated with the physiologically
stored traumatic memory there occurs (1) a
restoration of the neural balance and (2) a
reversal of the neural pathology. This allows
information processing to proceed to resolution, with a concomitant alteration of picture,
cognition, and anxiety level and a cessation of
intrusive symptomatology. Congruent with this
notion is the finding that repeated, low-voltage
current has a seemingly permanent effect on
the synaptic potential and appears to affect
memory (Barrioneuvo,
Schottler, & Lynch,
1980). The neural bursts evoked by the repeated saccades used in the EMD procedure
may entail this process.
It is possible that rhythmic, multi-saccadic
eye movements represent the brains automatic
inhibitory (or excitation-releasing) mechanism.
For example, it is possible that unconscious
material surfacing during dreaming is partially
desensitized by rapid eye movements (REM).
Congruent with this hypothesis are the results
of a study by Lavie, Hefez, Halperin, and
Enoch (1979) in which combat veterans suffering from PTSD revealed a longer latency to
enter REM sleep and spent less time in REM
sleep than did a control group. Thus it is
possible that anxiety and rapid eye movements
are reciprocally inhibitory.

Conclusions
The case described in this paper is typical of
the single-session desensitization
treatments
carried out by the author. Follow-up measures
obtained as many as 12 months later have
revealed that the memories continue to be

Eye

Movement

desensitized, cognitions restructured, the pictures remain altered or difficult to retrieve, and
the therapeutic effects on presenting complaints persist.
Multiple sessions have been necessary for
some combat veterans and for one sexual cult
victim who had been abused over a seven-year
period. Nevertheless, one-to-three individual
traumatic memories can be treated in a single
session which, for many PTSD victims may be
sufficient to eliminate the pronounced symptomatology. It must be emphasized, however,
that while the present description contains
sufficient information to desensitize approximately 60-70% of PTSD-related
traumatic
memories, specialized and intensive training is
necessary to approach the highest success
rates.
The EMD procedure is novel and still in the
process of refinement. To increase its credibility in the therapeutic community it is necessary
that the successes be independently replicated.
The outlook is promising in that therapists in
both the United States and Israel, having been
instructed in the procedure,
appear to be
obtaining comparable results. Published reports are expected to appear in the coming
year.

References
American
Psychiatric
Association.
(1980). Diagnostic and
slatistical manual of mental disorders (3rd ed.). Washington: APA.
Barrionuevo,
G., Schottler,
F., & Lynch, G. (1980). The
effects of repetitive low frequency stimulation
on control
and potentiated
synaptic responses in the hippocampus. Life Sciences, 27, 2385-2391.
Burgess, A. W., & Holmstrom,
L. L. (1985). Rape trauma

Desensitization

217

syndrome
and post traumatic stress response.
In A. W.
Burgess
(Ed.),
Rape and sexual assauh: A research
handbook. New York: Garland.
DeFazio, V.. Rustin, S., & Diamond, A. (1975). Symptom
development
in Vietnam veterans. Americnn Journal of
Orthopsychiatry, 43, 6W6.53.
Fairbank,
J. A.. & Brown. T. (1987). Current behavioral
approaches
to the treatment
of post-traumatic
stress
disorder.
Behavior Therapist, 3, 57-U.
Fairbank,
J. A., & Keane, T. M. (1982). Flooding
for
combat-related
stress disorders:
Assessment
of anxiety
reduction
across traumatic
memories.
Behavior Therapy, 13, 499-510.
Fairbank,
J. A., & Nicholson,
R. A. (1987). Theoretical
and empirical
issues in the treatment
of post-traumatic
stress disorder in Vietnam veterans. Journal of Clinical
Psychology, 43, 4&55.
Figley, C. R. (1978). Psychosocial
adjustment
among
Vietnam veterans. In C. R. Figley (Ed.), Stress disorders
among Viemam veterans: Theory, research, and treatmenr. New York: Brunner/Mazel.
Horowitz,
IM. J., & Becker,
S. S. (1972). Cognitive
response to stress: Experimental
studies of a compulsion to repeat trauma.
In H. G. Holt & E. Peterfreund
(Eds.), Psychoanalysis & Conremporary Sciences (Vol.
1). New York: Macmillan.
Janoff-Bulman,
R. (1985). The aftermath
of victimization:
Rebuilding
shattered
assumptions.
In C. R. Figley
(Ed.). Trauma and ifs wake. New York: Brunner/Mazel.
Keane: T. M., Zimering,
R. T., & Caddell, J. M. (1985).
A behavioral
formulation
of post-traumatic
stress disorder in Vietnam veterans. Behavior Therapist. 8. 9-12.
Keane, T. M., Fairbank,
J. A., Caddell, J. M.. Zimering,
R. T.. & Bender. M. A. (1985). A behavioral
annroach
to assessing and treating post-traumatic
stress dis&der in
Vietnam veterans. In C. R. Figley (Ed.), Trauma and ia
wake. New York: Brunner/Mazel.
Lavie, P., Hefez, A., Halperin,
G., & Enoch. D. (1979).
Long-term
effects of traumatic
war-related
events on
sleep. American Journal of Psychiatry, 136. 17.5-178.
Pavlov, I. P. (1927). Conditioned reflexes (G. V. Anrep,
Trans.).
New York: Liveright.
Shapiro,
F. (1989).
Efficacy
of the eye movement
desensitization
procedure
in the treatment
of traumatic
memories.
Journal of Traumatic Stress, 2. 199-223.
Thyer, B. A., Papsdorf, J. D., Davis. R., & Vallecorsa.
S.
(1984). Autonomic
correlates
of the subjective
anxiety
scale. Journal of Behavior Therapv
. _ & Experimemal
Psychiatry, 15. j-7.
Wolpe, J. (1982). The practice of behavior therapy. New
York: Pergamon
Press.