ORIGINAL ARTICLES

The Role of the h b i c System in

Phenomena of Temporal Lobe Epilepsy
Pierre Gloor, MD, PhD, Andre Olivier, MD, PhD, Luis F. Quesney, MD, PhD, Frederick Andermann, MD,
and Sandra Horowitz, M D

Experiential phenomena occurring in spontaneous seizures or evoked by brain stimulation were reported by 18 of
29 patients with medically intractable temporal lobe epilepsy who were investigated with chronic, stereotaxically
implanted intracerebral electrodes. The phenomena mainly consisted of perceptual (visual or auditory) hallucina-
tions or illusions, memory flashbacks, illusions of familiarity, forced thinking, or emotions. Experiential phenomena
did not occur unless a seizure discharge or electrical stimulation involved limbic structures. For such phenomena to
occur, seizure discharge or electrical stimulation did not have to implicate temporal neocortex. This was true even
for perceptual experiential phenomena. Many experiential responses elicited by electrical stimulation, particularly
when applied to the amygdala, were not associated with electrical afterdischarge. Limbic activation by seizure
discharge or electrical stimulation may add an affective dimension to perceptual and mnemonic data processed by
the temporal neocortex, which may be required for endowing them with experiential immediacy.
Gloor P, Olivier A, Quesney LF, Andermann F, Horowitz S: T h e role of the limbic system in experiential
phenomena of temporal lobe epilepsy. Ann Neurol 12:129-144, 1982

Almost 50 years ago, in 1933, while operating on a
conscious patient with intractable seizures, Penfield
for the first time evoked, by electrical stimulation of
the temporal cortex, a memory “flashback,” much to
his own and the patient’s surprise. In one of his last
books, The Mystery of the M i n d , Penfield [39]recalled
this earlier observation and wrote: “I was incredu-
lous. O n each subsequent occasion I marvelled . . . I
was astonished each time my electrode brought forth
such a response. How could it be? This had to do
with the mind! I called such responses ‘experiential’
and waited for more evidence.” That “psychic”
phenomena could occur in the course of an epileptic
seizure had been known since the last century, when
Hughlings Jackson [22-2 51 described the so-called
intellectual aurae or dreamy states. Such “psychic”
or “experiential” phenomena activated by epileptic
discharge arising in the temporal lobe may occur
as complex visual or auditory o r combined audi-
tory-visual hallucinations o r illusions, memory
“flashbacks,” erroneous interpretations of the pres-
ent in terms of the past (e.g., as an inappropriate
feeling of familiarity o r strangeness, the “deji vu”
and “jamais vu” phenomena), or as emotions, most
commonly fear. Penfield and associates [34, 36, 40,
411 called these phenomena “experiential,” an ap-
propriate term considering the fact that to the af-
fected patients they often assume an astonishingly
vivid immediacy, which they liken to that of actual
events. Nevertheless, the patients are never in doubt
that these phenomena occur incongruously, that is,
out of context, as if they were superimposed upon
the ongoing stream of consciousness, with the ex-
ception of fear, which is sometimes interpreted as
fear of an impending attack. This insight clearly dis-
tinguishes these phenomena from psychotic halluci-
nations and illusions.
We have had the opportunity to confirm many
of Penfield’s observations, but the experiential
phenomena he reported appeared to us to have an
anatomical substrate within the temporal lobe dif-
ferent from what he described: they depended on
limbic rather than on temporal neocortical activation.
Few detailed studies have been carried out on expe-
riential responses to brain stimulation since the pub‘
lication of Penfield and Perot’s paper in 1963 [41],
with the exception of those by Weingarten et a1 [45]
and by Halgren et a1 [ 181. Some of their observations
were similar to ours inasmuch as they suggested a
role of the limbic system in the elaboration of these
responses. However, in contrast to our studies, in
which both neocortical and limbic structures of the
temporal lobe were explored by electrical recordings
and stimulations, those workers investigated only the

From the Department of Neurology and Neurosurgery, McGill
University and the Montreal Neurological Institute, Montreal,
Que, Canada.
Received June 29, 1981, and in revised form Oct 26. Accepted for
publication Nov 7, 1981.
Address reprint requests to Dr Gloor, Montreal Neurological In-
stitute, 3801 University St, Montreal, Que, Canada H3A 2B4.

0364-5 134/82/080129-16$01.25 @ 1982 by the American Neurological Association 129

limbic c o m p o n e n t s , m a k i n g it impossible to deter-
m i n e whether the limbic or neocortical c o m p o n e n t s
of the temporal lobe are more important for t h e
elaboration of experiential p h e n o m e n a .
Patient Population and Methods
The present study is based on observations made in 35 pa-
tients in whom intracerebral depth electrodes were
stereotaxically implanted. These patients suffered from
medically intractable cerebral seizures for which a focal
origin was suspected but could not be clearly identified
with the usual scalp and sphenoidal electroencephalo-
graphic (EEG) recordings [ 141. Most of them had EEG evi-
dence of bilateral independent temporal epileptiform ab-
normalities, and in 2 9 of them, all or nearly all the seizures
were shown by depth electrode recordings to originate
from one temporal lobe [16]. In 4 patients the seizures
arose from either the frontal o r occipital lobe, and in 2 their
origin could not be identified with certainty. In most pa-
tients both temporal and frontal lobes were explored
stereotaxically, but in a few, unilateral temporal and occip-
ital o r temporal and frontal explorations were performed.
An example of the stereotaxic approach most frequently
used is shown in Figures 1 and 5. Three thin electrode
strands, each containing 10 contacts 5 m m apart, were in-
troduced horizontally on each side through the second
temporal convolution. The two deepest contacts of each
strand reached one of the limbic structures of the temporal
lobe, while the more external ones sampled the activity of
the temporal neocortex both at the surface and within the
deep and superficial sulci (mostly the depth of the sylvian
fissure and the superior temporal sulcus). T h e two deepest
contacts of the anterior and intermediate depth electrode
strands were located in the amygdala and the anterior por-
tion of the hippocampus, respectively. T h e three deepest
contacts of the posterior depth electrode strands sampled
130 Annals of Neurology Vol 12 No 2 August 1982
Fig I . Most commonly used arrangement for depth electrode
implantations. Each horizontally inserted electrode array con-
tains 10 contacts 5 mm apart. Contact I i n each array is
deepest. Contacts I and 2 of LA and RA rest i n the left and
right amygdala, respectively; contacts 1 and 2 of LB and R B
i n the left and right anterior hippocampus; and contact 1 of
LC and RC in the left and right parahippocampal gyrus and
contacts 2 and 3 of LC and RC i n the body of the left and
right hippocampus. (LFM, RFM = left and right superior
and mesial frontal electrodes; LFO, RFO = lejt and right or-
bital frontal electrodes.)
the activity of the parahippocampal gyrus and the adjacent
hippocampus. Frontal and occipital electrodes were also in-
serted horizontally (Fig 1).
T h e intracerebral EEG was recorded virtually around the
clock for two to five weeks. Patients spent the day in the
EEG laboratory. At night and on weekends their EEG and
behavior were monitored with the automatic computer-
assisted 16-channel cable telemetry and audio-video re-
cording systems designed by Ives et al [20,21]. T h e aim of
these prolonged recordings was to document a number of
spontaneous seizures sufficient to identify, as accurately as
possible, the site (or sites) of onset of the patient's habitual
attacks.
In addition, in a single day-long session, electrical stimu-
lations were applied through pairs of adjacent contacts of
all electrode strands in an attempt to elicit the patient's
warning o r other fragments of the clinical seizure. This
often provided further confirmation of the location of the
seizure focus. Stimulation parameters were as follows.
Square-wave bipolar symmetrical pulses of 0.5 msec dura-
tion were delivered by a constant-current Nuclear Chicago
Stimulator. All contact pairs were first stimulated at a low,
usually subthreshold, intensity of 500 PA. Later they were
restimulated at 1 and 2 mA, and more rarely at 3 and 4 mA. Table 1 . Experiential Illusions and Hallucinations Observed
Patients were asked to report what they experienced dur- with Stereotaxic Exploration of the Temporal Lobes
ing the stimulation. Positive reports were validated
No. of No. of
whenever feasible by warning without stimulation or
Experience Observations Patients
stimulation without warning. The latter was applied while
the patient was engaged in a conversation in order to dis- Visual illusions 9 3
tract his or her attention. In this situation the patient had Elementary visual 15” 3
no way of knowing when the stimulator would be turned hallucinations (phosphenes)
on. Very few patients reported false-positive responses, Complex visual 18 5
i.e., responses that failed to recur on stimulation without hallucinations
warning. Rarely did a patient report a response to warning Auditory illusions 0 0
without stimulation. Elementary auditory 0 0
T h e EEG was recorded continuously during stimulation hallucinations
sessions. The occurrence or absence of electrical afterdis- Complex auditory 3 2
charge was noted. Afterdischarges short enough to last for hallucinations
only the initial seconds of amplifier saturation immediately Olfactory hallucinations 2 1
after the offset of srimulation may sometimes have been Familiarity (deja vu) 23 4
missed. Such amplifier blockage could only affect the two Unfamiliarity (jamais vu) 0 0
electrodes through which stimulation had been applied,
Memory recall 19 5
however, and would not have prevented detection of a
Forced thinking 10 2
spread of afterdischarge to structures as close as 5 mm to
the site of stimulation. Fear >49 7
Anticonvulsant medication was gradually reduced when Anger 1” 1
the patients entered the study. By the time electrical Irritation >3 1
stimulation was carried out, such medication had been to- Emotional distress 6“ 3
tally withdrawn in most patients. (depression, guilt, etc)
Far-away feeling >3 1
Results Feeling of someone 1 1
Incidence and Types of Experiential Phenomena being nearby
Among the 3 5 patients studied in this manner, 18 Pleasant emotion 0 0
( 5 2 % ) exhibited experiential phenomena of one kind Sexual emotion 0 0
or another. In 4 patients these phenomena occurred Thirst lo* 2
Hunger 0 0
during spontaneous seizures only, in 8 they were
elicited with electrical stimulation only, and in 6 they Feeling of bodily 2 1
distortion
were observed both during spontaneous seizures and
Strange, indescribable 2 2
with electrical stimulation. If only the 29 patients feeling (mental)
are considered in whom a temporal origin of seizures Floating sensation 7 1
was demonstrated, the incidence of experiential (excitement? startle?)
phenomena rises to 62%. Usually, but not always,
~ ~

“All induced by electrical stimulation.

the experiential phenomena elicited by electrical
stimulation had previously occurred in the patients’
spontaneous seizures. The few patients in whom the
seizures were proved not to be of temporal lobe ori-
gin reported no experiential phenomena during
spontaneous seizures or in response to brain stimula-
tion.
Table 1 lists the experiential phenomena observed
in this study o r known to occur in temporal lobe
epilepsy. Fear was the most commonly encountered
experiential phenomenon. Illusions of familiarity
(the “deja vu” phenomenon), memory recalls, and
complex visual hallucinations were next. (Elementary
visual hallucinations were seen only with electrical
stimulations and were undoubtedly caused by inad-
vertent excitation of fibers of the optic radiation in
the depth of the temporal lobe. We did not consider
them to be experiential.) Emotional distress (in the
form of depression, guilt, or less definable unpleasant
bAngry mood and facial expression, no aggression.
‘In 1 instance, may have been caused by strong nausea
dNine of the 10 observations were in 1 patient.
emotional states), forced thinking, and complex au-
ditory hallucinations were less common. One com-
plex auditory hallucination was combined with a
complex visual hallucination. Neither auditory illu-
sions nor elementary auditory hallucinations were
reported. Anger was observed only once; it consisted
merely of an angry mood without aggressive behav-
ior. Pleasant emotions, sexual feelings, and hunger
were never reported, but thirst occurred in 2 pa-
tients. Some experiential phenomena were difficult
to classify, such as a feeling of bodily distortion, a
strange, indescribable feeling in the mind, or a
“floating sensation” which the patient likened at one

Gloor et al: Limbic Experiential Phenomena 131

time to the excitement felt when watching a football
game and at another time to a startle.
On e difficulty encountered in classifying these
phenomena was that what the patient considered a
unified experience sometimes had to be entered into
two o r three categories, for often a perceptual expe-
riential phenomenon was associated with an emotion
or with memory recall. Thus, a single complex ex-
perience was often listed in several of the categories
enumerated in Table 1.
Several illustrative case reports follow.
Patient 1
Fear, complex vi.tual hallucination with memory recall and
negative affectitie connotation evoked by limbic (particularly
amygdaloid) seizure discharge and stimulation
A 19-year-old woman had seizures that started with a feel-
ing of intense fear followed by loss of consciousness and
automatism in which she acted as if she were in the grips of
the most intense terror. She let out a terrifying scream and
her facial expression and bodily gestures were those of
someone having a horrifying experience. She was able to
recall her fear, but had no recollection of acting it out in the
later part of her seizures. The fear was associated with a
feeling of numbness in both legs. At a younger age she had
reported seeing crocodiles trying to bite her legs. This hal-
lucination was no longer a feature of her seizures when she
came under o u r care.
In the depth EEGs the seizures started in the limbic
structures of the right temporal lobe, particularly the
amygdala. Only one seizure began in the left hippocampus.
All seizures in which the patient lost consciousness in-
volved bilateral spread to temporal and frontal neocortex.
Numerous small seizures consisting only of fear, some-
times with startle, involved the right-sided limbic and tem-
poral neocortical structures diffusely o r were confined to
the right hippocampus, amygdala, and deep anterior tem-
poral neocortex. Electrical stimulation of the right amyg-
dala induced intense fear. Less intense fear was evoked
from left amygdaloid stimulation delivered without warn-
ing. It was not followed by afterdischarge. In addition to
this fear, the patient had a feeling of someone being
nearby. She then distinctly saw a person who seemed to be
standing in the sun. Upon questioning she identified this
person as a former boyfriend. She denied that she would
experience fear if she were to meet him but admitted that
she would be annoyed. Upon stimulation of the left hip-
pocampus, which produced a local afeerdischarge without
spread, she saw someone unknown 10 her sitting in the
grass, and experienced only slight fear.
Patient 2
Complex visual hallucination, forced thinking, or memwy recall
evoked by limbic Jeizure discharge and limbic stimulation
A 35-year-old woman suffered from seizures that were
ushered in by an epigastric sensation and an experience
she described as seeing or thinking of two comic strip
132 Annals of Neurology Vol 12 No 2 August 1382
characters, “Philomene” and ‘yacquot.” She told us that as a
child she had often looked at this comic strip with her
mother. Many seizures then proceeded to staring, unre-
sponsiveness, irrelevant stereotyped speech, and au-
tomatism, sometimes followed by a generalized convulsion.
Figure 2 shows the EEG of a minor seizure with gradual
buildup of the discharge in the left hippocampus and sub-
sequent spread to the neighboring parahippocampal gyrus
and minimally to the amygdala but not to the temporal
neocortex. At the onset of this discharge the patient felt
nothing, but soon experienced her habitual aura of seeing
or thinking of the two comic strip characters. This experi-
ence was also elicited by left amygdaloid stimulation with-
out afterdischarge.
Patient 3
Complex auditory and visual hallucinations or memory recall
evoked by limbic (particularly amygdaloid) stimulation
A man aged 32 years had seizures starting with a rising
epigastric sensation followed by loss of contact and com-
plex automatisms, sometimes leading to a generalized con-
vulsion. Depth EEGs revealed that his seizures began in the
right temporal lobe and predominantly involved its limbic
components.
Electrical stimulation of the right amydala at 1 mA
caused the patient to say: “It feels like a dream.” When
asked to describe it he said: “Some people talking in my
back. I can hear their voices, but I don’t understand what
they say.” There were two voices, both male. He later
thought this was more like a memory of a previous experi-
ence than a dream. No afterdischarge occurred.
Stimulation of the left hippocampus at 1 mA made him
say: “What’s going on here? I have a feeling of people
coming back. It is this Indian guy who was here yesterday.”
When asked whether h e could see him, h e said it was a
feeling, but he was able to describe the Indian’s clothing:
“ H e is wearing a dark suit.” Afterdischarge was noted in
the left hippocampus, amygdala, parahippocampal gyrus,
and adjacent deep temporal neocortex.
Upon stimulation at 1.5 mA in the left amygdala the pa-
tient reported another “dream”: “It feels like someone
talking to a child.” T h e voice was female (he later thought it
could have been his wife). He could not understand what
the voice said. He thought this experience was in his mem-
ory. No afterdischarge occurred. Stimulation at the same
intensity 5 mm more laterally in the left amygdala pro-
duced another ”dream”: “I see a guy in a T V commercial, I
believe it is a Coca-Cola commercial. I think the guy is Sid-
ney Poitier [an entertainer]. H e is wearing a brown suit.”
There was no afterdischarge.
Patient 4
lllusion of familiarity and prescience, emotional distresJ evoked
by seizure discharge originating in limbic structures and by
limbic stimulation
A 32-year-old man exhibited two types of seizures. In the
first, which originated in the left temporal lobe, there was
no warning, loss of awareness proceeded to automatism
No clinical signs, patient feels nothing
F i g 2. (Patient 2) Spontaneous mirror seizure daring which
the patient sees OY i s forced t o think of two comic strip charac-
ters. The seizure discharge starts in the lefi hippocampus
(LB 1) andspreads to the lefi parahippocampalgyrus ILC 1)
and minimally to the left amygdala ( L A I ) . (Abbreviations
same as in Figure 1.)
followed by postictal aphasia. T h e second, more frequent
type originated in the right temporal lobe. T h e patient had
a sudden unpleasant churning sensation in his stomach, a
pounding headache, and a sudden feeling of "deja vu" as-
sociated with the illusion that he could predict what would
happen next. He became anguished, looked depressed, and
often remained depressed after the seizures. O n e such sei-
Seeing.or thinking of "Philornene and Jacquot"
zure was recorded with the 16-channel cable telemetry
system and on audio-video tape (Fig 3). The patient pressed
the "seizure button" and a nurse came in with a flashlight.
She asked him whether [he attack was like the preceding
ones, to which he replied: "Yes, and you are coming in right
now, it is all part of it; I mean as soon as I pressed it [i.e.,
the seizure button], I almost did not, because I knew you
were going to come in. It is as if I were reliving all rhat is
happening now. I just knew that you were going to come in
with the flashlight on. In a minute there will be more
people as if it all happened before, just reliving all this. T h e
more I know what is going to happen the farther it goes and
the dizzier I feel. . . it makes me feel strange o r weird." H e
then suddenly stopped speaking. Figure 3 shows that the
seizure discharge started in the right hippocampus, then
Gloor et al: Limbic Experiential Phenomena 133
spread to involve the right amygdala and the right parahip-
pocampal gyrus with only modest spread to the right tem-
poral neocortex. It remained confined to the right temporal
lobe while the patient described his experiential illusions to
the nurse. As he stopped speaking, the seizure discharge in
the right temporal lobe ceased and was immediately re-
placed by widespread ictal discharge in the left temporal
lobe which later became bilateral.
The illusions of familiarity and prescience together with
the associated emotional distress, stomach sensation, and
headache were reproduced with stimulation of the right-
sided limbic structures but not with neocortical o r left-
sided brain stimulation. The stimulations evoking these
experiential illusions were followed by afterdischarges in-
volving the right-sided limbic structures and temporal
neocortex.
Patient 5
Complex visual hallucinations with memory recall and fear,
combined auditory-visual hallucination, illusion of familiarity,
feeling of bring unwelcome, guilt, and olfactory hallucination
evoked by limbic stimulations
A 22-year-old man had minor seizures in which he experi-
enced a dreamlike state (his own words) with “deja vu” as-
sociated with a feeling of “suffering great embarrassment”
and of being forced “into another time and place” to the
point of “losing awareness of everyone and everything.”
Amnesia and ambulatory automatism sometimes accom-
panied these attacks. He also had seizures without an aura
that involved postural movements of the extremities fol-
lowed by a tonic-clonic convulsion. This patient was re-
134 Annals of Neurology Vol 12 No 2 August 1982
F i g 3. (Patient 4)Spontaneous seizure associated with illusion
of familiarity and of prescience, starting in the right hippo-
campus (RB 1) and involving simultaneously the right amyg-
dala (RA 1) and the right parahippocampal gyrus (RC
11. The seizure discharge later spreadJ t o involve the right tem-
poral neocortex but still predominates in the limbic structures.
During this discharge the patient i s able t o describe his expe-
riential illusion t o the nurse lfor details see text). The patient
stops speaking at the time when, as shown in the lower part of
the figure, the seizure discharge on the right comes t o an end
and i s immediately followed by widespread seizure discharge in
the left temporal lobe initiated by profound voltage depression.
The discharge ultimately becomes bilateral. (LFS, RFS = left
and right superior frontal electrodes [contact 1 most medial];
other abbreviations same as in Figure 1 .)
markable because an astonishingly rich variety of expe-
riential responses was elicited with stimulations in limbic
areas in either one of the temporal lobes. No such re-
sponses were evoked by temporal neocortical stimulations.
When the right amygdala was stimulated with a 1 m A
current, he experienced something that he found difficult
to describe but finally likened to a feeling of falling into
water. No afterdischarge occurred. Stimulation was re-
peated without warning, causing the patient to say “Now!”
and quite excitedly: “Could you do it again?” This short
stimulation was not followed by afterdischarge. Stimulation
was repeated without warning. The patient immediately
opened his mouth with an astonished look on his face, sat
up, and said that now he knew what is was: it was the feel-
ing of being at a picnic in Brewer Park in Ottawa. “A kid
 \
RA2-RA3-
RA3-RA4 -------
RA4-RA5-
RA5-RA6-
RA7-RA8 W W ~
LAl-LA2- J V - - ~
LA2-LA3-
LA3-LA4
LA4-LA5--
- “Yes” -
- - .
-
unresponsive. Dale, looks frightened -Experiential
c -
Stim 2MA RAl-RA2
LA~-LA~W~W-
LA8-LA9 --
F i g 4. (Patient 5 ) Right amygdaloid stimulation (MI-RA2)
at 2 mA eliciting a complex experiential response (for descrip-
tion see text). Widespread afterdischarge involves limbic and
neocortical structures of the right temporal lobe. (Abbreviations
same as i n Figure 1 .)
was coming up to m e to push m e into the water. It was a
certain time, a special day during the summer holidays and
the boy was going to push me into the water. 1 was pushed
down by somebody stronger than me. I have experienced
that same feeling when I had ‘petit mals’ before.” Again no
afterdischarge was noted. When questioned, h e said that
this had been a true event in his life which occurred when
he was about 8 years old, probably shortly before his sei-
zures started. A “big fellow” had pushed his head under the
water at that time. When questioned whether he actually
saw himself being threatened by the “big fellow” h e said
no, but it was a feeling as if he were there and was being
chased. Other unpleasant o r fearful events in his life had
also become part of his “petit mals.”
Later in the day the right amygdala was stimulated again
at a higher intensity of 2 mA. An afterdischarge was elicited
and spread widely within the temporal lobe, as shown in
Figure 4. Immediately upon stimulation the patient said
“yeah,” and repeated this, but did not reply to questioning.
His face turned pale and he looked frightened o r sad. T h e
physician observing him grasped him by the arm and
clapped his hands. T h e patient was startled and made a
frightened exclamation, but still did not reply. A few sec-
onds later h e made a motion over his stomach with his left
hand and said: “It was one of those feelings, a feeling of
being someplace very far away, definitely noon. It is an at-
mosphere I often experience during my ‘petit mal’ attacks.
I t recalls to mind the day in the country with Tracy [a girl
who lived next door to him] and brother Jamie. It was very
spooky, but it was so far away. It was out by the sea and
high up o n a cliff, a feeling as if I were going to fall. It was a
1’
11
c_x_
--
------.-
scary feeling. W e are there, a world within that world, all of
us were there. It is so real, yet so artificial.” He was
nauseated and looked upset. H e then told us that Tracy’s
parents often took him to this place when they were in the
country. H e first denied but later claimed that this was a
real memory, although he could not identify the place to
which they had gone. When questioned whether he had
been aware that, just after stimulation, he was spoken to,
grabbed by the arm, and startled by a hand clap, he replied
in the affirmative, but when asked why he had not an-
swered he said: “Because I was there.” He expressed his
amazement that “this machine” could reactivate such expe-
riences.
Later during the day the right amygdala was again stimu-
lated with a still higher intensity of 3 mA. He again became
nauseated, felt that he was somewhere in the country with
Tracy at a place where he had been before, and felt that it
was dark and raining. He was extremely frightened and
pale, and pleaded not to repeat the stimulation. Afterdis-
charge was present in the limbic structures and the
neocortex of the right temporal lobe.
When the left amygdala was stimulated at 1 mA, the
patient after some hesitation said he had a feeling of dis-
comfort, “as if I were not belonging here.” He likened it to
being at a party and not feeling welcome. There was no
afterdischarge.
Stimulation of the left amygdala at 2 mA produced a
vague sensation, “something I experienced before,” but no
afterdischarge. Stimulation 5 mm more laterally, at the lat-
eral edge of the left amygdala, caused the patient to exclaim
with an exasperated voice: “Now what is it?” He smelled
something and, motioning with his hands as if searching for
the right expression, he added “Everything . . . it is some-
thing experienced definitely before.” No afterdischarge oc-
curred. Left amygdaloid stimulation repeated with 3 mA
produced an olfactory sensation, “something to d o with
animals,” which the patient claimed he had experienced
before.
Gloor et al: Limbic Experiential Phenomena 135
 Exasperatedvoice,
olfactory hallucination,
deja vu (no AD)
No clinical
No clinical
responses(AD)
response
(no AD)
Stimulation of the right hippocampus at 1 and 2 mA pro-
duced no response, but at 3 mA the patient looked hesitant
and said “it was fear and anxiety, like you are demanding to
hand in a report that was due 2 weeks ago . . . as if I were
guilty of some form of tardiness.” Afterdischarge was noted
in the region of the hippocampus and deep temporal
neocortex. Stimulation in the left hippocampus produced
no definite responses.
Stimulation of the right parahippocampal gyrus at 1 mA
elicited a “deji vu” (the patient’s own words) and an after-
discharge in the right parahippocampal gyrus and hip-
pocampus. Stimulation applied later to the same area with 2
mA caused the patient to say: “Yep, yep, yep, I am balanc-
ing on the edge of a fountain. I have often experienced this
in ‘petit mals.’ It is like I am in an old storybook. I am afraid
to fall into the fountain.” He smiled. When asked why h e
did so while claiming to b e afraid, h e said “because I have
experienced this so often.” Afterdischarge occurred in the
right parahippocampal gyrus and right hippocampus with
some spread to the right amygdala. Stimulation repeated
without warning elicited an identical experiential response.
Several hours later the right parahippocampal gyrus was
stimulated again, this time with 3 mA. T h e patient smiled
and reported that he saw and heard a musical instrument, “a
guitar.” His initial smile turned into an expression of un-
happiness with facial pallor. Afterdischarge was present in
the right parahippocampal gyrus and hippocampus with
some spread to the right amygdala and right temporal
neocortex.
In this patient the responses combined different expe-
riential qualities: complex visual experiences, memory
recall, “d6ja vu,” fear, guilt, olfactory hallucinations, and at
one time a combined auditory-visual hallucination. Even
though the details of the experiential contents varied from
stimulation to stimulation, the experiences evoked by re-
peated stimulations at the same site in the right amygdala
had a common feature. Indeed, with all three right amyg-
daloid stimulations, delivered at 1, 2, and 3 mA, respec-
136 Annals of Neurology Vol 12 No 2 August 1982
Complex visual hallucination,
I
memoryrecall, fear (AD)
Deja vu (no AD)
Feelingreminiscent of earlier stimulation-
evoked experiential response (no AD)
II “The very beginningof it”(no AD)
No clinical
response (AD)
No clinical responses (no AD)
F i g 5 . (Patient 5 ) Topographical distribution of responses ob-
tained with electrical stimulations upplied to adjacent pairs of
contacts along the LA and RL4rowJ of electrodes (see Fig 1).
The deepest contacts of these electrode arrays are located i n the
left and right amygdala; the more external contacts sample ac-
tivity recorded from temporal white matter and neocortex.
Height ofthe vertical lines roughly indicates intensity of re-
sponses elicited by electrical stimulation. ( A D = afterdis-
charge; LT = left; R T = right.j
tively, the diverse experiential phenomena had one com-
mon denominator: fear related to water, such as fear of
being pushed into the water (at 1 mA), fear of falling from a
cliff at the seaside (at 2 mA), and fear of being in the dark
while it is raining (at 3 mA). In addition, stimulation of the
parahippocampal gyrus on the same side with 2 mA elicited
an experiential response in which fear of falling into water
from the edge of a fountain was again a prominent feature.
This common element of fear related to water with other-
wise diverse experiential responses elicited by right-sided
limbic stimulations was totally absent from the experiential
responses elicited from left-sided limbic structures. T h e
time sequence in which these stimulations were applied
cannot explain this experiential feature common to several
right-sided limbic responses, since they were given hours
apart and were separated from each other by stimulations
in which other experiential phenomena had been elicited.
There was thus some kind of topographical specificity to
this feature of fear of water.
All the experiential responses elicited in this patient de-
pended upon activation of limbic and not neocortical
structures, as diagrammed in Figure 5 , which summarizes the
responses obtained with a 2 mA stimulation intensity along
the left and right anterior temporal depth electrode arrays.
T h e height of the vertical lines is roughly proportional to
the degree of intensity of the subjective experience re-
ported by the patient when stimulations were applied
through adjacent pairs of electrodes. A gradient of inten-
sity peaking in the limbic structures is evident on the right
side. Thus, stimulation of the right amygdala induced the
vivid experience of being out in the country by the sea.
Stimulation at the same intensity 5 mm more IateralIy at the
lateral edge of the amygdala evoked a similar but milder
response. Stimulation another 5 mm more laterally caused
the patient to say “the very beginning of it,” and more lat-
erally still there were no clinical responses. O n the left side
too, experiential responses were only obtained with
stimulations applied to limbic structures. No correlation
was found between elicitation of experiential responses and
that of afterdischarges (labeled A D in Figure 5): with
stimulations applied to limbic structures, experiential re-
sponses were obtained with o r without afterdischarges,
while neocortical stimulations, even when followed by af-
terdischarges, never evoked any clinical responses.
Characteristics of Experiential Phenomena
Even though the experiential qualities of the
responses obtained in these patients were often
quite overwhelming to them, difficulties arise in
categorizing the experiences. Most of the time the
patients described what they felt in visual terms.
Nevertheless, it was often difficult to be certain
whether they actually saw what they were describing
o r were only describing an inner experience that re-
minded them of something with a predominantly vis-
ual experiential content, whether a memory was
reactivated in a particularly vivid way, or whether a
thought forced itself upon their mind which they
chose to describe in terms suggesting that it was a
visual experience (e.g., Patient 2). An instructive
example was an experiential response obtained in a
36-year-old woman (A. S.) suffering from temporal
lobe seizures with automatism, who, upon stimula-
tion in the left hippocampus, pointed with her right
hand in front of her and said: “I was going to this
place, it is like a meeting area. The gentleman who
keeps all the tickets should be there.” Her behavior
and description of the experience suggested that this
was a complex visual hallucination. However, an
element of memory was involved, although in fact
she had never been to this place. For years she had
wanted to go to a hockey game. She had recently ob-
tained tickets for a game through a family member
who knew “a gentleman” working at the hockey
arena, but her hospitalization prevented her from
fulfilling this wish. It seems that this memory and the
ultimately frustrated desire to attend a hockey game
somehow entered into the elaboration of this expe-
riential response elicited by hippocampal stimulation.
In this case the visual perceptual content of the ex-
perience was linked to a not very intense affect, the
frustrated desire to attend a hockey game. Such an
association of affect with an experiential phenome-
non related to perception and memory was common:
few of the observed perceptual or mnemonic phe-
nomena seemed to be devoid of it. On the other
hand, one may wonder whether an emotional re-
sponse such as fear or emotional distress may some-
times merely represent an interpretation of an un-
pleasant visceral sensation. Of the more than 49
instances of fear, 27 were associated with a visceral
sensation, usually referred to the abdomen and less
commonly to the chest.
The experiential phenomena of temporal lobe
epilepsy, especially those with a perceptual content,
are often described (even sometimes by the patients
themselves) as dreamlike. Yet there are important
differences between them and most dreams. An
evolving “story” never seems to be part of experien-
tial phenomena evoked by temporal lobe discharge.
They are fragmentary and often have the quality of a
vivid “flashback” to a specific event in the past with-
out the full sequential reenactment of the event.
They could not be likened to the replay of a vid-
eotape. Rather, “flashbacks” with a perceptual con-
tent can be likened to sudden “snapshots” or, at the
most, brief snippets of a film strip, and the same
applies to experiential phenomena for which no evi-
dence of evocation of a memory can be obtained.
Anatomical Substrate of Experiential Phenomena
With regard to the anatomical substrate of experien-
tial phenomena, our results differ from those re-
ported by Penfield and his associates [34,41]. Table 2
lists the location of the spontaneous seizure dis-
charges or the afterdischarges elicited by electrical
stimulation associated with experiential illusions and
hallucinations in our 35 patients. In 37 instances the
discharges were confined to limbic structures, some-
times only to one o r two of them. In just 2 instances
Table 2. Location of Seizure or Afterdischarges
Associated with Experiential Phenomena
No. of
Location Observations
Amygdala 5
Hippocampus 5
Parahippocampal gymsa 1
Temporal neocortex 2
Amygdala + hippocampus 15
Amygdala + hippocampus + 1l h
parahippocampal gyrus
Amygdala + hippocampus + >49‘
parahippocampal gyrus +
temporal neocortex
“Not explored in 4 patients.
bIncludes 2 with further spread to neocortex after experiential re-
sponse.
+
‘Includes 1 amygdala neocortex, 1 hippocampus + neocortex,
with no other limbic involvement.
Gloor et al: Limbic Experiential Phenomena 137
 Without Afterdischarge 26 elementary visual hallucinations (phosphenes) most
Amygdala 44
/ /,Limbic 10 likely caused by stimulation of optic radiation fibers.
‘With Afterdischarge 18
‘Limbic and 8
With 2 of these phosphenes some fear was experi-
Neocortical enced, probably caused by spread of current to adja-
,Without Atterdischarge 6 cent limbic structures since upon their stimulation
Hippocampus 26 /,Limbic 7 stronger fear was evoked. Among the limbic struc-
‘With Afterdischarge 20
tures, the amygdala was the one from which the
‘Limbic and 13
Neocortical largest number of experiential responses, 44, was
,Without Afterdischarge 5* obtained, and in more than half of these, 26, there
Parahippacampal
12 ,Limbic 2 was no afterdischarge. Hippocampal stimulations
Gyrus
\With Afterdischarge 7 produced experiential responses less often, 26 times,
“Limbic and 5
Neocortical
and only 6 times without afterdischarge. Stimulations
,Without Afterdischarge 6 of the parahippocampal gyrus produced subjective
Temporal Neocortex
and WhiteMatler ‘* responses 12 times. Of these responses, 5 were not
\With Afterdischarge 0 associated with afterdischarge, but they were all
* Elementary visual hallucinations (except lor I instance in each group) elementary visual hallucinations (phosphenes) un-
doubtedly caused by spread of current to the adja-
F i g 6. Temporal lobe structures from uibicb experientiai re- cent optic radiation. In the 45 instances in which
sponses were elicited by electrical stimulaiion with and without limbic stimulations produced afterdischarges, 19
afterdiscbarge. remained confined to limbic structures.
These data show that in all instances but one, expe-
riential responses were produced by limbic and not by
was the temporal neocortex alone involved, and not neocortical or temporal white matter stimulation. The
on the lateral surface but in the depth. In more than likelihood with which such responses could be elic-
49 instances, the limbic structures as well as the tem- ited had an anteroposterior gradient, the amygdala
poral neocortex participated in t h e discharge. Thus, being the site from which they were most commonly
experiential phenomena typically were associated evoked. This was particularly evident for those re-
with seizure discharges or afterdischarges that in- sponses elicited without afterdischarge. The likeli-
volved either the limbic structures alone or limbic hood of evoking experiential responses further de-
and neocortical structures. creased as one moved from the hippocampus to the
The sites of electrical stimulation that produced parahippocampal gyrus. The data summarized in Fig-
experiential responses, either with or without detect- ure 6 clearly show that experiential responses could
able afterdischarge (Fig 6), far more frequently lay in occur in the absence of afterdischarge and that spread
the limbic system than in the temporal neocortex or of afterdischarge from limbic structures to the tem-
white matter. In fact, the 6 responses to temporal poral neocortex was not required for their evocation.
neocortical or white matter stimulation listed in Fig- Even complex visual hallucinations, which one would
ure 6 probably comprise only one true experiential tend to attribute to activation of temporal neocortex,
response, a “deja vu” illusion, the other 5 being appeared to depend upon activation of limbic struc-

Table 3. Complex Visual Hallucinations (5 Patients)

Limbic and
Temporal Temporal
Neocortex Neocortex
Parahippo-
Means of Amyg- Hippo- campal Limbic Deep and Deep and
Elicitation dala campus Gyrus Diffuse Deep Superficial Deep Superficial
Initial seizure 0 3 0 1 0 0 6 3
or after-
discharge
Secondary spread 1 0 1 0 0 2 0 0
Stimulation site
Without 4 0 0 0 0 0 0 0
after dis ch arge
With after- 4 5 2 0 0 0 0 0
discharge

138 Annals of Neurology Vol 12 No 2 August 1982

Table 4. Fear (7 Patients)
Parahippo-
Means of Amyg- Hippo- campal
Elicitation dala campus Gyrus
Initial seizure 3 5 1 11
or after-
discharge
Secondary spread 0 0 0 0
Stimulation site
Without 14 4 0
afterdischarge
With after- 8 9 5 0
discharge
tures (Table 3). Complex visual hallucinations could
occur with limbic discharges alone. Furthermore,
there were no instances in which temporal neocorti-
cal discharge without limbic participation was asso-
ciated with a complex visual hallucination. Finally, all
complex visual hallucinations produced by electrical
stimulation were elicted by stimulating limbic and
not neocortical structures. Amygdaloid stimulations
in particular often elicited complex visual hallucina-
tions without afterdischarge. The experience of fear
showed a similar close relationship to activation of
limbic structures of the temporal lobe (Table 4 ) , as
was expected.
Unlike Penfield and Perot [41], we did not find
that experiential responses were more likely to be
elicited from right rather than left temporal stimula-
tion. The former evoked experiential responses 45
times, the latter 48 times. This result is surprising
since right-sided seizures featuring experiential
phenomena were more common than left-sided ones
(26 versus 9). It reflects the fact that, of the 18 pa-
tients featuring experiential phenomena, 12 had a
predominantly right-sided seizure disorder. Fur-
thermore, intracarotid Amytal tests indicated that
among the 6 patients with a predominantly left-sided
seizure disorder, the right hemisphere was dominant
for speech in one, and in another speech was rep-
resented on both sides.
Discussion
Anatomical Correlations and Functional Anatomy
Our observations on 35 epileptic patients studied
with stereotaxic exploration of the temporal lobe
provide prima facie evidence that unless limbic
structures are activated, either in the course of a
spontaneous seizure or through artificial electrical
stimulation, experiential phenomena d o not occur.
This was true not only for emotional responses, in
Limbic and
Temporal Temporal
Neocortex Neocortex
Limbic Deep and Deep and
Diffuse Deep Superficial Deep Superficial
0 0 1 <7
0 2 0 0
0 2 0 0 0
0 0 0 0
which one would expect this to be the case, but also
for complex visual and auditory hallucinations, acti-
vation of memory recall, and illusions of familiarity.
Wieser [46] also noted that “psychic” phenomena
tend to occur when seizure discharges involve limbic
structures. However, in his experience complex hal-
lucinations were regularly associated with discharge
involving large parts of the temporal lobe, a finding at
variance with our own observations.
Penfield and co-workers [34, 36, 40, 411, in the
light of their findings, attributed experiential phe-
nomena to neocortical mechanisms of the tempo-
ral lobe, with the possible exception of fear. This is
what would be expected on theoretical grounds.
Present knowledge about the functions of the pri-
mate temporal neocortex suggests that complex vis-
ual and auditory hallucinations, and perhaps memory
recall and the illusion of familiarity as well, should be
elicitable from temporal neocortex rather than from
limbic structures, since temporal neocortex is in-
volved in higher auditory and visual perception and
memory and probably also in the matching of current
with past experience involving these modalities [ 5 , 6 ,
28-33,471. Penfield and Perot’s maps [41]show that
in the human temporal neocortex, the areas from
which complex auditory responses could be elicited
on electrical stimulation are anatomically segregated
from those from which complex visual responses
could be evoked. These results are in agreement with
the known representation of higher visual and audi-
tory perceptual functions in the subhuman primate
temporal lobe [ 5 , 6 , 29, 32, 33, 471 and therefore
appear to reflect the anatomical organization of
higher auditory and visual perceptual functions in the
human brain.
Thus, our findings that all experiential responses,
including perceptual ones, were elicited from limbic
and not from temporal lobe stimulation, and that
Gloor et al: Limbic Experiential Phenomena 139
likewise their appearance in spontaneous seizures
was linked to activation of limbic rather than neocor-
tical structures, were highly unexpected. The inci-
dence of experiential responses in patients with proved
temporal lobe epilepsy in our study was, however,
62%, considerably higher than in Penfield’s se-
ries, for which an incidence of approximately 12%
can be estimated by combining the figures given by
Mullan and Penfield [34]and by Penfield and Perot
[411. The discrepancy between our findings and
those of Penfield cannot be attributed to selection
error. We had no bias in favor of limbic as opposed to
neocortical stimulations or recordings in our study.
In Penfield’s studies, however, neocortical stimula-
tions were much more common than limbic ones.
Admittedly, with our approach only a relatively nar-
row sector of neocortex was accessible to stimulation
and recording: the middle and anterior portions of the
second temporal convolution and cortex lying subja-
cent to it within sulci of the temporal lobe (see Fig 5).
This limitation, however, cannot be responsible for
our results. In a recent series of 195 patients with
temporal lobe epilepsy who, with 1 exception, had
not been explored preoperatively with depth elec-
trodes and in whom both cortical and limbic struc-
tures of the temporal lobe were stimulated under
local anesthesia during craniotomy using the method
employed by Penfield, we again found that experien-
tial responses were more commonly evoked by lim-
bic than by neocortical stimulations. Surface neocor-
tical stimulations elicited experiential responses in
only 1 patient, in whom complex auditory hallucina-
tions of hearing music were elicited by stimulating
the first temporal convolution. Deep neocortical
stimulations elicited an elementary auditory halluci-
nation in one patient and fear in another. In contrast,
limbic stimulations produced combined complex vis-
ual and auditory hallucinations, an elementary visual
hallucination, a visual illusion, and an elementary au-
ditory hallucination, each in a different patient; ol-
factory hallucinations and illusion. of familiarity in 2
patients, respectively; fear in 7 patients; and a pleas-
ant emotion in 1 patient. We conclude that experien-
tial phenomena, including perceptual ones, are more
likely to occur in response to limbic than to temporal
neocortical stimulation or seizure discharge, in spite
of the fact that this is unexpected on theoretical
grounds.
O n e could argue, as Halgren et a1 [18] do, that
limbic stimulations or limbic seizure discharges may
produce experiential phenomena because of some
spread of excitation to neocortex, and that the reason
the experiential phenomena were so easily elicited in
our study, by either epileptic discharge or electrical
stimulation in limbic structures, was that these areas
are highly excitable. This explanation is not sup-
ported by our data on the occurrence of afterdis-
140 Annals of Neurology Vol 12 No 2 August 1982
charges, which show that experiential responses can
occur even when electrical stimulation fails to elicit a
detectable afterdischarge or when afterdischarges
remain strictly confined to limbic structures. Of all
the experiential responses obtained by limbic stimu-
lation in our series (excluding the elementary visual
hallucinations listed in Fig 6), 34 (43%) were not as-
sociated with a detectable afterdischarge. The corre-
sponding figure in the study by Halgren et a1 [IS] is
only 7.995, possibly because their stimulation inten-
sities were higher than ours. The lack of afterdis-
charges with stimulations eliciting experiential re-
sponses was particularly striking for the amygdala:
more than half, i.e., 26 (59%) of the experiential re-
sponses evoked from the amygdala were not fol-
lowed by afterdischarge. Also, in 10 (55.5%) of the
responses to amygdaloid stimulation associated with
afterdischarges, no spread occurred from limbic
structures to neocortex. We cannot rule out the pos-
sibility that some kind of activation of neocortical
neurons, which could not be identified by gross
recording methods, may have taken place. Against
this possibility is the fact that no experiential
phenomena were observed in the large number of in-
stances in which electrical stimulation of temporal
neocortex produced afterdischarges.
Anatomical studies [27, 441 have shown that the
flow of information from the temporal neocortical
areas subserving higher visual and auditory percep-
tual functions is mainly directed toward the limbic
structures of the temporal lobe. Accordingly, it is not
surprising to find that stimulation at the “limbic re-
ceiving end” of these connections can elicit responses
that reflect the function of the areas from which these
connections originate. The observation that such re-
sponses can be more easily elicited by stimulating at
the limbic than at the neocortical “end” of this system
suggests that limbic activation may be essential for
bringing to a conscious level percepts elaborated by
the temporal neocortex. One may conjecture that
whatever we experience with our senses, particularly
in the visual and auditory modalities, even after it has
been elaborated as a percept in temporal neocortex,
must ultimately be transmitted to limbic structures in
order to assume experiential immediacy. Attaching
some affective or motivational significance to a per-
cept may be the specific limbic contribution to this
process. This may be the precondition for the per-
cept to be consciously experienced or recalled and
may imply that all consciously perceived events must
assume some kind of affective dimension, if only ever
so slight. The fullblown emotional responses that
sometimes occur with massive limbic activation dur-
ing a seizure discharge may represent an extreme
form of activation of limbic affective mechanisms that
are not commonly involved at this intensity in every-
day living. This concept is not entirely new. Jasper
and Rasmussen [26], discussing their observations on
deep temporal lobe stimulations in human beings,
speculated that “it may require some ‘body’ and ‘vis-
cera’ to make experience memorable.” The view pro-
posed here reemphasizes the close anatomical and
functional relationship that exists between temporal
neocortex and the limbic system [13, 15, 27, 441
without invalidating the well-supported evidence that
each component has its own contribution to make to
the perceptual, mnemonic, and affective mechanisms
involved in subjective experience.
The much smaller number of auditory than visual
experiential phenomena observed in our series may
reflect the anatomical [27] and functional [7, 121 evi-
dence that the visual modality has privileged access to
the primate amygdala.
Nature of Evoked Experiential Phenomena
Experiential responses evoked from the human tem-
poral lobe are experienced subjectively and com-
municated verbally by the patient. They are therefore
not amenable to the same objective analysis as, for
example, a finger movement elicited by motor cortex
stimulation. Since the reports given by the patients
often astound us by their variety and apparent fanci-
ful content, one may ask to what extent they repre-
sent true cerebral responses, comparable, for ex-
ample, to phosphenes elicited by stimulation of the
primary visual cortex. The fact that experiential re-
sponses can be evoked by stimulation without warn-
ing and not by warning without stimulation as well as
their dependence on a specific anatomical substrate
should remove all doubts concerning their true cere-
bral origin. O n e may nevertheless still wonder
whether the patient’s verbal report represents a
sufficiently accurate description of what he or she
actually experiences, or whether a rather vague sen-
sation is being elaborated on. This is difficult to
answer. If patients commonly embroidered their ex-
periences, we would expect them more often to de-
scribe them in the form of a “story” rather than as a
vivid fragment devoid of an evolving “story line.”
Even though the experiential content varies from pa-
tient to patient, the phenomena are described in very
similar terms, as is evident from comparing the de-
scriptions of responses in individual patients re-
ported by Penfield and Perot [41] with each other
and with our own. We believe there is little embroi-
dering when patients describe these phenomena.
The experiential phenomena evoked by temporal
lobe discharge or stimulation differ in one impor-
tant respect from those evoked from primary motor
or sensory areas of cortex. As Halgren et a1 [18]
have rightly pointed out, the responses are “indi-
vidualized.” Halgren and associates [ 181 proposed
that the types of mental phenomena elicited by lim-
bic stimulation are related more to the particular per-
son who is being stimulated than to where in the lim-
bic system stimulation is applied. They tentatively
concluded that stimulation (and, by inference, spon-
taneous epileptic discharge) in the human limbic
system cannot activate a genuine brain mechanism
represented there, but that inactivation of such a
mechanism by electrical stimulation or seizure dis-
charge, possibly through interference with an inhib-
itory influence normally exerted by the hippocampus
on other parts of the brain, allows suppressed emo-
tional tendencies or neurotic concerns to surface into
consciousness. A similar view was also put forward by
Ferguson et a1 [ 111. According to this hypothesis, the
responses to limbic stimulation are produced by a
“paralyzing” interference with brain function similar
to the way aphasia can be produced by electrical
stimulation or a seizure discharge involving speech
cortex. We are unconvinced of this.
The fact that experiential responses are “indi-
vidualized” presents no difficulty in interpreting
them as positive cerebral responses. If the principal
function of the temporal cortex and limbic system is
to encode an individual’s personal life experience and
to relate it to associated motivational and affective
states [13, 15, 441, then one would predict that, un-
like responses elicited from the “hard wired” primary
motor and sensory areas, those obtained from neocor-
tical or limbic temporal lobe stimulation should
reflect individual past experiences. Penfield [37, 381
termed the temporal cortex “uncommitted” because
at birth it is ready to accept the imprint of an indi-
vidual’s life history. The same applies to limbic
structures. They were the first parts of the brain in
vertebrate evolution which, even in forms as low as
fishes [2, 431, made it possible to relate an indi-
vidual’s past experience to motivational mechanisms
and to encode this association in the brain [131. The
fact that stimulating “whom” seems more important
than stimulating “where” in the limbic system be-
comes understandable, because the responses reflect
at the same time the functional role of the stimulated
area and the patient’s past individual experience. In
the sense that the latter is likely to be an important
ingredient in the makeup of personality, we would
agree with Halgren and associates [18] and Ferguson
et a1 [ l l ] that some relationship may exist between
the type of experiential responses elicited by tem-
poral lobe stimulation or seizure discharge and a
patient’s personality. However, one must also con-
sider the possibility that temporal limbic dysfunction
caused by epileptogenic pathology may be responsi-
ble both for the patient’s personality structure and for
the experiential responses evoked by stimulation or
seizure discharges. Bear and Fedio’s [3, 41 data in
humans and Adamec’s [ 1) experimental observations
on cats with kindled seizures support this viewpoint.
Additional arguments make it unlikely that ex-
Gloor et al: Limbic Experiential Phenomena 141
periential phenomena result from epileptic or
stimulation-induced inactivation of limbic structures.
It is well known that ictal inactivation of limbic
structures produces automatism, confusion, and am-
nesia [9, 10, 171, which continue without any detect-
able break into the period of postictal depression.
Experiential phenomena d o not extend into the
postictal period, except sometimes those of an emo-
tional nature. The experiential phenomena reported
by rhe patients are therefore more likely to represent
positive than negative effects of temporal lobe
stimulation or seizure discharge.
In addition, the responses to limbic stimulation
conform closely to what one would predict should
result from activation of the perceptual, mnemonic,
and affective mechanisms represented in the tem-
poral lobe. This observation is particularly true for
Penfield and Perot’s [41]stimulation maps and less so
for our and Halgren et al’s [18]findings, which lack
fine topographical differentiation. It remains difficult,
however, to envision how a crude pattern of stimula-
tion such as is imposed by an electric current or sei-
zure discharge could produce the differentiated re-
sponses represented by the experiential phenomena
reported in this study. Penfield [35, 36, 38, 411
speculated that experiential responses represented a
stimulation-induced reactivation of “the record of the
stream of each individual’s consciousness.” He as-
sumed that the vivid details which often characterize
such experiential responses are evidence that all the
details of a previous experience are recorded in the
brain and can be “replayed” in their original sequence
when epileptic discharge or electrical stimulation
reactivates certain circuits in the temporal cortex.
However, hallucinations of scenes never experienced
before, y e t at times nevertheless related to some re-
membered event (such as in our Patient A. S.) may
also be evoked by such stimulations or epileptic dis-
charge and can be as vivid and detailed as those of a
true memory flashback. It therefore seems unlikely
that electrical stimulation or seizure discharges in-
volving temporolimbic circuits replay a faithful record
of past events. The fragmentary character of these
experiential phenomena also argue against such an
interpretation.” Perhaps it is rather that temporal
epileptic discharge or electrical stimulation, by a pro-
“That temporal lobe stimulation may cause a patient to recall a
melody which he may be able to hum at a normal speed [35,411 is
not evidence enough to suggest that a faithful record of a segment
of the past stream of consciousness is replayed in its original se-
quence and tempo. Such a forward marching in time is peculiar to
flashbacks involving music and is lacking in visual experiential
phenomena or even in auditory ones described as hearing some-
one speak. Music and musical memory utterly depend upon the
brain’s ability to reconstruct the temporal sequencing of auditory
signals. Being reminded of a tune and then proceeding to hum it
does not necessarily imply that the brain faithfully replays a
specific old memory.
142 Annals of Neurology Vol 12 No 2 August 1982
cess as y e t unknown, activates neuronai networks
that vividly bring forth the elementary and ultimately
affective impact of a given experience, its existential
context and immediacy in the Proustian sense, as op-
posed to a detailed record of an unfolding story or
the recollection of some self-directed action or men-
tal effort which Penfield and Perot [41] found unac-
countably missing among their observations. Such
experiential “flashbacks” occur at times in the course
of ordinary living in normal persons. Odor stimuli are
particularly apt to activate vivid feelings of finding
oneself in an almost forgotten place from one’s re-
mote past [ 8 ] ,providing a further indication that lim-
bic circuits are involved in their evocation.
A final point deserves discussion. It may be argued
that, since experiential phenomena could only be
elicited in persons with temporal lobe epilepsy, the
anatomical channels mediating these responses may
differ between their brains and the brains of persons
without epilepsy. To accept this viewpoint, one
would have to postulate that recurrent epileptic dis-
charge (or some other mechanism peculiar to the
epileptic brain) would be capable of remodeling the
pattern of functional representation in the brain. No
evidence exists from studies on clinical or experi-
mental epilepsy that such a remodeling may take
place in the epileptic brain. The argument that the
functional organization of the temporal lobe may in
some way be altered in temporal lobe epilepsy also
runs counter to the evidence from studies in humans
and subhuman primates [6,7,13, 14,28-33,44,47],
which indicate that temporolimbic circuits are in-
volved in the same perceptual, mnemonic, and affec-
tive mechanisms that become activated in the course
of an experiential response elicited in a temporal lobe
epileptic. Finally, it is not strictly true that experien-
tial responses can be elicited only from the brains of
persons with temporal lobe epilepsy. While the
human limbic system has never been stimulated in a
normal individual, Sem-Jacobsen and Torkildsen [42]
as well as Ishibashi et a1 [19] reported obtaining
experiential responses similar to those described in
this paper by stimulating the temporal lobes of
schizophrenics. Nevertheless, the finding that such
responses are most readily evoked from temporal
lobes affected by epileptogenic pathology may be
explained by some “epileptic facilitation” of tem-
porolimbic circuits by presently unknown mecha-
nisms.
This paper is based on the text of the First Wilder Penfield Memo-
rial Lecture of the Western EEG Society, presented by P. Gloor
under the title “Psychic phenomena of temporal lobe epilepsy:
nature’s experiment on human experiential mechanisms” on Feb-
ruary 20, 1981, in Reno, Nevada.
References
1. Adamec RE: Normal and abnormal limbic system mechanisms
of emotive biasing. In Livingston KE, Hornykiewicz 0 (eds):
Limbic Mechanisms. The Continuing Evolution of the Limbic
System Concept. New York and London, Plenum, 1978, pp
405-455
2. Atema J, Todd JH, Bardach JE: Olfaction and behavioral
sophistication in fish. In Pfaffmann C (ed): Olfaction and Taste
111. New York, Rockefeller University Press, 1969, pp 241-
251
3. Bear DM: Temporal lobe epilepsy-a syndrome of sensory-
limbic hyperconnection. Cortex 15:357-384, 1979
4. Bear DM, Fedio P: Quantitative analysis of interictal behavior
in temporal lobe epilepsy. Arch Neurol 34:454-467, 1977
5. Chow E Anatomical and electrophysiological analysis of
temporal neocortex in relation to visual discrimination learn-
ing in monkeys. In Delafresnaye JF (ed): Brain Mechanisms
and Learning. Oxford, England, Blackwell, 1961, pp 449-469
6. Dewson JH 111, Cowey A: Discrimination of auditory se-
quences by monkeys. Nature 222:695-697, 1969
7. Downer JL d e C: Changes in visual gnostic functions and
emotional behavior following unilateral temporal pole damage
in the “split brain” monkey. Nature 191:50-51, 1961
8. Engen T, Ross BM: Long-term memory of odors with and
without verbal descriptions. J Exp Psychol 100:221-227,
1973
9. Feindel W: Response patterns elicited from the amygdala and
deep temporoinsular cortex. In Sheer D (ed): Electrical
Stimulation of the Brain: An Interdisciplinary Survey of
Neurobehavioral Integrative Systems. Austin, TX, University
of Texas Press, 1961, pp 519-532
10. Feindel W, Penfield W Localization of discharge in temporal
lobe automatism. Arch Neurol Psychiatry 72:605-630, 1954
11. Ferguson SM, Rayport M, Gardner E, Kass W, Weiner H,
Reiser MF: Similarities in mental content of psychotic states,
spontaneous seizures, dreams, and responses to electrical
brain stimulation in patients with temporal lobe epilepsy.
Psychosom Med 31:479-498, 1969
12. Geschwind N: Disconnection syndromes in animals and man.
Brain 88:237-294, 585-644, 1965
13. Gloor P: Temporal lobe epilepsy: its possible contribution to
the understanding of the functional significance of the amyg-
dala and of its interaction with neocortical-temporal mecha-
nisms. In Eleftheriou BE (ed): The Neurobiology of the
Amygdala. New York and London, Plenum, 1972, p p 423-
457
14. Gloor P: Contributions of electroencephalography and elec-
trocorticography to the neurosurgical treatment of the epilep-
sies. In Purpura DP, Penry JK, Walter R D (eds): Neurosurgi-
cal Management of the Epilepsies. Adv Neurol 8:59-105,
1975
15. Gloor P: Inputs and outputs of the amygdala: what the amyg-
dala is trying to tell the rest of the brain. In Livingston KE,
Hornykiewicz 0 (eds): Limbic Mechanisms. The Continuing
Evolution of the Limbic System Concept. New York and
London, Plenum, 1978, pp 189-209
16. Gloor P, Olivier A, Ives J: Prolonged seizure monitoring with
stereotaxically implanted depth electrodes in patients with
bilateral interictal temporal epileptic foci: how bilateral is
bitemporal epilepsy? In Penry JK, Wada J (eds): Advances in
Epileptology: 10th Epilepsy International Symposium. New
York, Raven, 1980, pp 83-88
17. Gloor P, Olivier A, Ives J: Loss of consciousness in temporal
lobe seizures: observations obtained with stereotaxic depth
electrode recordings and stimulations. In Canger R, Angeleri
F, Penry JK (eds): Advances in Epileptology: 11th Epilepsy
International Symposium. New York, Raven, 1980, pp 349-
353
18. Halgren E, Walter RD, Cherlow DG, Crandall PH: Mental
phenomena evoked by electrical stimulation of the human
hippocampal formation and amygdala. Brain 101:83- 117,
1978
19. Ishibashi T, Hori H, Endo K, Sato T: Hallucinations pro-
duced by electrical stimulation of the temporal lobes in
schizophrenic patients. Tohoku J Exp Med 82:124-139,
1964
20. Ives JR, Gloor P: A long term time-lapse video system to
document the patient’s spontaneous clinical seizure syn-
chronized with the EEG. Electroencephalogr Clin Neu-
rophysiol 45:412-416, 1978
21. Ives JR, Thompson CJ, Gloor P: Seizure monitoring: a new
tool in electroencephalography. Electroencephalogr Clin
Neurophysiol 41:422-427, 1976
22. Jackson JH: Notes on cases of epileptic seizures: intellectual
warnings of epileptic seizures. Medical Times and Gazette
2:700-702, 1876. Reprinted in Taylor J (ed): Selected Writ-
ings of John Hughlings Jackson on Epilepsy and Epileptiform
Convulsions. London, Hodder and Stoughton, 1931, pp
274-275
23. Jackson JH: Lectures on the diagnosis of epilepsy, Lecture 111.
Medical Times and Gazette 1:141-143, 1879. Reprinted in
Taylor J (ed): Selected Writings of John Hughlings Jackson on
Epilepsy and Epileptiform Convulsions. London, Hodder and
Stoughton, 1931, pp 295-307
24. Jackson JH: On right or left-sided spasm at the onset of
epileptic paroxysms, and on crude sensation warnings and
elaborate mental states. Brain 3: 192-306, 1880-8 1
25. Jackson J H , Stewart P: Epileptic attacks with a warning of a
crude sensation of smell and with the intellectual aura (dreamy
state) in a patient who had symptoms pointing to gross organic
disease of the right temporo-sphenoidal lobe. Brain 22:534-
549, 1899
26. Jasper H H , Rasmussen T: Studies of clinical and electrical re-
sponses to deep temporal stimulation in man with some con-
sideration of functional anatomy. Res Pub1 Assoc Res Nerv
Ment Dis 36:316-334, 1958
27. Jones EG, Powell TPS: An anatomical study of converging
sensory pathways within the cerebral cortex of the monkey.
Brain 93:793-820, 1970
28. Kimura D: Right temporal lobe damage. Arch Neurol
8:264-271, 1963
29. Massopoust LC Jr, Barnes HW, Verdura J: Auditory fre-
quency discrimination in cortically ablated monkeys. J Aud
Res 5:85-93, 1965
30. Milner B: Psychological defects produced by temporal lobe
excisions. Res Pub1 Assoc Res Nerv Ment Dis 36:244-257,
1958
3 1. Milner B: Visual recognition and recall after right temporal
lobe excision in man. Neuropsychologia 6: 191-209, 1968
32. Mishkin M: Visual mechanisms beyond the striate cortex. In
Russell R (ed): Frontiers in Physiological Psychology. New
York, Academic, 1966, pp 93-119
33. Mishkin M: Cortical visual areas and their interactions. In
Karczmar AG, Eccles JC (eds): Symposium on the Brain and
Human Behavior. New York, Heidelberg, and Berlin,
Springer, 1972, pp 187-208
34. Mullan S, Penfield W: Illusions of comparative interpretation
and emotion. Arch Neurol Psychiatry 8 1:269-284, 1959
35. Penfield W: Memory mechanisms. Arch Neurol Psychiatry
67:178-191, 1952
36. Penfield W: The permanent record of the stream of con-
sciousness. In Proceedings of the 14th International Congress
on Psychology, Montreal. Acta Psychol (Amst) 11:47-69,
1954
37. Penfield W: Conditioning of the uncommitted cortex for lan-
guage learning. Brain 88:787-798, 1965
Gloor et al: Limbic Experiential P h e n o m e n a 143
38. Penfield W: Engrams in the human brain. Proc R SOCMed
61:831-840, 1968
39. Penfield W: The Mystery of the Mind. A Critical Study of
Consciousness and the Human Brain. Princeton, NJ, Prince-
ton University Press, 1975
40. Penfield W, Jasper H: Epilepsy and the Functional Anatomy
of the Human Brain. Boston, Little, Brown, 1954, pp 113-
143, 438-469
41. Penfield W, Perot P: The brain’s record of auditory and visual
experience. A final summary and discussion. Brain 86:595-
696, 1963
42. SemJacobsen CW, Torkildsen A: Depth recording and
electrical stimulation in the human brain. In Ramsey ER,
O’Doherty DS (eds): Electrical Studies on the Unanesthetited
Brain. New York, Hoeber, 1960, pp 275-290
43. Todd JH, Atema J, Bardach JE: Chemical communication in
social behavior of a fish, the yellow bullhead (Zrtahrus
natalis). Science 158:672-673, 1967
144 Annals of Neurology Vol 12 No 2 August 1982
44. Turner BH, Mishkin M, Knapp M: Organization of the amyg-
dalopetal projections from modality-specific cortical associa-
tion areas in the monkey. J Comp Neurol 191:515-543, 1980
45. Weingarten SM, Cherlow DG, Halgren E: The relationship of
hallucinations to the depth structures of the temporal lobe.
In Sweet WH, Obrador S, Martin-Rodriguez JG (eds):
Neurosurgical Treatment in Psychiatry, Pain and Epilepsy.
Baltimore, University Park Press, 1976, pp 553-568
46. Wieser HG: Ictal ”psychical phenomena” and stereo-
electroencephalographic findings. In Lechner E, Aranibar A
(eds): EEG and Clinical Neurophysiology: Proceedings of
the Second European Congress on EEG and Clinical
Neurophysiology, Salzburg, Austria (International Congress
Series N o 526). Amsterdam, Oxford, and Princeton, Excerpta
Medica, 1979, pp 62-76
47. Weiskrantz L: Experiments on the r.n.s. (real nervous system)
and monkey memory. Proc R SOCLond [Biol] 171:335-352,
1968