INVITED REVIEW
Electrical Stimulations of the Human Insula: Their Contribution to the
Ictal Semiology of Insular Seizures
Laure Mazzola,*†‡ François Maugui
ere,†§k and Jean Isnard†§k
*
Department of Neurology, University Hospital of St-Etienne, France; †Team “Central Integration of Pain”, Lyon Neuroscience Research Center, INSERM U
1028, CNRS UMR 5292, Lyon, France; ‡Jean Monnet University, St-Etienne, France; §Department of Functional Neurology and Epilepsy, Neurological Hospital,
Hospices Civils de Lyon, Lyon, France; and kClaude Bernard University Lyon 1, Lyon, France.
Introduction: Stereotactic stimulations of the insular cortex
through intracranial electrodes aim at characterizing the
semiology of insular seizures. These stimulations, carried out in
the context of Stereo-Electro-Encephalography (SEEG) during
presurgical monitoring of epilepsy, reproduce the ictal symptoms
observed during the development of insular seizures.
Methods: The authors reviewed the results of insular
stimulations performed in 222 patients admitted between
1997 and 2015 for presurgical SEEG exploration of atypical
temporal or perisylvian epilepsy. Stimulations (50 Hz, trains of 5
seconds, pulses of 0.5 ms, intensity 0.2–3.5 mA) were carried out
using transopercular electrodes implanted orthogonal to
midsagittal plane.
Results: Out of a total of 669 stimulations, 550 were clinically
eloquent in the absence of any postdischarge (237 and 313,
respectively, in the right and left insulae). Somatosensory
responses (61% of evoked sensations) including pain and visceral
sensations (14.9%) were the most frequent, followed by auditory
I t is now well recognized that the ictal manifestations of insular
seizures are polymorphic, mostly consisting of (1) somatosensory
symptoms, often affecting a large body region, which may present
as mere paresthesiae, thermal sensation, or acute access of pain; (2)
episodes of acute dyspnea with a sensation of strangulation; or (3)
visceral sensations from nausea to irrepressible vomiting. This bulk
of knowledge partly comes from a long experience of focal
electrical stimulations delivered to the insular cortex in the context
of presurgical evaluation of epilepsy.
If anecdotal examples of functional evaluation of the
human brain based on electrical stimulation have been reported
as early as the beginning of the nineteenth century,1,2 the
implementation of this procedure during surgical intervention
and its clinical rationale in epilepsy surgery were developed
only in the first half of the twentieth century by Penfield and his
team,3,4 who carefully collected subjective reports and observ-
able clinical symptoms evoked by the application of electric
stimuli during operations carried out under local anesthesia in
conscious patients. The initial purpose was to localize the
epileptogenic cortex by triggering symptoms mimicking, more
The authors have no funding or conflicts of interest to disclose
Address correspondence and reprint requests to Jean Isnard, MD, PhD,
Department of Functional Neurology and Epilepsy, Hospital for Neurology
and Neurosurgery Pierre Wertheimer, Hospices Civils de Lyon, 59, Boulevard
Pinel, 69677 Bron CEDEX, France; e-mail: jean.isnard@chu-lyon.fr.
Copyright Ó 2017 by the American Clinical Neurophysiology Society
ISSN: 0736-0258/17/3404-0307
DOI 10.1097/WNP.0000000000000382
clinicalneurophys.com Journal of Clinical Neurophysiology Volume 34, Number 4, July 2017
sensations (8%), vestibular illusions (7.5%), speech impairment
(5%), gustatory, (2.7%), and olfactory (1%) sensations. Although
these responses showed some functional segregation (in
particular a privileged pain representation in the postero-
superior quadrant), there was a clear spatial overlap between
representations of the different modalities.
Conclusions: Symptoms evoked by insular stimulations are
multiple. None of them can be considered as absolutely
specific to the insular cortex, but the occurrence in given seizure
of a somatosensory symptom such as pain or of a laryngeal
spasm associated with vestibular, auditory, aphasic, or olfacto-
gustatory symptoms points to a discharge development in the
insular cortex, which is the only cortical region where
stimulations demonstrate such a multimodal representation.
Key Words: Insular epilepsy, Electrical stimulation, Insular
cortex, Functional anatomy.
(J Clin Neurophysiol 2017;34: 307–314)
or less completely, the clinical symptomatology of spontaneous
seizures. Progressively, the notion emerged that this procedure
was usable for an individual functional mapping of the cortical
regions to be spared by a surgical resection and avoiding
postsurgical deficits. In this context, Penfield performed an
important number of insular stimulations after temporal lobec-
tomies or, more exceptionally, removals of the outer part of the
fronto-parietal operculum, mostly when these procedures had
failed to make the patients seizure-free, thus justifying
a reintervention.5
From this experience, Penfield drew two major conclusions
that are fully relevant today: firstly, there was some similarity
between some of the symptoms evoked by insular stimulations and
those usually considered as manifestations of temporal lobe
seizures, thus entailing the risk of confusing insular with temporal
seizures; secondly, the wide range of symptoms evoked by insular
stimulations suggested that the singularity of the insular cortex was
to be multifunctional. The era of cortical stimulations performed
within the constraints of the operative room and a limited time-
frame ended with the possibility to implant stereotactis depth
electrodes that permit to perform, and if needed to repeat,
stimulations at distance from the intervention in fully awake
patients.6,7 However, for decades, this revolution did not bring any
new information regarding the effects of insular stimulations,
because implantation of electrodes in the insular cortex was
considered too dangerous because of the density of blood vessels
in the lateral fissure. The history of insular stimulations was thus
307
 L. Mazzola, et al.
interrupted, and the question of insular epilepsy forgotten until the
late nineties when the safety of transopercular stereotactic
implantation of electrodes in the insula could be demonstrated.8,9
The exceptional abundance of insular responses to stimula-
tions, of which more than 80% are clinically eloquent, particu-
larly when delivered through electrodes implanted perpendicular
to the insular plane, suggests that the insular cortex is one of the
most eloquent cortical regions. Furthermore, the diversity of
evoked symptoms is such that the insular cortex is likely to
contribute to integration of multimodal inputs.
Nowadays, exploration by electrical stimulations remains
essential to understanding the physiological role of the insular
cortex, and to characterize the semiology of insular seizures by
reproducing the symptoms that reflect the development of
a seizure discharge in this cortical region.
PATIENTS AND METHODS
Patients
For this study, we reviewed the insular stimulation data
gathered from 222 patients (107 women and 114 men, mean age:
35.5 years, range: 20–59 years) who underwent a Stereo-Electro-
Encephalography (SEEG) exploration of the perisylvian and
insular cortex in our Epilepsy Department between March 1997
and April 2015, aiming at localizing their epileptogenic area
before surgery. The choice of SEEG targets was based on video-
EEG recordings of seizures, interictal fluorodeoxyglucose PET,
interictal and ictal Single Photon Emission Computed Tomogra-
phy, and brain MRI data. Electrical stimulation of the cortex is
a routine procedure to evaluate the epileptic threshold, and to
provide a functional map in the explored areas. All patients were
fully informed of the purpose and risks of the SEEG procedure,
and gave their written consent.
Electrodes Stereotactic and Insular Sites Location
The stereotactic implantation followed the procedures
described in our previous studies.10,11 Electrodes were implanted
perpendicular to the midsagittal plane, and were left in place
chronically up to 15 days. The electrodes had a diameter of
0.8 mm and contained anywhere from 5 to 15 recording contacts.
308 Journal of Clinical Neurophysiology Volume 34, Number 4, July 2017
Electrical Stimulations of the Insula
Contacts were 2 mm long and separated from one another by
1.5 mm. A cerebral angiogram was performed in stereotactic
conditions using an X-ray source 4.85 m away from the patient’s
head to eliminate the linear enlargement because of X-ray
divergence. The stereotactic coordinates12 of each electrode were
calculated preoperatively on the individual cerebral MRI pre-
viously enlarged at scale 1. Cerebral MRI and angiographic
images were superimposed to avoid any risk of vascular injury
during implantation.
From 1997 to 2008, a postimplantation frontal X-ray at scale
1 was performed and superimposed on individual T1 weighted
brain MRI to check for the final position of each electrode with
respect to the targeted anatomical structures, and to individualize
contacts located in the insula. Since 2009, MRI-compatible
electrodes have been used to confirm in individual brain MRIs
the position of contacts in the insular cortex.
All the 669 insular stimulation sites were localized using the
Talairach’s space using x, y, and z coordinates for medio-lateral,
rostro-caudal, and vertical axis, respectively (Fig. 1A). In this
stereotactic space, x ¼ 0 is the coordinate of the sagittal
interhemispheric plane, y ¼ 0 is that of the frontal plane passing
through the vertical anterior commissure, and z ¼ 0 is that of the
horizontal plane passing through the anterior and posterior
commissure plane. The anterior and posterior commissure
distance has been normalized for each patient. The stereotactic
coordinates of each stimulation site were those of the point
located halfway between the two adjacent contacts used for
bipolar stimulation (i.e., the center of the sphere of neural
elements activated by electrical stimulation).
Stimulation Paradigm
Electrical stimulations were produced by a current-regulated
neurostimulator designed for a safe diagnostic stimulation of the
human brain,13 according to the routine procedure used in our
department to map functionally eloquent and epileptogenic
areas.10,14,15 Square pulses of current were applied between
two adjacent contacts (bipolar stimulation). Only contacts located
in the gray matter were used for stimulation. Stimulations were
applied at 50 Hz, with pulse duration of 0.5 ms, train duration of
5 seconds, and intensity between 0.2 and 3.5 mA. These
parameters were used to avoid any tissue injury16 (charge density
FIG. 1. A, Location of all 669 insular
sites stimulated according to Talairach’s
stereotactic coordinates is represented by
open circles. Grey circles indicate sites
that were implanted more than once. All
sites are projected onto a 3D
representation of insula. Because of
interindividual variability of insula
anatomy, some contacts are projected
slightly outside of the virtual limit of the
reconstructed insula, although they were
actually located in the insula individually.
B, White circles show the spatial
distribution of the 119 contacts where
electrical stimulation did not evoke any
clinical sensation.
clinicalneurophys.com
 Electrical Stimulations of the Insula L. Mazzola, et al.

FIG. 2. Percentage of the Different

Types of Responses Obtained During the
550 Clinically Eloquent Insula
Stimulations.

per square pulse , 55 mC/cm2). Stimulus intensity was raised
from 0.2 mA by steps of 0.4 mA until any sensation was
obtained, and to a maximum of 3.5 mA. The stimulation
threshold was defined as the minimal intensity necessary to
evoke a response. No stimulation was delivered at suprathreshold
values. Our stimulation paradigm, using low stimulation thresh-
old, along with the bipolar mode of stimulation using adjacent
contacts, ensured a good spatial specificity to stimulate the target
structure.17
Collection and Processing of Data
During the stimulation session, patients were sitting in bed
and were asked to relax. Subjective reports and clinical
observations were collected immediately after each electrical
stimulation. EEG data and videos recorded during the stimula-
tions were analyzed retrospectively to characterize the type of
evoked sensations. Electrical stimulations applied within or
around a brain lesion, or inducing an after-discharge, were
excluded from analysis. A multivariate logistic regression was
used to compare the location of contacts evoking each type of
sensation (according to their x, y, and z coordinates) with the
location of all other stimulated contacts, including the non-
eloquent ones.
insular stimulation were somato-sensory sensations (n ¼ 335)
that represented 61% of all evoked sensations, including
paresthesiae (n ¼ 214; 39%), thermal sensations (n ¼ 64;
11.7%), and pain (n ¼ 57; 10.4%). Visceral sensations were
evoked in 14.9% (n ¼ 82), consisting of constrictive sensations
in the throat or abdomen (n ¼ 41), viscero-vegetative sensations
(n ¼ 27), viscero-psychic symptoms combining a visceral
sensation with a feeling of anxiety or fear (n ¼ 14). Then, in
decreasing order of frequency came auditory sensations (n ¼ 44;
8%), vestibular illusions (n ¼ 41; 7.5%), speech impairment (n ¼
27; 5%), gustatory (n ¼ 15; 2.7%) and olfactory (n ¼ 6; 1%)
sensations.
No significant differences in the type of evoked sensations
were noted between right- or left-sided insular stimulations (x2
test with Yates correction: P ¼ 0.10). The average stimulation
threshold was 1.8 6 0.9 mA, and did not vary significantly
according to the location of the stimulation site in the insula
(correlations between intensities threshold and x, y, or z
coordinates: R2 , 0.002), nor to the type of evoked sensation
(P ¼ 0.60).
Mean Talairach’s stereotactic coordinates of insula contacts
where the different types of sensations were evoked by electrical
stimulation, are shown in Table 1.

TABLE 1. Mean Talairach’s Stereotactic Coordinates (6SD) of

RESULTS Insula Contacts Where the Different Type of Sensations Were
Six hundred sixty-nine electrical stimulations were delivered Evoked by Electrical Stimulation
in the insular cortex of the 222 included patients. The locations of X Y Z
all stimulation sites are shown in Fig. 1A. One hundred nineteen
Somato-sensory 37.1 6 3.7 27.3 6 9.8 9.2 6 8
of these insular stimulations did not produce any clinical
Thermal 37.6 6 4 29.2 6 8 9.5 6 7.9
sensation. Although the locations of these “noneloquent” sites Pain 35.2 6 4.2 28.2 6 10.3 10.6 6 7.4
were widely distributed, there was a clear antero-posterior Viscero-vegetative 39.5 6 2.3 24 6 5.8 4.8 6 5.4
gradient, and sites where no clinical sensations were evoked Constrictive sensation 38.3 6 4 20.8 6 9.4 7.5 6 6.2
were significantly more anterior than the eloquent ones (for y Anxiety 39.8 6 3.7 0.9 6 6.7 3.6 6 6.8
coordinates: P , 0.0001; OR ¼ 1.1 [1.07–1.13]; there was no Vestibular 36.3 6 3 28.6 6 9.5 4.2 6 7.4
significant difference for the x and z coordinates) (Fig. 1B). Gustatory 36.3 6 4.3 24.9 6 2 212.1 6 4.8
The other 550 stimulations (82.2%) produced a clinical Olfactory 38.8 6 3.5 27.3 6 3.4 7.8 6 7.9
response, 237 (43%) during right and 313 (57%) during left Auditory 36.6 6 3.1 221.2 6 11.3 7.2 6 4.3
Dysarthria 37.8 6 3.7 25.6 6 10 5.9 6 9
insula stimulations. Percentages of each type of evoked sensa-
No clinical response 38 6 3.7 3.6 6 10.7 4.9 6 7.4
tions are illustrated in Fig. 2. Most of the clinical responses to

clinicalneurophys.com Journal of Clinical Neurophysiology Volume 34, Number 4, July 2017 309
 L. Mazzola, et al.
Somato-sensory Responses
Sites where stimulation evoked this type of sensation were
broadly distributed over the insular cortex, although they were
preferentially located in the postero-superior part of the insula
(for y coordinates: P ¼ 0.03; OR ¼ 0.98 [0.96–0.99]; for z
coordinates: P ¼ 0.002; OR ¼ 1.05 [1.02–1.09]) (Fig. 3). Body
regions involved through somato-sensory manifestations could
vary widely, from very restricted regions (such as nose, eye, ear),
to very large areas (for some patients half the body or even the
whole body). A combination of two or several body regions
could also be observed, such as the face and hand, or thorax and
foot. The median size of skin territories involved by somato-
sensory sensations obtained after insula stimulation was 3% of
the total skin surface (range 0.1%–100%), which is larger than
after the secondary (1%) or primary (0.5%) somatosensory areas
stimulation.11 When sensations involved the midline part of the
body (face or trunk), they were mostly bilateral (68.2%) and less
frequently contralateral (24.2%) or ipsilateral (7.6%) to stimula-
tion. Evoked responses affecting limbs were predominantly
contralateral to stimulation (94%) but could also be bilateral
(3%) or ipsilateral (3%).
Somato-sensory nonthermal and nonpainful sensations (n ¼
214; 39%) were reported as neutral or unpleasant sensations of
tingling, light touch, or slight electric current. Locations of
stimulation sites where they were obtained were largely distrib-
uted over the posterior three quarters of insular cortex (Fig. 3A).
Temperature sensations represented 11.7% (n ¼ 64) of insula
stimulations, and were evoked by stimulation through contacts
located in the median part of the insula, mostly around the central
sulcus (Fig. 3B). Warm sensations were more frequent (9.7%)
310 Journal of Clinical Neurophysiology Volume 34, Number 4, July 2017
Electrical Stimulations of the Insula
than cold sensations (3.8%). Painful sensations were obtained in
10.4% (n ¼ 57) in the posterior two-thirds of insula (Fig. 3C),
mostly in its postero-superior part where the threshold intensity
of pain responses was the lowest. All patients who reported
a painful sensation also had spontaneous behavioral manifes-
tations of pain, including facial expressions or verbal complaints.
The descriptive terms used by the patients for qualifying their
pain were: burning, painful sensation of electricity or electric
shock, stinging, painful pins and needles, crushing or cramp
sensation without visible movement or muscle contraction. Mean
visual analogic scale scores for pain were 6.7/10 (range 4–9/10).
In spite of usually large pain projection areas on the body
surface, a somatotopic organization of painful responses was
found in the caudal insula when pain was restricted to a single
somatotopic region (face, upper limb, or lower limb; 51%, n ¼
29), the face area being rostral to the upper limb area, and the
latter being located above the lower limb representation.
Visceral Sensations
Viscero-sensitive sensations represent the second major
group of symptoms that were produced by stimulation of the
insular cortex, as they were evoked in 14.9% (n ¼ 82), consisting
of a constrictive sensation in the throat or thoraco-abdominal
region (n ¼ 41), viscero-vegetative sensations (n ¼ 27), and
viscero-psychic symptoms (n ¼ 14) (Fig. 4). They were obtained
from contacts located significantly more anteriorly than other
stimulation sites (for y coordinate: P , 0.0001; OR ¼ 1.045; IC
1.02–1.07), especially around, or rostral to, central insular sulcus
(Fig. 4). Constrictive sensations (Fig. 4A) were located in the
pharyngo-laryngeal, retrosternal or abdominal region, and were
FIG. 3. Distribution of the 335 insular
contacts where stimulation evoked
somatosensory responses. A,
Somatosensory nonpainful nonthermal
sensations are represented in light blue
(B), thermal sensations in medium blue,
and (C) painful sensations in deep blue.
clinicalneurophys.com
 Electrical Stimulations of the Insula
described as unpleasant sensations of constriction, from a simple
discomfort to a frightening sensation of strangulation. Viscero-
vegetative sensations (Fig. 4B) included nausea (n ¼ 10),
salivation (n ¼ 5), facial blush (n ¼ 5), fainting fit (n ¼ 3),
dyspnea (n ¼ 2), urge to urinate (n ¼ 1), and sweaty hands (n ¼
1). Viscero-psychic symptoms (Fig. 4C) consisted of a visceral
sensation with a feeling of anxiety or fear (n ¼ 14), and
especially thoracic or abdominal constriction/heaviness associ-
ated with anxiety (n ¼ 7/14). It ranged from mild anxiety to real
panic, and in some cases (7/14) anxiety could appear quite
isolated.
Other Types of Evoked Sensations
Gustatory sensations were evoked in 2.7% of insula
stimulations (n ¼ 15) (Fig. 5). They were exclusively obtained
after the stimulation of a relatively restricted zone of insula, on its
medium-upper part, corresponding to the posterior short gyrus
(Fig. 5A). Taste sensations were described as “nasty” or
“unpleasant” (n ¼ 9), or could not be clearly identified (n ¼ 6).
Six olfactory sensations were evoked by insular stimulation
(1%). Patients described odors as unpleasant or as indefinable
during the three others. Olfactory sensations were located in the
nasal cavity and were not lateralized. Spatial distribution of
stimulation sites producing olfactory sensations is illustrated in
Figure 5A.
Vestibular sensations were reported in 41 of the 550
eloquent insular stimulations (7.5%). Sites were mainly located
in the medium and posterior-superior part of the insula (Fig. 5B)
(for y coordinates: P ¼ 0.01, OR ¼ 0.9 [0.93–0.99]; for z
clinicalneurophys.com Journal of Clinical Neurophysiology Volume 34, Number 4, July 2017
L. Mazzola, et al.
coordinates: P , 0.0001, OR ¼ 0.84 [0.80–0.89]). Most of the
vestibular sensations evoked by electrical stimulations were
described as a feeling of body motion (39/41, 95.1%), or as an
illusion that the visual environment was moving (2/41, 4.9%).
They were more frequently perceived as an illusion of translation
(14/41, 34.1%) than as an illusion of rotation (9/41, 21.9%).
Translations were described as feelings of levitation and flying
(5/14), of body rising (3/14), or the feeling of falling down (6/
14), backwards, or laterally. The side of falling sensation was not
related to the side of insular stimulation (whatever was the
stimulated side, the feeling of fall could be backwards, left- or
right-sided). Rotating sensations in the horizontal plane around
the patient’s body axis (yaw plane) were the most common (7/9),
and could be clockwise (5/7), or counterclockwise (2/5). Clock-
wise responses were induced either by right- (3/5) or left-sided
(2/5) insular stimulations. Roll plane illusions (rotating sensa-
tions in the frontal plane around an antero-posterior axis)
occurred less frequently (n ¼ 2). Often, patients could not
classify precisely the vestibular sensation (18/41, 43.9%), and
reported indefinable feelings of “head spinning” (n ¼ 14),
“instability” (n ¼ 3), and one patient described the “loss of
visual landmarks.”
Forty-four auditory sensations were evoked (8%) from
stimulation sites located in the very posterior and inferior part
of the insula (Fig. 5C) (for y coordinates: P , 0.0001; OR ¼
0.81 [0.76–0.86]). Most often, they were reported as unpleasant
(69.7%) or neutral (30.3%). Most of them were a simple auditory
hallucination (81.8% [n ¼ 36]), ranging from whistling (n ¼ 11),
buzzing (n ¼ 9), or rhythmic auditory perception (like “heart
beat”) (n ¼ 7) to alteration of sounds (“sounds seemed more
FIG. 4. Location of the 82 contacts
where visceral sensations were obtained.
Constrictive sensations are represented
in light pink (A), viscero-vegetative
sensations in deep pink (B), and viscero-
psychic symptoms in old pink (C).
311
 L. Mazzola, et al.
deep”) (n ¼ 9). In 18.2% (n ¼ 8), the sensation was more
complex, associating auditory perception to a feeling of vibra-
tion, pressure or paresthesiae in the ears. Auditory sensations
were mostly perceived in the contralateral ear to the stimulation
side (57.6%), but could be bilateral (36.4%) and exceptionally
ipsilateral (6%).
Finally, 27 speech disturbances were obtained (5%) con-
sisting of speech arrest (n ¼ 18) or slowing down (n ¼ 2), slurred
speech (n ¼ 4), or lowering of voice intensity (n ¼ 3). They
could be evoked in both nondominant (51.8%) and dominant
hemispheres for language (48.2%). The spatial distribution of
sites where they were evoked is represented in Figure 5D
showing a widespread distribution, although the middle-upper
part and the posterior-inferior part of the insula were pre-
dominantly involved.
No visual (except visual illusions of tilt or body translation
reported to vestibular responses), motor or déj
a vu sensations
were evoked by insula stimulation. No systematic and precise
analysis of cardiac rythm variation was performed. However, no
variation of heart rate was evidenced by visual analysis of the
EKG traces, including stimulations, for which palpitations were
reported by the patient.
DISCUSSION
Electrical stimulations of the cerebral cortex are able to
evoke clinical responses that mimic symptoms occurring at the
onset or during the spreading of epileptic discharges. Thus, they
directly contribute both to the interpretation of ictal symptoms in
312 Journal of Clinical Neurophysiology Volume 34, Number 4, July 2017
Electrical Stimulations of the Insula
FIG. 5. Distribution of contacts where:
(A) olfacto-gustatory sensations (red
circle for taste, yellow circles for smell
sensations); (B) vestibular sensations
(orange circles); (C) auditory sensations
(green circles); and (D) speech
disturbances (violin circles) were
obtained.
terms of their localizing value and to the delineation of the
epileptogenic focus when they reproduce clinical symptoms
identical to those experienced by patients during their spontane-
ous seizures. Furthermore, they stimulations permit the assess-
ment of the excitability of the explored cortex by the
measurement of the threshold intensity needed to trigger an
epileptic postdischarge, or a complete electro-clinical seizure.
Ictal semiology analysis based on correlations between the
ictal symptoms and concomitant seizure discharges recorded by
intracranial electrodes suffers from the limited spatial sampling
of invasive recordings. Therefore, linking clinical symptoms to
the presence of an ictal discharge in a given cortical area does not
provide the absolute guarantee that these symptoms are not
generated in unexplored cortical regions. On the contrary, the
localizing value of symptoms produced by direct cortical
stimulation in the absence of any postdischarge is highly reliable,
in particular when the stimulation is delivered between two
neighboring contacts along a single intracortical SEEG electrode.
Intra- and interindividual reproducibility of symptoms evoked by
electrical stimulations establishes a causal link between a clinical
symptom and a specific cortical region. This is why electrical
stimulations are essential for understanding ictal clinical symp-
toms and assigning them a localizing value, especially in the
insula which is a multifunctional cortex where a large variety of
evoked responses can be obtained.
In this study, we report on 550 clinical responses evoked by
669 insula stimulations. Insula can thus be considered as a very
“eloquent” structure since 82.2% of electrical stimulations elicited
a clinical response. Interestingly, sites where no clinical response
could be evoked were located mostly in the anterior part of the
clinicalneurophys.com
 Electrical Stimulations of the Insula
insula. This may be because of the fact that this part of the insula
has been assigned to cognitive18,19 and emotional functions20,21
that may not be apparent without specific protocols. In agreement
with available literature on insular stimulation,10,22–25 a great
number of very different symptoms were obtained, making the
insular cortex a singularly polymodal area. However, most of the
evoked responses have, in common, to be related to body
sensations, usually with neutral or unpleasant value, and no clear
hemispheric specialization. Although responses to stimulation
showed some functional segregation in the insula, there is a clear
spatial overlap between the representations of the different types of
responses. This spatial overlap may be the anatomical substratum
for functional integration of multiple physiological functions.
Symptoms evoked by insular stimulation can be split into
major symptoms, which include somato-sensory and visceral
responses representing the two most frequent types of evoked
responses (75.9%) and minor symptoms, corresponding to the
other 24.1%.
Major Symptoms: Somato-sensitive and
Visceral Responses
Somato-sensory Responses
This is the most frequent type of evoked sensations after
insula stimulation. The sites from where they are obtained are
largely distributed, but with clear posterior predominance,
especially for pain responses, as already reported.14,25,26 Cortical
stimulations show that, in spite of anatomical contiguity and some
similarity between their cyto-architectures, the posterior granular
insular cortex cannot be viewed as an extension of the parietal
opercular cortex, or secondary somato-sensory cortex (SII),
which became buried in the depth of the lateral fissure with the
phylogenetic development of the suprasylvian operculum. The
main argument in favor of distinct functionality between the two
regions is that responses to stimulation of the SII area are purely
somatosensory,11 while there is a large overlap between somato-
sensory, viscero-sensitive, vestibular, and olfacto-gustatory rep-
resentations in the postcentral insular cortex (Figs. 3–5). Some
features distinguish insular somato-sensory responses (SSR) from
those obtained after stimulation of primary somato-sensory cortex
(SI): (1) the size of skin territories involved by SSR is larger after
insular stimulation (limbs, body half); (2) pain is not observed
after stimulation of the primary cortex; and (3) the lateralization
of SSR to insular stimulation can be bilateral or even ipsilateral to
the stimulation side, while SSR to SI are strictly contralateral to
the stimulus, except for perioral sensations that may be poorly
lateralized. However, the distinction between SSR evoked by
opercular (SII) and insular stimulations, as well as that between
SII and insular ictal somatosensory symptoms, is less straightfor-
ward. For instance, stimulations can produce pain in both areas
with the same frequency (about 10%), with the same intensity (4–
9/10 on a visual analogic scale) at the same stimulus threshold,
and 36% of SII SSR are bilateral versus 30% of insular SSR; only
SSR strictly ipsilateral to stimulation are produced exclusively in
the insula.11 In patients with painful seizures explored by SEEG,
ictal pain can be reproduced by insular or SII stimulation, and the
insula and SII regions are systematically involved at onset of
seizures.27
clinicalneurophys.com Journal of Clinical Neurophysiology Volume 34, Number 4, July 2017
L. Mazzola, et al.
Visceral Sensations
They represent the second major group of symptoms
produced by stimulation of the insular cortex (14.9%). They
are predominantly obtained by stimulation of the medium and
anterior insula, corresponding to the location of dysgranular
cortex.28 The percentage of viscero-sensitive responses is
probably underestimated, because the emergence of Sylvian
vessels makes implantation of electrodes in the anterior part of
insula scarcer. This interpretation is supported by the observa-
tions of Penfield and Faulk5 who noted the high frequency of
visceral responses to stimulation of the antero-inferior insular
quadrant that they could explore after surgical removal of the
temporal operculum. Patients reported these symptoms as
identical to their ictal aura, although they were undergoing
surgery for temporal lobe epilepsy. Our experience of temporo-
insular SEEG explorations is that hippocampal discharges almost
always spread into the insular cortex, which is probably
responsible for the genesis of these ictal symptoms,29 often
associated with anxiety. Besides viscero-sensitive functions, the
anterior part of the insular cortex is also implicated in the control
of viscero-motor functions such as vomiting (f) or alteration of
gastric motility.5 Only 7.4% of evoked responses consisted in
constrictive sensations in the pharyngo-laryngeal, retrosternal, or
abdominal regions. However, this type of sensation, and
especially laryngeal constriction (described as unpleasant sensa-
tions, from a simple discomfort to a frightening sensation of
strangulation) is in our experience, much more frequent during
insular seizures than after insular stimulation.
Thus, a bipolar organization of major insular functions can
be evidenced through direct cortical stimulations: a posterior part
made of a granular neocortex, assigned to somato-sensory
functions and singularly to pain perception; and an anterior part,
made of dysgranular cortex, assigned to visceral sensitivity and
viscero-motor functions.
Minor Symptoms
These minor symptoms include:
1. Olfacto-gustatory sensations often described as unpleasant
without being clearly identified.
2. Vestibular sensations predominantly described as a feeling of
body motion such as feelings of levitation or falling down, or
as the feeling that the visual environment was moving.
3. Auditory responses generally reported as unpleasant auditory
hallucination, ranging from whistling, buzzing, or rhythmic
auditory perception, perceived in the contralateral ear to
stimulation side, but possibly bilateral, and exceptionally
ipsilateral. The sensation could be more complex, associating
auditory perception to a feeling of vibration, pressure, or
paresthesiae into the ears.
4. Speech disturbances, consisting of speech arrest, slurred
speech, or lowering of voice intensity that could be evoked
in both nondominant and dominant hemispheres for language.
Although vestibular sensations represent 7.5% of our
responses to insula stimulation, they are rarely reported by
patients with insular epilepsy. This could be because of the fact
that vestibular symptoms may be overshadowed by other more
313
 L. Mazzola, et al.
bothersome sensations. Another possibility could be that the
epileptic insula is different from the physiologic insula, as
reported by Pugnaghi et al.,24 that showed different rates and
types of evoked responses depending if stimulation was done on
a “physiological” insula versus an “epileptic” insula. Another
explanation could be that the vestibular sensations produced by
insular stimulation, which are similar to those reported after
stimulation of first and second temporal gyrus or inferior parietal
cortex,30 result from a dysfunction of the vestibular cortical
network. One may speculate that vestibular sensations produced
by a focal stimulation result from a subtle mismatch between
a preserved processing of visual and proprioceptive inputs and
distorted vestibular information, which is unlikely to be repro-
duced by an ictal insular discharge.
Contrary to Afif et al.23 and Pugnaghi et al.24 who
reported motor responses to insular stimulations, we
never elicited motor manifestations despite a greater
number of insular stimulations. This may be because of
methodological differences, since these authors used
oblique transfrontal or transparietal electrodes
(whereas, we used only electrodes implanted orthogo-
nal to the midsagittal plane) and higher maximal
stimulation intensities (up to 5 mA for Pugnaghi
et al.24). These differences could explain a greater
spatial diffusion of electrical stimulations in their
studies to striatum in depth, or to frontal operculum
in surface, that both can evoke motor symptoms.
When considering the semiology of insular seizures, none of
the major or minor symptoms evoked by stimulation is absolutely
specific to an insular origin. Data from stimulation studies reveal
that all types of responses described in this report may have an
insular origin, a notion that has remained underrecognized for
decades since the pioneer work of Penfield and collaborators, but
the occurrence of any of them in isolation during a seizure is not
specific of an insular discharge per se. For example, pain can be
induced after stimulation of the internal parietal operculum (SII
area), speech disturbances or auditory responses can be evoked
after stimulation of frontal or temporal operculum, respectively.
Only a sequence associating several of the minor and major
symptoms identified by stimulations, especially when associated
with a sensation of laryngeal constriction, strongly suggests an
insular origin of seizures.
REFERENCES
1. Aldini. An account of the galvanic experiments performed by John
Aldini, on the body of a malefactor executed at Newgate. With a short
view of some experiments which will be described in the author’s New
York. London, United Kingdom: Cuthell and Martin, 1803 In press.
2. Rolando L. Saggio sulla vera struttura del cervello dell’uomoe degl’a-
nimali e sopra le funzione del sistema nervoso, 1809. Reprinted as
Saggio sulla vera struttura del cervello dell’uomoe degl’animali e sopra
le funzione del sistema nervoso. Bologna: Arnaldo Forni Editore, 1974.
3. Penfield W, Jasper H. Epilepsy and the functional anatomy of the human
brain. Boston: Little Brown, 1954.
4. Penfield W, Rasmussen A. The cerebral cortex of man. New York:
McMillan Co, 1957; 11–65.
5. Penfield W, Faulk ME Jr. The insula; further observations on its
function. Brain 1955;78:445–470.
314 Journal of Clinical Neurophysiology Volume 34, Number 4, July 2017
Electrical Stimulations of the Insula
6. Bancaud J, Angelergues R, Bernouilli C, et al. Functional stereotaxic
exploration (stereo-electroencephalography) in epilepsie [in French].
Rev Neurol (Paris) 1969;120:448.
7. Lesser RP, L€uders H, Klem G, et al. Extraoperative cortical
functional localization in patients with epilepsy. J Clin Neurophysiol
1987;4:27–53.
8. Guénot M. Indications and risk of neurosurgical techniques in the adult
presenting with drug-resistant partial epilepsy (radiosurgery included) [in
French]. Rev Neurol (Paris) 2004;160 Spec No 1:5S185–5S194.
9. Bourdillon P, Ryvlin P, Isnard J, et al. Stereo-electroencephalography
(SEEG) is a safe procedure, including for insular implantations. World
Neurosurg 2017;99:353–361.
10. Isnard J, Guenot M, Sindou M, Mauguiere F. Clinical manifestations of
insular lobe seizures: a stereo-electroencephalographic study. Epilepsia
2004;45:1079–1090.
11. Mazzola L, Isnard J, Mauguiere F. Somatosensory and pain
responses to stimulation of the second somatosensory area (SII) in
humans. A comparison with SI and insular responses. Cereb Cortex
2006;16:960–968.
12. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain.
New York: Thieme, 1988.
13. Babb TL, Mariani E, Seidner KA, et al. A circuit for safe diagnostic
electrical stimulation of the human brain. Neurol Res 1980;2:181–
197.
14. Mazzola L, Isnard J, Peyron R, Mauguiere F. Stimulation of the human
cortex and the experience of pain: wilder Penfield’s observations
revisited. Brain 2012;135:631–640.
15. Mazzola L, Lopez C, Faillenot I, Chouchou F, Mauguiere F, Isnard J.
Vestibular responses to direct stimulation of the human insular cortex.
Ann Neurol 2014;76:609–619.
16. Gordon B, Lesser RP, Rance NE, et al. Parameters for direct cortical
electrical stimulation in the human: histopathologic confirmation.
Electroencephalogr Clin Neurophysiol 1990;75:371–377.
17. Nathan SS, Sinha SR, Gordon B, Lesser RP, Thakor NV. Determination
of current density distributions generated by electrical stimulation of the
human cerebral cortex. Electroencephalogr Clin Neurophysiol
1993;86:183–192.
18. Mayer JS, Bittner RA, Nikolic D, Bledowski C, Goebel R, Linden DE.
Common neural substrates for visual working memory and attention.
Neuroimage 2007;36:441–453.
19. Soros P, Marmurek J, Tam F, Baker N, Staines WR, Graham SJ.
Functional MRI of working memory and selective attention in vibro-
tactile frequency discrimination. BMC Neurosci 2011;8:48.
20. Lamm C, Singer T. The role of anterior insular cortex in social emotions.
Brain Struc Func 2010;214:579–591.
21. Kurth F, Zilles K, Fox PT, Laird AR, Eickhoff SB. A link between the
systems: functional differentiation and integration within the human
insula revealed by meta-analysis. Brain Struct Funct 2010;214:519–534.
22. Nguyen DK, Nguyen DB, Malak R, et al. Revisiting the role of the insula
in refractory partial epilepsy. Epilepsia 2009;50:510–520.
23. Afif A, Minotti L, Kahane P, Hoffmann D. Anatomofunctional
organization of the insular cortex: a study using intracerebral electrical
stimulation in epileptic patients. Epilepsia 2010;51:2305–2315.
24. Pugnaghi M, Meletti S, Castana L, et al. Features of somatosensory
manifestations induced by intracranial electrical stimulations of the
human insula. Clin Neurophysiol 2011;122:2049–2058.
25. Stephani C, Fernandez-Baca Vaca G, Maciunas R, Koubeissi M, Luders
HO. Functional neuroanatomy of the insular lobe. Brain Struct Funct
2011;216:137–149.
26. Ostrowsky K, Magnin M, Ryvlin P, Isnard J, Guenot M, Maugui
ere F.
Representation of pain and somatic sensation in the human insula:
a study of responses to direct electrical cortical stimulation. Cereb Cortex
2002;12:376–385.
27. Montavont A, Maugui
ere F, Mazzola L, et al. On the origin of painful
somatosensory seizures. Neurology 2015;84:594–601.
28. Morel A, Gallay MN, Baechler A, Wyss M, Gallay DS. The human
insula: architectonic organization and postmortem MRI registration.
Neuroscience 2013;236:117–135.
29. Isnard J, Guénot M, Ostrowsky K, Sindou M, Maugui
ere F. The role
of the insular cortex in temporal lobe epilepsy. Ann Neurol
2000;48:614–2330.
30. Kahane P, Hoffmann D, Minotti L, Berthoz A. Reappraisal of the human
vestibular cortex by cortical electrical stimulation study. Ann Neurol
2003;54:615–624.
clinicalneurophys.com