Download The influence of current direction on phosphene

Clinical Neurophysiology 112 (2001) 2015–2021
www.elsevier.com/locate/clinph
The influence of current direction on phosphene thresholds evoked by
transcranial magnetic stimulation
Thomas Kammer a,*, Sandra Beck a, Michael Erb b, Wolfgang Grodd b
a
Department of Neurobiology, Max-Planck-Institute for Biological Cybernetics, Spemannstrasse 38, D-72076, Tübingen, Germany
b
Section Experimental MR of CNS, Department of Neuroradiology, University of Tübingen, Tübingen, Germany
Accepted 10 September 2001
Abstract
Objectives: To quantify phosphene thresholds evoked by transcranial magnetic stimulation (TMS) in the occipital cortex as a function of
induced current direction.
Methods: Phosphene thresholds were determined in 6 subjects. We compared two stimulator types (Medtronic-Dantec and Magstim) with
monophasic pulses using the standard figure-of-eight coils and systematically varied hemisphere (left and right) and induced current direction
(latero-medial and medio-lateral). Each measurement was made 3 times, with a new stimulation site chosen for each repetition. Only those
stimulation sites were investigated where phosphenes were restricted to one visual hemifield. Coil positions were stereotactically registered.
Functional magnetic resonance imaging (fMRI) of retinotopic areas was performed in 5 subjects to individually characterize the borders of
visual areas; TMS stimulation sites were coregistered with respect to visual areas.
Results: Despite large interindividual variance we found a consistent pattern of phosphene thresholds. They were significantly lower if the
direction of the induced current was oriented from lateral to medial in the occipital lobe rather than vice versa. No difference with respect to
the hemisphere was found. Threshold values normalized to the square root of the stored energy in the stimulators were lower with the
Medtronic-Dantec device than with the Magstim device. fMRI revealed that stimulation sites generating unilateral phosphenes were situated
at V2 and V3. Variability of phosphene thresholds was low within a cortical patch of 2 £ 2 cm 2. Stimulation over V1 yields phosphenes in
both visual fields.
Conclusions: The excitability of visual cortical areas depends on the direction of the induced current with a preference for latero-medial
currents. Although the coil positions used in this study were centered over visual areas V2 and V3, we cannot rule out the possibility that
subcortical structures or V1 could actually be the main generator for phosphenes. q 2001 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Human visual cortex; TMS; Phosphenes; Threshold; Functional magnetic resonance imaging; Retinotopic map
1. Introduction
The excitatory effect of transcranial magnetic stimulation
(TMS) allows us to determine thresholds of cortical excitability in individual subjects. This is a standard procedure in
the motor cortex, as motor output can be assessed by the
compound muscle action potential (cMAP). More recently,
there has been growing interest in thresholds within the
visual system. Stimulating the occipital pole of the brain
causes the perception of a light flash, i.e. a phosphene
(Barker et al., 1985; Meyer et al., 1991). Thresholds of
phosphenes were measured quantitatively for the first time
within a pathophysiological framework. In patients suffering from migraine, Aurora et al. (1998) found lower thresh* Corresponding author. Tel.: 149-7071-601593; fax: 149-7071601577.
E-mail address: [email protected] (T. Kammer).
olds for phosphene induction than in normal volunteers. In
contrast, Afra et al. (1998) found no difference in phosphene
thresholds between normal volunteers and migraine
patients.
The controversial results have been attributed to methodological differences (Aurora and Welch, 1999; Mulleners
et al., 1999; Schoenen et al., 1999). Indeed, it is not yet clear
which conditions are best suited to eliciting the perception
of phosphenes and reproducibly determining phosphene
thresholds.
Effects of TMS depend on the direction of the induced
current, which is opposite to the current direction in the coil.
In the first systematic study of phosphenes, Meyer et al.
(1991) mentioned a preferential current direction for the
excitation of the visual cortex. Using a round coil and a
monophasic pulse, they found that currents passing the occipital cortex from lateral to medial to evoke stronger phos-
1388-2457/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 1388-245 7(01)00673-3
CLINPH 2001645
2016
T. Kammer et al. / Clinical Neurophysiology 112 (2001) 2015–2021
phenes than currents flowing in the opposite direction. The
same directional preference was reported by Amassian et al.
(1994) for the visual extinction effect. An induced current
passing the occipital pole from left to right disturbed the
perception of visual stimuli in the right visual hemifield
and therefore, seemed to preferentially stimulate the left
hemisphere. The opposite was found to be true for currents
flowing from right to left: they affected perception predominantly in the left visual hemifield. In the motor cortex, a
similar phenomenon is well known. Greater excitability of
the corticospinal motor system is found with monophasic
currents passing through the precentral gyrus from posterior
to anterior, than with a current direction from anterior to
posterior (Brasil-Neto et al., 1992; Mills et al., 1992;
Niehaus et al., 2000). We recently quantified motor threshold differences for both of these current directions.
(Kammer et al., 2001).
The aim of the present study was to quantify the differences in phosphene thresholds with different current directions. Furthermore, we asked how reliable the determination
of phosphene thresholds is. In addition we compared thresholds measured with two different stimulator types. Finally,
we identified the areas in the occipital lobe where the center
of the coil was placed during the threshold measuring procedures. To that end we first identified the borders of visual
areas with functional magnetic resonance imaging (fMRI)
and then used a stereotactic positioning system to project the
position of the coil onto the corresponding cortical surface
of the subject.
2001). Current directions were described for the initial
fast rising phase of the monophasic current pulse induced
in the brain (Kammer et al., 2001). In the Medtronic-Dantec
stimulator current directions were controlled with a dedicated switch. When the switch was set to ‘normal’ and the
coil handle pointed to the left, the induced current direction
in the brain was right to left. With the same coil orientation,
the ‘backward’ position of the switch resulted in an induced
current direction from left to right. With the Magstim stimulator the orientation of the coil handle to the left resulted in
an induced current direction from left to right. To flip
current direction to right–left the orientation of the coil
handle had to be flipped 1808.
2.2. Phosphene thresholds
Phosphene thresholds at the occipital pole were measured
separately for each hemisphere and current direction. The
coil handle was oriented horizontally. First, a stimulation
site was determined as follows: position of the coil was
moved step by step while the subject looking at a computer
screen (21 00 , observer distance 57 cm, visual field 368 £ 288,
background intensity 0.5 cd/m 2) was stimulated with a
suprathreshold intensity until he or she observed a sharply
delineated phosphene clearly restricted to the contralateral
visual field. At this coil position 10 different stimulator
intensities (step size 2%) were tested 10 times in a randomized order. Coil position was continuously monitored and
kept constant. After each stimulation the subject reported
the presence or absence of a phosphene. Using the Boltz-
2. Methods
2.1. Subjects and setup
Six healthy subjects (age 21–37 years, 4 males, two
females) participated in the study after giving their written
informed consent. The study was approved by the local
internal review board of the Medical Faculty, University
of Tübingen.
Subjects were stimulated either with a Medtronic-Dantec
Magpro stimulator (Skovlunde, Denmark) in the monophasic mode (maximal rate 0.33 Hz) or with a Magstim 200
stimulator (Whitland, Dyfed, UK, maximal rate 0.2 Hz).
Both stimulators were fitted with the standard figure-ofeight coil. The Medtronic-Dantec coil (MC-B70) is angled
1408. The two windings each have a diameter of 24–96 mm
and a mean diameter of 60 mm (as measured by X-ray).
They overlap by 38 mm. The Magstim coil (P/N 9790) is
not angled. The two windings each have a diameter of 56–
91 mm and a mean diameter of 74 mm. They are arranged
adjacent to each other and do not overlap. In the experiment
the coil was fixed on a tripod. Coil position in relation to the
head was monitored and registered continuously in all 6
degrees of freedom – 3 translational and 3 rotational –
with a custom-made positioning system (Kammer et al.,
Fig. 1. Example for phosphene threshold measurements and Boltzmann fits
(subject MV, right hemisphere, stimulator: Medtronic-Dantec). On the
abscissa, stimulus intensity is given in percent of maximal output. The
ordinate depicts the number of phosphenes perceived out of 10 stimulations. Open symbols indicate the induced current direction right–left, filled
symbols left–right. Sigmoidal functions were fitted to the data using the
Boltzmann equation. Thresholds were defined as the half-maximal value (5/
10) of the respective function (dashed line). Measurements were performed
3 times per current direction. For each measurement a new stimulation site
was determined first. The stimulation sites for these measurements are
shown in Fig. 5, upper right (right hemisphere).
T. Kammer et al. / Clinical Neurophysiology 112 (2001) 2015–2021
mann equation a fit for the response rates was calculated
with the phosphene threshold defined as the half-maximal
value of the fitted function (Fig. 1). For each hemisphere and
current direction, threshold measurements were performed
at least 3 times at slightly different stimulation sites that
fulfilled the above-mentioned criterion.
2017
induced currents from right to left, green ones from left to
right.
3. Results
3.1. Phosphenes
2.3. Retinotopic maps
Retinotopic maps were delineated using fMRI in 5 out of
6 subjects on a 1.5 T scanner (Siemens Magnetom Vision,
Erlangen, Germany). Eccentricity and polar angle were
mapped using an expanding checkerboard ring and a rotating checkerboard disk segment, respectively, flickering at
4 Hz (stimulation software courtesy of R. Goebel, Maastricht, The Netherlands). Functional images were sampled
in 16 slices oriented parallel to the calcarine fissure
(TR ¼ 2000 ms, TE ¼ 45 ms, flip angle ¼ 908, imaging
matrix ¼ 64 £ 64, voxel size ¼ 3 £ 3 £ 3 mm 3). Eight
cycles with a duration of 9 min were completed for each
parameter (eccentricity as well as polar angle). In addition
a high quality T1-weighted anatomical three-dimensional
(3D) scan was measured (TR ¼ 9:7 ms, TE ¼ 4 ms,
TI ¼ 300 ms, flip angle ¼ 88, 192 sagittal slices, matrix
256 £ 256, voxel size ¼ 1 £ 1 £ 1 mm 3). Data were
analyzed with BrainVoyager (Brain Innovation, R. Goebel,
Maastricht, The Netherlands). After transformation into 3D
space, motion correction, and spatial and temporal filtering,
eccentricity and polar maps were calculated using cross
correlation to a hemodynamically corrected reference function (Goebel et al., 1998). The phase lag indicating eccentricity or polar angle was color coded. A 3D mesh of the
cortical surface of each individual subject’s hemisphere
was reconstructed based on a segmentation of the white
matter in the T1-weighted anatomical scan. This mesh
was inflated and functional retinotopic maps were
projected onto the mesh. The borders between the retinotopic areas V1, V2 and V3 were delineated manually on
the basis of the polar maps (field-sign map, Sereno et al.,
1995). The surface meshes were refolded in order to
inspect the individual topography of retinotopic areas in
situ (see Fig. 5).
TMS evoked phosphenes in all subjects. Before the first
threshold session the subjects completed at least one training session to become familiar with the setup. During that
session they drew the margins of the evoked phosphenes
directly on the screen with a digital drawing device. The
phosphenes always appeared to be gray or white. They
looked like clouds or bubbles and were only transient. The
forms observed were similar to those described previously
(Kammer, 1999). Depending on the stimulation site they
were situated in one of the lower quadrants of the visual
field or bilaterally in both lower quadrants. Only rarely did
the phosphenes extend into the corresponding upper quadrant just at the horizontal border. In the present study we
restricted the stimulation sites to those evoking unilateral
phosphenes only. They were generally observed parafoveally with a typical horizontal extension of 5–158 in the
visual field. In a few cases they reached the center of fixation.
3.2. Phosphene thresholds
Phosphene thresholds varied systematically with stimulator type and current direction. This is evident in the raw data
of Fig. 2. Although the interindividual variability reached up
to 21% of maximal output intensity, an induced current
oriented from lateral to medial (right–left for the right hemisphere and left–right for the left hemisphere) invariably
yielded lower thresholds than a current oriented from medial
to lateral. Overall, threshold values expressed as a percen-
2.4. Phosphene stimulation sites
The coordinates of a set of surface points measured with
the positioning system were used to establish a coordinate
transformation matrix to the 3D skull and cortex mesh with
a surface matching procedure (BrainVoyager). Using this
transformation matrix the position and orientation of the
TMS coil was visualized on the reconstructed cortical
surface of the individual subject (see Fig. 5). Each pin
symbol represents the position of the midpoint of the
inner coil surface during one threshold measurement. At
this site, the maximum electric field is induced by the
coil. Current directions were color coded. Red pins code
Fig. 2. Phosphene thresholds of 6 subjects. On the abscissa different stimulation conditions are given. Induced current directions are abbreviated: r–l,
right–left; l–r, left–right. Phosphene threshold values are depicted as a
percentage of the maximal output of the stimulator used. Each symbol
represents the mean value of at least 3 measurements in one subject.
2018
T. Kammer et al. / Clinical Neurophysiology 112 (2001) 2015–2021
Fig. 3. Mean values of relative phosphene thresholds (^SD) for all stimulation conditions. The threshold values were normalized with respect to the
square root of the maximal stored energy of the stimulator type. These
normalization factors were: Medtronic-Dantec 17.32, Magstim 26.83
(Kammer et al., 2001). See text for results of ANOVA.
tage of maximal output intensity were lower with the
Magstim stimulator than with the Medtronic-Dantec stimulator.
Since the two machines differ in the amount of energy
stored (Medtronic-Dantec: 300 J, Magstim: 720 J, cf.
Kammer et al., 2001), we normalized the data with respect
to the square root of the energy stored, which is approximately proportional to the induced electric field. These
normalized data (shown in Fig. 3) were subjected to a 3way analysis of variance (ANOVA) with the factors stimulator type (Medtronic-Dantec versus Magstim), hemisphere
(right versus left), and current direction (latero-medial
versus medio-lateral). Thus, the factor current direction
was standardized with respect to symmetrical orientation
toward the midsagittal plane. Significant main effects of
stimulator type (Fð1; 5Þ ¼ 87:97; P , 0:0002) and current
direction (Fð1; 5Þ ¼ 78:36; P , 0:0003) were obtained.
Thresholds were higher with the Magstim stimulator
(mean ratio Magstim/Medtronic-Dantec 1.24) and with an
induced current flowing from medial to lateral (mean ratio
medio-lateral/latero-medial 1.19). No significant difference
was found for the factor hemisphere (Fð1; 5Þ ¼ 1:39;
P ¼ 0:29), and no significant interaction occurred.
shown in two subjects. In all 5 subjects where MR measurements were performed, the stimulation sites representing the
midpoint of the figure-of-eight coil were located over the
dorsal surface of the occipital lobe in large clusters at the
right and left hemisphere within areas of about 2 £ 2 cm 2,
sparing a zone about 1–1.5 cm to the left and right of the
midline. As we restricted our study to unilateral phosphenes
we took no measurements at the area along the midline. We
performed the threshold measurement for a given stimulator, hemisphere, and current direction at least 3 times,
searching for a new ‘optimal’ position for phosphene
generation within every single measurement session. Since
threshold values with each particular configuration were
highly reproducible (Figs. 1 and 4) these lateral ‘patches’
with an extension of 2 £ 2 cm 2 seem to be the regions with
the lowest thresholds for phosphenes restricted to one visual
field.
In all subjects the stimulated areas corresponded to the
dorsal parts of the extrastriate areas V2 and V3 of the left
and right hemispheres as characterized by fMRI (see Fig. 5).
Due to the restriction to unilateral phosphenes, thresholds
were not measured at stimulation sites directed towards V1.
However, in every subject phosphenes could be evoked
within this midline zone whenever the maximum of the
electric field evoked was oriented toward the dorsal part
of V1.
3.5. Vertical current directions
Given the previous reports of Meyer et al. (1991) and
Amassian et al. (1994) we only applied horizontal currents
in the main study. In 3 subjects we also tested phosphene
thresholds for vertical current directions in a separate
session using the Medtronic-Dantec stimulator. Phosphenes
in the right visual field were chosen using positions selected
individually for each of the 4 current directions in the left
3.3. Reproducibility of phosphene thresholds
Determination of phosphene threshold was highly reproducible within a single session. For each stimulator type,
current direction, and hemisphere, at least 3 independent
measurements were performed. Threshold values for these
repetitions with each configuration varied between 0.29 and
4.45% of maximal stimulator intensity with a median variation of 1.69% (Fig. 4).
3.4. Stimulation site
In Fig. 5 coil positions for threshold measurements are
Fig. 4. Reproducibility of phosphene thresholds. For every configuration
(current direction, hemisphere, and stimulator type) maximal differences of
threshold values determined in at least 3 independent measurements were
calculated in percent of maximal stimulator output. In the histogram the
median value for these maximal threshold differences (1.69%) is marked as
a line.
T. Kammer et al. / Clinical Neurophysiology 112 (2001) 2015–2021
2019
Fig. 5. Stimulation sites for threshold determination. The midpoint of the figure-of-eight coil is depicted as a pin over the reconstructed cortical surface of the
occipital pole in two subjects (left: KP, right: MV). Induced current direction is right–left at the red pins and left–right at the green pins. The colored areas on
the cortical surfaces indicate visual areas as determined by fMRI. Polar maps projected onto the inflated brain surfaces were used to mark the borders between
the areas V1 and V3a. Then the areas were color coded as indicated in the lower right brain. The scale bar of 10 mm is plotted onto the plane of the occipital
pole. V1d, dorsal; V1v, ventral.
hemisphere (right–left, up–down, left–right, and down–up).
No difference in phosphene threshold was found in any of
the subjects when we compared the two vertical directions
up–down and down–up. The pattern for horizontally
oriented currents was the same as in the main study, with
lower thresholds for currents oriented latero-medially than
for those oriented in the opposite direction. The threshold
level for vertically oriented currents was found to lie
between the values for latero-medial and medio-lateral
currents (Fig. 6).
4. Discussion
Thresholds for phosphenes induced by a monophasic
TMS pulse depend on the current direction. A current
passing the occipital lobe from medial to lateral must be
stronger to reach the phosphene threshold than a current
passing the occipital lobe from lateral to medial. This finding is independent of hemisphere and stimulator type. It
corresponds to the qualitative finding described by Meyer
et al. (1991). Much as in our previous study in the motor
system (Kammer et al., 2001), the normalized thresholds are
higher with the Magstim stimulator than with the Medtronic-Dantec stimulator. This could be due to different pulse
risetimes (Barker et al., 1991, cf. Fig. 1 in Kammer et al.,
2001) or to differences in coil geometries. Results of a
modeling approach (Thielscher and Kammer, unpublished
data) suggest that the latter contributes most to the observed
differences.
2020
T. Kammer et al. / Clinical Neurophysiology 112 (2001) 2015–2021
Fig. 6. Comparison of horizontal and vertical current directions. Normalized threshold values of 3 subjects and mean values are shown. Induced
current directions in the left hemisphere are indicated on the abscissa.
A prerequisite for a quantitative approach is a method of
accurately determining phosphene thresholds. In previous
studies, fine tuning in steps of 1% of maximal output intensity was used (Afra et al., 1998; Aurora et al., 1998) but no
control for statistical fluctuations of perception around the
threshold level was performed. Stewart et al. (2001) used a
rather coarse step size of 5% of maximal output intensity. In
the present study, we determined the phosphene threshold
by the classical psychophysical approach using the method
of constant stimuli with a step size of 2%, and fitted the data
to a sigmoidal function in order to determine the threshold
level for half-maximal responses. Since our impression was
that subjects tend to bias the report depending on the
previous trial we completely randomized the stimulation
intensities. With this method we reached a median variability of only 1.7% (Fig. 4). The exact stimulation site does not
seem to be critical. For every new measurement we searched
for a new stimulation site where subjects reported the
percept of a strong phosphene with suprathreshold stimulation. These stimulation sites with the low variability in
phosphene thresholds were distributed within an area of
2 £ 2 cm 2 at the surface of the occipital part of the hemisphere (Fig. 5). This is in the same range as the area of
stimulation sites over the motor cortex evoking cMAPs
with amplitudes larger than two-thirds of the maximum
(‘top-third-positions’, Classen et al., 1998).
The finding of a preferred current direction for the occipital lobe from lateral to medial also parallels the scenario in
the motor cortex, where thresholds are lowest with currents
from posterior to anterior (Brasil-Neto et al., 1992; Kammer
et al., 2001; Mills et al., 1992; Niehaus et al., 2000). Thus, a
current direction preference seems to be a general property
of a cortical region. Since in the case of phosphene thresholds the horizontally oriented currents pass the occipital
pole perpendicular to the main axis of the line formed by
the interhemispheric cleft, a common feature for the optimal
current orientation might be the perpendicularity to the main
axis of the underlying gyrus. However, the anatomical
substrate for this current direction preference has not yet
been found. The general property of a current direction
preference was recently confirmed in a third cortical region:
the prefrontal cortex (Hill et al., 2000).
In which brain structures are phosphenes evoked? Our
stereotactic approach coregistering TMS stimulation sites
with retinotopic maps derived by fMRI allows us to identify
the visual areas where the maximum peak of the electric
field under the figure-of-eight coil is situated (Fig. 5). But
we cannot be sure that the perception of a phosphene is
caused by the activity of the cortical structure located
where the electric field reaches maximum strength. In a
separate study (Kammer et al., 2000) we mapped the
whole occipital area and always found bilateral phosphenes
in a vertical zone around the interhemispheric cleft with the
maximum of the electric field over V1. Phosphenes could be
evoked from sites up to 5 cm lateral to the interhemispheric
cleft, suggesting that striate areas as well as extrastriate
areas are indeed the target structures. This is in line with
Cowey and Walsh (2000). However, neither Cowey and
Walsh (2000) nor we observed qualitative differences
between phosphenes from extrastriate target sites (V2/V3)
and phosphenes from striate sites. Such a difference might
be expected since differences have been observed with
direct electrical stimulation of the cortical surface (Lee et
al., 2000; Penfield and Rasmussen, 1950). Two alternative
explanations have to be considered (Fig. 7). The target
structure for TMS could be the optical radiation (Marg
and Rudiak, 1994) whose fibers are situated under extrastriate cortical areas before terminating in V1 (Talairach and
Tournoux, 1988). In this scenario, higher field intensities are
required to reach the deeper structures than to stimulate
Fig. 7. Diagram of possible target structures for phosphene generation. A
horizontal section through the right occipital pole is shown (modified from
Talairach and Tournoux, 1988). Visual areas V1–V3 are color coded similar to Fig. 5. The fibers of the optical radiation terminate in the dorsal (V1d)
and ventral part (V1v) of the area striata in the calcarine sulcus. The dorsal
portion of the fibers passes V2 in the white matter. The arrows indicate the
two horizontal current directions investigated.
T. Kammer et al. / Clinical Neurophysiology 112 (2001) 2015–2021
cortical areas at the convexity of a gyrus, since the intensity
drops exponentially as a function of coil distance (Epstein et
al., 1990). Thus, cortical sheets covering the optical radiation, e.g. V2, are exposed to higher fields as well. It remains
to be explained why this cortical exposure, which presumably leads to a stimulation of neurons in these cortical areas,
does not interfere with the subcortical effect. The second
scenario would be that because of the coil position extrastriate areas are stimulated first, but do not result directly in
a phosphene percept. Activity of the extrastriate areas
evokes activity in V1 via the massive backprojection fibers
(Felleman and Van Essen, 1991; Hupe et al., 1998). The
phosphene percept finally results from V1 activation. This
would explain the uniformity of phosphene percepts. Moreover, it fits with the observation that no phosphenes could be
evoked in a blindsight patient with a lesion in V1 (Cowey
and Walsh, 2000).
In conclusion, phosphenes evoked by monophasic TMS
pulses share the direction preference known from the motor
cortex with a preference for induced currents from lateral to
medial. Therefore, we recommend using this current direction rather than vertical ones (Stewart et al., 2001) to determine phosphene thresholds.
Acknowledgements
We would like to thank our subjects for their participation
in the study, Anja Höpfner and Hans-Günther Nusseck for
technical support, and Kuno Kirschfeld for fruitful discussions.
References
Afra J, Mascia A, Gerard P, Maertens de Noordhout A, Schoenen J. Interictal cortical excitability in migraine – a study using transcranial
magnetic stimulation of motor and visual cortices. Ann Neurol
1998;44:209–215.
Amassian VE, Maccabee PJ, Cracco RQ, Cracco JB, Somasundaram M,
Rothwell JC, Eberle L, Henry K, Rudell AP. The polarity of the induced
electric field influences magnetic coil inhibition of human visual cortex:
implications for the site of excitation. Electroenceph clin Neurophysiol
1994;93:21–26.
Aurora SK, Welch KMA. Phosphene generation in migraine. Ann Neurol
1999;45:416.
Aurora SK, Ahmad BK, Welch KMA, Bhardhwaj P, Ramadan NM. Transcranial magnetic stimulation confirms hyperexcitability of occipital
cortex in migraine. Neurology 1998;50:1111–1114.
Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic stimulation of
human motor cortex. Lancet 1985;I:1106–1107.
Barker AT, Garnham CW, Freeston IL. Magnetic nerve stimulation: the
effect of waveform on efficiency, determination of neural membrane
time constants and the measurement of stimulator output. Electroenceph clin Neurophysiol 1991;43(Suppl):227–237.
Brasil-Neto JP, Cohen LG, Panizza M, Nilsson J, Roth BJ, Hallett M.
Optimal focal transcranial magnetic activation of the human motor
cortex: effects of coil orientation, shape of induced current pulse, and
stimulus intensity. J Clin Neurophysiol 1992;9:132–136.
2021
Classen J, Knorr U, Werhahn KJ, Schlaug G, Kunesch E, Cohen LG, Seitz
RJ, Benecke R. Multimodal output mapping of human central motor
representation on different spatial scales. J Physiol (Lond)
1998;512:163–179.
Cowey A, Walsh V. Magnetically induced phosphenes in sighted, blind and
blindsighted observers. NeuroReport 2000;11:3269–3273.
Epstein CM, Schwartzberg DG, Davey KR, Sudderth DB. Localizing the
site of magnetic brain stimulation in humans. Neurology 1990;40:666–
670.
Felleman DJ, Van Essen DC. Distributed hierarchical processing in the
primate cerebral cortex. Cereb Cortex 1991;1:1–47.
Goebel R, Khorramsefat D, Muckli L, Hacker H, Singer W. The constructive nature of vision – direct evidence from functional magnetic resonance imaging studies of apparent motion and motion imagery. Eur J
Neurosci 1998;10:1563–1573.
Hill AC, Davey NJ, Kennard C. Current orientation induced by magnetic
stimulation influences a cognitive task. NeuroReport 2000;11:3257–
3259.
Hupe JM, James AC, Payne BR, Lomber SG, Girard P, Bullier J. Cortical
feedback improves discrimination between figure and background by
V1, V2 and V3 neurons. Nature 1998;394:784–787.
Kammer T. Phosphenes and transient scotomas induced by magnetic stimulation of the occipital lobe: their topographic relationship. Neuropsychologia 1999;37:191–198.
Kammer T, Erb M, Beck S, Grodd W. Multimodal mapping of the visual
cortex: comparison of functional MRI and stereotactic TMS. Eur J
Neurosci 2000;12(Suppl 11):192.
Kammer T, Beck S, Thielscher A, Laubis-Herrmann U, Topka H. Motor
thresholds in humans. A transcranial magnetic stimulation study
comparing different pulseforms, current directions and stimulator
types. Clin Neurophysiol 2001;112:250–258.
Lee HW, Hong SB, Seo DW, Tae WS, Hong SC. Mapping of functional
organization in human visual cortex – electrical cortical stimulation.
Neurology 2000;54:849–854.
Marg E, Rudiak D. Phosphenes induced by magnetic stimulation over the
occipital brain: description and probable site of stimulation. Optom Vis
Sci 1994;71:301–311.
Meyer BU, Diehl RR, Steinmetz H, Britton TC, Benecke R. Magnetic
stimuli applied over motor cortex and visual cortex: influence of coil
position and field polarity on motor responses, phosphenes, and eye
movements. Electroenceph clin Neurophysiol 1991;43(Suppl):121–
134.
Mills KR, Boniface SJ, Schubert M. Magnetic brain stimulation with a
double coil: the importance of coil orientation. Electroenceph clin
Neurophysiol 1992;85:17–21.
Mulleners WM, Chronicle EP, Vredeveld JW, Koehler PJ. Cortical excitability in migraine. Ann Neurol 1999;45:415–416.
Niehaus L, Meyer BU, Weyh T. Influence of pulse configuration and direction of coil current on excitatory effects of magnetic motor cortex and
nerve stimulation. Clin Neurophysiol 2000;111:75–80.
Penfield W, Rasmussen T. The cerebral cortex of man: a clinical study of
localization and function, New York, NY: Macmillan, 1950.
Schoenen J, Maertens de Noordhout A, Afra J. Reply. Ann Neurol
1999;45:417.
Sereno MI, Dale AM, Reppas JB, Kwong KK, Belliveau JW, Brady TJ,
Rosen BR, Tootell RBH. Borders of multiple visual areas in humans
revealed by functional magnetic resonance imaging. Science
1995;268:889–893.
Stewart LM, Walsh V, Rothwell JC. Motor and phosphene thresholds: a
transcranial magnetic stimulation correlation study. Neuropsychologia
2001;39:415–419.
Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain,
Stuttgart, NY: Thieme, 1988.