\

PERGAMON Neuropsychologia 26 "0888# 014Ð024

A primer of magnetic stimulation as a tool for neuropsychology

Vincent Walsh\ Matthew Rushworth0
Department of Experimental Psychology\ University of Oxford\ South Parks Rd\ Oxford OX0 2UD\ UK

Received 29 June 0887^ accepted 09 July 0887

Abstract

Transcranial magnetic stimulation "TMS# o}ers the neuropsychologist a {virtual lesion| method of investigating the e}ects of
cortical dysfunction[ Classical neuropsychology relies on patients with irreversible\ and often di}use brain lesions and these factors
place limitations on the inferences that can be drawn about normal brain function[ Thus the neuropsychologist is constrained by the
extent to which the damaged brain undergoes reorganisation and by the inability to address questions regarding the timing of
cognitive functions[ TMS can disrupt cognitive functions for a few tens of milliseconds "although some e}ects of TMS can be seen
for longer#\ with spatial resolution in the order of a centimetre and therefore allows one to study the role of brain areas without the
masking e}ects of cortical reorganisation[ The spatial and temporal resolutions are not unique to TMS but because TMS can be
used as a temporary interference technique\ it has a functional resolution with which one can address questions beyond the range of
other neuroimaging and patient studies[ Here we outline how TMS produces transitory {lesion| e}ects\ examine how the e}ects of
stimulation spread in depth and breadth across the cortex and discuss the principles of the use of TMS in neuropsychology[ Finally\
we also itemise some issues of safety[ Þ 0888 Elsevier Science Ltd[ All rights reserved[

0[ Introduction structure of the system*say that it is broadly modular

If one were to provide a crude list of the goals of
cognitive neuropsychology it would probably read some!
thing like {where< what< and when<|[ Neuropsychology\
and especially so with the use of PET and fMRI\ has
been very successful at proposing candidate answers for
the {where<| question] one can localise some visual func!
tions to regions of the extrastriate cortex ð74Ł\ language
functions to regions of the temporal and frontal cortices
ð53Ł and memory and decision making functions to pre!
frontal regions ð18\ 47Ł*the list could go on[ But local!
isation of function is not the main or even a principal
concern of many neuropsychologists and functional dis!
sociations can occur without an appreciable di}erence
between lesions[ The {what<| question is more central to
neuropsychology] i[e[\ what are the components of the
task being analysed<
To answer this question\ neuropsychologists have the
problem of making statements about function in the
intact brain\ inferred from studies of injured brains[ Shal!
lice discusses two possible approaches to this question]
{The _rst is to make an assumption about the overall
 Corresponding author[ Tel[] 90754 160215^ fax] 90754 209336^
e!mail] vincent[walshÝpsy[ox[ac[uk[
0
Correspondence can be addressed to Mr M[ Rushworth[ E!mail]
matthew[rushworthÝpsy[ox[ac[uk[
in Marr|s sense[ One then asks what neuropsychological
_ndings tell us about its detailed functional architecture!
[ [ [ The second approach ðasksŁ whether neu!
ropsychological _ndings tell us anything about the types
of basic elements out of which cognition might be con!
structed| ðRef[ 15\ p[ 10Ł[ Modern experimental psycho!
logical theory does not just consider the basic
organisation of a system\ i[e[\ its modularity^ it also con!
siders the ordering\ interaction and evolution through
time of the system|s basic elements[ Because of its time
limited\ reversible\ and repeatable nature TMS allows
neuropsychologists to frame questions about these
important issues] the ordering of component processes
ð04\ 60Ł\ their interaction ð71Ł\ and their evolution through
time\ particularly as a result of learning ð20\ 70Ł[
Discussion of the technology of modern cognitive neu!
roscience often centres on questions of spatial and tem!
poral resolution[ It can be argued\ however\ that TMS has
a co`nitive resolution that distinguishes it as an essential
weapon in the neuropsychologist|s contemporary
armoury[ It is becoming increasingly clear that it is
necessary also to consider the functional inte`ration of
di}erent brain areas in order to understand how even
simple cognitive tasks are performed[ It is possible that
it is only from the perspective of functional integration
that we can understand the role of cortical areas\ such
as the prefrontal cortex\ that lie at some distance from

9917Ð2821:87:, ! see front matter Þ 0888 Elsevier Science Ltd[ All rights reserved[
PII] S 9 9 1 7 Ð 2 8 2 1 " 8 7 # 9 9 9 7 6 Ð 5
015 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024
primary sensory or motor areas[ The temporal resolution
and reversibility of TMS mean that it is well suited for
investigating functional interactions and changes in func!
tional interaction during the course of learning[ Other
brain mapping techniques record brain activity which is
correlated with some behavioural event[ But the cor!
relations do not show that an area is necessary for a
particular function[ TMS\ on the other hand\ because it
is an interference technique\ can be used to establish the
necessity of a brain region for cognitive processes[
There are some practical limitations to working with
patients that preclude the use of experimental techniques
that have been fruitful in other ares of psychology^ as we
shall see\ TMS does not su}er from the same limitations[
Many psychological studies use reaction time "RT# as the
dependent variable and from this inferences can be drawn
about the order in which certain functions occur ð52\
67Ł\ the resources required for each additional processing
component added to a task "e[g[\ the e}ect of adding
more items to a visual search display*ðRef[ 79Ł\ the cost
of switching from one type of stimulus or task to ano!
ther*ðRef[ 0Ł# or interactions between di}erent elements
of processing "e[g[\ the reversal of performance in visual
search*ðRef[ 12Ł#[ In many patients\ brain damage is
widespread and may not just a}ect the cognitive tasks of
interest but it may also a}ect basic processes involved in
response production[ In such cases it is di.cult to obtain
reliable and consistent baseline and task!related measures
of RT[ In cognitive tasks\ the limited interference caused
by single pulse TMS may be insu.cient to cause subjects
to make errors but it is compatible with RT paradigms
ð04\ 41\ 56Ł[
Practical considerations can also make it di.cult to
carry out psychophysical experiments with neu!
ropsychological patients^ again the designs can be too
subtle and too demanding to be pro_tably used with a
patient who has su}ered brain damage[ Nevertheless\ if
one wishes to make statements about the fundamental
elements of processing it is often necessary to do so by
probing only the most sensitive of mechanisms\ i[e[\ those
working at the limits of detection\ discrimination or
decision making[ It is already clear that TMS can be success!
fully combined with psychophysical methods ð32\ 49Ł[
1[ How does TMS work<
During transcranial magnetic stimulation a brief mag!
netic pulse is applied over the scalp of a subject\ at a
point which overlies a speci_c cortical area[ The pulse is
generated from a number of capacitors which discharge
a large brief current into a coil held above the subject|s
head[ The current generates a magnetic _eld below the
coil and this _eld passes\ unattenuated by the skin and
scalp\ into the cortex[ The e}ect of the magnetic _eld at
the cortex is to induce a current which results in neural
activity[ Figure 0 shows the cycle of events in a single
pulse stimulation[ The important points here are that a
large current "7 kA in the example shown# is required to
generate a magnetic _eld of su.cient intensity to stimu!
late the cortex and that the electric _eld induced in the
cortex is dependent upon the rate of change as well as
the intensity of the magnetic _eld[ To achieve this latter
requirement the current is delivered to the coil with a
very short rise time "approx[ 199 us# and the pulse has an
overall duration of approximately one millisecond[ The
induced _eld has two sources ð57Ł[ One is the induction
e}ect from the current in the coil "and this is what is
usually meant when discussing TMS#^ the other is an
accumulation of charge on the scalp or between the scalp
and the skull[
Magnetic stimulation can be applied either in single
pulse or repetitive pulse mode "rTMS#[ In single pulse
mode the stimulation is delivered at a precise point in
time during a task[ In multi pulse mode a repeated train
of pulses can be applied at rates of up to 49 Hz and for
durations of tens\ hundreds or thousands of ms[ Exper!
imentally they di}er in that rTMS is useful for localising
brain regions but does not possess the temporal res!
olution of single pulse TMS[ There is also some indication
that whereas single pulse TMS can cause subjects to
produce errors on sensory detection tasks it is far less
successful at doing so in more cognitive tasks[ For this
reason the study of language ð01Ł and memory ð29Ł is
usually carried out with rTMS[ However\ as we discuss
below\ there is a role for single pulse TMS in the study
of cognition when dependent variables other than error
rates are measured[
Many TMS studies have explored the e}ect that TMS
has on peripheral measures such as EMG recordings
taken from various muscles ð40\ 57Ł[ It soon became clear\
however\ that such peripheral EMG phenomena are the
consequence of an interaction with central\ cortical pro!
cessing ð06\ 22\ 23Ł[ This was graphically illustrated in an
experiment reported by Lemon et al[ ð36Ł in which TMS
pulses directed at the motor cortex were delivered at
di}erent stages of reaching movements[ The TMS e}ect
on EMG depended on the time of the pulse|s delivery[
The most critical times\ however\ did not correspond
to the times of greatest background muscular activity[
Similar task related changes in vulnerability to motor
cortical TMS have been recorded directly in the med!
ullary pyramid of the macaque\ upstream of the spinal
cord and muscles ð5Ł[ It seems likely\ then\ that TMS is
most e}ective when it coincides with a critical epoch
of cortical processing[ In the macaque\ the latencies of
lumbosacral axonal responses to scalp TMS are con!
sistent with a cortical site of action ð13Ł[ In human
subjects\ PET has been used to demonstrate an increase in
cortical synaptic activity\ as indexed by increased regional
cerebral blood ~ow "rCBF#\ in the cortex below the point
of stimulation ð50\ 51Ł[ Many of the e}ects of TMS are
 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024
Fig[ 0[ Sequence of events in TMS] delivery of a single pulse[ An
electrical current of up to 7 kA is generated by a capacitor and dis!
charged "top panel# into a circular\ or _gure!of!eight shaped\ coil which
in turn produces a magnetic pulse of up to 1 Tesla[ The pulse has a rise
time of approximately 199 us and a duration of 0 ms "1nd panel# and
due to its intensity and brevity changes at a rapid rate "panel 2#[ The
changing magnetic _eld generates an electric _eld "panel 3#\ resulting in
neural activity "panel 4#[ The net change in charge density in the cortex
is zero "panel 5#[ In addition to single pulse stimulation some simulators
can deliver trains of pulses up to a rate of 49 Hz[ Rapid rate stimulation
can induce seizures so there is a trade!o} between stimulus intensity
and the rate of repetition[ The exact area stimulated by the pulse and
the depth of stimulation depend on several factors including coil shape
and whether the pulse is monophasic or biphasic[ The details of the
electric current in the stimulating coil and the subsequent e}ects on
neural tissue are taken from Ref[ ð39Ł "with permission of The MagStim
Company# and are calculated for the MagStim 199 with a 69 mm
circular coil[
016
also consistent with interpretations and explanations
based on what is known about cortical functions[
The e}ects of TMS are not necessarily excitatory[ For
example\ subthreshold stimulation\ which does not a}ect
EMG\ can inhibit the e}ect of subsequent suprathreshold
stimulation ð35Ł[ Again the inhibitory e}ect of TMS
appears to have a cortical origin^ it has no measurable
e}ect in the cervical cord ð19Ł\ it has no e}ect on spinal
H re~exes ð35Ł\ but it is a}ected by cortical myoclonus
ð54Ł[ Such inhibitory subthreshold stimuli are associated
with a decrease in cortical synaptic activity\ as indexed
by rCBF decreases ð59\ 51Ł[
TMS applied to a single area can have either a positive
or a negative e}ect depending on the rate of stimulation[
Tarazona et al[ ð68Ł applied rTMS to the motor cortex at
either 0 or 09 Hz while subjects were learning a motor
sequence reaction time task[ Remarkably\ the 0 Hz stimu!
lation\ which also decreased cortical activity\ was cor!
related with a slight decrease in implicit learning\ but the
09 Hz stimulation\ which increased cortical excitability\
was correlated with an increase in implicit learning[
Few cognitive studies have exploited the distinct fac!
ilitatory and inhibitory e}ects of TMS[ Facilitatory
e}ects have mainly been limited to the production of
phosphenes or speci_c muscle contractions but it is of
more psychological interest\ for the present\ that TMS
disrupts the normal pattern of cortical processing[ The
pulse is applied to the cortex during performance of a
task and the intention is not to produce a particular
movement or perception\ but to delay or worsen per!
formance of the task at hand[ In the context of a task the
induced current operates as {neural noise|\ that is\ the
pulse adds random activity in the midst of organised
activity in the cortical region[ This neural noise serves to
delay or disrupt performance and it is in this sense that
TMS operates as a lesion[ There is of course no damage
to the cortex and no long term e}ect of the pulse "see
Safety\ below#[ There are other uses too of TMS and\ for
the sake of completeness\ Table 0 shows the variety of
dependent variables which can be used in TMS\ but for
the rest of this article\ when we refer to TMS\ unless
quali_ed\ we mean TMS in its disruptive\ or lesion mode[
2[ Spatial and temporal resolution
The spatial resolution of TMS is\ perhaps\ not estab!
lished as _rmly as that in imaging techniques such as PET
and fMRI[ Often\ of course\ analysis methods and the
averaging of several subjects| data mean that imaging
experiments do not\ in practice\ yield results at the limits
of the possible spatial resolution[ A full understanding of
the relationship between the extent of induced current
and the anatomical speci_city of e}ects awaits clari!
_cation[ However\ studies now routinely report the
ability of TMS to distinguish the e}ects of stimulating
017 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024

Table 0
An outline and examples of the kinds of experimental questions that can be asked with TMS

Dependent variable Example References

Muscle activity
Subjective self report
Assessed self report
Behavioural measures
Blood ~ow and brain activity
"0# In both young monkeys and humans there is a drop in the threshold e}ect ð16Ł
of TMS over motor cortex that is correlated with the development of _ne
_nger movements
"1# Changes in size:threshold of the motor cortical representations of muscles ð20Ł
during motor skill acquisition
"2# The electromyographic e}ect of TMS over the motor cortex is modulated ð15Ł
by motor imagery
"0# Facilitatory] phosphemes ð1Ł
"1# Disruptive] reports of perceptual e}ects during task performance ð24\ 25Ł
Depression studies ð17Ł
"0# Facilitatory] 09 Hz stimulation of motor cortex enhances implicit learning ð68Ł
of a motor sequence
"1# Interhemispheric facilitation] parietal stimulation increases ipsilateral tac! ð64Ł
tile sensitivity
"2# Disruptive] reaction time or error increases during task performance ð6\ 20\ 27\ 41Ł
Correlations between brain activity and TMS ð26\ 59\ 50Ł

scalp sites 0[4 cm ð60Ł or 0 cm apart ð8Ł[ In a study of
the latency and variability of motor evoked potentials
"MEPs# recorded in four di}erent muscles\ Brasil!Neto
et al[ ð8Ł were able to distinguish between the e}ects of
TMS on scalp positions 9[4Ð0 cm apart[ These levels of
resolution are of course much _ner than one could hope
to obtain in neuropsychological studies[
The current induced by TMS spreads beyond the
intended cortical target area[ With transcranial electrical
stimulation "TES# of su.cient intensity it is possible to
directly stimulate axons at a depth of several tens of ms
below the motor cortex ð13Ł[ Such direct or D responses
are also produced with TMS ð3\ 13\ 48Ł[ At the threshold
for their production\ however\ TMS direct responses
appear to be generated near to the cell soma\ at the axonal
hillock[ At such threshold levels intracortically mediated\
indirect or I responses predominate ð3\ 13\ 22\ 48Ł[ At
high levels of stimulation\ it is possible that the e}ect of
TMS is partly due to the direct stimulation of axons
rather than overlying cortex "e[g[\ ðRef[ 60Ł#[
To enhance the spatial speci_city of TMS\ _gure!of!
eight coils are available^ but even with these coils the
changes in activity in the underlying cortex can spread
across several centimetres ð37Ł[ Ilmoniemi et al[ ð26Ł mea!
sured EEG responses after applying TMS to the visual
or motor cortex and found that the resulting changes in
activity were maximal close to the point of stimulation
but also that the e}ect spread to adjacent areas within 4Ð
09 ms[ It is also clear that TMS can elicit activity in
distant areas that are anatomically interconnected with
the target region ð05Ł[ Ilmoniemi et al[ ð26Ł found that
within about 19 ms of stimulation\ EEG e}ects had
spread across the callosum to the homotopic regions of
cortex in the contralateral hemisphere "Fig[ 1#[ Paus et
al[ ð50\ 51Ł used PET to analyse the anatomical distri!
bution of the e}ects of TMS applied to the frontal eye
_elds[ The result was not only an increase in rCBF at
the stimulated site\ but also at several sites anatomically
connected to the frontal eye _elds[ The message here is
clear[ TMS does stimulate cortical areas but the e}ect is
not limited to the site at which the pulse is directed[ This
has implications for the choice of control conditions\
discussed below[
It is di.cult to estimate the length of time for which
a single TMS pulse interferes with cortical processing[
Suprathreshold TMS can delay movements for up to 049
ms ð07Ł\ but not all of the delay is due to the pulse|s e}ect
on the cortex[ The inhibitory or conditioning e}ect of
subthreshold TMS pulses on subsequent suprathreshold
TMS pulses is maximal when the inter!pulse interval is
less than just 5 ms ð35Ł[ Pulses between 0[2Ð0[8 Tesla
intensity have an e}ect on scalp EEG at any given pos!
ition for a time period somewhere between 19Ð29 ms ð26Ł[
Schluter et al[ ð60Ł compared the e}ect of TMS over three
di}erent sites in the region of the motor and the dorsal
premotor "PMd# cortices[ TMS over the most anterior
site disrupted performance when it was delivered 039 ms
after the presentation of the cue for movement[ TMS at
an adjacent site 0[4 cm away was most e}ective when it
was delivered 39 ms later[
The critical time for TMS delivery to a human brain
area appears to coincide with the time at which single
unit responses can _rst be recorded from the homologous
area of the macaque brain[ For example\ the critical per!
iod for human PMd TMS\ at 039 ms ð60Ł\ corresponds to
a time when more than 19) of macaque PMd neurons
have begun to respond in a movement selection task ð10Ł[
The critical time periods for TMS applied to V0 ð04Ł are
also comparable with macaque single unit data[
Some aspects of the critical time periods for TMS
 )
V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024 018

Fig[ 1[ The e}ects of TMS are not restricted to the site of stimulation but can be seen in anatomically connected regions as early as 19Ð29 ms after
stimulation when subjects are not performing any task[ Ilmoniemi et al[ ð26Ł\ recorded from visual and motor cortex after application of TMS at the
sites marked X[ The activity is greatest around the site of stimulation for the _rst few ms\ but in both cases the e}ects of TMS have been translated
to the homotopic regions of cortex in the hemisphere contralateral to stimulation[ Reproduced with the permission of the authors and Oxford
University Press[

appear anomalous[ In some cases the critical periods are
earlier than would be expected on the basis of human
scalp recorded event related potentials "ERPs#[ This may
be because ERPs only become recordable when large
populations of neurons are synchronously active and
ERP data usually refer to the point at which maximum
task!related activity is recorded[ Whether TMS yields a
more accurate estimation of the time course of events
remains to be explored but the indications to date are
that TMS timing data are closer to estimates based on
studies of single unit recordings ð4\ 04\ 10Ł[
The brevity of the critical period of TMS also needs to
be explained[ For example\ the distinct 39 ms period
during which human PMd\ but not primary motor cortex\
is vulnerable to TMS is di.cult to reconcile with the data
from the macaque that suggest that many cells in both
areas are concurrently active for a longer period of time
ð30\ 31\ 62\ 63Ł[ The same single unit studies suggest that
there is a limited period when just the PMd cells may be
encoding a task parameter[ It may be the case that a
cortical area is only vulnerable to TMS during the time
period that the area uniquely codes for a certain aspect
of task performance ð04Ł[
3[ Necessary and suf_cient control conditions
Clearly\ it is not enough to stimulate a region and
observe an e}ect^ one has to be sure that the e}ects are
either speci_c in time or in space or to the particular task
used[ In neuropsychological or primate lesion studies the
speci_city of the de_cits reported are ensured by control
tasks\ in which a lesion a}ects task A but not task B\ or
by lesion controls\ in which lesion X a}ects task A but
lesion Y does not\ or both*lesion X a}ects task A but
not B and lesion Y a}ects task B but not A[ All these
approaches are valid in TMS experiments[ The choice of
control site can be determined in one of three ways that
we shall call dissociation\ proximity\ and time[ The dis!
sociation method is of course already familiar to neu!

CMYK Page 018 )

029 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024
ropsychology*one simply selects two areas posited to
have di}erent functions and applies TMS to the two using
exactly the same logic as one would in a lesion study using
non!human primates "see ðRef[ 42Ł#[ For an example of
proximity\ suppose one wished to examine the e}ects of
parietal cortex stimulation on an eye movement task[ It
would be reasonable to stimulate a grid of points during
the task centred on\ say\ the P3 electrode site and compare
the site at which an e}ect was obtained with those sur!
rounding points at which TMS had no e}ect "see e[g[\
ðRef[ 43Ł#[ The proximity method results in a statement
to the e}ect that {area X is important for task A but
neighbouring sites are not|[ The proximity method is
particularly important where structural MRI scans of
individual subjects are not available to help localise areas
of cortex and it is also useful in pilot experiments[ It
may also appeal to those for whom the anatomy is of
secondary importance to the business of obtaining
informative functional dissociations[ The early work of
Amassian for example is uninformative with respect to
the localisation of TMS e}ects\ but the stimulus and
task speci_city of the e}ects of TMS helped to lay the
foundations for current experiments on visual perception
ð2Ł[
Among interference techniques\ dissociations in time
are unique to TMS[ A number of studies have contrasted
the time at which TMS has an e}ect at two brain areas
ð7\ 04\ 60Ł[ TMS can of course be used to assess the role
of functional areas at much longer time scales[ A recent
TMS study has exploited this aspect of TMS to inves!
tigate functional plasticity over periods of several
minutes[ Classen et al[ ð00Ł measured TMS!evoked
responses to stimulation of the cortical representation of
the thumb[ The direction of the elicited thumb movement
was consistent as shown in Fig[ 2[ Subjects were then
given a simple motor skill learning task "directed thumb
movements in a direction opposite to the direction of
TMS!evoked thumb movements#[ TMS was used after
training\ again to evoke directionally selective thumb
movements[ As a result of training\ cortical stimulation
now elicited thumb movements in the trained direction\
indicating that the organisation of thumb representation
underwent learning!related changes[ To produce this
e}ect training took place over periods between 4Ð29 min[
The direction of thumb movements elicited by TMS after
training was monitored to assess the time required for
the thumb representation to return to normal[ As Fig[ 2
shows\ 19 min after training the TMS!evoked movements
were predominantly in the trained direction[
Other studies of plasticity have demonstrated cortical
changes occurring over minutes ð75Ł\ hours\ months and
years ð02\ 03\ 21\ 34\ 45\ 70Ł[ For example\ Pascual!Leone
and Torres measured the extent to which the cortical
representation of the Braille reading _ngers were
expanded relative to non!Braille reading _ngers[ Som!
atosensory evoked potentials were signi_cantly larger
when recorded from stimulation of the representation of
the Braille _ngers than the non!Braille _ngers and the
representation of the Braille _ngers was also shown to be
expanded by 1Ð2 times in area[
The peripheral e}ects of TMS also need to be
considered[ The delivery of a magnetic pulse produces a
sharp audible crack and it is important to be sure that
this is not a distracting stimulus which could impair per!
formance[ TMS applied to the scalp is also likely to
stimulate facial muscles and so produce face twitches or
eye blinks[ A simple control for eye blink artefacts is to
run trials in which eye blinks are caused by stimulation
of the seventh cranial nerve[ Face twitches are likely to
be produced by stimulation at some of the control sites
mentioned above[
4[ Paradoxical lesions
Brain lesions can sometimes produce improvements
in some kinds of performance[ These e}ects are not as
common as the deleterious consequences but they can be
equally informative[ Sprague ð66Ł\ for example\ excised
the entire visual cortex from one hemisphere of the cat
and observed\ as one would expect\ a contralateral hemia!
nopia[ However\ removal of the superior colliculus con!
tralateral to the cortical lesion could restore some visual
orienting abilities[ Another example of paradoxical facili!
tation is the improvement in tactile sensitivity in the hand
ipsilateral to a parietal lobe lesion "ð64Ł^ see Kapur ð33Ł
for review#[ One reason for the di.culty in investigating
improvements in neuropsychological patients may be
that as the patient learns to use a damaged brain system
to deal with the world\ reorganisation and new strategies
cause any bene_ts of the damage to be masked[ The
disorganisation caused by TMS\ however\ is likely to
be too short for the brain to begin to reorganise and
behavioural facilitations are something which can be
studied intentionally rather than serendipitously[
Perhaps {paradoxical| is no longer the correct term for
these kinds of _ndings and it may be better to consider
them as revealing competing and mutually inhibitory sys!
tems in the brain[ In the visual system\ for example\ TMS
over area V4 impairs performance on a visual search
task which requires attention to motion but speeds up
performance on a search task that requires attention to
form and colour ð71Ł[ A reasonable hypothesis to generate
from this _nding is that areas necessary for colour and
motion perception compete for processing resources in
much the same way as stimuli are considered to compete
in some models of visual attention ð08Ł[
5[ Safety {beware of an optimist wielding a stimulating
coil| ðRef[ 39Ł
The use of TMS is rightly subject to approval by local
ethical committees and there are some precautions which
 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024 020

Fig[ 2[ The e}ects of training on cortical representation from Classen et al[\ ð00Ł[ TMS was used to evoke thumb movements "A# before training[
Subjects were then given a training task in which movements were made in the direction opposite to that evoked by TMS[ After a few minutes
training\ TMS applied to the same scalp location as before training\ elicited thumb movements in the trained rather than the pretrained direction[
The bottom trace shows that after training was completed there was a gradual return to the pretraining response to stimulation of motor cortex[
Reproduced with the permission of the authors and the American Physiological Society[

must be taken in all studies using the technique[ The
safety of single pulse stimulation is well established but
further precautions should be taken when using repetitive!
pulse TMS[ Repetitive!pulse TMS can cause seizures[
The magnetic _eld produced by stimulating coils can
cause a loud nose and temporary elevations in auditory
thresholds have been reported ð46Ł[ The use of ear plugs
is recommended in all experiments[ Some subjects may
experience headaches or nausea or may simply _nd the
face twitches and other peripheral e}ects of TMS too
uncomfortable[ Such subjects obviously should be
released from any obligation to continue the experiments[
More serious are the concerns that TMS may induce an
epileptic seizure[ There are a number of cases of epileptic
_ts induced by repetitive pulse TMS and caution is necess!
ary[ As a guide\ any subject with any personal or family
history of epilepsy or other neurological condition should
be precluded from partaking in an experiment which
does not involve investigation of that condition[ Pascual!
Leone et al[ ð46Ł assessed the safety of rTMS and noted
that seizures could be induced in subjects who were not
associated with any risk factors[ The paper presents some
guidelines for the use of rTMS and familiarity with this
paper should be a prerequisite of using rTMS[ The paper
is only one guide and it is not exhaustive[ Further\ the
study deals only with three sites of stimulation and
expresses pulse intensity as a percentage of motor thres!
hold[ Studies which apply rTMS to the prefrontal or
occipital visual areas cannot merely lift criteria from this
paper and assume they transfer to other conditions[ We
also recommend that anyone wishing to use rTMS visit
the TMS website "http]::pni[unibe[ch:maillist[htm#[ The
TMS community is constantly reviewing safety pro!
cedures and this website is a starting point for access to
sound information[
A more recent paper ð72Ł summarizes the consensus
that exists within the community[ The adverse e}ects
recorded include seizures "though these are rare#\ some
enhancement e}ects on motor reaction time and verbal
recall\ and e}ects on a}ect "some subjects have been
reported to cry following left prefrontal rTMS\ and
others to laugh#[ There is little information about poten!
tial longer term problems with rTMS but the issue cannot
be ducked[ If\ on the one hand rTMS is potentially useful
in the alleviation of depression ð17Ł it must be conceded
that rTMS can have longer term e}ects[ It would be
disingenuous to suggest that all long term e}ects are
likely to bene_cial rather than deleterious[ It should be
021 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024

Fig[ 3[ Applying TMS to a region of cortex can enhance or inhibit performance on di}erent tasks[ The top _gure shows six visual search tasks in
which subjects were required to detect the presence absence of a target[ The _gure below shows the e}ects of applying TMS to area V4[ In two tasks
"tasks a and b# there is little or no e}ect of TMS[ When TMS is applied to V4 during a search requiring attention to motion "tasks e and f# performance
is signi_cantly slower with TMS[ Tasks on which attention to attributes other than motion is required\ are facilitated by TMS over V4[ The dotted
line at 0 represents reaction times without TMS[ Solid lines show reaction times with TMS relative to without[ From ðRef[ 71Ł[

noted\ however\ that the improvements in mood as a simple precaution that may be taken is to prevent indi!
result of rTMS follow several sessions of magnetic stimu! vidual subjects from taking part in repeated experiments
lation and that the e}ects may be cumulative ð17Ł[ A over a short period of time[
 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024
As with the paper by Pascual!Leone et al[\ ð46Ł the use
of rTMS should follow a close reading of Wassermann|s
ð72Ł report[
References
ð0Ł Allport A\ Styles EA\ Hseih S[ Shifting intentional set] exploring
the dynamic control of tasks in Attention + Performance 04]
conscious and nonconscious information processing[ Cambridge\
MA] MIT Press[
ð1Ł Amassian VE\ Cracco RQ\ Maccabee PJ\ Cracco JB\ Rudell AP\
Eberle L[ Unmasking human visual perception with the magnetic
coil and its relationship to hemispheric asymmetry[ Brain Research
0882^594]201Ð5[
ð2Ł Amassian VE\ Maccabee PJ\ Cracco RQ\ Cracco JB\ Rudell AP\
Eberle L[ Measurement of information processing delays in human
visual cortex with repetitive magnetic coil stimulation[ Brain
Research 0882^594]206Ð10[
ð3Ł Amassian VE\ Stewart M\ Quirk GJ\ Rosenthal JL[ Physiological
basis of motor e}ects of a transient stimulus to cerebral cortex[
Neurosurgery 0876^19]63Ð82[
ð4Ł Ashbridge E\ Walsh V\ Cowey A[ Temporal aspects of visual
search studied by transcranial magnetic stimulation[ Neu!
ropsychologia 0886^24]0010Ð20[
ð5Ł Baker SN\ Olivier E\ Lemon RN[ Task!related variation in cort!
icospinal output evoked by transcranial magnetic stimulation in
the macaque brain[ Journal of Physiology "London# 0884^377]684Ð
790[
ð6Ł Beckers G\ Homberg V[ Cerebral visual motion blindness] tran!
sitory akinetopsia induced by transcranial magnetic stimulation of
human area V4[ Proc Roy Soc Lond B 0881^138]062Ð7[
ð7Ł Beckers G\ Zeki S[ The consequences of inactivating areas V0 and
V4 on visual motion perception[ Brain 0884^007]38Ð59[
ð8Ł Brasil!Neto JP\ McShane LM\ Fuhr P\ Hallett M\ Cohen LG[
Topographic mapping of the human motor cortex with magnetic
stimulation] factors a}ecting accuracy and reproducibility[ Elec!
troencephalography and clinical Neurophysiology 0881^74]8Ð05[
ð09Ł Chen R\ Corwell\ B\ Yaseen Z\ Hallett M\ Cohen LG[ Mechanisms
of cortical reorganization in lower!limb amputees[ Journal of
Neuroscience 0887^07]2332Ð49[
ð00Ł Classen J\ Liepert J\ Wise S\ Hallett M\ Cohen L[ Rapid plasticity
of human cortical movement representation induced by practice[
Journal of Neurophysiology 0887^68]006Ð12[
ð01Ł Claus D\ Weis M\ Treig T\ Lang C\ Eichorn KF\ Sembach O[
In~uence of repetitive magnetic stimuli on verbal comprehension[
Journal of Neurology 0882^139]038Ð49[
ð02Ł Cohen LG\ Bandinelli S\ Findley TW\ Hallett M[ Motor reor!
ganization after upper limb amputation in man[ Brain
0880^003]504Ð16[
ð03Ł Cohen LG\ Celnik P\ Pascual!Leone A\ Corwell B\ Faiz L\ Dam!
brosia J\ Honda M\ Sadato N\ Gerlo} C\ Catala MD\ Hallett M[
Functional relevance of cross!modal plasticity in blind humans[
Nature 0886^278]079Ð2[
ð04Ł Corthout E\ Uttle B\ Ziemann U\ Cowey A\ Hallett M[ Two
periods of processing in the "circum#striate visual cortex as revealed
by transcranial magnetic stimulation[ Neuropsychologia 0888^26[
ð05Ł Cracco RQ\ Amassian VE\ Maccabee PJ\ Cracco JB[ Comparison
of human transcallosal responses evoked by magnetic coil and
electrical stimulation[ Electroencephalography and Clinical Neu!
rophysiology 0878^63]306Ð13[
ð06Ł Day BL\ Dressler D\ Maertens de Noordhoot A\ Marsden CD\
Nakashima K\ Rothwell JC\ Thompson PD[ Electric and magnetic
stimulation of the human motor cortex] surface EMG and single
motor unit responses[ Journal of Physiology 0878^301]338Ð62[
ð07Ł Day BL\ Rothwell JC\ Thompson PD\ Maertens de Noordhout
022
A\ Nakashima K\ Shannon K\ et al[ Delay in the execution of
voluntary movement by electrical or magnetic brain stimulation in
intact man[ Brain 0878^001]538Ð52[
ð08Ł Desimone R\ Duncan J[ Neural mechanisms of selective visual
attention[ Annual Review of Neuroscience 0884^07]082Ð111[
ð19Ł Di Lazzaro V\ Restuccia D\ Oliviero A\ Pro_ce P\ Ferrara L\
Insola A\ et al[ Magnetic transcranial stimulation at intensities
below active motor threshold activates intracortical inhibitory cir!
cuits[ Exp Brain Res 0887^008]154Ð7[
ð10Ł Di Pellegrino G\ Wise SP[ A neurophysiological comparison of
three distinct regions of the primate frontal lobe[ Brain
0880^003]840Ð67[
ð11Ł Donders FC[ On the speed of mental processes[ In] Koster WG\
editor[ Attention and Performance[ v[ II[ Amsterdam] North
Holland\ pp[ 301Ð20[
ð12Ł Driver J\ McLeod P[ Reversing visual search asymmetries with
conjunctions of movement and orientation[ Journal of Exper!
imental Psychology] Human Perception and Performance
0881^07]11Ð22[
ð13Ł Edgeley SA\ Eyre JA\ Lemon RN\ Miller S[ Direct and indirect
activation of the corticospinal tract by electromagnetic and elec!
trical stimulation of the scalp in the macaque monkey[ Journal of
Physiology 0881^314]209Ð19[
ð14Ł Eyre JA\ Miller S\ Ramesh V[ Constancy of central conduction
delays during development in man] investigation of motor and
somatosensory pathways[ Journal of Physiology 0880^323]330Ð41[
ð15Ł Fadiga L\ Buccino G\ Craighero L\ Fogassi L\ Gellese V\ Pavesi
G[ Corticospinal excitability is speci_cally modulated by motor
imagery] a magnetic stimulation study[ Neuropsychologia 0888^26[
ð16Ł Flament D\ Hall EJ\ Lemon RJ[ The development of cortico!
motorneuronal projections investigated using magnetic brain
stimulation in the infant macaque[ Journal of Physiology
0881^336]644Ð57[
ð17Ł George MS\ Wassermann EM\ Williams WA\ Callahan A\ Ketter
TA\ Basser\ P\ Hallett M\ Post RM[ Daily repetitive transcranial
magnetic stimulation improves mood in depression[ Neuroreport
0884^5]0742Ð5[
ð18Ł Goldman Rakic PS[ The prefrontal landscape] implications of
functional architecture for understanding human mentation and
the central executive[ Roberts AC\ Robbins TW\ Weiskrantz\ L\
editors[ The Prefrontal Cortex[ OUP\ 0887[
ð29Ł Grafman J\ Pascual!Leone A\ Alway D\ Nichelli P\ Gomez!Tor!
tosa E\ Hallett M[ Induction of a recall de_cit by rapid rate trans!
cranial magnetic stimulation[ NeuroReport 0883^4]0046Ð59[
ð20Ł Grafman J\ Wasserman EM[ Neuropsychologia 0888^26[
ð21Ł Hall EJ\ Flament D\ Fraser C\ Lemon RN[ Non!invasive brain
stimulation reveals reorganised cortical outputs in amputees[ Neu!
roscience Letters 0889^005]268Ð75[
ð22Ł Hess CW\ Mills KR\ Murray NMF[ Responses in small hand
muscles from magnetic stimulation of the human brain[ Journal of
Physiology "London# 0876^277]286Ð308[
ð23Ł Hess CW\ Mills KR\ Murray NMF[ Responses in small hand
muscles from magnetic stimulation of the human brain[ Journal of
Physiology "London# 0876^277]286Ð308[
ð24Ł Hotson M\ Braun D\ Herzberg W\ Boman D[ Transcranial mag!
netic stimulation of extrastriate cortex degrades human motion
direction discrimination[ Vision Research 23]1004Ð12[
ð25Ł Hotson J\ Anand S[ The selectivity and timing of motion processing
in human temporo!parieto!occipital and occipital cortex] a trans!
cranial magnetic stimulation study[ Neuropsychologia 0888^26[
ð26Ł Ilmoniemi RJ\ Virtanen J\ Ruohonen J\ Karhu J\ Aronen HJ\
Naatanen R\ Katila T[ Neuronal responses to magnetic stimulation
reveal cortical reactivity and connectivity[ NeuroReport
0886^7]2426Ð39[
ð27Ł Jahanshahi M\ Dirnberger G[ The left dorsolateral prefrontal cor!
tex and random generation of responses] studies with transcranial
magnetic stimulation[ Neuropsychologia 0888^26[
023 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024
ð28Ł Jalinous R[ Technical and practical aspects of magnetic nerve
stimulation[ Journal of Clinical Neurophysiology 0880^7]09Ð14[
ð39Ł Jalinous R[ Guide to magnetic stimulation[ Whitland\ Wales] The
MagStim Company\ 0884[
ð30Ł Johnson PB\ Ferraina S\ Bianchi L\ Carminiti R[ Cortical networks
for visual reaching] physiological and anatomical organization of
frontal and parietal lobe arm regions[ Cerebral Cortex 0885^5]091Ð
07[
ð31Ł Kalaska JF\ Crammond DJ[ Cerebral cortical mechanisms of
reaching movements[ Science 0881^0408Ð12[
ð32Ł Kammer\ T[ Phosphenes and transient scotomas induced by mag!
netic stimulation of the occipital lobe] their topographic relation!
ship[ Neuropsychologia 0888^26[
ð33Ł Kapur N[ Paradoxical functional facilitation in brain!behaviour
research[ A critical review[ Brain 0885^008]0664Ð89[
ð34Ł Kew JJM\ Ridding MC\ Rothwell JC\ Passingham RE\ Leigh PN\
Sooriakumaran S\ et al[ Reorganization of cortical blood ~ow and
transcranial magnetic stimulation maps in human subjects after
upper limb amputation[ Journal of Neurophysiology 0883^61[
ð35Ł Kujirai T\ Caramia MD\ Rothwell JC\ Day BL\ Thompson BD\
Ferbert A\ et al[ Cortico!cortical inhibition in human motor cortex[
Journal of Physiology "London# 0882^360]490Ð19[
ð36Ł Lemon RN\ Johansson RS\ Westling G[ Corticopinal control dur!
ing reach\ grasp\ and precision lift in man[ Journal of Neuroscience
0884^04]5034Ð45[
ð37Ł Maccabee PJ\ Eberle L\ Amassian VE\ Cracco RQ\ Rudell A\
Jayachandra M[ Spatial distribution of the electric _eld induced
in volume by round and Figure {7| magnetic coils] relevance to
activation of sensory nerve _bres[ Electroencephalogr Clinical
Neurophysiology 0889^65]020Ð30[
ð38Ł Marg E[ Magnetostimulation of vision] direct non!invasive stimu!
lation of the retina and the visual brain[ Optometry and Vision
Science 0880^57]316Ð39[
ð49Ł Miller MB\ Fendrich R\ Eliassen JC\ Demirel S\ Gazzaniga MS[
Transcranial magnetic stimulation] delays in visual suppression
due to luminance changes[ NeuroReport 0885^6]0639Ð3[
ð40Ł Mills KR[ Magnetic brain stimulation] a tool to explore the action
of the motor cortex on single human spinal motor neurones[ Trends
in Neuroscience 0880^03]390Ð4[
ð41Ł Muri RM\ Rivaud S\ Gaymard B\ Ploner CJ\ Vermersch AI\ Hess
CW\ Pierrot!Deseilligny[ Role of prefrontal cortex in the control
of express saccades[ A transcranial magnetic stimulation study[
Neuropsychologia 0888^26[
ð42Ł O|Mara SM\ Walsh V[ The logic of and lessons from lesions[ In]
Lynch MA\ O|Mara SM\ editors[ Neuroscience Lab Fax\ Aca!
demic Press[ pp[ 56Ð68[
ð43Ł Oyachi H\ Ohtsuka K[ Transcranial Magnetic Stimulation of the
posterior parietal cortex degrades accuracy of memory!guided sac!
cades in humans[ Investigative Ophthalmology and Vision Science
0884^25]0330Ð8[
ð44Ł Pascual Leone A\ Gomez Tortosa E\ Grafman J\ Alway D\ Nichelli
P\ Hallett M[ Induction of visual extinction by rapid!rate trans!
cranial magnetic stimulation of parietal lobe[ Neurology
0883^33]383Ð7[
ð45Ł Pascual!Leone A\ Torres F[ Plasticity of the sensorimotor cortex
representation of the reading _nger in Braille readers[ Brain
0882^005]28Ð41[
ð46Ł Pascual!Leone A\ Houser CM\ Reese K\ Shotland LI\ Grafman J\
Sato S\ Valls!Sole J\ Brasil!Neto JP\ Wasserman EM\ Cohen LG\
Hallett M[ Safety of rapid!rate transcranial magnetic stimulation
in normal volunteers[ Electroencephalography and Clinical Neu!
rophysiology 0882^78]019Ð29[
ð47Ł Passingham RE[ The frontal lobes and voluntary action[ Oxford
University Press\ 0882[
ð48Ł Patton HD\ Amassian VE[ Single and multiple!unit analysis of
cortical stage of pyramidal tract activation[ Journal of Neuro!
physiology 0843^06]234Ð52[
ð59Ł Paus T\ Jech R\ Thompson CJ\ Comeau R\ Peters T\ Evans AC[
Dose!dependent reduction of cerebral blood ~ow during rapid!
rate transcranial magnetic stimulation of the human sensorimotor
cortex[ Journal of Neurophysiology 0887^68]0091Ð6[
ð50Ł Paus T\ Jech R\ Thompson CJ\ Comeau R\ Peters T\ Evans AC[
Transcranial magnetic stimulation during positron emission tom!
ography] a new method for studying connectivity of the human
cerebral cortex[ Journal of Neuroscience 0886^06]2067Ð73[
ð51Ł Paus T[ Imaging the brain before\ during and after transcranial
magnetic stimulation[ Neuropsychologia 0888^26[
ð52Ł Posner MI[ Chronometric explorations of Mind[ NJ] Hillsdale\
0867[
ð53Ł Price CJ\ Wise RJS\ Frackowiack RSJ[ Demonstrating the implicit
processing of visually presented words and pseudowords[ Cerebral
Cortex 0885^5]51Ð69[
ð54Ł Ridding MC\ Brown P\ Rothwell JC[ Ipsilateral cortico!cortical
inhibition in patients with cortical myoclonus[ Mov Disord
0883^8"1#]027[
ð55Ł Ridding MC\ Taylor JL\ Rothwell JC[ The e}ect of voluntary
contraction on cortico!cortical inhibition in human motor cortex[
Journal of Physiology "London# 0884^376]430Ð7[
ð56Ł Ro T\ Cheifet S\ Ingle H\ Shoup R\ Rafal R[ Localization of the
human frontal eye _elds and motor hand area with transcranial
magnetic stimulation and magnetic resonance imaging[ Neuro!
psychologia 0888^26[
ð57Ł Roth BJ[ A theoretical calculation of the electric _eld induced in
the cortex during magnetic stimulation[ Electroencephalography
+ Clinical Neurophysiology 0880^70]36Ð45[
ð58Ł Rothwell JC[ Evoked potentials\ magnetic stimulation studies\ and
event!related potentials[ Current Opinion in Neurobiology
0882^5]604Ð12[
ð69Ł Rushworth MFS\ Owen AM[ The functional organization of the
lateral frontal cortex] conjecture or conjuncture in the elec!
trophysiology literature< TICS^1]35Ð42[
ð60Ł Schluter ND\ Rushworth MFS\ Mills KR\ Passingham RE[ Signal\
set\ and movement related activity in human lateral premotor
cortex[ A study using transcranial magnetic stimulation[ Neuro!
psychologia 0888^26[
ð61Ł Schluter N\ Rushworth MFS\ Nixon PD\ Mills K\ Passingham
RE[ Temporary interference in human lateral premotor cortex
suggests dominance for the selection of movements] a study using
transcranial magnetic stimulation[ Brain^010]674Ð88[
ð62Ł Scott SH\ Kalaska JF[ Reaching movements with similar hand
paths but di}erent arm orientations[ I[ Activity of individual cells
in motor cortex[ Journal of Neurophysiology 0886^66]715Ð41[
ð63Ł Scott SH\ Sergio LE\ Kalaska JF[ Reaching movements with simi!
lar hand paths but di}erent arm orientations[ II[ Activity of indi!
vidual cells in dorsal premotor cortex and parietal area 4[ Journal
of Neurophysiology 0886^67]1302Ð15[
ð64Ł Seyal M\ Ro T\ Rafal R[ Increased sensitivity to ipsilateral
cutaneous stimuli following transcranial magnetic stimulation of
the parietal lobe[ Ann Neurology 0884^27]153Ð6[
ð65Ł Shallice T[ The neuropsychology of mental structure[ Cambridge
University Press\ 0877[
ð66Ł Sprague JM[ Interaction of cortex and superior colliculus in
mediation of visually guided behaviour in the cat[ Science
0855^043]0433Ð6[
ð67Ł Sternberg S[ The discovery of processing stages] extensions of
Donder|s method[ In] Koster WG\ editor[ Attention and Per!
formance[ v[ II[ Amsterdam] North!Holland\ 0858]168Ð204[
ð68Ł Tarazona F\ Tormos JM\ Canete C\ Catala MD\ Pascual!Leone A[
Modulation of motor cortical excitability by repetitive transcranial
magnetic stimulation in~uences implicit learning[ Society for Neu!
roscience Abstracts 0866^12]0853^654[1[
ð79Ł Treisman A\ Gelade G[ A feature integration theory of attention[
Cognitive Psychology 0879^01]86Ð025[
ð70Ł Walsh V\ Ashbridge E\ Cowey A[ Cortical plasticity in perceptual
 V[ Walsh\ M[ Rushworth : Neuropsycholo`ia 26 "0888# 014Ð024
learning demonstrated by transcranial magnetic stimulation[
Neuropsychologia 0887^25]34Ð8[
ð71Ł Walsh V\ Ellison A\ Battelli L\ Cowey A[ Task!speci_c impair!
ments and enhancements induced by magnetic stimulation of
human visual area V4[ Proceedings of the Royal Society Lond B
0877^B154]426Ð32[
ð72Ł Wassermann EM[ Risk and safety of repetitive transcranial mag!
netic stimulation] report and suggested guidelines from the inter!
national workshop on the safety of repetitive transcranial magnetic
stimulation\ June 4Ð6\ 0885[ Electroencephalography and Clinical
Neurophysiology 0887^087]0Ð05[
024
ð73Ł Zangemeister WH\ Canavan AGM\ Hoemberg V[ Frontal and
parietal transcranial magnetic stimulation "TMS# disturbs pro!
gramming of saccadic eye movements[ Journal of the Neurological
Sciences 0884^022]31Ð41[
ð74Ł Zeki SM[ A vision of the brain[ Blackwell Science Publications\
0882[
ð75Ł Ziemann U\ Corwell B\ Cohen LG[ Modulation of plasticity in
human motor cortex after forearm ischaemic nerve block[ Journal
of Neuroscience 0887^07]0004Ð12[