PHYSIOLOGICAL RESEARCH ISSN 0862-8408
© 2005 Institute of Physiology, Academy of Sciences of the Czech Republic, Prague, Czech Republic Fax +420 241 062 164
E-mail: physres@biomed.cas.cz http://www.biomed.cas.cz/physiolres




Physiol. Res. 54: 123-128, 2005


The Influence of Low-frequency Left Prefrontal Repetitive
Transcranial Magnetic Stimulation on Memory for Words but
Not for Faces

L. ŠKRDLANTOVÁ1,3, J. HORÁČEK1,2,3, C. DOCKERY1, J. LUKAVSKÝ4,
M. KOPEČEK1,2,3, M. PREISS1,3, T. NOVÁK1,3, C. HÖSCHL1,2,3

1Prague Psychiatric Centre, 2Third Medical Faculty of Charles University, 3Centre of
Neuropsychiatric Studies, 4Institute of Psychology, Czech Academy of Sciences, Prague,
Czech Republic

Received July 12, 2004
Accepted October 6, 2004



Summary
Brain imaging studies suggest localization of verbal working memory in the left dorsolateral prefrontal cortex (DLPFC)
while face processing and memory is localized in the inferior temporal cortex and other brain areas. The goal of this
study was to assess the effect of left DLPFC low-frequency repetitive transcranial magnetic stimulation (rTMS) on
verbal recall and face recognition. The study revealed a significant decrease of free recall in word encoding under rTMS
(110 % of motor threshold, 0.9 Hz) in comparison with sham stimulation (p=0.03), while no significant difference was
found with facial memory tests. Our findings support the essential role of the left DLPFC in word but not facial memory
and confirm the content specific arrangement of cortical areas involved in semantic memory. As a non-invasive tool,
rTMS is useful for cognitive brain mapping and the functional localization of the category specific memory system.


Key words
Face memory • Verbal memory • Repetitive transcranial magnetic stimulation • rTMS


Introduction

Transcranial magnetic stimulation (TMS) is a
newly developing, non-invasive method that induces
functional changes in a relatively small area of the human
cerebral cortex. The principle of TMS is derived from the
Faraday’s flux-cutting law. Short electric current transits
in the coil produce a temporary magnetic field. This field
passes over the medium, i.e. cranium and is turned back
into an electric field, which induces transmembrane
potentials and leads to the depolarization of neuron
membranes (Barker et al. 1985).
TMS is used to affect different cognitive
functions by inducing artificial cortical lesions and thus it
can provide insight into cortical connectivity and
localization. TMS, applied in cognitive brain mapping, is
used as single pulse (spTMS), or rapid-rate pulse (up to
50 Hz) called repetitive TMS (rTMS). The spTMS
applied in specific synchronization with the cognitive
process can disrupt it and produce a so-called virtual
lesion lasting for up to one second. The effect of rTMS
depends on the power of the magnetic field, the

124 Škrdlantová et al. Vol. 54


localization of the stimuli and the frequency. High-
frequency rTMS (>1 Hz) increases the cortical
excitability and results in the long-term enhancement of
synaptic transfer. Conversely, low-frequency rTMS
(≤1 Hz) inhibits the cortical excitability and leads to
weakening of the transfer at the synapses (Chen et al.
1997, Pascual-Leone et al. 1994, Wassermann and
Lisanby 2001, Romero et al. 2002). In comparison with
spTMS, the rTMS method allows an increase or decrease
of cortical activity for the entire duration of the
stimulation or longer.
Currently, TMS has been used in various
experiments to study memory encoding and retrieval (for
review Mull and Seyal 2001, Grafman and Wasserman
1999, Pascual-Leone and Hallett 1994) and its results
correspond with the data from neuroimaging studies.
The investigation of structure-function
relationships by means of functional neuroimaging
brought evidence for the vital role of the frontal cortex in
semantic memory. The anatomical specialization of
semantic memory for different types of stored
information (words, faces, animals) has been proposed
(Fletcher and Henson 2000, Kelley et al. 1998). The
cortical information-modality specialization is a
challenging concept in brain mapping and theory of
information processing. A review of human functional
imaging studies indicates that verbal memory tasks
activate foremost the left dorsolateral prefrontal cortex,
while face encoding and consolidation involves the basal
frontotemporal cortex (Fletcher and Henson 2000,
Cabeza and Nyberg 2000).
The primary goal of our study was to evaluate
the effect of low-frequency rTMS localized on the
DLPFC (the dominant hemisphere) on the encoding of
semantic memory for faces and words. With regard to the
cortical inhibitory effect of low-frequency rTMS and the
cortical specialization for different types of memory we
tested the following hypothesis: low-frequency rTMS on
left DLPFC in comparison with sham (inactive)
stimulation decreases the memory encoding of verbal but
not face information.

Methods

Subjects
The study was a sham-controlled trial in which
all participants completed two memory test examinations.
Ten right-handed volunteers (aged 23.8±2.3 years, mean
± SD), five males (24.4±3.2) and five females (23.2±1.1),
carefully screened for neurological, psychiatric, or
physical illness or head injury were involved into the
study. All subjects had no previous history of and were
not taking any medication. Motor thresholds, measured in
percent of Magstim output, were 54.8±8.51 % for the
whole group (59±12.4 % for males and 51.2±2.5 % for
females). Mean education level in our group was
16.8±1.2 years (16.6±1.1 for males and 17.0±1.4 for
females). The Institutional Ethics Committee approved
the study and an informed consent form was signed by
each subject.

Cognitive testing
The tests were conducted under randomized
assignment to sham or experimental rTMS. The inter-
testing interval within the sham and experimental
condition was at least one week, with each session
occurring approximately at the same daytime.
Active and sham memory testing had two parts:
the word memory (recall) test and the face recognition
memory test. Both subtests were in two equivalent
versions (verified previously on a large sample of 101
subjects) to avoid the influence of learning in the second
(re-) testing. The subtests were administered in random
order with a half hour interval between the two. All tests
were administered on a computer, which was placed at
eye level 1 m in front of the subjects while the rTMS was
applied. The word-memory test explored verbal memory
by presenting a series of 25 target real words, with 4 to 6
letters per word (for example: rose, mother, table) in
white sans-serif font on a black screen in Czech. An
interstimulus interval (frequency) was 4.5 s (word
presentation for 3 s then 1.5 s of blank screen).
Immediately after this encoding period (and rTMS) the
subjects were asked to recall the words, which they
viewed during the presentation and the number of correct
words was counted as the result. The face recognition test
investigated memory for faces by consecutively
presenting 25 target photographs of young Caucasian
male faces in identical apparel on a monitor with the
same interstimulus interval as in the word test. After the
presentation the participants were tested for recognition
by being prompted to choose the already presented face
from its pair (a same-sized, but novel face), which were
side by side on the screen and the exposure time was
done by the click for next pair.

Repetitive transcranial magnetic stimulation (rTMS)
The rTMS application for each subtest started

2005 Influence of rTMS on Memory for Words and Faces 125

one minute before the test began and continued
throughout the acquisition phase (192 magnetic pulses
per one subtest). A high-speed Magstim Super Rapid
Stimulator (Magstim, Whitland, UK) with a continuous
air cooling figure-of-eight coil and a 70 mm diameter
span for each wing was used for the rTMS
administration. Motor threshold (MT) for each participant
was measured with an EMG Neurosign 400 at the
beginning of the first session. The intensity of the motor
threshold was established by the lowest strength needed
to elicit five motor evoked potentials within 10 trials. The
left DLPFC stimulation site was defined as the region
5 cm rostral in the same sagittal plane as the optimal site
for MT production in the right abductor pollicis brevis
(Rossini 1994).
The active stimulation condition was defined as
low-frequency rTMS with a stimulation intensity that was
110 % of motor threshold. The frequency of rTMS was
0.9 Hz over DLPFC (192 magnetic pulses per one
subtest). This stimulation procedure was within safety
guidelines (Wasserman 1998). The sham stimulation was
defined with a coil diverted by 90 degrees and a
stimulation intensity of 50 % of output of stimulator with
the same frequency, duration and position as in the active
rTMS. This method of sham stimulation was established
as adequate in the previous studies (Loo et al. 2000).
Immediately following the word or face stimuli
presentation the rTMS was finished and free recall of
words or recognition of faces was evaluated. The
investigator responsible for test presentation was blind to
the rTMS condition.

Statistical analysis
Due to the sample size and non-normal
distribution (Kolmogorov-Smirnov test) we used the non-
parametric Wilcoxon matched pairs test for within subject
comparison based on 10 differences in the sham and
active rTMS conditions in both the face-recognition and
the word-recall subtests. Regarding the a priori
hypothesis that low frequency would decrease the
memory recall we used one-tailed p values (confidence
interval, C.I. = 95 %).

Results

In our sample we did not find any significant
difference between results of recognition of faces in the
sham (median 22.50, C.I. 95 % = 24.0 - 25.11) and active
(median 22.50, C.I. 95 % = 20.42 - 23.58) stimulation
conditions (sum of signed ranks, W´= –2.0, p>0.5,
Fig. 1.). In the free recall of words we found significant
worsening under active (median 11.00, C.I. 95 % = 9.11 -
13.89) rTMS in comparison with sham conditions
(median 13.00, C.I. 95 % = 11.78 -16.02) (sum of signed
ranks, W = –18.0, p = 0.03, Fig. 1.).



Fig. 1. The influence of low-frequency 0.9 Hz rTMS (ACTIVE) in
comparison with control rTMS (SHAM) on memory for words and
faces. The active 0.9 Hz rTMS during encoding decreased word
recall (Wilcoxon matched pairs test, p ≤ 0,05, marked by *) but
not face recognition (p = n.s.).


Discussion

In our study we used low-frequency rTMS to
inhibit the cortical activity during memory encoding. The
inhibitory effect of low-frequency rTMS is hypothetically
attributed to the transsynaptic activation of GABAergic
inhibitory interneurons, the recurrent inhibition of the
targeted neurons through axonal collaterals (Pascual-
Leone et al. 1994, 1998), and long-term depression
phenomenon (Chen et al. 1997) or “disfacilitation” by
interference within the rTMS frequency and axonal inputs
with specific firing rate (Romero et al. 2002). Although
the inhibitory mechanism of low-frequency rTMS is not
fully elucidated, it produces a temporally distinct
decrease of cortical activity useful for evaluating the role
of a specific cortical area involved in cognition.
The negative results of low-frequency rTMS on
face recognition are in accordance with the neuroimaging
studies. It was shown that the fusiform gyrus shows a
greater response to faces than to other stimuli (Puce et al.
1995, 1996, Kanwisher et al. 1996, 1997). Memory
encoding of faces activates the left inferior temporal
cortex and bilateral mediotemporal cortex and other
regions distributed in the left and right dorsolateral as
orbitofrontal and premotor cortex (Kelly et al. 1996,
Haxby et al. 1996, 2000ab). In the face learning task, the

*
co
rr
ec
ta
ns
w
er
s
(m
ed
ia
n)
0
5
10
15
20
25
words
ACTIVE
words
SHAM
faces
ACTIVE
faces
SHAM
co
rr
ec
ta
ns
w
er
s
(m
ed
ia
n)

126 Škrdlantová et al. Vol. 54


sensory information is processed in a series of areas of
cortex including the unimodal association visual areas in
the inferotemporal cortex specifically concerned with
face recognition and is also conveyed to the hippocampus
and the remainder of the mediotemporal lobe (Martin et
al. 1996). Additionally, studies involving cortical lesions
support the importance of the temporal and parietal
cortex for the memory involved in face matching
(Ojemann et al. 1992, Sergent et al. 1992). The role of
prefrontal cortex in face encoding and recognition is less
understood and hemisphere lateralization seems to be
strongly dependent on the materials being encoded.
Encoding of unfamiliar faces produced right-dorsal
frontal activation, whereas encoding of words produced
left-lateralized homologous activation (Kelley et al.
1998). Some studies have found that the encoding of
faces also activates left prefrontal cortex (Haxby et al.
1996). Our findings support the evidence that (in contrast
to word encoding) left DLPFC specialization does not
play an essential role in face encoding and recognition
and that different brain areas (such as inferotemporal,
mediotemporal, parietal or right frontal) are more
responsible for this function.
Neuroimaging studies focused on verbal free
memory recall tasks have demonstrated that the left
dorsolateral frontal cortical areas (area 45, 46 and 9) are
involved in the encoding and recall of verbal information,
in addition to contributing to speech (Nyberg et al. 1996,
Kelley et al. 1998, Cabeza and Nyberg, 2000, Kim et al.
1999, Paulesu et al. 1993). Our finding of decreased word
encoding with 0.9 Hz rTMS over the left DLPFC are in
concordance with these neuroimaging studies and also
with the TMS “virtual” lesion trials supporting the role of
the left DLPFC in verbal memory function (Ferbert et al.
1991, Mull and Seyal 2001, Grafman and Wassermann
1994).
Our study has some methodological limitations.
The comparison of recognition of faces with the free-
recall task for words in this study cannot be taken as fully
equivalent. However the combination of these paradigms
is a way to balance the difficulty of both subtests and to
keep the same rTMS duration for interference with
memory and the same interstimulus interval (4.5 s) for
face and word presentation during the encoding. Based on
our preliminary experiment the word recognition
procedure similar to the face recognition was rejected for
its simplicity within the 2-min rTMS paradigm.
Nevertheless, due to the control (sham) condition and the
within subject design it is possible to compare the
influence of rTMS on the memory for faces and words,
albeit the word recall was more difficult as indicated by
the lower number of correct answers when compared to
face recognition.
Some of the inconsistency with previous studies
could result from the lack of synchronization of the
stimuli and the target objects as was established in the
spTMS studies. Our study was based on the rTMS design
which, in contrast to the spTMS, creates the spatially
demarcated period of decreased cortical activity during
task performance. It was presumed that stimulation at
regular intervals throughout the presentation of the target
stimuli was sufficient to reveal deficits, if indeed the left
DLPFC was involved in the processing of either type of
the test objects. The synchronization of the TMS pulses
and presentation of the target stimuli may contribute to a
more reliable effect of TMS on time convey interval and
sequence of memory encoding.
Our results support the theories of specialization
of the implicated cortical network and the role of left
DLPFC in verbal memory encoding but not in face
encoding and recognition. As a non-invasive tool, rTMS
is useful for cognitive brain mapping and the functional
localization of the category specific memory system.
These results appear to be of potential value in the future
detection of memory deficits associated with the
pathology of specific diseases.

Acknowledgements
This work was supported by the research projects VZ J
13/98: 111200005 MSMT Czech Republic, CNS
LN00B122 MSMT Czech Republic and grant NF/7578-3
MZ Czech Republic. Thanks to S. Veselková for assisting
with the study and P. Čelakovský M.D., CSc. for helping
with several aspects of the research process.

References

BARKER AT, JALINOUS R, FREESTON IL: Non-invasive magnetic stimulation of human motor cortex. Lancet
1 (8437): 1106-1107, 1985.
CABEZA R, NYBERG L: Imaging cognition II: an empirical review of 275 PET and FMRI studies. J Cogn Neurosci
12: 1-47, 2000.

2005 Influence of rTMS on Memory for Words and Faces 127

CHEN R, CLASSEN J, GERLOFF C, CELNIK P, WASSERMANN EM, HALLETT M: Depression of motor cortex
excitability by low-frequency transcranial magnetic stimulation. J Neurophysiol 80: 2870-2881, 1997.
FERBERT A, MUSSMANN N, MENNE A, BUCHNER H, HARTJE W: Short-term memory performance with
magnetic stimulation of the motor cortex. Eur Arch Psychiatry Clin Neurosci 241: 135-138, 1991.
FLETCHER PC, HENSON RN: Frontal lobes and human memory. Insights from functional neuroimaging. Brain 24:
849-881, 2000.
GRAFMAN J, WASSERMANN E: Transcranial magnetic stimulation can measure and modulate learning and
memory. Neuropsychologia 37: 159-167, 1999.
HAXBY JV, UNGERLEIDER LG, HORWITZ B, MAISOG JM, RAPOPORT SI, GRADY CL: Face encoding and
recognition in the human brain. Proc Natl Acad Sci USA 93: 922-927, 1996.
HAXBY JV, HOFFMAN EA, GOBBINI MI: The distributed human neural system for face perception. Trends Cogn
Sci 6: 223-233, 2000a.
HAXBY JV, PETIT L, UNGERLEIDER LG, COURTNEY SM: Distinguishing the functional roles of multiple regions
in distributed neural systems for visual working memory. Neuroimage 11,: 380-391, 2000b.
KANWISHER N, CHUN MM, MCDERMOTT J, LEDDEN P: Functional imaging of human visual recognition. Cogn
Brain Res 5: 55-67, 1996.
KANWISHER N, MCDERMOTT J, CHUN MM: The fusiform face area: a module in human extrastriate cortex
specialized for face perception. J Neurosci 17: 4302-4311, 1997.
KELLEY WM, MIEZEN FM, MCDERMOTT KB, BUCKNER RL, RAICHLE ME, COHEN NJ, OLLINGER JM,
AKBUDAK E, CONTURO TE, SNYDER AZ, PETERSEN SE: Hemispheric specialization in human dorsal
frontal cortex and medial temporal lobe for verbal and nonverbal memory encoding. Neuron 20: 927-936,
1998.
KIM JJ, ANDREASEN NC, O'LEARY DS, WISER AK, BOLES PONTO LL, WATKINS GL, HICHWA RD: Direct
comparison of the neural substrates of recognition memory for words and faces. Brain 122: 1069-1083, 1999.
LOO CK, TAYLOR JL, GANDEVIA SC, MCDARMONT BN, MITCHELL PB, SACHDEV PS: Transcranial
magnetic stimulation (TMS) in controlled treatment studies: are some "sham" forms active? Biol Psychiatry
47: 325-331, 2000.
MARTIN A, WIGGS CL, UNGERLEIDER LG, HAXBY JV: Neural correlates of category-specific knowledge.
Nature 379: 649-652, 1996.
MULL BR, SEYAL M: Transcranial magnetic stimulation of left prefrontal cortex impairs working memory. Clin
Neurophysiol 112: 1672-1675, 2001.
NYBERG L, MCINTOSH AR, CABEZA R, HABIB R, HOULE S, TULVING E: General and specific brain regions
involved in encoding and retrieval of events: what, where, and when. Proc Natl Acad Sci USA 93: 11280-1285,
1996.
OJEMANN JG, OJEMANN GA, LETTICH E: Neuronal activity related to faces and matching in human right
nondominant temporal cortex. Brain 115: 1-13, 1992.
PASCUAL-LEONE A, HALLETT M: Induction of errors in a delayed response task by repetitive transcranial magnetic
stimulation of the dorsolateral prefrontal cortex. NeuroReport 5: 2517-2520, 1994.
PASCUAL-LEONE A, VALLS-SOLE J, WASSERMANN EM, HALLETT M: Responses to rapid-rate transcranial
magnetic stimulation of the human motor cortex. Brain 117: 847-858, 1994.
PASCUAL-LEONE A, TORMOS JM, KEENAN JP, TARAZONA F, CANETE C, CATALA MD: Study and
modulation of cortical excitability with transcranial magnetic stimulation. J Clin Neurophysiol 15: 333-343,
1998.
PAULESU E, FRITH CD, FRACKOWIAK RSJ: The neural correlates of verbal component of working memory.
Nature 362: 342-344, 1993.
PUCE A, ALLISON T, GORE JC, MCCARTHY G: Face-sensitive regions in human extrastriate cortex studied by
functional MRI. J Neurophysiol 74: 1192-1199, 1995.

128 Škrdlantová et al. Vol. 54


PUCE A, ALLISON T, ASGARI M, GORE JC, MCCARTHY G: Differential sensitivity of human visual cortex to
faces, letter strings, and textures: a functional magnetic resonance imaging study. J Neurosci 16: 5205-5215,
1996.
ROMERO JR, ANSCHEL D, SPARING R, GANGITANO M, PASCUAL-LEONE A: Subthreshold low frequency
repetitive transcranial magnetic stimulation selectively decreases facilitation in the motor cortex. Clin
Neurophysiol 113: 101-107, 2002.
ROSSINI PM: Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and
procedures for routine clinical application: Report of IFCN Committee. Electroencephalogr Clin Neurophysiol
91: 79-92, 1994.
SERGENT J, OHTA S, MACDONALD B: Functional neuroanatomy of face and object processing. Brain 115: 15-36,
1992.
WASSERMANN EM: Risk and safety of repetitive transcranial magnetic stimulation: report and suggested guidelines
from the International Workshop on the Safety of Repetitive Transcranial Magnetic Stimulation, June 5-7,
1996. Electroencephalogr Clin Neurophysiol 108: 1-16, 1998.
WASSERMANN EM, LISANBY SH: Therapeutic application of repetitive transcranial magnetic stimulation: a review.
Clin Neurophysiol 112: 1367-1377, 2001.


Reprint requests
Dr. J. Horáček, Prague Psychiatric Centre, Ústavní 91, 181 00 Prague 8, Czech Republic. E-mail:
horacek@pcp.lf3.cuni.cz