Neuroscience Letters 482 (2010) 40–44
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/neulet
Exercising during learning improves vocabulary acquisition:
Behavioral and ERP evidence
Maren Schmidt-Kassowa,b,∗, Anna Kulkab, Thomas C. Gunterc, Kathrin Rothermichb, Sonja A. Kotzb
aInstitute of Medical Psychology, Johann Wolfgang Goethe University, Frankfurt, Germany
bMax Planck Institute for Human Cognitive and Brain Sciences, Research Group Neurocognition of Rhythm in Communication, Leipzig, Germany
cMax Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany
a r t i c l e i n f o
Received 23 March 2010
Received in revised form 18 June 2010
Accepted 29 June 2010
a b s t r a c t
Numerous studies have provided evidence that physical activity promotes cortical plasticity in the adult
brain and in turn facilitates learning. However, until now, the effect of simultaneous physical activity
(e.g. bicycling) on learning performance has not been investigated systematically. The current study
aims at clarifying whether simultaneous motor activity influences verbal learning compared to learning
in a physically passive situation. Therefore the learning behavior of 12 healthy subjects (4 male, 19–33
years) was monitored over a period of 3 weeks. During that time, behavioral and electrophysiological
in vocabulary tests when subjects were physically active during the encoding phase. Thus, our data
indicate that simultaneous physical activity during vocabulary learning facilitates memorization of new
© 2010 Elsevier Ireland Ltd. All rights reserved.
Marcus Tullius Cicero (106 B.C.–43 B.C.) used to say that it is exer-
cise alone that supports the spirits, and keeps the mind in vigor. Most
likely he would not have suspected that his statement is still rele-
vant more than 2000 years later. In fact, the influence of exercise
not only on physical health but also on cognition is a topic that
has re-entered the limelight in the last 10 years. The application of
new neuropsychological methods enables to track neuroplasticity
changes as a function of exercise. As a consequence, several animal
tic plasticity and in particular on the genesis of new neurons in the
adult mammalian brain [30,31,14,15,9,29,2]. In addition, there is
cumulating evidence on a biochemical level that physical exercise
which in turn should mediate the effects of exercise and cognition.
Recently, several studies have confirmed the close relationship
between physical activity and cognitive abilities not only in ani-
mals but also in humans (e.g. [4,3,11]). Thus, regular exercise has
been shown to prevent cognitive decline in elderly [4,11] which
ory tasks, reaction time, or vocabulary measurements in physically
active elderly compared to same age non-active participants .
∗Corresponding author at: Institute of Medical Psychology, Heinrich-Hoffmann-
Strasse 10, 60528 Frankfurt am Main, Germany. Tel.: +49 69 6301 6308;
fax: +49 69 6301 7606.
E-mail addresses: firstname.lastname@example.org,
email@example.com (M. Schmidt-Kassow).
However, studies on the relationship of exercise and cognition
in elderly vary in duration, type of physical activity, and inten-
sity. Nonetheless, they all indicate that physical activity positively
influences cognition [11,16] as shown in functional magnetic res-
onance imaging (MRI) and event-related potentials (ERP) studies.
For instance, Hillman et al. [13,12] showed that the amplitude of
the P3b is increased in physically active compared to non-active
participants indicating efficient allocation of attentional resources
mer group . On the other hand, MRI studies provide evidence
for increase in prefrontal and temporal gray matter volume .
However, not only the aging, but also the young brain profits from
itive correlation of exercise, learning, and intelligence in children
as well as young adults [24,19,12,27].
ical activity pushes brain activity, which in turn makes it an ideal
candidate to improve learning. However, studies investigating the
effect of different learning situations (physically active vs passive)
systematically are rather rare. One of them has been published
recently by Winter et al. . The authors investigated whether
physical activity prior to vocabulary learning accelerates the learn-
ing process in young athletic men. They demonstrated that short
but intensive training prior to a learning session results in the best
learning outcome. Hence, the study of Winter et al. convincingly
probed the effect of a single physical intervention on verbal learn-
ing. In the current experiment we aim to extend the results of
Winter et al. by (i) verifying whether the described effect is specific
0304-3940/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved.
M. Schmidt-Kassow et al. / Neuroscience Letters 482 (2010) 40–44
to athletic men, by (ii) looking at the long-term effect of regular
physical activity (in contrast to a single bout), by (iii) investigat-
ing whether physical activity during learning does also accelerate
the learning process, and by (iv) conducting a cross-language N400
priming experiment prior to and after the training to track for
changes in brain plasticity.
We address these aims for the following reasons: firstly, the
beneficial effect of the physical intervention as reported by Win-
ter et al. may be specific to young athletic men as they may be
particularly motivated to pass the physical intervention. In turn,
motivation and not exercise may be the primary factor that medi-
ates the learning outcome. Less athletic participants, however,
should be less motivated to pass a physical training. To test the
effect of exercise and not of motivation (which has also been
shown to be a modulating factor in learning ) we decided that
subjects should not be particularly interested in physical activity
but should not be obese. Secondly, we aimed to investigate the
long-term effect of regular physical activity and tracked partici-
pants’ learning curve over 3 weeks. Referring to the first research
question our assumptions were twofold: If the positive effect of
physical activity as described by Winter and colleagues is sim-
ply a matter of motivation due to the new environment (physical
activity prior to learning), this effect should diminish after a few
training sessions. If the benefit of physical activity is due to medi-
ating physiological factors, as e.g. increased regional blood flow
 or higher levels of learning mediating hormones (e.g. ),
the beneficial effect of learning should be stable during the 3
Thirdly, we were interested in the effect of simultaneous physi-
acute secretion of learning mediating hormones, and hence simul-
taneous learning may be maximally efficient. On the one hand,
it is conceivable that simultaneous movement and learning may
interfere with the encoding of new vocabulary.
Lastly, we were interested in the neural substrates underlying
different training effects. Thus, we conducted an N400 priming
experiment prior and after the whole learning period. Beyond
behavioral performance (vocabulary tests), ERPs permit to track
cognitive processes as they unfold in time and thus they iden-
tify mechanisms underlying the beneficial effects of simultaneous
physical activity on vocabulary learning.
These four issues are addressed in the current experiment that
combines electrophysiological and behavioral measurements as a
function of the learning outcome.
In the current experiment we tested 12 participants (8 female)
with a mixed factorial design and asked them to learn French
vocabulary. Participants were selected thoroughly to control for
confounding factors. Hence, all participants were native German
were left-handedness, practicing of any endurance sport including
We controlled very carefully for participants’ previous knowledge
of French. In doing so, we consequently excluded all participants
that had previously learned French at school or in private lessons.
Further exclusion criteria were extensive journeys to France (more
than 2 weeks).
group (no physical intervention). The respective group conditions
will be explained in more detail in the following. Participants in
both groups were paralleled according to age (mean Spinning: 25.2
years; mean Passive: 25.1 years), gender (i.e., 2 male and 4 female
participants in each group), and working memory capacity, i.e., all
participants were low-span readers according to the Reading Span
Test (equal or lower than 3.0). The Reading Span Test was orig-
inally developed by Daneman and Carpenter . For the current
experiment we used a German adaptation of the Reading Span Test
At the beginning of the study, all of the participants were
instructed to avoid changes in their physical activity level for the
duration of the experiment (3 weeks).
Within the 3 weeks, participants were asked to learn 80 French
words (40 nouns and 40 verbs). Participants underwent 3 indi-
vidual trainings sessions per week resulting in 9 learning sessions
in total. Each learning session lasted 30min during which partici-
pants listened to the 80 words twice. Thus, all 80 vocabulary pairs
were presented in French–German order before they were pre-
sented again in German–French order. Within the French–German
and the German–French block the order of vocabulary pairs was
randomized for each learning session. Within the 3 weeks of train-
ing participants listened each item 18 times in total. Stimuli were
presented auditorily via headphones (Sennheiser HD 202). The
loudness level was adjusted to the individual preference and kept
constant across all learning sessions. Both French and German
items were spoken by a female German native speaker, who was
a non-professional speaker but had a phonetic–linguistic back-
ground. All stimuli were normalized to an intensity level of 75dB
using the software PRAAT. The stimulus onset asynchrony (SOA)
of French–German vocabulary pairs amounted to 2s. The SOA
between successive vocabulary pairs was 6s. Thus, the onset of the
next item was maximally predictable. We excluded action verbs
to ensure that better performance of the cycling group was not
semantically induced .
In the Spinning group participants were instructed to cycle in
jects cycled at a speed of 60 rounds per minute (RPM), a pace that is
usually recommended to beginners in fitness centres. To acquaint
participants with this tempo, 42 sinusoidal tones with a frequency
of 0.5Hz were presented before the actual vocabulary presentation
was started. In the Spinning group we controlled for the partici-
130/78; SD: 13.4/11.3) three times in each training session. Partic-
ipants were instructed to exercise at a medium exertion level, i.e.,
they should breathe a little faster and feel a little warmer.
In the Passive group participants listened to the same acoustic
listening to the to-be-memorized words, imitating a traditional
classroom situation. The environment was kept constant across
was equipped identically except for the bicycle in the room of the
After every third learning session participants performed a
vocabulary test to assess their learning progress. Here, all par-
ticipants (Spinning and Passive group) were sitting at a table and
listened to all of the French items while they were asked to write
down the German translation. Response time was limited to 8s.
Participants performed three vocabulary tests in total.
to track for changes in brain plasticity. We conducted a cross-
language N400 priming ERP (event-related potentials) experiment
 prior to and after the 3-week training to obtain a sensitive mea-
surement to track learning differences between groups that may
not be detected by behavioral measures alone. Here, participants
were tested in a dimly illuminated sound-attenuating booth, were
and blink as little as possible. They listened to French–German and
formed a lexical decision task on the second word of each word
pair. In order to permit the lexical decision task we created French
and German pseudowords that corresponded in syllable number
M. Schmidt-Kassow et al. / Neuroscience Letters 482 (2010) 40–44
Fig. 1. Behavioral performance plotted separately for each week.
and complexity. During the experiment, each trial was introduced
by a visual cue (asterisk) on the centre of a computer screen. After
between the first and the second word in each trial was kept con-
stant, namely 500ms. Immediately after the offset of the second
stimulus, participants were asked to perform the lexical deci-
sion. The next trial started 500ms after the participant’s response
(button press, counterbalanced for correct and incorrect). All 320
experimental pairs (40 per condition) were presented auditory via
two loud speakers in pseudorandomized order. The experimen-
tal trials were presented in eight blocks of approximately 5min
each. After the fourth block participants were offered a break.
An N400 effect was expected for French–German mismatching,
e.g. *gateau–Hund (English translation: cake–dog), compared to
French–German matching, e.g. chien–Hund (English translation:
dog–dog) word pairs that should be modulated as a function of
learning type (Spinning vs Passive). The N400 is a negative compo-
nent which is most pronounced at centro-parietal electrodes and
typically elicited around 400ms after the presentation of a critical
item . This component is affected by a large number of manipu-
lations including frequency of a particular item, close probability,
repetition, and semantic priming. If learning was successful, the
acoustic presentation of the French word should prime the corre-
sponding German item and hence reduce the N400. Mismatching
word–word pairs on the other hand should induce a larger N400
as participants’ lexical retrieval was misleading. Hence the differ-
ence in the electrophysiological response to matching compared
to mismatching items should increase as a function of successfully
We computed two-tailed t-tests for each vocabulary test to test
for group differences in performances at test day. We did not find
a significant main effect of group, probably due to a lack of power,
however, planned comparisons showed that members of the Spin-
ning group performed significantly better than members of the
Passive group at each day of testing (week 1: Spinning vs Pas-
sive: t(1,5)=6.47, p=.001; week 2: Spinning vs Passive: t(1,5)=4.46,
p<.01; week 3: Spinning vs Passive: t(1,5)=3.95, p=.01), see Fig. 1.
The EEG was recorded from 29 scalp sites by means of Ag/AgCl
electrodes mounted in an elastic cap (Electro-Cap Inc., n.d.) accord-
as ground, the left mastoid as on-line reference (recordings were
were kept below 5k?. In order to control for eye movements,
a horizontal and a vertical EOG was recorded. EEG and EOG sig-
nals were digitized on-line with a sample frequency of 250Hz. An
anti-aliasing filter of 67.5Hz was applied during recording.
Individual EEG recordings were scanned for artifacts such as
electrode drifting, amplifier blocking, muscle artifacts, eye move-
ments, or blinks by means of a rejection algorithm as well as on
basis of visual inspection. Epochs lasted 200ms before onset of
the second item up to 1500ms after the critical item. All contam-
inated trials were rejected and the remaining trials (i.e., session
1: French–German match: 28.3; French–German mismatch: 29.8;
26.6) were averaged per participant, condition, and electrode site
with a 200ms pre-stimulus baseline. For graphical display only,
data were filtered off-line with a 7Hz low pass filter. All following
statistical evaluations were carried out on unfiltered ERP data.
The ERP results (Fig. 2) further support that cycling dur-
ing encoding enhances vocabulary learning. We conducted a
repeated-measures ANOVA with the within-subject factor con-
dition (match/mismatch) and the between-subject factor group
(Spinning/Passive) for each experimental session. Both ANOVAs
included right centro-parietal electrodes (C4, CP6, P4, CZ, PZ) in
a time window from 350 to 750ms measured from the onset of the
second word in pair.
In the first experimental session, both groups (Spinning and Pas-
sive) failed to show an N400 effect in response to mismatching
as compared to matching word pairs as revealed by a non-
significant main effect for condition (F(1,10)=.99, p>.3). This result
ensures that participants’ performance was at the same level at
the beginning of the learning period. Hence, the first session was
an additional control condition and we checked each data set care-
has shown an N400 prior to the respective training. In fact we had
to exclude one potential participant on the basis of this first EEG
However, in the second session, after three weeks of vocabu-
effect (mean: −4.35?V) over central and right hemispheric elec-
trode sites in response to the prime-target mismatch condition
(*gateau–Hund) compared to the Passive group (mean: −3.21?V).
Here, the omnibus ANOVA resulted in a significant main effect for
dition effect between the Spinning and the Passive group yielded a
significant effect for condition only for the Spinning group (F(1,5):
22.97, p<.01) but not for the Passive group (F(1,5): 5.85, p=.06) in
the post-training session.
In the current study we investigated whether long-term simul-
taneous physical activity positively influences the memorization
of foreign language vocabulary and whether better performance
is associated with plasticity changes as evidenced by changes in
the N400 amplitude. This ERP component is sensitive to learn-
ing induced changes in cortical plasticity [25,17] and proficiency
in a second language [18,1]. We hypothesized that members of
the Spinning group should benefit from the simultaneous physical
activity as reflected in performance and ERP responses. Our results
are in line with this hypothesis. Members of the Spinning group
showed significantly better performance in the vocabulary tests at
each testing day. Furthermore they showed a larger N400 effect
as compared to those participants who learned vocabulary in the
physically passive condition.
The current results are very promising given the fact that a pos-
itive influence of simultaneous physical activity on learning has
never been demonstrated before. Although previous studies pro-
vided evidence for a beneficial effect of high-intensity physical
activity on the subsequent learning outcome, it has never been
further pushes or hinders mnemonic functions. Here, we provide
first evidence that simultaneous physical activity positively influ-
ences the memorization of new vocabulary, a result that goes hand
in hand with plasticity changes as evidenced in an enhanced N400
amplitude compared to the control group. Moreover, in the cur-
rent study we included female and male participants and we paid
attention that participants were not particularly athletic. We thus
M. Schmidt-Kassow et al. / Neuroscience Letters 482 (2010) 40–44
Fig. 2. N400 elicited by the second word in the match and the mismatch condition splitted by group and experimental session. Waveforms show the average for matching
and mismatching items from 200ms prior to the item onset up to 1500ms prior to the 3 weeks training (right hand site) and after the 3 weeks training (left hand site).
and athletic men to young adults. This implies that motivational
aspects, such as the motivation to complete a 30min bout of exer-
cise, cannot completely explain the positive influence of physical
activity on learning. We found significant differences between the
two groups even after three weeks of learning, i.e., after nine learn-
ing sessions. This persistent finding suggests that the novelty of
the new learning experience alone cannot account for the benefi-
cial effect of sport. We therefore exclude pure motivational aspects
(i.e., due to the novel learning condition) as a primary cause for this
improved learning behavior.
Admittedly, our sample size is rather small. Thus, further exper-
iments are inevitable to substantiate this initial effect. In addition,
the modulating biochemical factor that drives improved perfor-
mance needs to be explored. In this context, the brain derived
neurotrophic factor (BDNF) has been extensively discussed as it
induces changes in cortical plasticity and is involved in mnemonic
processes [10,7]. Thus, enhanced BDNF blood serum levels during
learning predict better performance in subsequent test sessions
. On the basis of the current study we cannot conclude how
physiological dynamics improve performance, but we currently
address this open issue by tracking participants’ BDNF level in
serum in our lab.
ences of simultaneous bicycling positively on vocabulary learning
as evidenced by brain imaging and behavioral data. Although
the underlying physiological mechanism still need to be further
addressed, we conclude that simultaneous physical activity boosts
learning capacities even in young adults that are on the peak of
their cognitive health .
The authors would like to thank Heike Boethel, Nadine Klip-
phahn, and Kathrin Müller for help in data acquisition. The first
author was supported by a grant from the German Research Foun-
dation (DFG SCHM 2693/1-1).
M. Schmidt-Kassow et al. / Neuroscience Letters 482 (2010) 40–44
 R.P. Alvarez, P.J. Holcomb, J. Grainger, Accessing word meaning in two lan-
guages: an event-related brain potential study of beginning bilinguals, Brain
Lang. 87 (2003) 290–304.
 B.R. Christie, B.D. Eadie, T.S. Kannangara, J.M. Robillard, J. Shin, A.K. Titterness,
Exercising our brains: how physical activity impacts synaptic plasticity in the
dentate gyrus, Neuromolecular Med. 10 (2008) 47–58.
D.X. Marquez, L. Hu, A.F. Kramer, Aerobic exercise training increases brain
G.J. Jerome, D.X. Marquez, S. Elavsky, Cardiovascular fitness, cortical plasticity,
and aging, Proc. Natl. Acad. Sci. U.S.A. 101 (2004) 3316–3321.
guage Comprehension, Palgrave MacMillan, San Diego, 2004, pp. 187–206.
 M. Daneman, P.A. Capenter, Individual differences in working memory and
reading, J. Verbal Learn. Verbal Behav. 19 (1980) 450–466.
 M.F. Egan, M. Kojima, J.H. Callicott, T.E. Goldberg, B.S. Kolachana, A. Bertolino,
E. Zaitsev, B. Gold, D. Goldman, M. Dean, B. Lu, D.R. Weinberger, The BDNF
val66met polymorphism affects activity-dependent secretion of BDNF and
human memory and hippocampal function, Cell 112 (2003) 257–269.
 M. Endres, K. Gertz, U. Lindauer, J. Katchanov, J. Schultze, H. Schröck, G. Nick-
enig, W. Kuschinsky, U. Dirnagl, U. Laufs, Mechanisms of stroke protection by
physical activity, Ann. Neurol. 54 (2003) 582–590.
 K. Fabel, G. Kempermann, Physical activity and the regulation of neurogenesis
in the adult and aging brain, Neuromolecular Med. 10 (2008) 59–66.
 A.R. Hariri, T.E. Goldberg, V.S. Mattay, B.S. Kolachana, J.H. Callicott, M.F. Egan,
D.R. Weinberger, Brain-derived neurotrophic factor val66met polymorphism
affects human memory-related hippocampal activity and predicts memory
performance, J. Neurosci. 23 (2003) 6690–6694.
 C.H. Hillman, K.I. Erickson, A.F. Kramer, Be smart, exercise your heart: exercise
effects on brain and cognition, Nat. Rev. Neurosci. 9 (2008) 58–65.
 C.H. Hillman, E.M. Snook, G.J. Jerome, Acute cardiovascular exercise and exec-
utive control function, Int. J. Psychophysiol. 48 (2003) 307–314.
 C.H. Hillman, E.P. Weiss, J.M. Hagberg, B.D. Hatfield, The relationship of age and
cardiovascular fitness to cognitive and motor processes, Psychophysiology 39
 G. Kempermann, Why new neurons? Possible functions for adult hippocampal
neurogenesis, J. Neurosci. 22 (2002) 635–638.
 G. Kempermann, S. Jessberger, B. Steiner, G. Kronenberg, Milestones of neu-
ronal development in the adult hippocampus, Trends Neurosci. 27 (2004)
 A.F. Kramer, K.I. Erickson, Capitalizing on cortical plasticity: influence of phys-
ical activity on cognition and brain function, Trends Cogn. Sci. 11 (2007)
 J. McLaughlin, L. Osterhout, A. Kim, Neural correlates of second-language word
learning: minimal instruction produces rapid change, Nat. Neurosci. 7 (2004)
 K.J. Midgley, P.J. Holcomb, J. Grainger, Language effects in second language
learners and proficient bilinguals investigated with event-related potentials,
J. Neurolinguist. 22 (2009) 281–300.
 A.C. Pereira, D.E. Huddleston, A.M. Brickman, A.A. Sosunov, R. Hen, G.M. McK-
hann, R. Sloan, F.H. Gage, T.R. Brown, S.A. Small, An in vivo correlate of
exercise-induced neurogenesis in the adult dentate gyrus, Proc. Natl. Acad. Sci.
U.S.A. 104 (2007) 5638–5643.
 S. Pietropaolo, Y. Sun, R. Li, C. Brana, J. Feldon, B.K. Yee, The impact of volun-
Brain Res. 192 (2008) 42–60.
 R.T. Pivik, R.J. Broughton, R. Coppola, R.J. Davidson, N. Fox, M.R. Nuwer, Guide-
lines for the recording and quantitative analysis of electroencephalographic
activity in research contexts, Psychophysiology 30 (1993) 547–558.
 S.-A. Rüschemeyer, M. Brass, A.D. Friederici, Comprehending prehending: neu-
 T.A. Salthouse, J.R. Nesselroade, D.E. Berish, Short-term variability in cognitive
performance and the calibration of longitudinal change, J. Gerontol. B Psychol.
Sci. Soc. Sci. 61 (2006) P144–P151.
 B.A. Sibley, J.L. Etnier, The relationship between physical activity and cognition
in children: a meta-analysis, Pediatr. Exerc. Sci. 15 (2003) 243–256.
 M. Stein, T. Dierks, D. Brandeis, M. Wirth, W. Strik, T. Koenig, Plasticity in the
adult language system: a longitudinal electrophysiological study on second
language learning, Neuroimage 33 (2006) 774–783.
 K. Steinhauer, Electrophysiological correlates of linguistic processing during
reading locally ambiguous relative clauses, Master’s Thesis, Free University
 S. Stroth, S. Kubesch, K. Dieterle, M. Ruchsow, R. Heim, M. Kiefer, Physical
fitness, but not acute exercise modulates event-related potential indices for
executive control in healthy adolescents, Brain Res. 1269 (2009) 114–124.
 W.-T Tseng, N. Schmitt, Toward a model of motivated vocabulary learning: a
structural equation modeling approach, Lang. Learn. 58 (2008) 357–400.
 M. Uda, M. Ishido, K. Kami, M. Masuhara, Effects of chronic treadmill running
on neurogenesis in the dentate gyrus of the hippocampus of adult rat, Brain
Res. 1104 (2006) 64–72.
 H. van Praag, G. Kempermann, F.H. Gage, Running increases cell proliferation
and neurogenesis in the adult mouse dentate gyrus, Nat. Neurosci. 2 (1999)
 H. van Praag, Neurogenesis and exercise: past and future directions, Neuro-
molecular Med. 10 (2008) 128–140.
incremental exercise to exhaustion in humans, Brain Res. 1121 (2006) 59–65.
 B. Winter, C. Breitenstein, F.C. Mooren, K. Voelker, M. Fobker, A. Lechtermann,
K. Krueger, A. Fromme, C. Korsukewitz, A. Floel, S. Knecht, High impact running
improves learning, Neurobiol. Learn. Mem. 87 (2007) 597–609.
ical activity and cognitive decline in elderly women: women who walk, Arch.
Intern. Med. 161 (2001) 1703–1708.