R

Asd

CD

h

••••

a

ARR1AA

KAEESTE

1

e

2T

h0

Behavioural Brain Research 281 (2015) 267–275

Contents lists available at ScienceDirect

Behavioural Brain Research

jou rn al hom epage: www.elsev ier .com/ locate /bbr

esearch report

cute aerobic exercise enhances attentional modulation ofomatosensory event-related potentials during a tactileiscrimination task

hristina Popovich, W. Richard Staines ∗

epartment of Kinesiology, University of Waterloo, 200 University Ave. W, Waterloo, Ontario N2L 3G1, Canada

i g h l i g h t s

Exercise modulates neural processing associated with selective tactile attention.ERPs were enhanced following an acute bout of aerobic exercise.Aerobic exercise facilitated gating of task-irrelevant tactile stimuli.Relevant sensory signals were enhanced at later stages of processing.

r t i c l e i n f o

rticle history:eceived 10 October 2014eceived in revised form5 December 2014ccepted 19 December 2014vailable online 27 December 2014

eywords:ttentionvent-related potentialsRPsomatosensory cortexactilexercise

a b s t r a c t

Neuroimaging research has shown that acute bouts of moderate intensity aerobic exercise can enhanceattention-based neuronal activity in frontal brain regions, namely in the prefrontal cortex (PFC), as wellas improve cognitive performance. The circuitry of the PFC is complex with extensive reciprocal cor-ticocortical and thalamocortical connections, yet it remains unclear if aerobic exercise can also assistattentional control over modality-specific sensory cortices. To test this, we used a tactile discriminationtask to compare tactile event-related potentials (ERPs) prior to and following an acute bout of moderateintensity aerobic exercise. We hypothesized that exercise preceding performance of the task would resultin more efficient sensory gating of irrelevant/non-attended and enhancement of relevant/attended sen-sory information, respectively. Participants received vibrotactile stimulation to the second and fifth digiton the left hand and reported target stimuli on one digit only. ERP amplitudes for the P50, P100, N140and long latency positivity (LLP) were quantified for attended and non-attended trials at FC4, C4, CP4and P4 while P300 amplitudes were quantified in response to attended target stimuli at electrodes FCZ,CZ and CPZ. Results showed no effect of attention on the P50, however, both P100 and LLP amplitudeswere significantly greater during attended, task-relevant trials, while the N140 was enhanced for non-attended, task-irrelevant stimuli. Moreover, unattended N140 amplitudes over parietal sites contralateral

to stimulation were significantly greater post-exercise versus pre-exercise, while LLP modulation variedwith greater unattended amplitudes post-exercise over frontal sites and greater attended amplitudespost-exercise over parietal sites. These results suggest that a single session of moderate intensity aerobicexercise facilitated the sensory gating of task-irrelevant tactile stimuli so that relevant sensory signalscould be enhanced at later stages of somatosensory processing in healthy young adults.

. Introduction

Selective attention is commonly referred to as a higher-orderxecutive process that requires conscious allocation of cognitive

∗ Corresponding author at: Department of Kinesiology, University of Waterloo,00 University Ave W, Waterloo, Ontario N2L 3G1, Canada.el.: +1 519 888 4567x37756; fax: +1 519 885 0470.

E-mail address: rstaines@uwaterloo.ca (W.R. Staines).

ttp://dx.doi.org/10.1016/j.bbr.2014.12.045166-4328/© 2014 Elsevier B.V. All rights reserved.

© 2014 Elsevier B.V. All rights reserved.

resources toward relevant information in the presence of envi-ronmental distracters. Attention-related neuroimaging studieshave shown that attending to task-relevant sensory informationcan activate a widespread neural network consisting of corticaland subcortical structures [1,2], as well as enhance neuronalresponses in modality-specific sensory cortices [3–5]. It is thought

that attentional control is mediated by the prefrontal cortex (PFC)which operates as an inhibitory-control filtering mechanism, sup-pressing irrelevant sensory information so that relevant sensorysignals can be amplified and passed on for further processing

2 ural B

ifirmidratcopssstntsa

balMcaeusbiacmhogemooomsaosfoauipTscoio

eaaea

68 C. Popovich, W.R. Staines / Behavio

n modality-specific brain regions [6,7]. Indeed, neuroimagingndings in healthy adults have shown attention-related neu-onal enhancement of task-relevant tactile stimuli in frontal andodality-specific somatosensory regions [8,9], while studies

n patients with focal frontal lobe lesions have found that PFCamage produces disinhibition of distractor stimuli during aange of sensorimotor and cognitive processes [10]. Furthermore,

recent imaging study in healthy young adults showed thatransient disruption to neuronal activity in dorsolateral PFC viaontinuous theta burst stimulation (cTBS), produced disinhibitionf task-irrelevant stimuli at early and later stages of somatosensoryrocessing during a tactile discrimination task [11]. A follow-uptudy in PFC patients using the same tactile discrimination taskhowed similar aberrations in the neuronal profile of some of theame early and late somatosensory ERPs found following cTBS tohe dorsolateral PFC [12]. Collectively, these studies support theotion that the PFC has a critical role in mediating attentional con-rol over incoming sensory information by suppressing irrelevantensory signals so that attended task-relevant information may bemplified in modality-specific sensory regions.

Notably, a growing body of evidence suggests that a singleout of acute aerobic exercise improves cognitive performancend produces neurophysiological changes, particularly in frontalobe regions, during tasks requiring selective attention [13–15].

oreover, studies examining acute exercise-induced effects onognition report that the greatest increases in neuronal activityre found in frontal regions following moderate bouts of aerobicxercise when tasks requiring higher-order attentional control aretilized [16–19]. These exercise effects on attentional processingupport the inverted U-shape arousal hypothesis first proposedy Yerkes and Dodson (1908), which states that as arousal states

ncrease with physical exertion, cognitive performance improves ton optimal point after which further increases in physical exertionause decreased arousal levels resulting in decrements in perfor-ance [20]. However, the majority of human neuroimaging studies

ave employed psychological tasks tailored to examine the effectsf exercise on PFC function (i.e. the Erikson Flanker, Stroop, oro/nogo tasks). As a result, it remains unclear if acute aerobicxercise modulates cortical activity downstream from the PFC inodality-specific sensory regions. Investigating the effects of aer-

bic exercise on frontoparietal interactions is an important avenuef the exercise and cognition research to pursue since the circuitryf the PFC is complex with reciprocal corticocortical and thala-ocortical connections involved in modulating modality-specific

ensory regions via attentional mechanisms. Determining whethercute bouts of aerobic exercise can improve attentional regulationver modality-specific sensory cortices provides an additional per-pective regarding how exercise may transiently improve cognitiveunction perhaps via more efficient modulation between cortic-cortical networks. Notably, an EEG study performed in soccerthletes and non-athletes using a lower limb somatosensory stim-lation oddball task showed that the athletic group demonstrated

ncreased P300 amplitudes and decreased latencies over central-arietal electrode sites compared to the non-athletic group [21].hese findings imply that long-term physical activity that requireselective attention and skilled motor responses may induce plastichanges in somatosensory processing during the execution of goal-riented behaviors [21]. However, no exercise intervention wasmplemented in this study, thus a causal link between the effectsf exercise on somatosensory processing cannot be determined.

The purpose of the present study was to use EEG and a well-stablished tactile discrimination task to examine whether an

cute bout of moderate intensity aerobic exercise, would enhancettention-based modulation of somatosensory ERPs generated atarly and later stages of somatosensory processing in healthy youngdults. Based on the findings of Bolton and Staines [11], our first

rain Research 281 (2015) 267–275

hypothesis was that early and later somatosensory ERPs wouldbe modulated by attentional relevance. Specifically, we hypoth-esized that an acute bout of moderate intensity aerobic exercisepreceding performance of a tactile discrimination task would resultin more efficient sensory-gating of irrelevant/non-attended andenhancement of relevant/attended sensory information. Therefore,improvements in sensory-gating would be observed as increasedneural suppression of task-irrelevant stimuli and enhancement ofrelevant sensory information post-relative to pre-exercise.

2. Methods

2.1. Participants

EEG was collected from 16 healthy participants (aged 21–28,6 males). One participant was excluded due to the absence ofclearly defined somatosensory event-related potentials of interest(i.e. P50, P100, N140, LLP components). The final sample consistedof 9 females and 6 males (mean age = 25.2 years). Participants wereself-reported right-handed individuals with no medical history ofany major neurological illness, and no heart or blood pressureconditions that could be exacerbated with aerobic exercise. Eachparticipant provided informed consent and confirmed that theywere physically able to engage in 20 min of moderate intensityaerobic activity on a recumbent bicycle. Experimental procedureswere approved by the University Of Waterloo Office Of ResearchEthics.

2.2. Tactile Odd-Ball Task

Subjects were seated in a sound-attenuating booth (Indus-trial Acoustics, 120A, NY), facing a blank computer screen andinstructed to look directly ahead throughout testing. Tactile stimuliwere delivered via 2 blunt plastic probes contacting the fingertipsof the second and fifth digits on the left hand as depicted in Fig. 1.These probes (approximately 1 cm diameter) were vibrated usingpiezo-electric actuators at a rate of 25 Hz for 125 ms during eachstimulus. Vibrotactile stimulation was delivered by digitally gen-erated waveforms converted to an analog signal (DAQCard 6024E,National Instruments, Austin, TX, USA) and then amplified (Bryston2B-LP, Peterborough, ON, Canada). The amplitude of vibration wasset so that minimal values (i.e. small amplitude targets) exceededsensory threshold for all subjects. These stimuli were presentedrandomly to each finger (but never simultaneously) with randominterstimulus intervals in the range of 500–1500 ms. An oddballparadigm was employed whereby 17% of the trials for each fingerconsisted of the deviant stimulus. These deviant stimuli repre-sented the targets to which attention was paid and the amplitudeof these targets was set to 25% the standard (non-target) ampli-tude (i.e. deviant stimuli, 105% of perceptual threshold; standardstimuli, 120% perceptual threshold). The two experimental con-ditions required subjects to either (a) attend to the second digit(Attend D2; Attend Index) on the left hand while ignoring all stimulidelivered to the fifth digit, or (b) attend to the fifth digit (Attend D5;Ignore Index) on the left hand while ignoring all stimuli to the sec-ond digit. Subjects reported target stimuli on the attended fingerby pressing a button with the right hand. Subjects were instructedto gently rest their hand on the probes to ensure consistent handpressure throughout testing. Headphones delivering white noisewere worn throughout the experiment to block sound from thevibration device. Trials requiring attention to either D2 or D5 were

randomly presented to subjects in 3 min blocks followed by a restperiod of approximately 1 min. There were a total of 6 testingblocks with 3 blocks attending to D2 and 3 blocks attending to D5.Approximately 600 standard (non-target) stimuli were applied

C. Popovich, W.R. Staines / Behavioural Brain Research 281 (2015) 267–275 269

rotac

tc

2

att(mawicpacbaswdd

2

itu(SswatfiSc(iaEva

Fig. 1. Experimental paradigm. This figure depicts how participants received vib

o each digit over the entire experiment with each testing blockonsisting of approximately 100 standard and 20 target stimuli.

.3. Exercise protocol

Upon task completion, participants were seated comfortably on cycle ergometer while a measure of their resting heart rate wasaken using a Polar heart rate monitor. Participants were instructedo begin pedaling on the ergometer to reach their target heart rate60% of age-predicted maximum heart rate; [220 − age] × 0.60), and

aintain this exercise intensity for the remainder of the 20 minerobic session. For each participant, the ergometer resistance levelas always set to its minimum value (level 1). To ensure that partic-

pants adhered to the exercise protocol, frequent heart rate monitorhecks were recorded, as well as two self-reported measures of thearticipant’s perceived rating of exertion were recorded after 10nd 20 min of exercise using the 10 point Borg Scale. After the exer-ise session, participants were seated back into the sound proofooth where electrode impedances were checked to ensure thatll recorded sites were below 5 k� and a final heart rate mea-ure was recorded to ensure that participants’ heart rates wereithin 10% their original resting heart rate (approximately 5 minuration). Participants were then instructed to repeat the tactileiscrimination task again.

.4. Recording and analysis

EEG data was recorded from 32 electrode sites according to thenternational 10–20 system for electrode placement and referencedo bilateral mastoids. All channel recordings had impedance val-es below 5kohms. EEG data were amplified (20,000×), filteredDC-200 Hz, 6 dB octave roll-off) and digitized (1000 Hz, SynAmps2,can 4.5, Compumedics Neuroscan, Charlotte, NC) before beingtored for off-line analysis. Somatosensory ERPs to tactile stimuliere averaged relative to a 100 ms pre-stimulus baseline for each

ttention condition. Data were band-pass filtered (1–30 Hz) andrials with artifacts (i.e. eye blinks, muscle activity) were identi-ed by visual inspection and were excluded from further analysis.ince stimulation of the fifth digit did not result in consistentlylear ERP components, analysis was restricted to non-target stimulii.e. large amplitude vibrations) delivered to the second digit (i.e.ndex finger). Several somatosensory ERP components were evalu-

ted in this study focusing on the peak amplitude for each definedRP component. Somatosensory ERPs were measured from indi-idual participant averages for each task condition. Mean ERPmplitudes were computed for each subject within specified time

tile stimulation to their second digit (D2) or the fifth digit (D5) on the left hand.

windows selected around the post stimulus latencies of eachERP component of interest: P50 (40–70 ms), P100 (85–125 ms),N140 (125–165 ms), LLP (175–250 ms), and the P300 (300–600 ms).Clearly defined components and peaks were required for inclu-sion. Separate two-way ANOVAs with factors attention (Attendindex, Ignore index) and time (Pre-Exercise, Post-Exercise) werecalculated for each component of interest at the electrode sitesas follows: P50 (C4, CP4), P100 (P4, CPZ, P3), N140 (FC4, CP4, P4),LLP (FC4, CP4, P4) and P300 (FCZ, CZ, CPZ). Pre-planned contrastswere performed to test the hypotheses that ERP modulation wouldshow greater suppression of task-irrelevant distractor stimuli (i.e.Ignore Index conditions) and greater enhancement of task-relevantstimuli (Attend Index conditions), post-exercise relative to pre-exercise.

The P50 ERP component has been shown to be generated in theprimary somatosensory cortex (SI) [22], while studies examiningearly somatosensory evoked potentials (SEPs) typically show morerobust over central-parietal electrodes contralateral to stimulationin SI areas [23,24], thus the P50 was analyzed at electrode sitesC4 and CP4 positioned over SI, contralateral to vibrotactile stim-ulation. We analyzed the P100 component at electrode sites P4,CPZ, and P3 since the P100 amplitude has known neural genera-tors in bilateral secondary somatosensory cortices (SII) [24], withmaximal activation typically seen in contralateral posterior pari-etal sites, but with enhancements also extending to ipsilateral andfrontal sites [24–26]. For the later components, both the N140 andthe LLP have shown increased activation in frontal regions duringattentionally-demanding tasks resulting in continued engagementof modality-specific cortices [27], thus we analyzed these ERPs atfrontoparietal electrode sites contralateral to stimulation (FC4, CP4,P4). Lastly, attention-based modulation of the P300 componenttypically occurs at fronto-central electrodes sites in response totask-relevant target stimuli [28], thus for this component, we ana-lyzed electrode sites FCZ, CZ, CPZ in response to target stimuli. Here,we time-locked and averaged somatosensory ERPs to the targetstimuli (i.e. smaller vibrations) delivered to the index finger duringthe Attend Index conditions only, and a one-tailed paired t-test wasperformed to test whether exercise would increase attention-basedmodulation of this somatosensory ERP.

Behavioral performance was evaluated by determining thenumber of targets hit relative to the overall number of targets thatwere presented for each subject. This success rate was expressedas a percentage and compared using a two-way repeated measuresANOVA with factors attention (Attend Index, Ignore Index), and

time (Pre-Exercise, Post-Exercise). The significance level was set atp ≤ 0.05 for all comparisons.

270 C. Popovich, W.R. Staines / Behavioural Brain Research 281 (2015) 267–275

Fig. 2. Grand averaged waveforms. Grand average waveforms all for conditions are shown for: parietal electrode sites contralateral to vibrotactile stimulation (CP4, P4)and a contralateral frontal site (FC4). (B) Tactile ERP traces of interest (P100, N140, and LLP) for electrode sites overlying frontal and somatosensory regions contralateral tos ditiona

3

3

tiPanwppeF

timulation. Black solid and dashed traces show Attend Index and Ignore Index connd Ignore Index conditions, respectively, following exercise.

. Results

.1. Early ERP components: P50 & P100

Fig. 2 shows waveforms at electrode sites contralateral to vibro-actile stimulation (i.e. FC4, CP4, and P4) with the components ofnterest specified (i.e. the P50, P100, N140, and LLP). Results for the50 amplitude revealed no main effects or significant interactiont any of the electrode sites analyzed. Results for the P100 compo-ent showed a main effect of attention whereby P100 amplitudesere enhanced by task-relevant stimuli (i.e. Attend Index) com-

ared to task-irrelevant stimuli (i.e. Ignore Index) (P4: F1,14 = 14.0,

= 0.001; CPZ: F1,14 = 6.73, p = 0.02; P3: F1,14 = 9.38, p = 0.01). How-ver, no other P100 results reached statistical significance (refer toig. 4A).

s, respectively, prior to exercise. Gray solid and dashed traces show Attend Index

3.2. Later ERP components: N140, LLP, & P300

A main effect of attention was found for the N140 compo-nent at all electrode sites analyzed whereby N140 amplitudeswere increased during task-irrelevant conditions (i.e. when tac-tile stimuli were delivered to the index and attention wasdirected toward another spatial location - Ignore Index con-dition) (FC4: F1,14 = 14.35, p = 0.002; CP4: F1,14 = 9.58, p = 0.01;P4 F1,14 = 8.49, p = 0.01). A significant attention*time interac-tion was also found for the N140 amplitude at P4 (F1,14 = 9.96,p = 0.01), and a trend toward significance was found at CP4

(F1,14 = 3.85, p = 0.07). Pre-planned contrasts at these electrodesites revealed that the amplitude of the N140 component wassignificantly greater post-relative to the pre-exercise sessionwhen stimuli delivered to the index finger were irrelevant

C. Popovich, W.R. Staines / Behavioural Brain Research 281 (2015) 267–275 271

F arget

c ed traG ctivel

(F

aetCasCpaetpat

ig. 3. P300 Grand averaged waveforms. Grand average waveforms time-locked to tomponent at midline electrodes FCZ, CZ, CPZ for all conditions. Black solid and dashray solid and dashed traces show Attend Index and Ignore Index conditions, respe

P4: F1,14 = 18.46, p = 0.001; CP4: F1,14 = 4.62, p = 0.05) (refer toig. 4B).

A main effect of attention was found for the LLP componentt all electrode sites measured whereby LLP amplitudes werenhanced by task-relevant stimuli (i.e. Attend Index) compared toask-irrelevant stimuli (i.e. Ignore Index) (FC4: F1,14 = 5.67, p = 0.03;P4: F1,14 = 11.45, p = 0.01; P4: F1,14 = 10.23, p = 0.01). Significantttention × time interactions were also found at all electrodeites analyzed for the LLP component (FC4: F1,14 = 7.40, p = 0.02;4: F1,14 = 6.56, p = 0.02; CP4: F1,14 = 8.11, p = 0.01; P4: F1,14 = 5.02,

= 0.04). Pre-planned contrasts also showed that at electrodes CP4nd P4, the LLP amplitude was enhanced post- relative to the pre-xercise session when attention was directed toward relevant tac-

ile stimuli (CP4: F1,14 = 6.67, p = 0.02; P4: F1,14 = 5.56, p = 0.03). Pre-lanned contrasts revealed that at frontal electrode site FC4, the LLPmplitude was decreased post-exercise versus pre-exercise duringhe task-irrelevant condition (F1,14 = 4.77, p = 0.05) (Fig. 4C/D).

stimuli (i.e. smaller vibrations) delivered to the index finger are shown for the P300ces show Attend Index and Ignore Index conditions, respectively, prior to exercise.

y, following exercise.

Lastly, analysis of the P300 amplitude revealed no statistical dif-ferences between the P300 amplitude when attention was directedtoward relevant tactile stimuli post- versus pre-exercise at anyelectrode sites analyzed (FCZ: t14 = 0.362, p > 0.05; CZ: t14 = 0.35,p > 0.05; CPZ: t14 = 0.36, p > 0.05). Results for the P300 latencyalso revealed no statistical differences pre- versus post-exerciseduring the Attend Index conditions (FCZ; t14 = 0.1, p > 0.05; CZ:t14 = 0.36, p > 0.05; CPZ: t14 = 0.17, p > 0.05) (refer to Fig. 3 for P300traces).

3.3. Behavioral results

Behavioral analysis showed a main effect of attention with suc-

cess rates being higher when attention was directed toward theindex finger versus away from it suggesting that participants wereslightly less accurate in detecting target stimuli presented to thepinky finger (F1,14 = 9.66, p = 0.01) (Fig. 5).

272 C. Popovich, W.R. Staines / Behavioural Brain Research 281 (2015) 267–275

Fig. 4. Group ERP means. Group means for A) P100, B) N140, and C/D) LLP, ERP components at electrodes of interest (FC4, CP4, P4). Black solid and outlined bars representgroup data for the Attend Index and the Ignore Index condition prior to exercise, respectively. Gray solid and outlined bars represent group data for the Attend Index and theIgnore Index condition following exercise, respectively. Error bars show SEM, * denotes significance p < 0.05.

Fig. 5. Group means for behavioral data. Behavioral performance was evaluated bydetermining the number of targets hit relative to the overall number of targets thatwere presented for each subject, and this success rate was expressed as a percentage.Black solid and outlined bars represent group data for the Attend Index and theAttend Index condition prior to exercise, respectively. Gray solid and outlined barsrepresent group data for the Attend Index and the Ignore Index condition followingexercise, respectively. Error bars show SEM, * denotes significance p < 0.05.

4. Discussion

Evidence suggests that acute aerobic exercise selectively up-regulates excitability in frontal lobe regions, thereby promotinggreater top-down attentional control during executive functioningtasks [13–15]. However, the effects of acute exercise in corti-cal areas downstream from the PFC in modality-specific corticesare not well understood. This study examined how an acutebout of moderate intensity aerobic exercise modulated neuronalactivity in somatosensory cortices using a tactile discriminationtask. Our ERP results, particularly for the LLP component, sug-gest that an acute bout of moderate intensity aerobic exercisefacilitated the sensory-gating role of the PFC by suppressingneuronal responses to unattended, task-irrelevant stimuli atfrontal regions and amplifying attended, task-relevant signals atmodality-specific somatosensory regions contralateral to stim-ulation. Furthermore, our N140 results suggest that an acutebout of moderate intensity aerobic exercise may improve selec-tive attentional processing by enhancing involuntary shifts ofattention by showing greater enhancement of the N140 compo-nent to task-irrelevant stimuli post-relative to the pre-exercisesession. Our findings are in line with current literature examin-ing the inverted U-shape relationship between aerobic exerciseand cognitive function by demonstrating that a single bout of

moderate intensity exercise modulates frontoparietal interac-tions during a tactile discrimination task and facilitates selectiveattentional processing of tactile information in healthy youngadults.

ural Br

4s

[btvemi

ttrtdsopuueiPmd[[uowPfiE

seguEbePttmiimmhSNsaKoetiataop

C. Popovich, W.R. Staines / Behavio

.1. Neural mechanisms of attention-based modulation inomatosensory cortices

The P100 component is modulated by vibrotactile stimulation29] and attentional processes [30–32], but it is thought to haveilateral neural generators in SII regions [33,34]. Selective atten-ion studies have reported increased P100 amplitudes to attendedersus unattended tactile stimuli with effects being greater thanarlier ERP responses generated in SI [11,30,31,35], but these effectsay also be task-specific and/or due to inter-individual differences

n P100 modulation [29].Using the same tactile discrimination task as reported in

his study, Bolton and Staines [11] revealed no modulation ofhe P50 component, but increased P100 amplitudes to task-elevant versus task-irrelevant stimuli. It was also found thatransient disruption to the dorsolateral PFC via cTBS producedisinhibition of the P100 component to task-irrelevant stimuli,uggesting that this region plays an important role in the gatingf task-irrelevant tactile information during selective attentionalrocessing [11]. Our results were similar to previous findingssing the same task, in that there was no attention-related mod-lation of the P50 component, but the P100 amplitude wasnhanced to attended, task-relevant versus unattended, task-rrelevant tactile stimuli [11]. Failure to drive modulation of the50 component using tactile stimulation paradigms is not uncom-on, and task-difficulty may be an important factor required for

riving modulation at early stages of somatosensory processing22]. Based on previous work using the same tactile paradigm11,12,35], it seems likely that the cognitive demand of this partic-lar task may not be sufficient for driving attentional modulationf the most primary measures of somatosensory processing. Thereere also no exercise-related changes in the modulation of the

100 amplitude when attention was directed toward the indexnger (i.e. no differences Attend Index Pre-Exercise versus Post-xercise).

The N140 component has been observed over frontal electrodeites contralateral to stimulation and along midline and ipsilat-ral sites [26], but is thought to have contributions from bilateralenerators in SII regions with the hemisphere contralateral to stim-lation online earlier [36]. Although the neural generators of thisRP are not well understood, Allison et al. [23] suggested thatilateral activation of the frontal lobes may contribute to the gen-ration of the frontal N140. The frontal N140 may also reflectFC activation and reciprocal neural communication with pos-erior and subcortical structures [26]. If this supposition is true,hen the N140 component would be an important electrocortical

arker of PFC activation during situations when relevant sensorynformation must be maintained and retrieved to execute behav-oral goals [23,26]. However, unlike earlier modality-specific ERPs,

odulation of the N140 component may involve activation ofultiple cortical generators, thus making direct inferences about

ow attention modulates this component is somewhat nebulous.everal EEG studies investigating attentional modulation of the140 component have found that the presentation of transient

timuli at a to-be-ignored spatial location can involuntarily directttention toward that unattended, task-irrelevant side [37–40].ida et al. [38,40] compared the influence of deviant stimulin attentional modulation of the P100 and N140 somatosensoryvoked potential when deviants were embedded in an oddballask to when they were presented in isolation. Subjects werenstructed to either count the number of deviants presented, press

button indicating the detection of deviants (active attention

asks), or to ignore all tactile events and read a book (passivettention task). Results showed that attentional enhancementf N140, but not the P100, was greater when deviants wereresented in isolation versus when they were embedded in an

ain Research 281 (2015) 267–275 273

oddball sequence which was not the case for the P100 compo-nent. The authors concluded that isolated deviants capture moreattentional resources relative to when they are embedded inan oddball discrimination task. Adler et al. [37] complementedthese findings by showing that N140 enhancement was great-est for distracter stimuli presented to the to-be-ignored spatiallocation but only when the discrimination task involving the to-be-attended finger was relatively easy. It was suggested that thepresentation of distracter stimuli at an unattended spatial loca-tion can involuntarily pull attention toward that body locationunder low, but not high, cognitive load demands. Similarly, Boltonand Staines [11,35] found enhanced N140 amplitudes to task-irrelevant versus relevant tactile stimuli presented to the indexfinger using the same tactile discrimination task adopted in thisstudy and surmised that their findings may relate to the inter-pretation proposed by Adler and colleagues [37]. In our study,we showed consistent results for the N140 component acrosssensorimotor electrode sites contralateral to vibrotactile stimula-tion (i.e. CP4, P4), whereby the N140 amplitude was statisticallygreater post-exercise relative to pre-exercise, when attention wasdirected away from the index finger and toward another spa-tial location (i.e. Ignore Index). These findings suggest that theability to involuntarily shift attention toward task-irrelevant tac-tile stimuli was more efficient following a single bout of acutemoderate intensity aerobic exercise. More efficient involuntaryshifts in attention toward and away from distracter stimuli in theenvironment may have, in turn, facilitated sensory processing oftask-relevant stimuli observed at later stages of somatosensoryprocessing.

The LLP has multiple neural generators from broadly distributedlocations, and is often seen as a sustained positivity occurringapproximately 200–500 ms post-stimulus [24]. Several tactile ERPstudies have reported variable attention-related effects on LLPmodulation [11,12,26,32,35]. These discrepancies are likely dueto differences in attentional requirements or demands of thetask employed. The precise role of this later positivity remainsunclear; however, several attention-based tactile ERP studies haveimplied that the LLP may share functional similarities to the P300component [25,26]. The P300 component is typically observedduring higher-order cognitive functions and is thought to haveroles in selective attention and resource allocation, whereby theP300 amplitude is proportional to the amount of attentionalresources engaged in processing a given stimulus [41]. EEG stud-ies in healthy young and older adults using the same tactilediscrimination task have reported increased LLP amplitudes dur-ing attended, task-relevant stimuli compared to non-attended,irrelevant tactile stimuli [11,35]. Notably, ERP data from PFCpatients using the same tactile paradigm revealed aberrations,particularly, in the neuronal profile of the LLP component, withincreased amplitudes elicited by task-irrelevant tactile stimuli[12]. This finding along with cTBS work by the same authorssuggests that the PFC plays an integral role in sensory gat-ing of early (i.e. P100) and later (LLP) stages of somatosensoryprocessing in order to ensure that relevant sensory signals areamplified in sensory-specific cortices [11,12]. In our study, theamplitude of the LLP was suppressed at contralateral frontal andipsilateral centroparietal sites, post-relative to pre-exercise, duringthe task-irrelevant condition, and was enhanced over modality-specific somatosensory cortices contralateral to stimulation. If theLLP arises from frontal generators and shares functional simi-larities with the P300 component, then our results suggest thatan acute bout of moderate intensity exercise may have facili-tated the sensory gating role of frontal networks by suppressingirrelevant sensory signals so that relevant tactile information could

be passed on and amplified in modality-specific somatosensorycortex.

2 ural B

4

ldOtOpeasoghs

cEPoa[chutmtuOt8tcutttpites[essaeneadeHtttccspiod

[

[

[

[

[

[

74 C. Popovich, W.R. Staines / Behavio

.2. Exercise-related benefits on cognition

Acute aerobic exercise has been shown to induce neuromodu-atory changes in PFC activity and improve cognitive performanceuring tasks requiring higher-order attentional control [13–15].ur study supports this idea and provides novel information about

he effects of exercise on cortical areas downstream from the PFC.ur most interesting results occurred for the N140 and LLP com-onents, whereby an acute bout of moderate intensity aerobicxercise potentially assisted involuntary shifts of attention towardnd away from distractors in the environment so that relevant sen-ory signals could be enhanced at later stages of sensory processingver modality-specific somatosensory cortices. These findings sug-est that an acute bout of moderate intensity aerobic exercise inealthy young adults facilitated selective attentional processing ofomatosensory information.

In the present study, an acute bout of moderate intensity exer-ise did not drive significant modulation of early somatosensoryRPs of interest (i.e. the P50/P100 components), nor of the later300 component. EEG studies have shown that the amplitudesf these earlier somatosensory ERPs are typically enhanced whenttention is directed toward task-relevant sensory information5,11,22,32,35]. However, failure to drive modulation of the P50omponent is not uncommon in tactile literature and seems to beeavily dependent on the attentional demand of the cognitive tasksed [22]. Previous findings using the same tactile discriminationask as reported here have also failed to drive attention-related P50

odulation [11,35]. Thus, we have reason to infer that our chosenask may not be attentionally demanding enough to drive mod-lation at the most primary stage of somatosensory processing.ur behavioral results further support this supposition since par-

icipants performed the task with success rates of approximately0%, suggesting that the attentional task was relatively easy. Fur-hermore, the absence of exercise-induced modulation of the P300omponent may also be a result of the tactile discrimination tasksed. The majority of studies reporting exercise-induced effects onhe P300 component in healthy young adults have used attentionalasks including: the flanker, the Stroop, or a go/nogo inhibitory con-rol task [16–19]. These tasks reliably elicit P300 responses, whilerevious studies using sensory oddball tasks report mixed exercise-

nduced effects on the P300 component [42–45]. The observationhat P300 amplitudes were reduced post-exercise relative to pre-xercise in our healthy young adults is contrary to most exercisetudies using attentional tasks requiring greater cognitive demand16–19] and the argument could be made that learning or practiceffect may account for this result. However, behavioral resultsuggest that no change in cognitive performance across exerciseessions, thus we have reason to believe that the reduction in P300mplitude post-exercise is not a reflection of practice or learningffects. Instead, it is plausible that this sensory oddball task mayot be an advantageous paradigm for examining exercise-relatedffects on the P300 component in healthy young adults. Lastly, thebsence of discernible ERPs from stimulation of the fifth digit hin-ered our ability to examine whether exercise-induced attentionalffects would extend to a different spatial location of the same hand.owever, this absence of neuronal modulation following vibrotac-

ile stimulation of this digit is not unusual. The main purpose ofhis condition was to direct attention to a different spatial loca-ion than the index finger and in this way served as an attentionalontrol. Previous work using the same task also failed to elicit dis-ernible ERPs from the fifth digit [11,12,35], while a recent fMRItudy reported an absence in the hemodynamic responsiveness,

articularly, of the fifth digit to vibrotactile stimulation, suggest-

ng that there may be digit-specific differences in the activationf neuronal responses [46]. Furthermore, electrophysiology recor-ings in owl and squirrel monkeys have shown that the neuronal

[

rain Research 281 (2015) 267–275

representation of the fifth digit in area 3b of SI is the smallest whencompared to the remaining four digits [47].

5. Conclusion

Our results suggest that an acute bout of moderate inten-sity aerobic exercise facilitated selective attentional processingof somatosensory information by improving: i) the efficiency ofinvoluntary attentional shifts toward task-irrelevant environmen-tal distracters over modality-specific somatosensory regions, andii) attentional control of somatosensory input at later stages of sen-sory processing over frontoparietal regions. These findings implythat an acute bout of moderate intensity exercise is also capableof influencing attentional regulation of somatosensory informa-tion perhaps via enhanced top-down attentional control overmodality-specific somatosensory cortices. Future studies will useTMS neuroimaging techniques better suited to examining the directeffect of exercise on modulating frontoparietal interactions.

Acknowledgements

This work was supported by funding to WRS from the Natu-ral Sciences and Engineering Research Council of Canada (NSERC),the Canada Research Chairs program, the Canada Foundation forInnovation and the Ontario Research Fund. CP was supported bygraduate scholarship funds from the Ontario Graduate Scholarship(OGS) program.

References

[1] Corbetta M. Frontoparietal cortical networks for directing attention and theeye to visual locations: identical, independent, or overlapping neural systems.Proc Natl Acad Sci USA 1998;95:831–8.

[2] Hopfinger JB, Buonocore MH, Mangun GR. The neural mechanisms of top-downattentional control. Nat Neurosci 2000;3:284–91.

[3] Motter BC. Focal attention produces spatially selective processing in visual cor-tical areas V1, V2, and V4 in the presence of competing stimuli. J Neurophysiol1993;70:909–19.

[4] Petkov CI, Kang X, Alho K, Bertrand O, Yund EW, Woods DL. Attentional mod-ulation of human auditory cortex. Nat Neurosci 2004;7:658–63.

[5] Popovich C, Staines WR. The attentional-relevance and temporal dynamics ofvisual-tactile crossmodal interactions differentially influence early stages ofsomatosensory processing. Brain Behav 2014;4:247–60.

[6] Skinner JE, Yingling CD. Regulation of slow potential shifts in nucleus reticu-laris thalami by the mesencephalic reticular formation and the frontal granularcortex. Electroencephalogr Clin Neurophysiol 1976;40:288–96.

[7] Yingling CD, Skinner JE. Selective regulation of thalamic sensory relaynuclei by nucleus reticularis thalami. Electroencephalogr Clin Neurophysiol1976;41:476–82.

[8] Dionne JK, Meehan SK, Legon W, Staines WR. Crossmodal influences insomatosensory cortex: interaction of vision and touch. Hum Brain Mapp2010;31:14–25.

[9] Staines WR, Graham SJ, Black SE, McIlroy WE. Task-relevant modulation ofcontralateral and ipsilateral primary somatosensory cortex and the role of aprefrontal-cortical sensory gating system. Neuroimage 2002;15:190–9.

10] Knight RT, Staines WR, Swick D, Chao LL. Prefrontal cortex regulates inhi-bition and excitation in distributed neural networks. Acta Psychol (Amst)1999;101:159–78.

11] Bolton DAE, Staines WR. Transient inhibition of the dorsolateral prefrontal cor-tex disrupts attention-based modulation of tactile stimuli at early stages ofsomatosensory processing. Neuropsychologia 2011;49:1928–37.

12] Bolton DAE, Staines WR. Attention-based modulation of tactile stimuli: a com-parison between prefrontal lesion patients and healthy age-matched controls.Neuropsychologia 2014;57:101–11.

13] Ekkekakis P. Illuminating the black box: investigating prefrontal cortical hemo-dynamics during exercise with near-infrared spectroscopy. J Sport ExercPsychol 2009;31:505–53.

14] Endo K, Matsukawa K, Liang N, Nakatsuka C, Tsuchimochi H, Okamura H, et al.Dynamic exercise improves cognitive function in association with increasedprefrontal oxygenation. J Physiol Sci 2013;63:287–98.

15] Yanagisawa H, Dan I, Tsuzuki D, Kato M, Okamoto M, Kyutoku Y, et al.Acute moderate exercise elicits increased dorsolateral prefrontal activa-

tion and improves cognitive performance with Stroop test. Neuroimage2010;50:1702–10.

16] Kamijo K, Hayashi Y, Sakai T, Yahiro T, Tanaka K, Nishihira Y. Acute effects ofaerobic exercise on cognitive function in older adults. J Gerontol B Psychol SciSoc Sci 2009;64:356–63.

ural Br

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

C. Popovich, W.R. Staines / Behavio

17] Kamijo K, Nishihira Y, Hatta A, Kaneda T, Wasaka T, Kida T, et al. Differen-tial influences of exercise intensity on information processing in the centralnervous system. Eur J Appl Physiol 2004;92:305–11.

18] Kamijo K, Nishihira Y, Higashiura T, Kuroiwa K. The interactive effect of exerciseintensity and task difficulty on human cognitive processing. Int J Psychophysiol2007;65:114–21.

19] Hillman CH, Pontifex MB, Raine LB, Castelli DM, Hall EE, Kramer AF. The effectof acute treadmill walking on cognitive control and academic achievement inpreadolescent children. Neuroscience 2009;159:1044–54.

20] Tomporowski PD. Effects of acute bouts of exercise on cognition. Acta Psychol(Amst) 2003;112:297–324.

21] Iwadate M, Mori A, Ashizuka T, Takayose M, Ozawa T. Long-term phys-ical exercise and somatosensory event-related potentials. Exp Brain Res2005;160:528–32.

22] Schubert R, Ritter P, Wüstenberg T, Preuschhof C, Curio G, Sommer W,et al. Spatial attention related SEP amplitude modulations covary withBOLD signal in S1–a simultaneous EEG-fMRI study. Cereb Cortex 2008;18:2686–700.

23] Allison T, McCarthy G, Wood CC. The relationship between human long-latencysomatosensory evoked potentials recorded from the cortical surface and fromthe scalp. Electroencephalogr Clin Neurophysiol 1992;84:301–14.

24] Hämäläinen H, Kekoni J, Sams M, Reinikainen K, Näätänen R. Humansomatosensory evoked potentials to mechanical pulses and vibration: con-tributions of SI and SII somatosensory cortices to P50 and P100 components.Electroencephalogr Clin Neurophysiol 1990;75:13–21.

25] Desmedt JE, Robertson D. Differential enhancement of early and late compo-nents of the cerebral somatosensory evoked potentials during forced-pacedcognitive tasked in man. J Physiol 1977;271:761–82.

26] Desmedt JE, Tomberg C. Mapping early somatosensory evoked potentials inselective attention: critical evaluation of control conditions used for titratingby difference the cognitive P30, P40, P100 and N140. Electroencephalogr ClinNeurophysiol 1989;74:321–46.

27] Pasternak T, Greenlee MW. Working memory in primate sensory systems. NatRev Neurosci 2005;6:97–107.

28] Pfefferbaum A, Ford JM, Weller BJ, Kopell BS. ERPs to response production andinhibition. Electroencephalogr Clin Neurophysiol 1985;60:423–34.

29] Goff GD, Matsumiya Y, Allison T, Goff WR. The scalp topography of humansomatosenosry and auditory evoked potentials. Electroencephalogr Clin Neu-rophysiol 1977;42:57–76.

30] Desmedt JE, Huy NT, Bourguet M. The cognitive P40, N60 and P100 compo-nents of somatosensory evoked potentials and the earliest electrical signs of

sensory processing in man. Electroencephalogr Clin Neurophysiol 1983;56:272–82.

31] Josiassen RC, Shagass C, Roemer RA, Slepner S, Czartorysky B. Early cognitivecomponents of somatosensory event-related potentials. Int J Psychophysiol1990;9:139–49.

[

ain Research 281 (2015) 267–275 275

32] Eimer M, Forster B. The spatial distribution of attentional selectivity intouch: evidence from somatosensory ERP components. Clin Neurophysiol2003;114:1298–306.

33] Gu Q. Neuromodulatory transmitter systems in the cortex and their role incortical plasticity. Neuroscience 2002;111:815–35.

34] Mima T, Nagamine T, Nakamura K, Shibasaki H. Attention modulates bothprimary and second somatosensory cortical activities in humans: a magne-toencephalographic study. J Neurophysiol 1998;80:2215–21.

35] Bolton DAE, Staines WR. Age-related loss in attention-based modulation oftactile stimuli at early stages of somatosensory processing. Neuropsychologia2012;50:1502–13.

36] Frot M, Mauguière F. Timing and spatial distribution of somatosensoryresponses recorded in the upper bank of the sylvian fissure (SII area) in humans.Cereb Cortex 1999;9:854–63.

37] Adler J, Giabbiconi CM, Müller MM. Shift of attention to the body location of dis-tracters is mediated by perceptual load in sustained somatosensory attention.Biol Psychol 2009;81:77–85.

38] Kida T, Nishihira Y, Wasaka T, Nakata H, Sakamoto M. Passive enhancementof the somatosensory P100 and N140 in an active attention task using deviantalone condition. Clin Neurophysiol 2004;115:871–9.

39] Kida T, Nishihira Y, Wasaka T, Nakata H, Sakamoto M. Differential modula-tion of temporal and frontal components of the somatosensory N140 and theeffect of interstimulus interval in a selective attention task. Cogn Brain Res2004;19:33–9.

40] Kida T, Wasaka T, Nakata H, Akatsuka K, Kakigi R. Active attention modulatespassive attention-related neural responses to sudden somatosensory inputagainst a silent background. Exp Brain Res 2006;175:609–17.

41] Donchin E, Coles MGH. Is the P300 component a manifestation of contextupdating? Behav Brain Sci 1988;11:357–74.

42] Grego F, Vallier JM, Collardeau M, Bermon S, Ferrari P, Candito M, et al. Effectsof long duration exercise on cognitive function, blood glucose, and counterreg-ulatory hormones in male cyclists. Neurosci Lett 2004;364:76–80.

43] Nakamura Y, Nishimoto K, Akamatu M, Takahashi M, Maruyama A. The effectof jogging on P300 event related potentials. Electromyogr Clin Neurophysiol1999;39:71–4.

44] Polich J, Lardon MT. P300 and long-term physical exercise. ElectroencephalogrClin Neurophysiol 1997;103:493–8.

45] Yagi Y, Coburn KL, Estes KM, Arruda JE. Effects of aerobic exercise and genderon visual and auditory P300, reaction time, and accuracy. Eur J Appl PhysiolOccup Physiol 1999;80:402–8.

46] Schweizer R, Voit D, Frahm J. Finger representations in human primary

somatosensory cortex as revealed by high-resolution functional MRI of tactilestimulation. Neuroimage 2008;42:28–35.

47] Merzenich MM, Nelson RJ, Kaas KH, Stryker MP, Jenkins WM, Zook JM, et al.Variability in hand surface representations in areas 3b and 1 in adult owl andsquirrel monkeys. J Comp Neurol 1987;258:281–96.