lable at ScienceDirect

Psychology of Sport and Exercise 12 (2011) 509e514

Contents lists avai

Psychology of Sport and Exercise

journal homepage: www.elsevier .com/locate/psychsport

Fencing expertise and physical fitness enhance action inhibition

John S.Y. Chan a, Alan C.N. Wong a, Yu Liu b, Jie Yu c, Jin H. Yan c,*

aDepartment of Psychology, The Chinese University of Hong Kong, Hong Kong, ChinabDepartment of Psychology, Tsinghua University, Beijing, ChinacDepartment of Sports Science and Physical Education, The Chinese University of Hong Kong, Hong Kong, China

a r t i c l e i n f o

Article history:Received 19 August 2010Received in revised form15 April 2011Accepted 22 April 2011Available online 6 May 2011

Keywords:Fencing performanceInhibitory controlMotor skillPhysical fitness

* Corresponding author. Psychomotor Laboratory, DPE, The Chinese University of Hong Kong, Shatin, NT,2609 6094; fax: þ852 2603 5781.

E-mail address: jhyan@cuhk.edu.hk (J.H. Yan).

1469-0292/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.psychsport.2011.04.006

a b s t r a c t

Objective: This study investigated the effects of fencing expertise and physical fitness on the inhibitorycontrol of fencers and non-fencers.Design: This study used a 2� 2 factorial design. Fencers and non-fencers both in low-fit and averagely-fitsubgroups were compared in reaction times (RT) and accuracy in simple reaction time (SRT) and go/no-go reaction time (go/no-go RT) tasks.Method: The participants were 30 fencers (aged 18e26) and 30 non-fencers (aged 19e25), each havinga different fitness level. With a standard computer keyboard, each participant performed an SRT task byresponding to all stimuli. In the go/no-go RT task, each participant responded only to the go signals whilewithholding their response to the no-go signals.Results: There were no significant differences between the participants with different levels of fitness orfencing expertise in SRT, go/no-go RT, omission error and commission error. However, an interaction offitness and fencing expertise on commission error was found (p< .05). Averagely-fit fencers committeda similar number of errors to the averagely-fit non-fencers, but the high-fit fencers committed signifi-cantly fewer errors compared to the high-fit non-fencers (p< .05).Conclusions: Fencing experience and physical fitness facilitate a person’s ability to withhold action whennecessary. The interactive nature of aerobic fitness and sport expertise on action inhibition suggests thatcognitive control benefits most from the combination of physical and mental training compared to wheneach is administered singly.

� 2011 Elsevier Ltd. All rights reserved.

Physical activities and exercise benefit physical and cognitivewell-being (Hassmén, Koivula, & Uutela, 2000; Lee, Hsieh, &Paffenbarger, 1995). Among a wide range of cognitive capabilities,executive functions supported by the frontal and the prefrontalareas of brain, receive particular and substantial benefits fromphysical exercise (Colcombe & Kramer, 2003; Kramer et al., 1999).Consistent with this view, habitual sport or exercise participantsare presumably expected to show better cognitive capabilities thantheir non-exercising counterparts (Ozel, Larue, & Molinaro, 2004).

Improvements in cognitive functioning are associated with thelength of sport participation (Brisswalter, Collardeau, & René, 2002).Improved signal detection and perceptual abilities were observedafter a single session of aerobic exercise (Gliner, Matsen-Twisdale,Horvath, & Maron, 1979; Lybrand, Andrews, & Ross, 1954). Suchenhancing effects from acute exercise are similar to the state-

epartment of Sports Science/Hong Kong, China. Tel.: þ852

All rights reserved.

dependent transient alternations of cognition that return to thebaseline level after a short period of time. In contrast, the enhancingeffects of chronic exercise are more long-lasting than those of acuteexercise. In addition to the effects of exercise duration, acute andchronic exercises produce different levels of improvement incognitive functioning. Chronic exercise results in a greater cognitiveenhancement (Etnier et al.,1997; Thomas, Landers, Salazar, & Etnier,1994).

Cognitive abilities improve after chronic involvement in sports.For example, elite basketball, volleyball and water-polo playersshowed better performances in a series of perceptual-motor skillsthan did the novice players (e.g., perceptual speed, attention, andestimation of object speed and direction) (Kioumourtzoglou,Kourtesses, Michalopolou, & Derri, 1998). Theories about cogni-tive changes after exercise or sports training have been proposed.It is plausible that long-term involvement in sports alters neuralactivities or brain structures and functions. In an imaging study,increased brain volume both in the white and the gray regions wasobserved among older adults after aerobic training (Colcombeet al., 2006). In addition, it has been speculated that aerobic

Table 1Demographic information of the participants in four experimental groups.

Groupa Age Fitness scores

Averagely-fit non-fencers 20.87 (1.46) 43.24 (3.92)High-fit non-fencers 20.07 (1.10) 51.51 (7.01)Averagely-fit fencers 20.53 (2.85) 45.87 (4.23)High-fit fencers 21.07 (2.58) 53.83 (2.96)

Note. Fitness level is indexed by the questionnaire estimated maximal oxygenuptake, VO2max (mL kg�1min�1). SD is presented in parentheses.

a N¼ 15, comparable gender composition for each group.

J.S.Y. Chan et al. / Psychology of Sport and Exercise 12 (2011) 509e514510

exercises can influence neural functions and cognitive capabilitiesby altering the plasticity of a person’s neural system (Kramer &Erickson, 2007).

In addition to the length of sport participation, cognitive abilitiesare enhanced as a result of the specific demands of certain sports. Forexample, action inhibition, a subset of executive control, is better inplayers of some sports than it is in those of other sports. This showsthe relation between a given sport and specific cognitive abilities.Kida, Oda, and Matsumura (2005) indicated that the responseinhibition was different between baseball batters of varying skilllevels, but not among tennis players of various skill levels. Hence,a differential improvement of cognitive abilities in different sportswas observed. Skilled baseball batters showed stronger responseinhibition ability in conflicting conditions in a go/no-go task than theamateur and novice players. In addition to baseball players, profes-sional fencers also showed stronger action inhibition than did thenovices in another study (Di Russo, Taddei, Apnile, & Spinelli, 2006).

We assume that the improved action inhibition is the result offrequent and intense conflict control developed by fencers andbaseball batters in the training and subsequent competitions.Specific sport-related demands play a critical role in enhancingcognitive abilities. Fencers are required to make as few errors aspossible while judging accurately and acting swiftly. Fencers arechallenged by their opponents’ feints or fakes in training andcompetitions. Sport-specific experiences lead to better cognitiveabilities. For instance, elite fencers can better adapt to conflictingsituations and control actions in tasks which require a high degreeof concentration or attention. Similarly, baseball batters hold backtheir movements unless the approaching balls go to the designatedarea for batting. However, faking is uncommon in tennis. Tennisplayers require action inhibition less frequently than fencers orbaseball batters. Tennis players have to return the ball as accuratelyand as promptly as possible. The nature of this sport providesrelatively fewer opportunities for tennis players to improve actioninhibition. Thus, it is possible that frequent experience of certaincognitive demands improves those cognitive abilities.

Although previous research shows the differences in cognitiveability between athletes and non-athletes and claims that sporttraining is the cause, the effect of a potential confounding variablesuch as fitness level has not been examined. A recent ERP studyreported that physical fitness could modulate the neural-electricactivities for executive control after acute exercise, thus compli-cating the interpretation of previous results (Stroth et al., 2009).Specifically, professional athletes have higher fitness levels thannon-athletes. It is likely that fitness rather than exercise per se leadsto superior cognitive functioning. Previous studies suggested thatfitness plays a role in modulating action-inhibitory control amongelderly people (Bixby et al., 2007; Kamijo et al., 2009). Elderlypeople with a higher fitness level lose less brain tissue in thefrontal, parietal and temporal cortices than their less fit counter-parts. The results suggested that fitness serves as a protectivefunction for people against neural degenerations (Raz, 2002). Thus,the different cognitive capabilities observed in athletes and non-athletes may be caused by various levels of physical fitness, sportexperiences, or the interaction of both factors. This issue, however,has not been addressed in the literature. Without knowing theinteractive effects of fitness levels and sports expertise on cognitivefunctioning, we cannot know whether the conclusions drawn fromprevious studies about cognitive enhancement from exercise orsport training are valid.

The purpose of this study is to shed light on the relation betweenphysical fitness and sport experience and their collective effects onthe enhancement of cognitive capability. Although independenteffects of fitness and sport experience were found in previousresearch, the relationship between physical fitness and sport

experience has not been investigated. Thus, we examined the effectsof physical fitness and sport experience on the executive functioningof fencers and non-fencers. We specifically examined the action-inhibitory ability of fencers and non-fencers with high and normalfitness levels. To accomplish this goal, we used a go/no-go reactiontime (RT) task to examine inhibitory control. We expected thatparticipants with better inhibitory control would commit fewererrors in the task. In addition, we used a simple RT task as a controlmeasure of motor speed. This test could provide evidence that thedifferences observed in the go/no-go RT task were not explained bythe differences in motor abilities among participants. Moreover, weexamined the effects of practice and fatigue on performance toexclude potential confounding impacts on our results.

We had three specific hypotheses. Firstly, although a positiverelationship between fitness level and executive function has beenfound among the older adult population, such an effect has not beenobserved among young adults in behavioral studies (Chodzko-Zajko,1991).We expected no significant difference betweenparticipants ofdifferent fitness levels in the measure of response accuracy in thego/no-go RT task. Secondly, we hypothesized that the fencers wouldcommit fewer errors than the non-fencers. Fencers could betterinhibit their actions than non-fencers (Di Russo et al., 2006). Finally,previous studies suggested that both fitness and sport training arerelated to the enhancement of executive control. We thereforehypothesized that there would be an interactive effect of fencingexpertise and fitness level on inhibitory control.

Method

Participants

There were a total of 60 participants: 30 fencers (15 females and15 males, with 5 years or more fencing experience, practiced 5e6times a week, M¼ 20.63 years, SD¼ 2.11 years); and 30 collegestudents (15 females and 15 males, no prior experience in fencing,M¼ 20.63 years, SD¼ 2.11 years). The gender ratio was the sameboth for the fencers and the non-fencers. All subjects had normal orcorrected-to-normal vision and had no history of psychiatric andneurological disorders. Fencers and non-fencers were divided intotwo subgroups on the basis of a median split according to theirquestionnaire estimated VO2max. The fencers’ fitness level, indexedby the estimated VO2max, was not significantly different from that ofthe non-fencers’ (p¼ .13). The high-fit subjects had a significantlyhigher estimated VO2max than the averagely-fit subjects (p< .001).Informed consent approved by the Research Ethical Committee ofthe Chinese University of Hong Kong was obtained from eachsubject. Each subject received HK$50 for participation. Table 1shows the demographics of each group.

Stimuli

Fig. 1 shows the stimuli for both RT tasks. White visual stimuli(10 cm� 10 cm) were presented at the center of the screen against

Fig. 1. Stimuli used in the (a) simple reaction time (RT) and (b) go/no-go task.

J.S.Y. Chan et al. / Psychology of Sport and Exercise 12 (2011) 509e514 511

a black background. Instead of using those in a typical go/no-gotask, shape stimuli were used in the simple RT task to preventsubjects from being confused about the two tasks. Being compat-ible to those used in Di Russo et al.’s (2006) study, stimuli for the go/no-go task were of two similar line patterns. Rossi, Zani, Taddei, andPesce (1992) suggested that the performance differences betweenfencers and non-fencers in executive control tasks were moresignificant when the stimuli were highly similar. This was the casebecause some fencers might rely on visual stimulation for much ofthe time. The use of highly similar visual stimuli could yielda robust between-group difference in the go/no-go tasks.

Procedure

All participants performed the simple RT and the go/no-go RTtasks after completing the fitness questionnaires. In the two RTtasks, each participant was seated about 80 cm in front of a 15-inchcomputer screen. The order of the two RT tasks was counter-balanced across the participants. Twenty practice trials were givenprior to each RT task to allow the participants to familiarizethemselves with the tasks. The participants were given shortbreaks between consecutive experimental blocks. The participantshad to respond by pressing the key “B”with the index finger of theirpreferred hand.

Simple reaction time (SRT) taskThree blocks of 100 trials were included. The trial order was

randomized within each block. The participants were required torespond to all different stimuli as quickly as possible. Four shapes ofstimuli appeared on the same proportion of trials. Between thefixation and the target stimuli, delays of variable time (500, 1000,1500, 2000, 2500 ms) were inserted to prevent subjects fromanticipating. These five intervals of delay were randomly presentedon an equal proportion of trials. A trial began with the appearanceof a fixation for 500 ms followed by a variable delay period andthen the shape stimulus.

Go/no-go reaction time (go/no-go RT) taskFour blocks of 100 trials were included. The trial order was

randomized within each block. The participants were required torespond to the go signal as quickly and accurately as possible whilewithholding a response to the no-go signal. One of the line patternswas the go stimulus for half of the participants whereas the otherline pattern was the go stimulus for the other half of the partici-pants. Both response accuracy and speed were emphasized. Beforethe task, the participants were notified that 75% of the trials werego trials and the rest were no-go trials in a block, thereby elicitinga covert tendency to make a go response among subjects. Thisprocedure was effective in increasing inhibitory demands(Ciesielski, Harris, & Coffer, 2004). A trial began with the appear-ance of a fixation lasting for 500 ms followed by the go or no-gosignal. A 3-s time-out period was adopted in each trial, so thatthe next trial would be presented if there was no response after theappearance of go or no-go signals for 3 s.

Fitness questionnairesInformation about each participant’s age, gender, height and

weight was collected. A modified version of the Physical ActivityRating (PA-R) (Jackson et al., 1990) and the Perceived FunctionalAbility (PFA) (George, Stone, & Burkett, 1997) questionnaires weregiven. The aerobic fitness, indexed by VO2max, of each subject wasestimated. Instead of a traditional maximal graded exercise test, thenon-exercise formula provided by Bradshaw et al. (2005) was usedto estimate VO2max; this is because it can promisingly predict theactual VO2max (R¼ .93) and requires no specific machines toperform it. The non-exercise formula is: VO2max (mL kg�1min�1)¼48.073þ (6.178� gender; female ¼ 0, male¼ 1)� (0.246� age)�(0.619 � BMI)þ (0.712� PFA)þ (0.671� PA-R), where VO2max isthe estimated maximal oxygen uptake, BMI is the body mass index(height divided by weight squared), PFA is the Perceived FunctionalActivity Rating (ranging from 2 to 26) and PA-R is the PhysicalActivity Rating (ranging from 0 to 10). The fitness test results wereused to separate the averagely-fit and high-fit fencers from thenon-fencers; this made up a total of four experimental groups(averagely-fit non-fencers, high-fit non-fencers, averagely-fitfencers, and high-fit fencers).

Statistical analyses

Reaction time (RT) was the interval between the stimulus onsetand the response onset. Trials with incorrect responses or RT withthree SDs below or above the mean RT of each participant wereexcluded. In the go/no-go task, any responses in the no-go trialswere recorded as commission errors. No-responses in go trials wererecorded as omission errors. Two-way (fitness level and fencingexpertise) analysis of variance (ANOVA) was conducted for simpleRT, go/no-go RT, omission and commission errors in go/no-go taskto determine differences and interaction effects among groups withdifferent fitness levels and fencing expertise. Further, we examinedthe effects of fatigue and practice on performance. Data from thefirst and second halves of the experiment were separated.A 2� 2� 2 ANOVA (fitness� expertise� time) with time (first orsecond half as a repeated measure factor) was performed on allbehavioral indices.

Results

Simple RT, go/no-go RT, commission errors and omission errorsin go/no-go tasks were examined. The results were summarized inTable 2. For simple RT, the subjects’ fitness level had no significantmain effect, F (1, 56)¼ .154, p¼ .70, hp2¼ 0. Simple RT was notstatistically different between high-fit and averagely-fit subjects. In

Table 2Results of behavioral measures.

Groupa SRT Go/no-go RT CE OE

Averagely-fit non-fencers 311.63 (65.92) 429.98 (81.79) 5.93 (4.82) 5.80 (4.43)High-fit non-fencers 313.99 (49.41) 430.22 (67.59) 7.33 (5.02) 6.60 (8.41)Averagely-fit fencers 293.08 (29.43) 451.59 (39.33) 6.60 (3.92) 4.80 (1.42)High-fit fencers 300.37 (37.55) 448.85 (63.50) 3.47 (3.46) 5.53 (3.31)

Note. CE and OE are commission and omission errors committed in go/no-go tasks, respectively. Both SRT and go/no-go RT are in the unit of millisecond. SD is presented inparentheses.

a N¼ 15, comparable gender composition for each group.

J.S.Y. Chan et al. / Psychology of Sport and Exercise 12 (2011) 509e514512

addition, there was a non-significant main effect of fencing exper-tise, F (1, 56)¼ 1.71, p¼ .20, hp2¼ .03. The fencers did not have anyadvantage in motor speed over the non-fencers. The interactioneffect of fitness level and fencing expertise was non-significant,F (1, 56)¼ .04, p¼ .84, hp2¼ 0.

Go/no-go RT came from the responses in go trials. Subjects withhigher fitness were not statistically different from those withaverage fitness in terms of go/no-go RT, F (1, 56)¼ .01, p¼ .94,hp

2¼ 0. A non-significant main effect of fencing expertise on go/no-go RT was found, F (1, 56)¼ 1.44, p¼ .24, hp2¼ .03. An interactioneffect of two independent variables was statistically non-significanton go/no-go RT, F (1, 56)¼ .01, p¼ .93, hp2¼ 0.

Commission errors occurred by responding to no-go trials.Averagely-fit and high-fit subjects committed a comparablenumber of errors, F (1, 56)¼ .59, p¼ .44, hp2¼ .01. Whether theparticipants had fencing training or not, the amount of errorscommitted was not significantly different, F (1, 56)¼ 2.03, p¼ .16,hp

2¼ .03. The interaction of fitness level and fencing expertise wasstatistically significant, F (1, 56)¼ 4.07, p< .05, hp2¼ .07. Analysesshow that averagely-fit fencers and averagely-fit non-fencers madea similar number of errors, t (28)¼ .70, p¼ .68. However, the high-fit fencers committed significantly fewer errors than the averagely-fit non-fencers, t (28)¼ 4.60, p< .05 (Fig. 2).

A number of go trials missed by the participants with differentfitness levels was not significantly different, F (1, 56)¼ .34, p¼ .56,hp

2¼ .01. Omission errors were not related to fencing expertisebecause the fencers and the non-fencers committed a similarnumber of omission errors in the go trials, F (1, 56)¼ .62, p¼ .43,

Fig. 2. Commission error results of different experimental groups in go/no-go tasks.Error bars represent standard errors. High-fit fencers committed significantly fewererrors than high-fit non-fencers, whereas averagely-fit fencers and averagely-fit non-fencers committed a comparable amount of errors.

hp2¼ .01. A non-significant interaction of fitness level and fencing

expertise was found, F (1, 56)¼ 0, p¼ .98, hp2¼ 0.To better understand the interaction of fitness and skill level,

two groups of fencers were further divided into four groups, two ofwhich had more fencing experience than the others (15 subjects inhigh-skill and low-skill groups, respectively; mean trainingtime¼ 13.45 years vs 7.36 years). Groups containing players withdifferent skill levels were comparable in fitness. An additional 2� 3ANOVA (fitness and skill levels) showed no significant main effectsof fitness on all behavioral results (SRT: F (1, 54)¼ .32, p¼ .58,hp

2¼ .01; go/no-go RT: F (1, 54)¼ .01, p¼ .93, hp2¼ 0; omissionerror: F (1, 54)¼ .15, p¼ .70, hp

2¼ 0; commission error:F (1, 54)¼ .07, p¼ .79, hp2¼ 0). The main significant effect of skilllevel can only be observed in commission error in the go/no-go task(SRT: F (2, 54)¼ 1.07, p¼ .35, hp2¼ .04; go/no-go RT: F (2, 54)¼ .70,p¼ .50, hp2¼ .03; omission error: F (2, 54)¼ .53, p¼ .59, hp2¼ .02;commission error: F (2, 54)¼ 8.79, p< .001, hp2¼ .25). Interactionsof fitness and skill level on the behavioral measures were non-significant (SRT: F (2, 54)¼ .17, p¼ .85, hp

2¼ .01; go/no-go RT:F (2, 54)¼ .19, p¼ .83, hp2¼ .01; omission error: F (2, 54)¼ .27,p¼ .76, hp2¼ .01; commission error: F (2, 54)¼ .82, p¼ .45, hp2¼ 0).Notably, in the post-hoc tests, the errors committed by the moreskilful were significantly less than those of the less skilful and thenon-fencers (p< .001). The less skilful players made a similarnumber of commission errors as the non-fencers (p¼ .21) (Fig. 3).Long-term sport training is thus thought to be essential to ensureprominent enhancement in cognition.

There were no significant differences on all dependent variablesbetween the first and the second half of the experiment. No

Fig. 3. Effects of different fitness and skill levels on commission error in go/no-gotasks. Error bars represent standard errors. Errors committed by high-skill fencerswere fewer than low-skill and non-fencers.

J.S.Y. Chan et al. / Psychology of Sport and Exercise 12 (2011) 509e514 513

significant interactions among time, fencing expertise and/orfitness were observed on all behavioral measures (Table 3). Theresults suggest that the effects of practice and fatigue on perfor-mance were non-significant and that such effects do not vary withfencing expertise and/or fitness.

Discussion

Consistent with our hypotheses, there was no main effect offitness level on commission errors. The results suggest that youngadults with various fitness levels do not differ in their action inhi-bition. This finding is inconsistent with that of Stroth et al. (2009):namely, that physical fitness modulated executive functioning afteracute exercise. There are two possible explanations for thesediscrepancies. Firstly, physical fitness from acute and chronicexercise enhances cognitive abilities to different degrees. Cognitiveimprovements from acute exercise may be more closely related tophysical fitness than to chronic exercise. Individuals with a higherfitness level are more likely to have their cognition enhanced afteracute exercise than after chronic exercise when other variables areconstant. Secondly, the index of BMI was used as a fitness indicatorin that study. BMI might not be as precise as other indicators, forexample, VO2max; in truly reflecting a person’s aerobic capacity.

Interestingly, in accordance with one of our hypotheses,a significant interaction between fencing expertise and physicalfitness on commission error was observed. There was no significantdifference between fencers and non-fencers when they were in theaverage fitness level. In high fitness levels, the fencers’ actioninhibitionwas significantly better than that of the non-fencers (e.g.,fewer errors). This was the first publication reporting an interactionof fitness and fencing expertise. Enhancement in action inhibitionrequires both a high fitness level and an expertise in cognitivelydemanding situations. This observation is similar to the demands infencing training or competition. Physical, perceptual and psycho-logical factors are all important contributors for successful perfor-mances. Both fitness and expertise are crucial in determiningathletes’ performance in fencing (Roi & Bianchedi, 2008). Wechallenge previous results about the cognitive enhancement ofathletes. Our study suggests that it would be inappropriate tointerpret previous results without knowing the fitness status of theathletes. Thus, for research about cognitive improvements inparticipants with diverse activity levels (Kida et al., 2005;Kioumourtzoglou et al., 1998), physical fitness is an importantvariable for data interpretation. Participants with a comparablefitness level can be directly compared. Our findings suggest thatboth mental training and physical training may be essential forcognitive enhancement.

Churchill et al. (2002) suggested that long-term exercise inducesthe generation of new synapses, neurons, and the enhanced plas-ticity of glia and vascular systems. These improvements arepossibly responsible for perceptual-motor learning and cognitivefunctioning. In addition, animal research suggests that aerobic

Table 3Repeated ANOVA results on behavioral measures.

SRT Go/no-go RT CE OE

Time .95 (.33) .55 (.46) .03 (.87) .20 (.66)Time� fencing

expertise2.27 (.14) 3.58 (.06) .67 (.42) 0 (.95)

Time� fitness 1.08 (.30) 3.23 (.08) 1.31 (.26) 1.46 (.23)Time� fencing

expertise� fitness.13 (.72) 1.73 (.19) 2.86 (.10) 1.17 (.29)

Note. CE and OE are commission and omission errors committed in go/no-go tasks,respectively. F-Value and p-value are presented outside and inside parentheses,respectively.

exercises assist neuronal development and also contribute tohigher plasticity or adaptation of the brain to external stimulations(Colcombe et al., 2003; van Praag, Christie, Sejnowski, & Gage,1999). However, neural plasticity is limited for individuals whohave an average fitness level. Little cognitive enhancement occurswhen there is external stimulation. Neural plasticity and adaptationare more marked for high-fit individuals than for low-fit oraverage-fit individuals, thereby providing a greater opportunity forfurther brain development. Upon receiving specific stimulation, theneurons for inhibitory control develop, thereby consequentlycontributing to the improvement and enhancement of that cogni-tive ability. The type of cognitive ability which will be enhanceddepends upon the type of sport training a person engages in andtheir subsequent sporting experience. The mechanism for theinteraction of aerobic fitness and fencing expertise, however, is stillunclear. More research is required to shed light on this relationshipand to examine the relative neural plasticity.

Furthermore, additional measures in the experiment canpotentially narrow down possible explanations for the results ofcommission error in the go/no-go task. This is of the greatestinterest for the purposes of our study. Participants with differentfitness and fencing expertise showed similar response speeds, asreflected by simple and go/no-go RTs. Various commission errors ingo/no-go tasks were unlikely because of the fundamental motordifferences between different participants rather than the disparityof their higher-level cognitive functions. Because different subjectgroups committed a comparable amount of omission errors, it waspossible that the commission errors were not caused by the care-lessness of the subjects, but by the action inhibition. In addition,performance differences were non-significant between the firstand second half of the experiment. It was unlikely that fatigue andpractice affected the performance of both tasks.

Inconsistent with our hypotheses, the fencers did not showbetter action inhibition than the non-fencers. One may argue thatthe go/no-go task was not sufficiently difficult and challenging todistinguish the action inhibition between the fencers and the non-fencers. However, subjective reports from most participants indi-cated that the stimuli used in the go/no-go task were sufficientlysimilar and that both tasks were suitable and challenging for theparticipants.

This study has important practical and theoretical implications.Firstly, it is essential that athletes improve both fitness and sportexpertise to achieve maximal benefits for their cognitive-motorperformance. Fencing involves both aerobic and executive controldemands. A higher level of aerobic fitness and better action inhi-bition should be emphasized in fencing training. Non-athletesshould participate in exercise or activities that require bothaerobic and cognitive demands. Aerobically challenging andcognitively challenging exercises benefit senior citizens whoexperience cognitive declines (Colcombe & Kramer, 2003). Theo-retically, the present study facilitates the understanding of inhibi-tory control and the effects of experience and fitness on actioninhibition. In the literature, therewere a few reports concerning theinhibitory control of young athletes. It is now clearer that actioninhibition capability declines with aging. Training or exercisetraining should be used to slow down this decline and to preventseniors from suffering other cognitive dysfunctions.

In conclusion, this study provides preliminary results about theimportant relationship between action inhibition, physical fitness,and sport expertise. Aerobic fitness and sport expertise are neces-sary for achieving good cognitive control. Future research shouldexamine the persistence of the enhanced cognitive ability acrossa person’s life span. Whether the improved cognitive skills orabilities in childhood can persist into later life or prevent aging-related cognitive declines needs to be examined. In addition, the

J.S.Y. Chan et al. / Psychology of Sport and Exercise 12 (2011) 509e514514

mechanisms of the interactive effect of fitness and sport trainingare still not well understood. Brain-imaging studies will show howthe brain responds to the increased fitness levels and specificexpertise while also understanding the mechanisms for theimproved inhibitory control. By using structural and functionalMRI, we can examine the structural and functional differencesbetween individuals in different age groups and with differentfitness levels, and hence determine whether the differences arerelated to the action inhibition capacities.

Acknowledgements

The author thanks the participants for their time and cooper-ation. This work was supported in part by the Direct Grant andKnowledge Transfer Grant (FTP110121F18) of the Chinese Univer-sity of Hong Kong to J. H. Y and by the Department of Psychology ofthe Chinese University of Hong Kong to A. C. N. W.

References

Bixby, W. R., Spalding, T. W., Haufler, A. J., Deeny, S. P., Mahlow, P. T.,Zimmerman, J. B., et al. (2007). The unique relation of physical activity toexecutive function in older men and women. Medicine and Science in Sports andExercise, 39, 1408e1416.

Bradshaw, D. I., George, J. D., Hyde, A., LaMonte, M. J., Vehrs, P. R., Hager, R. L., et al.(2005). An accurate VO2max nonexercise regression model for 18e65-year-oldadults. Research Quarterly for Exercise and Sport, 76, 426e432.

Brisswalter, J., Collardeau, M., & René, A. (2002). Effects of acute physical exercisecharacteristics on cognitive performance. Sports Medicine, 32, 555e566.

Chodzko-Zajko, W. J. (1991). Physical fitness, cognitive performance, and aging.Medicine and Science in Sports and Exercise, 23, 868e873.

Churchill, J. D., Galvez, R., Colcombe, S., Swain, R. A., Kramer, A. F., &Greenough, W. T. (2002). Exercise, experience and aging brain. Neurobiology ofAging, 23, 941e955.

Ciesielski, K. T., Harris, R. J., & Coffer, L. F. (2004). Posterior brain ERP patternsrelated to the go/no-go task in children. Psychophysiology, 41, 882e892.

Colcombe, S. J., Erikson, K. I., Scalf, P. E., Kim, J. S., Prakash, R., McAuley, E., et al.(2006). Aerobic exercise training increases brain volume in aging humans.Journal of Gerontology: Medical Sciences, 61, 1166e1170.

Colcombe, S. J., & Kramer, A. F. (2003). Fitness effects on the cognitive function ofolder adults: a meta-analytic study. Psychological Sciences, 14, 125e130.

Colcombe, S. J., Kramer, A. F., Erickson, K. I., Scalf, P., McAuley, E., Cohen, N. J., et al.(2003). Cardiovascular fitness, cortical plasticity, and aging. Proceedings of theNational Academy of Sciences of the United States of America, 101, 3316e3321.

Di Russo, F., Taddei, F., Apnile, T., & Spinelli, D. (2006). Neural correlates of faststimulus discrimination and response selection in top-level fencers. Neurosci-ence Letters, 408, 113e118.

Etnier, J. L., Salazar, W., Landers, D. M., Petruzzello, S. J., Han, M., & Nowell, P. (1997).The influence of physical fitness and exercise upon cognitive functioning:a meta-analysis. Journal of Sport and Exercise Psychology, 19, 249e277.

George, J. D., Stone, W. J., & Burkett, L. N. (1997). Non-exercise VO2max estimation forphysically active college students. Medicine and Science in Sports and Exercise,29, 415e423.

Gliner, J. A., Matsen-Twisdale, J. A., Horvath, S. M., & Maron, M. B. (1979). Visualevoked potentials and signal detection following a marathon race.Medicine andScience in Sports and Exercise, 11, 155e159.

Hassmén, P., Koivula, N., & Uutela, A. (2000). Physical exercise and psychologicalwell-beings: a population study in Finland. Preventive Medicine, 20, 17e25.

Jackson, A. S., Blair, S. N., Mahar, M. T., Wier, L. T., Rossand, R. M., & Stuteville, J. E.(1990). Prediction of functional aerobic capacity without exercise testing.Medicine and Science in Sports and Exercise, 22, 863e870.

Kamijo, K., Hayashi, Y., Sakai, T., Yahiro, T., Tanaka, K., & Nishihira, Y. (2009). Acuteeffects of aerobic exercise on cognitive function in older adults. Journal ofGerontology: Psychology Sciences, 64, 356e363.

Kida, N., Oda, S., & Matsumura, M. (2005). Intensive baseball practice improves thego/no-go reaction time, but not the simple reaction time. Cognitive BrainResearch, 22, 257e264.

Kioumourtzoglou, E., Kourtesses, T., Michalopolou, M., & Derri, V. (1998). Differ-ences in several perceptual abilities between experts and novices in basketball,volleyball and water-polo. Perceptual and Motor Skills, 86, 899e912.

Kramer, A. F., & Erickson, K. I. (2007). Capitalizing on cortical plasticity: influence ofphysical activity on cognition and brain function. Trends in Cognitive Sciences, 11,342e348.

Kramer, A. F., Hahn, S., Cohen, N. J., Banich, M. T., McAuley, E., Harrison, C. R., et al.(1999). Ageing, fitness and neurocognitive function. Nature, 400, 418e419.

Lee, I. M., Hsieh, C. C., & Paffenbarger, R. S. (1995). Exercise intensity and longevity inmen. JAMA, 273, 1179e1184.

Lybrand, W., Andrews, T., & Ross, S. (1954). Systematic fatigue and perceptualorganization. American Journal of Psychology, 67, 704e707.

Ozel, S., Larue, J., & Molinaro, C. (2004). Relation between sport and spatial imagery:comparison of three groups of participants. The Journal of Psychology: Interdis-ciplinary and Applied, 138, 49e64.

Raz, N. (2002). Aging of the brain and its impact on cognitive performance. InF. I.M. Craik, &T. A. Salthouse (Eds.), The handbook of agingand cognition (pp.1e90).Mahwah, NJ, USA: Erlbaum.

Roi, G. S., & Bianchedi, D. (2008). The science of fencing: implications for perfor-mance and injury prevention. Sports Medicine, 38, 465e481.

Rossi, B., Zani, A., Taddei, F., & Pesce, C. (1992). Chronometric aspects of informationprocessing in high level fencers as compared to non-athletes: an ERPs and RTstudy. Journal of Human Movement Studies, 23, 17e28.

Stroth, S., Kubesch, S., Dieterle, K., Ruchsow, M., Heim, R., & Kiefer, M. (2009).Physical fitness, but not acute exercise modulates event-related potentialindices for executive control in healthy adolescents. Brain Research, 1269,114e124.

Thomas, J. R., Landers, D. M., Salazar, W., & Etnier, J. L. (1994). Exercise and cognitivefunction. In C. Bouchard, R. J. Shephard, & T. Stephens (Eds.), Physical activity,fitness, and health: International proceedings and consensus statement (pp.521e529). Champaign, IL, England: Human Kinetics Publishers.

van Praag, H., Christie, B. R., Sejnowski, T. J., & Gage, F. H. (1999). Running enhancesneurogenesis, learning, and long-term potentiation in mice. Proceedings of theNational Academy of Sciences of the United States of America, 96, 13427e13431.