Abstract:Objectives: Thisstudy investigated the effect of moderate to vigorous aerobic physical activityon the cognitive functioning of children.

The aim was to attempt to identify apositive relationship between physical activity and cognitive functioning. Thiswas done by testing the cognitive functioning of a group of thirty-ninechildren before and after exercise, as well as before and after a sedentaryperiod of the same length. Methods: The cognitivefunctioning of the participants was measured using the Trail Making Test. Thistest is made up of two parts, A and B, which measure different elements ofcognitive functioning. This was performed pre and post a moderate to vigorousforty minute session of aerobic physical activity. The Trail Making Test wasalso carried out pre and post a forty minute sedentary period in order tocompare results and to investigate the effects, if any physical activity had onthe performance of the test.Results: Physical activityappeared to have an effect on the performance of Part A of the Trail MakingTest.

This was due to a significant difference in the change in Part A timespost physical activity when compared with post sedentary. Physical activity wasnot found to have a significantly different effect on the performance of Part Bof the test when compared with the sedentary times recorded.Conclusion: Ourresults indicate that acute bouts of moderate to vigorous physical activityhave a positive effect on the areas of cognitive functioning involved in visualsearch and motor response.

 Introduction:Cognitive functioning (C.F.) can be defined as any “intellectual process by which one becomesaware of, perceives, or comprehends ideas.It involves all aspects of perception, thinking, reasoning,and remembering”(Mosby’s Medical Dictionary, 2009). Physical activity(P.A.) refers to bodily movements performed by muscles in the body which resultin the expenditure of energy (Bouchard et al, 1993).

The benefits of P.A. on general health and well-beinghas been well documented and researched. It has been proven to reduce the riskof illnesses such as cardiovascular disease, types of cancer, diabetes, anddepression (Bouchard & Shephard, 1994).

While the benefits of P.A. onhealth have been long accepted, its relationship with C.F. has been debated.Despite some disagreement over the effect that exercise has on the C.F.

ofindividuals, a huge amount of research in this field has produced positiveresults. An example of such a study is that conducted by the CaliforniaDepartment of Education in 2001. In a study of over 900,000 children betweenthe ages of 10 and 14, a positive relationship was found between P.

A. level andacademic achievement (CDE, 2001). Another study conducted in 2011 suggested that P.A.

may influence the manner in which the brain organises its attentional resources(Hillman et al, 2011). They found that exercise could have an impact on themanner in which the brain responds to external stimuli, allowing individuals tobe more efficient in terms of attention span. This would have a major impact onlearning, as it would increase the brains capacity for informational input,while reducing the energy expenditure.Though large numbers of studies have been conducted inthis area, it remains unclear as to which types of P.A. cause the greatesteffect, if any, on C.F.

In a 2009 study carried out by Pesce et al, the effectof different forms of exercise on recall memory in children were investigated.This study reported that acute bouts of vigorous P.A.

promoted the storing ofnew information into the long-term memory of the subjects, while bouts oflow-moderate exercise did not (Pesce et al, 2009).The area of fitness which appears to have the greatestimpact on C.F. from reviewing relevant literature is aerobic fitness. Childrenwith high levels of aerobic fitness were found to be more successful than theirless-fit counterparts when tested on their abilities to process and learn newinformation. This was done by providing information for the children to studyalone, before being asked to recall the information later (Raine et al, 2013).

It is without question that executive function plays akey role in academic performance. A study of 51 school children found thatthose with poor levels of executive function consistently produced lower gradesthan those with higher levels (St Clair-Thompson et al, 2006). Although theimportance of intellectual strengths of students clearly effect academicperformance, it can equally be argued that non-intellectual strengths such asmotivation and discipline play an integral role (Duckworth & Seligman,2006). If true, P.A. may play a major role in improving the academicperformances of students in schools, due to its ability to developself-discipline, teamwork, and potentially self-esteem (McCauley et al, 2005).

 Based on the most effective previous research carriedout in this area, we decided to test the effects of an acute bout ofmoderate-vigorous, aerobic physical activity on cognitive functioning in children.Cognitive functioning was measured using the Trail-Making Test. This test waschosen as it provides a measure of two domains C.F. separately, with visualsearch and motor-speed skills being tested in Part A, while mental flexibilityand set-switching are measured in Part B (Bowie & Harvey, 2006). The hypothesis for the study was that an acute bout ofmoderate-vigorous, aerobic P.

A. would have a positive effect on C.F. inchildren.

 Methodology: ParticipantsThirty-nine (22 females and 17 males) fourth class students with a mean age of9.6 years and standard deviation of 0.8 participated in the study. The studentswere from the West Dublin area. The participants were from a co-educationalschool which allowed for mixed gender testing. Participants were selected fromthe two fourth classes within the school.

None of the students in either of theclasses reported any forms of learning disabilities. The students were providedwith consent forms, which were to be signed by their parents or guardians toallow them to take part in the study. Students were also required to sign anassent form to acknowledge their own willingness to take part in the study.Participating students were all assigned a number.

This allowed for ease ofrecording results and only the testers were aware of which numbers correspondto each student. No initial cognitive or fitness testing was carried out on thestudents to allow for an unbiased selectionTrailMaking TestForour study, we used the Trail Making Test (TMT) in order to test cognitivefunction. The TMT has been shownto be a reliable and valid neuropsychological test (Tombaugh, 2004). Themost widely used version of the TMT comprises parts A and B. In part A, thesubject uses a pencil to connect a series of 25 encircled numbers in numericalorder. In part B, the subject connects 25 encircled numbers and letters innumerical and alphabetical order, alternating between the numbers and letters.

For example, the first number ”1” is followed by the first letter ”A,”followed by the second number ”2” then second letter ”B” and so on. Thiscontinues up to the number “13”, ensuring there are 25 encircled characters ineach test. The numbers and letters are placed in a semi-random fixed order, insuch a manner as to avoid overlapping lines being drawn by the examinee.

Each participant was given a “practice” Test A and Test B directly beforecarrying out the corresponding tests. These practise tests were a goodrepresentation of the actual TMT. The TMT measures a variety of cognitivefunctions within the person being tested.

Test A is generally presumed to be atest of visual search and motor speed skills, whereas Test B is considered alsoto be a test of higher level cognitive skills such as mental flexibility andset shifting. We tested the effect of physical activity on the performancelevels on the TMT.  Systemfor Observing Fitness Instruction Time (SOFIT)Tomeasure the intensity of the exercise we used the System for Observing FitnessInstruction Time (SOFIT) observation tool. The SOFIT observation tool has been shown to be a reliable and validmeans of measuring physical activity levels (Rowe et al., 2004). This involved monitoring a specific student’sactivity levels at specific intervals, for a determined amount of time, e.g.

Student 1 is monitored every minute for 5 minutes. Activity levels were given arating between 1 to 5 with each number indicating a different level ofintensity as outlined below:1.         Lying down 2.         Sitting3.         Standing 4.         Walking5.

         Vigorous.While one tester led the lesson, the other three were observing and recordingthe intensity level of a participant. After 5 minutes of observing we wouldselect a different participant from the group and repeat the process. We didthis throughout the lesson. ExercisePhysicalactivity in this study was considered to be any aerobic exercise performed at amoderate to vigorous intensity.

For this study, we wanted the participants toperform physical activity at a moderate to vigorous intensity for forty minutesin order to ensure the students were utilising their aerobic fitness. In orderto achieve this we compiled a lesson which consisted of three sections, awarm-up, circuit and a cool down.Warm-Up: The warm-up started with the participants running around in a squareand performing instructions given to them by the teacher. The instructionsincluded high knees, heel flicks, touching the ground with both hands, jumpingup as high as they could. Circuit: The circuit was 25 minutes long and was madeup of six sections: Interval sprints (half basketball court), jumping jacks,high knees, heel flicks, repeated standing jump (both feet together) anddistance running (80% of full basketball court repeated).

The participants weredivided amongst the stations and performed the task for thirty seconds with athirty second rest period between each station. Once all participants hadcompleted all the stations they were given a two-minute rest period. This wasrepeated three times.Cool-Down: Consisted of a light jog the full length of a basketball courtfollowed by a number of various dynamic stretches. ProcedureThedata for this research paper was collected over two days. The testing for was carried out as follows:Activegroup: Each student was brought to the designated testing room in groups of 4and completed the TMT. Once completed they returned to their classroom and thenext 4 students were brought to the same room. Once the whole class hadcompleted the TMT, the class as a whole were brought outside and performed 40minutes of vigorous aerobic PA.

SOFIT observation sheets were filled in duringactive time using method above. Following the exercise, the class were againtested with the TMT using the same procedure. Sedentarygroup: Each student was brought to the designated testing room in groups of 4and completed the TMT. Once the student completed the TMT they returned totheir classroom for 40 minutes. This 40-minute period was defined as sedentarytime where they engaged in the normal class routine. After each student hadspent the allocated time in the classroom, they were brought back to thetesting room to complete the TMT again.

This procedure was repeated for eachstudent.Day1Activegroup: Class ASedentarygroup: Class BDay2Activegroup: Class BSedentarygroup: Class A  StatisticalAnalysesDataanalysis was performed using the software Statistical Package for the SocialSciences (SPSS).Theprincipal modes of data analysis were one-way repeated measures analysis ofvariance (ANOVA) and paired sample t-tests.Theone-way repeated measures ANOVA was used to analyse the effect of physicalactivity (pre and post) on the TMT scores of each participant twice during thestudy. This provided us with descriptive statistics, a Wilk’s Lambda value ofsignificance, as well as pairwise comparisons.Thepaired sample t-tests were used to analyse the difference in times between thePre and Post scores for Active and Sedentary times.

This was carried out foreach of Parts A and B of the TMT separately. The total time taken to completeparts A and B was also analysed using this method.   Results: Resultsfrom SOFIT averaged for the group as 4.5, which falls between walking andvigorous category. A one-way repeated measures ANOVAwas conducted to compare scores on the effect of physical activity on Part A ofthe TMT at Time 1 (pre activity), Time 2 (post activity), Time 3 (presedentary), and Time 4 (post sedentary). The means and standard deviations arepresented in Table 1. There was a significant effect for time, Wilks’ Lambda =.432, F (3, 36) = 15.

8, p < .0005, multivariate partial eta squared = .57. Table 2 below shows the differencesbetween each of the Times for TMT Part A outlined above using the BonferroniPost-Hoc test.

Significant differences were found between Times 1 and 2, Times1 and 4, and Times 3 and 4 (all p<0.05). All other comparisons were found tobe insignificant.        Table 1Descriptive statisticsfor the effect of physical activity on TMT Parts A&B at Time 1, Time 2,Time 3 and Time 4 Descriptive Statistics   Test A Test B Pre Active Post Active Pre Sedentary Post Sedentary Pre Active Post Active Pre Sedentary Post Sedentary Mean (t) 48.

3079 35.0718 43.3582 37.1582 118.8895 97.1151 105.3964 91.5626 Standard Deviation 21.

48137 16.62194 15.73948 17.56802 51.43195 44.61589 40.00113 49.21863 Number of Participants 39 39 39 39 39 39 39 39   Table 2Pairwise comparisonsfor the effect of physical activity on TMT Parts A&B at Time 1, Time 2,Time 3 and Time 4 Pairwise Comparisons (significant values) Test A Pre Active Post Active Pre Sedentary Post Sedentary Test B   Pre Active Post Active Pre Sedentary Post Sedentary   Pre Active   0.

000 1.000 0.036 Pre Active   0.000 0.789 0.059   Post Active 0.000   0.

059 1.000 Post Active 0.000   1.000 1.000   Pre Sedentary 1.000 0.059   0.

005 Pre Sedentary 0.789 1.000   0.041   Post Sedentary 0.036 1.000 0.005   Post Sedentary 0.059 1.

000 0.041                  Figure 1 A one-way repeated measures ANOVAwas conducted to compare scores on the effect of physical activity on Part B ofthe TMT at Time 1 (pre activity), Time 2 (post activity), Time 3 (presedentary), and Time 4 (post sedentary). The means and standard deviations arepresented in Table 1 above. There was a significant effect for time, Wilks’Lambda = .553, F (3, 36) = 9.68, p < .0005, multivariate partial eta squared= .

48. Table 2 above shows the differencesbetween each of the Times for TMT Part B outlined above using the BonferroniPost-Hoc test. Significant differences were found between Times 1 and 2, andTimes 3 and 4 (all p<0.

05). All other comparisons were found to beinsignificant.          Figure 2   A paired-samples t-test wasconducted to evaluate the effect of physical activity on students’ scores inTMT part A. There was a statistically significant decrease in TMT part A scoresfrom Pre – Post Active Part A (M = 13.24, SD = 14.71) to Pre – Post SedentaryPart A (M = 6.

2, SD = 10.61), t (38) = 2.39, p <. 05 (two-tailed). The meandecrease in Pre – Post Part A scores was 7.04 with a 95% confidence intervalranging from 1.09 to 12.

99. The eta squared statistic (0.131) indicated amoderate effect size. A paired-samples t-test wasconducted to evaluate the effect of physical activity on students’ scores inTMT part B. There was not a statistically significant decrease in TMT part B scoresfrom Pre – Post Active Part B (M = 21.77, SD = 28.88) to Pre – Post SedentaryPart B (M = 13.

83, SD = 30.17), t (38) = 1.21, p >.

05 (two-tailed). Themean decrease in Pre – Post Part B scores was 7.94 with a 95% confidenceinterval ranging from -5.39 to 21.27. The eta squared statistic (0.

04)indicated a small effect size. A paired-samples t-test wasconducted to evaluate the effect of physical activity on students’ overallscores in TMT. There was not a statistically significant decrease in TMT scoresfrom Pre – Post Active Total (M = 35.

01, SD = 35.08) to Pre – Post SedentaryTotal (M = 20.03, SD = 32.11), t (38) = 1.9, p >. 05 (two-tailed).

The meandecrease in Pre – Post scores was 14.98 with a 95% confidence interval rangingfrom -0.95 to 30.9. The eta squared statistic (0.09) indicated a moderateeffect size.All results above can be seen intable 3 and 4 below.

 Table 3 Paired Sample Statistics   Mean Standard Deviation Pair 1 Difference between Pre and Post Active Test A   13.2362 14.70758 Difference between Pre and Post Sedentary Test A 6.2000 10.60885 Pair 2 Difference between Pre and Post Active Test B   21.7744 28.

88280 Difference between Pre and Post Sedentary Test B 13.8338 30.17413 Pair 3 Total Pre – Total Post (Active)   35.0105 35.08191 Total Pre – Total Post (Sedentary)   20.

0338 32.11204 Paired sample statistics for Pre –Post Active and Sedentary Parts A and B, and Total Pre – Total Post      Table 4Paired sample t-tests for Pre – PostActive and Sedentary Parts A and B, and Total Pre – Total Post Paired Sample T-Tests     Mean   Standard Deviation 95% Confidence Interval of the Difference Lower 95% Confidence Interval of the Difference Upper   t   df   Sig. (2-tailed)   Pair 1   7.03615   18.

35357   1.08661   12.98569   2.394   38   .

022   Pair 2   7.94051   41.12293   -5.39000   21.27103   1.206   38   .235   Pair 3   14.97667   49.

12639   -.94827   30.90160   1.904   38   .

065   Pair 1 Difference between Pre and Post Active Test A–Difference between Pre and Post Sedentary Test A Pair 2 Difference between Pre and Post Active Test B–Difference between Pre and Post Sedentary Test B Pair 3 Total Post Active – Total Post Sedentary    Discussion:The results of the study do not definitively supportthe hypothesis. The post scores were significantly lower in Parts A and B ofthe TMT after both exercise and sedentary time. This indicates that althoughthe participants improved every time they completed the test, it was notnecessarily due to P.A. The most obvious positive correlation between P.A. andC.

F. from our research was the effect P.A. appeared to have on the scoresobtained in Part A of the TMT. The average time taken to complete Part A was reducedto a greater extent after physical activity when compared with the postsedentary test. This can be seen from the pairwise comparisons in Table 1, aswell as in Pair 1 of the paired sample tests in Table 4. This indicates thatthe exercise bout improved the participant’s visual-search and motor speedskills.

This is in keeping with previous studies carried out on the effects ofacute P.A. on C.F. It has been found that acute exercise increases the speedwith which the body can physically respond during cognitive testing(Tomporowski, 2003).

This is potentially due to the discovery that exercisecauses an increase in the production of neurotransmitters, which aid inprocessing speed (McMorris et al, 2011). The intensity of the P.A. in our studymay also have played a role in the observed effect on Part A times. On averagethe participants scored a 4.5 out of 5 on the SOFIT observation tool,indicating students remained between walking and vigorous activity throughoutthe lesson. High intensity bouts of exercise such as this have been shown toimprove basic C.F.

such as choice-reaction times post-exercise due to increasedadrenaline levels in the blood (Brisswalter et al, 2002). The exerciseperformed, though quite high in intensity, did not take any of the participantsto exhaustion. This may also have benefited the performance of theparticipants, as prolonged moderate to vigorous exercise of up to 40 minuteswithout reaching exhaustion has been found to provide the optimum arousal forcontingent negative variation. This has been found to be involved with C.F’ssuch as expectancy and attention (Kamijo et al, 2004).Despite the apparent benefits of high intensity P.A.

on simple C.F. tasks, it has been found to be less effective on the performanceof more complex tests (Hogervorst et al, 1996). This is in keeping with theresults obtained in our study in relation to Part B of the TMT, which measuresmental flexibility and set-shifting. Our results showed no significantdifference in the effects of P.A. on the times for Part B when compared withthe sedentary scores, as can be seen above in Pair 2 of the paired sample testsin Table 4. This is in keeping with previous research carried out, whichsuggested that P.

A. had little or no effect on the set-shifting component ofexecutive function (Coles & Tomporowski, 2008).Although our results do not indicate a clear positiverelationship between P.A. and C.F., the study contained several limitations.Population size was one of these limitations, as only thirty-nine participantstook part in the study.

This may not have provided a full representation of theeffects of P.A. on C.

F. for the age being tested. Another limitation was theapparent effect of trial familiarisation on the study. Upon completing thestudy, each participant had completed the TMT four times. Although these testswere done across two days, the participants may have displayed improvements dueto repetition, and not necessarily as a result of P.A.

  ConclusionThe results of our study indicate that P.A. appears toeffect certain aspects of C.F., specifically visual search and motor response,while being ineffective to others such as set-shifting and mental flexibility.Although our results and previous studies providelittle support for a positive effect of acute P.A. on complex C.

F., it ispossible long term P.A.

programs could provide different findings. Long termexposure to P.A. has been found to improve C.F.

in elderly sufferers ofParkinson’s disease over time (Tanaka et al, 2009). Also, in a meta-analyticalstudy of 134 case studies based on the effects of P.A. on C.F., the effects ofboth acute and long term P.A. were investigated.

It was found that acute P.A.provided in general small, short term improvements on C.

F. However, long termexercise programs could potentially provide more significant, longer lastingeffects on C.F. (Etnier et al, 1997). Also, the effects of both acute and longterm P.

A. on C.F. may vary with the ages of the participants being tested. As a result, further research must be carried out onthe long term effects of P.A.

on C.F. Also, a wide range of ages must be testedto provide a more complete picture of the potential relationship. If adefinitively positive relationship were to be established, the repercussions onboth health and academic achievement in children could be enormous. It wouldalso put an onus on schools to improve activity levels, as no other institutionhas as much influence on the lives of children, irrespective of their economicbackground or family life (Resaland et al, 2015).