The study was conducted as a collaborative project between the Department of Physiology, Jawaharlal Institute of Postgraduate Medical Education & Research (JIPMER), and a rural residential school (Jawahar Navodaya Vidyalaya), in Pondicherry, India, from 28th March, 2012 to 30th June, 2013. The study was commenced after the approval from the JIPMER Scientific Advisory Committee and JIPMER Institute Ethics Committee for human studies.

Participants: Participants were recruited from Jawahar Navodaya Vidyalaya, Pondicherry studying in grade VII to XII in the age group of 12-17 years. All the participants were full time school residents, and belonged to lower socio-economic group. After explaining the study to the participants, their written informed assent and a written consent from their local guardian or parents were obtained. The study intervention was incorporated into the school curriculum as a part of the “National Adolescent Program”. Age of the participants was taken from the date of birth specified in their school records.

Students who volunteered for the study (n=451; boys: 255, girls: 196) underwent a general physical examination and were screened for any physical or mental ailments. Participants with history of previous or current neurological disorder, alcohol abuse, epilepsy, mental retardation or any drug intake were excluded from the study and 439 students (boys: 250, girls: 189) were recruited. The participants were randomized into two groups: Structured Physical Activity (SPA) group (n= 219: boys 124, girls 95) and unstructured physical activity (USPA) group (n=220; boys: 126, girls: 94). Based on their athletic training status, the participants in both groups were further subdivided into the following four sub-groups: SPA athlete (n = 39; boys: 22, girls: 17), SPA non-athlete (n = 180; boys: 102, girls: 78), USPA athlete (n = 40; boys: 23, girls: 17) and USPA non-athlete (n = 180; boys: 103, girls: 77). The flow of the study is given in [Table/Fig-1].

[Table/Fig-1]:

Flow Diagram of the study

The athlete students included in the study are those who were routinely participating in national level interscholastic sporting events that included football, volleyball, badminton, basketball, kho-kho, and kabaddi. They were routinely undergoing supervised physical conditioning training for a minimum of 25 hours a week.

Sample size: Sample size was estimated to detect the minimum difference of 10% of Letter cancellation test (time) between the groups at the 5% level of significance and 90% power. The sample size required for the study was 120. The sample size was further modified to 145 in each group with an expected dropout of 20%. We were able to achieve the sample size of n=139 in unstructured physical activity group and n=136 in structured physical activity group)

Randomization: Participants were age and gender stratified and then randomized using sequentially numbered, sealed, opaque envelope technique with block size 8 and allocation ration 1:1. The principal investigator (Dr. Vivek Kumar Sharma) enrolled the participants, and co-investigator (Dr. Senthil Kumar Subramanian) did the randomization and allocation of participants to the specific group. The outcomes were analysed by the principal investigator who was blinded to the randomization procedure.

Intervention: Structured physical activity was designed by following the physical activity guidelines for adolescents given by the World Health Organization (WHO) and the Department of Health, UK i.e., structured physical activity of 60 minutes per day to meet the goal of at least 30 minutes per day, three times a week, of vigorous intensity physical activity, muscle strengthening & stretching and bone strengthening exercises [1–3]. The details of the SPA schedule are enclosed in the Annexure: JNV module. Physical activity sessions were administered by the school physical education instructors. They submitted the attendance and progress report of the training sessions to the investigators every week. Sessions were administered in separate playgrounds for each group to minimize cross-contamination. In the SPA non-athletes group, the participants were given SPA training for duration of two hours a day, six days a week, for six months. In the SPA athletes group, the participants were given SPA training for the same duration apart from their routine sport specific physical conditioning training. The intensity of the exercise was monitored by perceived effort scale (0 is the level of sitting idle and 10 is maximal effort) or by randomly measuring the pulse rate during the physical activity sessions (50-70% of maximal heart rate moderate intensity, >70% vigorous intensity).

In the USPA group, the participants were given supervised but unstructured physical activity for duration of two hours a day, six days a week, for six months. They were free to choose any indoor or outdoor recreational activity as per their wish, on any given day. Indoor recreational activities include gymnasium, music, drawing, and craft. Outdoor recreational activities include football, basketball, kho-kho, badminton, basketball, skipping, and gardening. Athletes in the USPA group continued their sport-specific physical conditioning training and participated in any USPA of their liking on any given day, for the above-mentioned duration [14].

Parameters: We chose those paper and pencil tests that are commonly used in the literature so that the results are easy comparable with other studies [15] and for the ease of administration in those places where computer is still not available. One practice sesion was given to all the participants 3 days before the baseline recording as scores for NCT has been shown to significantly improve between the first and second administrations of the NCT. Students were asked to report to the school examination room at 9 AM, two hours after having a light breakfast on the day of the tests.

NCT consisted of following tests and administered in the same order as mentioned below:

This NCT was again administered to all the participants after six months of intervention. We have discussed the effect of age and sex on these parameters in our earlier publication [20].

Results were analysed only for those participants who had completed the respective intervention with more than 80% attendance (per-protocol analysis). Interventions were completed by 347 participants (boys: 190, girls: 157) grouped as follows: SPA-Non athlete (n = 136, boys: 77, girls: 59), SPA athlete (n = 36, boys: 19, girls: 17), USPA Non-athlete (n = 139, boys: 75, girls: 64), USPA athlete (n = 36, boys: 19, girls: 17). The data were recorded and statistically analysed using the SPSS software, version 19, by the principal investor (who was blinded to the group allocation). Between groups analysis was done using unpaired t-test for normally distributed data and Mann-Whitney U test for non-normally distributed data. Within group analysis was done using paired t-test. Effect of intervention was quantified using effect size Pearson ‘r’ correlation coefficient.

[Table/Fig-1] shows the flow of the study.

[Table/Fig-2] shows that subjects in both the groups have no significant difference in their baseline height, weight and age comparison. Randomization has ensured equal distribution of subjects according to their sex (boys and girls) and physical activity status (athletes and non-athletes). Further in sub-group analysis based on physical activity status, baseline height, weight and age of the subjects showed no significant difference in both the USPA and SPA groups.

[Table/Fig-3] shows that there was no baseline difference between groups in the tested neuro-cognitive parameters. There was significant improvement in LCT omission, LCT time, TTA, TTB, and RFFT (designs, rotations and perseverations) in both USPA and SPA groups after six months of respective interventions. On comparison of pre-post difference between the groups, SPA group showed significantly better improvements in LCT time, TTB and RFFT (designs, rotations and perseverations).

[Table/Fig-4] shows that there was no baseline difference of tested parameters in non-athletes of both groups. There was significant improvement in LCT omission, LCT time, TTA, TTB, and RFFT in both USPA and SPA non-athlete groups after six months of intervention. In addition, improvements seen in in LCT (time), TTB, RFFT patterns and perseverations in SPA non-athletes group, was significantly better when compared to that seen in USPA non-athletes.

[Table/Fig-5] shows that there was no baseline difference in tested parameters in athletes of both groups. There was significant improvement in LCT (omission), TTB & RFFT (designs, rotations and perseverations) in USPA athlete group after six months of intervention. Significant improvement in LCT (Omissions & time), TTB, & RFFT (designs, rotations, and perseverations) was observed in SPA athlete group after. In addition, improvements seen in LCT (time) and RFFT perseverations in SPA athletes group, was significantly better when compared to that seen in USPA athletes.

[Table/Fig-6] demonstrates that there was significantly higher percentage improvement in LCT, TTB and RFFT in SPA group than in USPA group. Further, on subgroup comparison there was significantly higher improvement in SPA group in LCT (time), TTB & RFFT (patterns) with a decrease in perseverations. In athlete groups also, significantly higher improvement was seen in SPA group in LCT (time) & RFFT (rotations) with a decrease in perseverations.

Our results show that there was significant improvement in all the tested parameters in both SPA and USPA groups after six months of intervention. Thus, our study findings confirm that physical activity training, both structured and unstructured, improves the perceptual cognitive abilities of adolescents [5,8–10,15,21]. Physical activity of at least two hours duration every day, for six months, improves the cognitive functioning of adolescents even when the physical activity is unstructured as in when the adolescents are let free to choose their activity of choice. Our results corroborate with previous studies which have also found similar effects on cognitive functioning of children and adolescents after aerobic exercise interventions [5,8–10,15,21]. However, to the best of our knowledge, there is no study conducted in which USPA has been compared with SPA in adolescents for the long duration of six months. Our findings demonstrate that there was overall significantly higher percentage improvements observed in SPA group subjects in the cognitive domains of attention, concentration, mental flexibility, and non-verbal fluency than USPA group subjects. Further, on sub-group analysis, we observed that there was significantly higher percentage improvements in SPA non-athlete group in LCT, TTB & RFFT when compared to that seen in USPA non-athlete group whereas, more improvement in SPA-athlete group was seen only in LCT and RFFT when compared to that seen in USPA-athlete group. These observations suggest that even if adolescents are undergoing physical conditioning athletic training, structuring of physical activity further improves their cognitive functioning. We hypothesize that SPA is better than USPA because it may be due to a higher number and intensity of cognitive loading components in the module that included yoga, moderate-high intensity endurance exercises and higher number of sports activities.

We hypothesize to explain our findings with the help of the following established mechanisms (that act as a link between physical activity and cognition):

Thus, long-term participation in both USPA and SPA trains the individual in executing complex motor movements and cognitively engaging tasks like goal-directed thinking, thereby influencing their brain executive functioning, however, the improvement is comparatively less with USPA which may involve more repetitive PA tasks such as walking and running.

The study was done in a residential school where all the students were provided balanced food and had equal opportunity for physical activity. This could have helped us to reduce the effect of cultural and socio-economic differences among the participants. In addition, it was possible for us to do uniform implementation and monitoring of SPA and USPA training in the school. However, we were not able to avoid cross contamination between the groups.

Except for some minor injuries on the field, that were attended to immediately by the resident nursing staff in the school, there were no serious adverse events during the trial period.

Dropout rate was nearly 20% in each group and the common reasons for dropping out were academic commitments, term exams, and subsequent holidays.

Our study demonstrates that WHO recommended structured physical activity training in adolescents provides additional benefit in their neurocognitive improvements when compared to unstructured physical activity training.