DISCUSSION

The time taken to trace the maze and the number of errors were significantly decreased after 30 days of yoga practice, whereas the control group did not show this change. Over 5 trials both yoga and control groups showed a significant decrease in time taken to complete the maze on Day 1. On Day 30 the control group continued to show a significant decrease with repeated trials, which. was not observed in the yoga group. However, the number of errors was significantly less at Trial 5 compared to Trial 1, for the yoga group on Day 1, but not on Day 30. The control group did not show any differences in number of errors over 5 trials.

Maze learning based on time taken, was improved in both groups on Day 1. On Day 30 the control group showed a decrease in time taken at Trial 5 compared to Trial 1. However, the yoga group showed no significant effect of repetition on Day 30, for time and errors. This may have been related to the fact that on Day 30 at Trial 1, both time and error scores were significantly lower for the yoga group compared to Trial 1 scores on Day 1, and there was not much scope for further reduction. These trends are evident from Fig. 1.

Also, on Day 1, the time taken to complete the maze came down from (mean) 97.3 sec to 35.2 sec in 5 trials in the yoga group, whereas in the control group it came down from 98.3 to 54.1 sec. Thus on Day 1 itself the yoga group had subjects who were fast learners, and therefore the two groups were not well matched for maze learning.

This decrease in time and error scores in the yoga group after 30 days of yoga practice compared with before may be related to this difference between the groups at baseline. At baseline the two groups did not take significantly different time to complete the maze. However, while the yoga group showed maze learning with reduced errors on Day 1, this change was not observed in the control group. This was another difference between the groups at baseline.

The effect of yoga on maze learning may also be related to factors such as motor performance, with improved motor performance seen as an increase in the speed of tracing the maze path. All subjects were right-hand dominant and used their right hand to trace the maze. Right hand activity in maze tracing involves more of left-hemispheric, abstract-conceptual strategy where the spatial configuration of the maze is learned as a series of left or right turns, with some interspersed up or down movements (9). In the present study, maze tracing may be believed to be carried out by left-right orientation. The decrease in time and error scores after yoga could also be related to better motor performance and improved left-right orientation, though baseline group differences prevent any definitive conclusion being made.

In summary, (1) The five trials on Day 1 and the five trials on Day 30 behave like 10 successive trials which lead to progressive improvement in both groups. (2) The yoga group has subjects who are fast learners compared to subjects in the control group. Hence the differences between the groups may be due to differences between the groups at baseline, rather than be the effect of yoga itself. To understand whether the effects of yoga training on maze learning observed in this study are related to baseline differences between groups, or to factors such as improved motor performance or left-right orientation, future studies may be planned with well matched groups and parameters selected to differentiate these contributing factors.