Mirror Hand Tracing Experiment, Results

Results:
The data of this experiment seem to have experienced a complementary balance in that one test subject performed the task terribly while another performed it with greater efficiency. The third test subject fell between the two. The data obtained from this experiment are summarized in the following graphs, and they suggest that practice effects definitely played a role in test subjects’ performance. The “Time of Completion” graph displays the difficulty test subjects had initially with navigating the test placed before them. Yet, after these subjects performed 14 tests with the opposite hand, and returned to their non-dominant hand, their performance was greatly improved. This suggests that the 14 dominant-hand tests improved the subjects’ ability to more quickly navigate the non-dominant tests later on.

The errors graph shows similar yet intriguingly different results. It appears that the subjects made many errors in their first 3 attempts at completing the trial with their non-dominant hand. When attempted with their dominant hand, the number of errors committed in their trials revealed an gradual decline of errors. When subjects returned to testing with their non-dominant hand, their testing error frequency began a rise back to previously attained non-dominant hand testing levels. While this result could be a confounding condition of test subject attrition, the data suggest that practice effects played less a role in error reduction than in time reduction.

In performing a statistical analysis of the data, the two-tailed t-test was chosen because its appropriateness, given the hypothesis that practice effects would not be observed with any statistical significance. The p-value for the “Time of Completion” data calculated to 0.2295. This p-value exceeds the alpha value of 0.05. What this means is that hand-to-hand transferable practice effects did not play a statistically significant role in improving test subjects’ abilities to complete the star-tracing task any quicker.

The p-value for the “Errors” data was 0.0007. This p-value undercuts the set alpha value of 0.05, indicating statistical significance in that dominant-hand to non-dominant-hand transferable practice effects were present.