Browsing by Author "Rahkila, Ville"
Now showing items 1-1 of 1
-
Rahkila, Ville (2024)The aim of this thesis was to examine the test-retest reliability of three outcome-based behavioural tasks - the reversal learning task, the weather prediction task, and the working memory task - with three measuring sessions. In addition, the aim was to determine the more reliable of the two learning tasks: the reversal learning task and the weather prediction task. Test-retest reliability refers to a task’s ability to produce reliable results when repeated. The repetition of a task has been found to cause a so-called retest effect which means the changing of test scores when the participants is exposed to an identical or similar test in the past. The magnitude of this retest effect depends, for example, on the temporal interval of the measurements, the design of the task, and the age and intelligence of the participant. The reversal learning task, the weather prediction task, and the working memory task are all utilized in health care and their test-retest reliability may have clinical implications. The participants were 40 healthy adults aged 18 to 35, and all of them were subjected to two of the behavioural tasks in three sessions with 4 to 14 day intervals. The data was analyzed with an intraclass correlation coefficient suitable for the determining of test-retest reliability and analyses of variance. The results of the analyses of variance were further analyzed with pairwise comparisons. The intraclass correlation coefficients of the behavioural tasks ranged from moderate to good. The reversal learning task’s analysis of variance revealed statistically significant variance in the outcome variables between sessions. Similarly, the reaction times of the weather prediction task had statistically significant variance between the sessions. The working memory task’s accuracy, on the other hand, had statistically significant variance between trial types. Reaction times of the working memory task had statistically significant variance between both the sessions and the trial types. Test-retest reliability is a complex issue which cannot be determined in a single study. That being said, the results of this study support the repeated clinical use of these instruments among healthy adults of 18 to 45 years of age and with the 4 to 14 day interval. When it comes to the learning tasks, the weather prediction task seemed to be the more reliable one, but the differences were small.
Now showing items 1-1 of 1