IOWA gambling task: A Turkish normative study and correlations with other executive measures

No Thumbnail Available
Gürvit, İbrahim Hakan
Şandor, Serra
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue

Iowa Gambling Task (IGT) was improved by Bechara and Damasio to assess decision making processes in daily life. Unlike the other tasks that assess executive functions, this task is spesifically used to evaluate decision making processes and to determine whether there may be lesions in orbitofrontal and ventromedial cortices. Our objective was to obtain normative values in a representative Turkish population and see if demographic variables such as gender, education and age influence performance, ascertain whether practice overtime with and repeated exposure to the test improves performance and finally whether the performance in a general IQ test and several executive function tests can predict better performance in this task. Study population consisted of 90 subjects (45 subjects for each sex) who were divided into 3 groups according to age and education; 20-39, 20-59, 60+ years of age and grade schooling, high schooling and college degree respectively. Initially, the subjects were given the IGT and the neuropsychological test battery: general IQ test (Alexander) and tests of executive function (Wisconsin Card Sorting Test [WCST], Tower of London Test [TOL], Stroop Test). In the first and fourth weeks after the initial administration of the tests, only IGT is the repeated. Multivariate ANOVA was used to eliminate the influence of gender, age and education and linear stepwise backward regression was applied to see whether any other test index predicted IGT performance in the first and second trial of the tests and finally repeated measures ANOVA (with gender as covariate) was used to clarify if the performance improved between the 2 halves and 2 sessions of the IGT. Education and age did not influence the IGT performance but men performed significantly better in various indices of IGT2 performance only. Nevertheless, rmANOVA showed that both sexes significantly improved their total score in IGT2 as compared to IGT1. Moreover, both sexes improved their performance in the 2nd half of the IGT2 but not in IGT1 as compared to their respective 1st halves. Finally, only a number of TOL indices correlated with IGT1 total score (total number of correct solutions, total number of moves [negative correlation] and time to initiate 1st move) and WCST indices with IGT2 (number of perseverative responses [negative correlation] and failure to maintain set score). These results replicate the previous findings that normal subjects improve their performance over time and males can perform better in some respects in this test and suggest that better planning in the 1st exposure and better mental flexibility in the 2nd may lead to better performance.