Sahar Zakeri - M.Sc. Student, Computational Neuroscience Laboratory, Department of Biomedical Engineering, Faculty of Electrical Engineering, Sahand University of Technology, Tabriz, Iran
Ataollah Abbasi - Assistant professor, Computational Neuroscience Laboratory, Department of Biomedical Engineering, Faculty of Electrical Engineering, Sahand University of Technology, Tabriz, Iran
Ateke Goshvarpour - Ph. D. Student, Computational, Neuroscience Laboratory, Department of Biomedical Engineering, Faculty of Electrical Engineering, Sahand University of Technology, Tabriz, Iran

The study of creativity and its effects on human life are growing and expanding. However, few studies have investigated the relationship between creativity and physiological changes. As galvanic skin response (GSR) analysis is often used to detect cognitive behavior, this paper presents a novel approach for distinction between creativity/normal states and high/low creative from GSR signals. To these ends, 21 linear and 3 nonlinear features of this signal were extracted. Torrance Tests of Creative Thinking (TTCT- Figural B), which consists of three tasks, was used. The GSR signals of 21 participants (12 men, 12 women and 21-12 years) were recorded. Applying Wilcoxon statistical test, significant differences between rest and each task of creativity were observed; however, better discrimination was performed between rest and the first task. To classify the extracted features, Support Vector Machine (SVM) was used. The results indicated that SVM in combination with linear features classification algorithm effectively reduced the complexity of the calculations then will be reach the maximum accuracy of 12.11% in separating rest from the second task. Also, maximum accuracy of 12..69 was observed in task 3 between high/low creative subjects. It has also shown that differentiation between the rest and creative states is better performed by linear analysis than that of nonlinear ones. In conclusion, the combination of the SVM classifier with linear and nonlinear features can be useful to show the relation between creativity and physiological changes.

Creativity, Galvanic Skin Response, Support Vector Machine, Torrance Tests of Creative Thinking