Cargando…
The Effect of Creative Tasks on Electrocardiogram: Using Linear and Nonlinear Features in Combination with Classification Approaches
Objective: Interest in the subject of creativity and its impacts on human life is growing extensively. However, only a few surveys pay attention to the relation between creativity and physiological changes. This paper presents a novel approach to distinguish between creativity states from electrocar...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Psychiatry & Psychology Research Center, Tehran University of Medical Sciences
2017
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5425352/ https://www.ncbi.nlm.nih.gov/pubmed/28496502 |
Sumario: | Objective: Interest in the subject of creativity and its impacts on human life is growing extensively. However, only a few surveys pay attention to the relation between creativity and physiological changes. This paper presents a novel approach to distinguish between creativity states from electrocardiogram signals. Nineteen linear and nonlinear features of the cardiac signal were extracted to detect creativity states. Method: ECG signals of 52 participants were recorded while doing three tasks of Torrance Tests of Creative Thinking (TTCT/ figural B). To remove artifacts, notch filter 50 Hz and Chebyshev II were applied. According to TTCT scores, participants were categorized into the high and low creativity groups: Participants with scores higher than 70 were assigned into the high creativity group and those with scores less than 30 were considered as low creativity group. Some linear and nonlinear features were extracted from the ECGs. Then, Support Vector Machine (SVM) and Adaptive Neuro-Fuzzy Inference System (ANFIS) were used to classify the groups. Results: Applying the Wilcoxon test, significant differences were observed between rest and each three tasks of creativity. However, better discrimination was performed between rest and the first task. In addition, there were no statistical differences between the second and third task of the test. The results indicated that the SVM effectively detects all the three tasks from the rest, particularly the task 1 and reached the maximum accuracy of 99.63% in the linear analysis. In addition, the high creative group was separated from the low creative group with the accuracy of 98.41%. Conclusion: the combination of SVM classifier with linear features can be useful to show the relation between creativity and physiological changes. |
---|