Authors:
Angel Alberto Jiménez Sarango
1
;
Andrés Patiño
1
;
María-Inés Acosta-Urigüen
1
;
Juan Gabriel Flores Sanchez
1
;
Priscila Cedillo
1
;
2
and
Marcos Orellana
1
Affiliations:
1
Laboratorio de Investigación y Desarrollo en Informática - LIDI, Universidad del Azuay, Av. 24 de mayo, Cuenca, Ecuador
;
2
Universidad de Cuenca, Cuenca, Ecuador
Keyword(s):
Stroop, Stress, Machine Learning, K-means, Clustering.
Abstract:
The Stroop test also called the colors and words test, is a widely used attention test to detect neuropsychological problems. Moreover, the stress test is a psychological instrument used to diagnose the level of stress and to identify the most common symptoms. This research aims to evaluate whether there is a relationship between the score of the Stroop test and the participant's level of stress. Data are collected through a web application, where participants answered the stress test and completed the Stroop test. Several variables were collected, such as the precision of each answer, the time spent, and demographic information. The machine learning technique called k-means was applied to process the collected data; the results include clusters of unlabeled data to find relationships. The main findings show that a person's stress level is directly linked to the number of correct answers obtained in the Stroop test; according to the clusters that show higher stress levels, the number
of correct answers decreased progressively.
(More)