Decisión Making, Stress Assessed by Physiological Response and Virtual Reality Stimuli / Toma de Decisiones, Estrés Valorado mediante la Respuesta Fisiológica y Estímulos en Realidad Virtual

Abstract Many decisions must be made under stress; therefore, stress and decision-making are intrinsically related not only at the behavioral level but also at the neural level. Additionally, virtual reality tools have been proposed as a method to induce stress in the laboratory. This review focuses on answering the following research question: Does stress assessed by physiological variables of a subject under virtual reality stimuli increase the chances of error in decision-making? The reviewed studies were consulted in the following databases: PubMed, IEEE Xplore, and Science Direct. The analysis of the consulted literature indicates that the stress induced in the laboratory using virtual reality tools and the physiological response of the central and autonomous nervous system are complementary subjects and allow the design of training and support systems for the decision-making process.

Resumen Muchas decisiones deben tomarse bajo estrés, por lo tanto, el estrés y la toma de decisiones están intrínsecamente relacionados, no solo a nivel conductual sino también a nivel neural. Además, las herramientas de realidad virtual se han propuesto como un método para inducir estrés en el laboratorio. El presente trabajo de revisión temática se centra en responder la siguiente pregunta de investigación: ¿El estrés evaluado mediante variables fisiológicas de un sujeto bajo estímulos de realidad virtual aumenta las posibilidades de error en la toma de decisiones? Los estudios revisados fueron consultados en las siguientes bases de datos: PubMed, IEEE Xplore y Science Direct. El análisis de la literatura consultada indica que el estrés inducido en el laboratorio, por medio de herramientas de realidad virtual, y la respuesta fisiológica del sistema nervioso central y autónomo son temas que se complementan y permiten el diseño de sistemas de soporte y entrenamiento para el proceso de toma de decisiones.

Study on the impact of partition-induced dataset shift on k-fold cross-validation.

Cross-validation is a very commonly employed technique used to evaluate classifier performance. However, it can potentially introduce dataset shift, a harmful factor that is often not taken into account and can result in inaccurate performance estimation. This paper analyzes the prevalence and impact of partition-induced covariate shift on different k-fold cross-validation schemes. From the experimental results obtained, we conclude that the degree of partition-induced covariate shift depends on the cross-validation scheme considered. In this way, worse schemes may harm the correctness of a single-classifier performance estimation and also increase the needed number of repetitions of cross-validation to reach a stable performance estimation.