Departamento de Estádistica, Universidad Carlos III de Madrid, Leganes, Spain.
Departamento de Ciencias de la Computación, Universidad Rey Juan Carlos, Móstoles, Spain.
Sci Rep. 2023 Aug 22;13(1):13689. doi: 10.1038/s41598-023-40455-4.
The assessment of cognitive functions is mainly based on standardized neuropsychological tests, widely used in various fields such as personnel recruitment, education, or health. This paper presents a virtual reality game that allows collecting continuous measurements of both the performance and behaviour of the subject in an immersive, controllable, and naturalistic experience. The application registers variables related to the user's eye movements through the use of virtual reality goggles, as well as variables of the game performance. We study how virtual reality can provide data to help predict scores on the Attention Control Scale Test and the Barratt Impulsiveness Scale. We design the application and test it with a pilot group. We build a random forest regressor model to predict the attention and impulsivity scales' total score. When evaluating the performance of the model, we obtain a positive correlation with attention (0.434) and with impulsivity (0.382). In addition, our model identified that the most significant variables are the time spent looking at the target or at distractors, the eye movements variability, the number of blinks and the pupil dilation in both attention and impulsivity. Our results are consistent with previous results in the literature showing that it is possible to use data collected in virtual reality to predict the degree of attention and impulsivity.
认知功能的评估主要基于标准化的神经心理学测试,这些测试广泛应用于人员招聘、教育或健康等各个领域。本文提出了一种虚拟现实游戏,它可以在沉浸式、可控和自然的体验中连续收集受试者的表现和行为数据。该应用程序通过使用虚拟现实护目镜记录与用户眼球运动相关的变量,以及游戏表现的变量。我们研究了虚拟现实如何提供数据来帮助预测注意力控制量表测试和巴瑞特冲动量表的分数。我们设计并测试了该应用程序。我们构建了一个随机森林回归模型来预测注意力和冲动量表的总分。在评估模型的性能时,我们发现它与注意力(0.434)和冲动(0.382)呈正相关。此外,我们的模型确定了最显著的变量是注视目标或分心物的时间、眼球运动的可变性、眨眼次数和瞳孔在注意力和冲动方面的扩张。我们的结果与文献中的先前结果一致,表明可以使用虚拟现实中收集的数据来预测注意力和冲动的程度。