Suppr超能文献

结合基于证据的游戏化和机器学习来评估学前儿童的执行功能:一项可行性研究。

Incorporating Evidence-Based Gamification and Machine Learning to Assess Preschool Executive Function: A Feasibility Study.

作者信息

Eng Cassondra M, Tsegai-Moore Aria, Fisher Anna V

机构信息

Department of Psychiatry & Behavioral Sciences, Stanford University, 1520 Page Mill Road Stanford, Stanford, CA 94304, USA.

Department of Psychology, Carnegie Mellon University, 5000 Forbes Avenue, 335I Baker Hall, Pittsburgh, PA 15213, USA.

出版信息

Brain Sci. 2024 Apr 30;14(5):451. doi: 10.3390/brainsci14050451.

Abstract

Computerized assessments and digital games have become more prevalent in childhood, necessitating a systematic investigation of the effects of gamified executive function assessments on performance and engagement. This study examined the feasibility of incorporating gamification and a machine learning algorithm that adapts task difficulty to individual children's performance into a traditional executive function task (i.e., Flanker Task) with children ages 3-5. The results demonstrated that performance on a gamified version of the Flanker Task was associated with performance on the traditional version of the task and standardized academic achievement outcomes. Furthermore, gamification grounded in learning science and developmental psychology theories applied to a traditional executive function measure increased children's task enjoyment while preserving psychometric properties of the Flanker Task. Overall, this feasibility study indicates that gamification and adaptive machine learning algorithms can be successfully incorporated into executive function assessments with young children to increase enjoyment and reduce data loss with developmentally appropriate and intentional practices.

摘要

计算机化评估和数字游戏在儿童时期变得越来越普遍,因此有必要对游戏化执行功能评估对表现和参与度的影响进行系统研究。本研究探讨了将游戏化和一种根据儿童个体表现调整任务难度的机器学习算法纳入传统执行功能任务(即侧翼任务)对3至5岁儿童的可行性。结果表明,游戏化版本的侧翼任务表现与传统版本任务的表现以及标准化学业成绩结果相关。此外,基于学习科学和发展心理学理论应用于传统执行功能测量的游戏化增加了儿童对任务的喜爱,同时保留了侧翼任务的心理测量特性。总体而言,这项可行性研究表明,游戏化和自适应机器学习算法可以成功纳入幼儿执行功能评估中,通过适合发展阶段且有针对性的实践增加趣味性并减少数据丢失。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/8cb4/11119088/add31d371180/brainsci-14-00451-g001.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验