Using ChatGPT is easy, using it effectively is tough? A mixed methods study on K-12 students’ perceptions, interaction patterns, and support for learning with generative AI chatbots

计算机科学 生成语法 感知 生成模型 聊天机器人 人机交互 人工智能 心理学 神经科学
作者
Maria Klar
出处
期刊:Smart Learning Environments [Springer Nature]
卷期号:12 (1) 被引量:40
标识
DOI:10.1186/s40561-025-00385-2
摘要

Abstract Generative AI (genAI) chatbots can be a powerful tool for learning, especially self-regulated learning, but they can also add more complexity to the learning process because of their numerous capabilities. Learners might not make use of these capabilities and fail to profit from them. One core genAI affordance is to adapt any learning material to the learner’s individual needs. To help learners make use of these adaptation capabilities, instructional prompts can help, but they can also have negative effects on cognitive load. This study investigated K-12 learners’ perceptions and interaction patterns with a chatbot, focusing on their use of content adaptation capabilities. In this experimental study with 106 secondary school students, a group receiving adaptation guidance (including brief instruction, prompt suggestions, and adjustable chatbot response length and language level) was compared to a control group without adaptation guidance. Results show that learners perceived chatbots as easy to use despite limited prior experience. Without guidance, they underutilized the chatbot’s adaptation capabilities. The experimental group used twice as many adaptation prompts (M = 6.0) compared to the control group (M = 2.5), without experiencing increased cognitive load. Interestingly, both groups showed similar knowledge gains and reported high satisfaction levels. These findings suggest that easily implementable interventions can enhance students’ use of genAI chatbot capabilities, potentially improving their self-regulated learning experiences. Future research should explore the long-term effects of adaptation guidance on learning outcomes and self-regulated learning skills.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
cc完成签到 ,获得积分10
刚刚
忧郁如柏完成签到,获得积分10
刚刚
1秒前
1秒前
1秒前
吴灵发布了新的文献求助10
1秒前
香蕉觅云的应助被就是现在采纳,获得10
1秒前
2秒前
canghong完成签到,获得积分10
2秒前
丘比特的应助被Xia采纳,获得10
2秒前
3秒前
英姑的应助被高锰酸钾采纳,获得10
3秒前
makabaka发布了新的文献求助30
4秒前
5秒前
5秒前
小短腿大脑袋完成签到,获得积分10
5秒前
11发布了新的文献求助10
5秒前
bkagyin的应助被ymx采纳,获得10
5秒前
YihanChen发布了新的文献求助10
5秒前
6秒前
深藏blue完成签到,获得积分10
6秒前
如月之恒完成签到,获得积分10
6秒前
安详的亦丝完成签到 ,获得积分10
6秒前
bkagyin的应助被CWFF采纳,获得10
6秒前
在左完成签到,获得积分10
6秒前
7秒前
洛莫发布了新的文献求助30
7秒前
默默发布了新的文献求助50
7秒前
niuniu发布了新的文献求助30
7秒前
汉堡包的应助被郑继庆采纳,获得10
8秒前
Zxx完成签到 ,获得积分10
9秒前
天天快乐的应助被赵念婉采纳,获得10
9秒前
妮妮发布了新的文献求助10
9秒前
谢大喵发布了新的文献求助10
10秒前
10秒前
乐乐发布了新的文献求助10
11秒前
11秒前
田彬杰发布了新的文献求助10
11秒前
12秒前
求学发布了新的文献求助10
12秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Energie-Contracting 1500
Klimaschutzrecht durch den Energiesektor im Mehrebenensystem – Ein Rechtsvergleich 1000
Sustainable Engineered Micro/Nanofibers and Hybrid Nanocomposites for Environmental Remediation 400
Ecotourism 400
Analytical Separation Science 300
Civilization and War 300
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8071431
求助须知:如何正确求助?哪些是违规求助? 9491539
关于积分的说明 21364989
捐赠科研通 7639691
什么是DOI,文献DOI怎么找? 3398869
关于科研通互助平台的介绍 2528684
邀请新用户注册赠送积分活动 2437214