Unpacking epistemic insights of artificial intelligence (AI) in science education: a systematic review
There is a growing application of Artificial Intelligence (AI) in K-12 science classrooms. In K-12 education, students harness AI technologies to acquire scientific knowledge, ranging from automated personalized virtual scientific inquiry to generative AI tools such as ChatGPT, Sora, and Google Bard...
| Main Authors: | , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Springer
2024
|
| _version_ | 1811139213677035520 |
|---|---|
| author | Cheung, KKC Long, Y Liu, Q Chan, H-Y |
| author_facet | Cheung, KKC Long, Y Liu, Q Chan, H-Y |
| author_sort | Cheung, KKC |
| collection | OXFORD |
| description | There is a growing application of Artificial Intelligence (AI) in K-12 science classrooms. In K-12 education, students harness AI technologies to acquire scientific knowledge, ranging from automated personalized virtual scientific inquiry to generative AI tools such as ChatGPT, Sora, and Google Bard. These AI technologies inherit various strengths and limitations in facilitating students’ engagement in scientific activities. There is a lack of framework to develop K-12 students’ epistemic considerations of the interaction between the disciplines of AI and science when they engage in producing, revising, and critiquing scientific knowledge using AI technologies. To accomplish this, we conducted a systematic review for studies that implemented AI technologies in science education. Employing the family resemblance approach as our analytical framework, we examined epistemic insights into relationships between science and AI documented in the literature. Our analysis centered on five distinct categories: aims and values, methods, practices, knowledge, and social–institutional aspects. Notably, we found that only three studies mentioned epistemic insights concerning the interplay between scientific knowledge and AI knowledge. Building upon these findings, we propose a unifying framework that can guide future empirical studies, focusing on three key elements: (a) AI’s application in science and (b) the similarities and (c) differences in epistemological approaches between science and AI. We then conclude our study by proposing a development trajectory for K-12 students’ learning of AI-science epistemic insights. |
| first_indexed | 2024-04-09T03:56:24Z |
| format | Journal article |
| id | oxford-uuid:dc14b2e1-9c65-4c4b-82f3-55b18030a902 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:02:31Z |
| publishDate | 2024 |
| publisher | Springer |
| record_format | dspace |
| spelling | oxford-uuid:dc14b2e1-9c65-4c4b-82f3-55b18030a9022024-04-30T16:12:33ZUnpacking epistemic insights of artificial intelligence (AI) in science education: a systematic reviewJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:dc14b2e1-9c65-4c4b-82f3-55b18030a902EnglishSymplectic ElementsSpringer2024Cheung, KKCLong, YLiu, QChan, H-YThere is a growing application of Artificial Intelligence (AI) in K-12 science classrooms. In K-12 education, students harness AI technologies to acquire scientific knowledge, ranging from automated personalized virtual scientific inquiry to generative AI tools such as ChatGPT, Sora, and Google Bard. These AI technologies inherit various strengths and limitations in facilitating students’ engagement in scientific activities. There is a lack of framework to develop K-12 students’ epistemic considerations of the interaction between the disciplines of AI and science when they engage in producing, revising, and critiquing scientific knowledge using AI technologies. To accomplish this, we conducted a systematic review for studies that implemented AI technologies in science education. Employing the family resemblance approach as our analytical framework, we examined epistemic insights into relationships between science and AI documented in the literature. Our analysis centered on five distinct categories: aims and values, methods, practices, knowledge, and social–institutional aspects. Notably, we found that only three studies mentioned epistemic insights concerning the interplay between scientific knowledge and AI knowledge. Building upon these findings, we propose a unifying framework that can guide future empirical studies, focusing on three key elements: (a) AI’s application in science and (b) the similarities and (c) differences in epistemological approaches between science and AI. We then conclude our study by proposing a development trajectory for K-12 students’ learning of AI-science epistemic insights. |
| spellingShingle | Cheung, KKC Long, Y Liu, Q Chan, H-Y Unpacking epistemic insights of artificial intelligence (AI) in science education: a systematic review |
| title | Unpacking epistemic insights of artificial intelligence (AI) in science education: a systematic review |
| title_full | Unpacking epistemic insights of artificial intelligence (AI) in science education: a systematic review |
| title_fullStr | Unpacking epistemic insights of artificial intelligence (AI) in science education: a systematic review |
| title_full_unstemmed | Unpacking epistemic insights of artificial intelligence (AI) in science education: a systematic review |
| title_short | Unpacking epistemic insights of artificial intelligence (AI) in science education: a systematic review |
| title_sort | unpacking epistemic insights of artificial intelligence ai in science education a systematic review |
| work_keys_str_mv | AT cheungkkc unpackingepistemicinsightsofartificialintelligenceaiinscienceeducationasystematicreview AT longy unpackingepistemicinsightsofartificialintelligenceaiinscienceeducationasystematicreview AT liuq unpackingepistemicinsightsofartificialintelligenceaiinscienceeducationasystematicreview AT chanhy unpackingepistemicinsightsofartificialintelligenceaiinscienceeducationasystematicreview |