Post #9: Teachable Agents
In researching for my final paper in this course this week, I came across an interesting piece of technology called "Teachable Agents". With this instructional technology, students assume the role of a teacher. They add information and create causal connections on a digital mind map which teaches the “agent” character in the simulation. Artificial Intelligence technologies then “enables TA to use the concept maps to answer questions, thereby providing interactivity, a model of thinking, and feedback” (Doris et al., 2010, p. 649). Here is a video of the technology in action:
I found this technology particularly intriguing because of its simplicity, and my thinking about how much my Grade 3 students would love to use it. They love feeling like experts, and I know they would be extremely motivated by the idea of being the "teacher", and having to understand content well enough to teach their "student". I think it's so neat how the technology makes them feel responsible for this cartoon student's learning, even though it is just a representation of their own understanding.
TA would also allow teachers to keep track of formative assessments. Students have to create a literal map of their thinking, then can quiz their agent or play games that test the agent's understanding. They are able to see where the agent lacks understanding, in turn monitoring their own learning and practicing self-regulation. The use of TA would follow regular in-class lessons about content, and could be used throughout a unit as a formative piece of assessment or as a summative piece.
Through TA, students form connections between concepts learned and receive instant feedback on their understanding. It is equipped with "a qualitative reasoning engine" which "uses path traversal algorithms that enable the agent to reason through causal chains in the concept map" (Biswas et al. 2005, as cited in Doris et al., 2010). They teach their agent about connections and information, assess them to see what they know then use deductive reasoning to fix gaps in understanding.
Although this technology is not widely used or tested yet, it got me thinking about the benefits of using an expert/teacher-centred learning experience with my own students. The students in my class thrive with responsibility and helping others, and I think using this type of model would be an excellent way to teach them self-regulation. If they don't create the connections and apply themselves to their learning, their "student" won't thrive. TA is absolutely something I can see myself implementing in my own classroom.
References
Doris B. Chin; Ilsa M. Dohmen; Britte H. Cheng; Marily A. Oppezzo; Catherine C. Chase; Daniel L. Schwartz. Preparing Students for Future Learning with Teachable Agents. Educational technology research and development 2010, 58 (6), 649–669. https://doi.org/10.1007/s11423-010-9154-5.
I found this technology particularly intriguing because of its simplicity, and my thinking about how much my Grade 3 students would love to use it. They love feeling like experts, and I know they would be extremely motivated by the idea of being the "teacher", and having to understand content well enough to teach their "student". I think it's so neat how the technology makes them feel responsible for this cartoon student's learning, even though it is just a representation of their own understanding.
TA would also allow teachers to keep track of formative assessments. Students have to create a literal map of their thinking, then can quiz their agent or play games that test the agent's understanding. They are able to see where the agent lacks understanding, in turn monitoring their own learning and practicing self-regulation. The use of TA would follow regular in-class lessons about content, and could be used throughout a unit as a formative piece of assessment or as a summative piece.
Through TA, students form connections between concepts learned and receive instant feedback on their understanding. It is equipped with "a qualitative reasoning engine" which "uses path traversal algorithms that enable the agent to reason through causal chains in the concept map" (Biswas et al. 2005, as cited in Doris et al., 2010). They teach their agent about connections and information, assess them to see what they know then use deductive reasoning to fix gaps in understanding.
Although this technology is not widely used or tested yet, it got me thinking about the benefits of using an expert/teacher-centred learning experience with my own students. The students in my class thrive with responsibility and helping others, and I think using this type of model would be an excellent way to teach them self-regulation. If they don't create the connections and apply themselves to their learning, their "student" won't thrive. TA is absolutely something I can see myself implementing in my own classroom.
References
Comments
Post a Comment