Teaching without learning : Is it OK With weak AI?
(2018) 19th International Conference on Artificial Intelligence in Education, AIED 2018 In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 10948 LNAI. p.241-245- Abstract
Two different learning models for a teachable agent were tested with respect to perceived intelligence, the protégé effect, and learning in Swedish grade 5 and 6 students. A strong positive correlation was found between perceived intelligence and the protégé effect, but no significant differences were found between the two different implementations of the learning algorithm. The results suggest that while the perceived intelligence of the agent relates to the induced protégé effect, this perceived intelligence did not correspond to the implemented learning model. This, in turn, suggest that a simple learning model can be sufficient for a teachable agent system, but more research is needed.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/2a19cb6e-a86e-4736-9c43-9e8d24f7475d
- author
- Månsson, Kristian LU and Haake, Magnus LU
- organization
- publishing date
- 2018-01-01
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Learning model, Learning outcome, Perceived intelligence, Protégé effect, Teachable agent
- host publication
- Artificial Intelligence in Education - 19th International Conference, AIED 2018, Proceedings
- series title
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
- volume
- 10948 LNAI
- pages
- 5 pages
- publisher
- Springer
- conference name
- 19th International Conference on Artificial Intelligence in Education, AIED 2018
- conference location
- London, United Kingdom
- conference dates
- 2018-06-27 - 2018-06-30
- external identifiers
-
- scopus:85049369052
- ISSN
- 1611-3349
- 0302-9743
- ISBN
- 9783319938455
- DOI
- 10.1007/978-3-319-93846-2_44
- language
- English
- LU publication?
- yes
- id
- 2a19cb6e-a86e-4736-9c43-9e8d24f7475d
- date added to LUP
- 2018-07-17 11:32:14
- date last changed
- 2024-06-24 17:01:42
@inproceedings{2a19cb6e-a86e-4736-9c43-9e8d24f7475d, abstract = {{<p>Two different learning models for a teachable agent were tested with respect to perceived intelligence, the protégé effect, and learning in Swedish grade 5 and 6 students. A strong positive correlation was found between perceived intelligence and the protégé effect, but no significant differences were found between the two different implementations of the learning algorithm. The results suggest that while the perceived intelligence of the agent relates to the induced protégé effect, this perceived intelligence did not correspond to the implemented learning model. This, in turn, suggest that a simple learning model can be sufficient for a teachable agent system, but more research is needed.</p>}}, author = {{Månsson, Kristian and Haake, Magnus}}, booktitle = {{Artificial Intelligence in Education - 19th International Conference, AIED 2018, Proceedings}}, isbn = {{9783319938455}}, issn = {{1611-3349}}, keywords = {{Learning model; Learning outcome; Perceived intelligence; Protégé effect; Teachable agent}}, language = {{eng}}, month = {{01}}, pages = {{241--245}}, publisher = {{Springer}}, series = {{Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}}, title = {{Teaching without learning : Is it OK With weak AI?}}, url = {{http://dx.doi.org/10.1007/978-3-319-93846-2_44}}, doi = {{10.1007/978-3-319-93846-2_44}}, volume = {{10948 LNAI}}, year = {{2018}}, }