Teachable agents: Difference between revisions

The educational technology and digital learning wiki
Jump to navigation Jump to search
mNo edit summary
No edit summary
 
(7 intermediate revisions by one other user not shown)
Line 1: Line 1:
{{stub}}
==Definition==
==Definition==
{{under construction}}
[[Intelligent agent]]s within computer-based environments are {{quotation | autonomous entities that can exist in complex, dynamic, and open environments}} (Maes, 1997). Agents can possess distinct behaviours  and goals that can be defined and adapted by the system {{quotation |as it learns appropriate “behavior” from the user and from other agents}} (Maes, 1997).


[[Learning by teaching]]
'''Teachable agents''' in education are [[pedagogical agent]]s and intelligent agents at once that through programming can be 'taught' to perform certain tasks within simulation-based environments to explore and solve problems. It is believed  that by allowing students to program computer agents students will benefit from the effects of [[learning by teaching]], [[discovery learning]] and more generally, [[project-oriented learning]].


{{quotationbox|


Programming intelligent agents requires several important processes: 1)  defining what the agent needs to know 2) defining a  representation of this knowledge 3) and programming this knowledge into the agent.  This process of evaluating the needs of the agent to perform a task can lead to deeper understanding of the domain the agent needs to  know about. 
Programming intelligent agents requires several important processes:  
*  Brophy S., Schwartz, D., Biswas, G., & Bransford, J. (1998, August). Learning Through Programmable Agents. Presented at Workshop on Pedagogical Agents, ITS '98, San Antonio, TX, August 1998.
# defining what the agent needs to know  
# defining a  representation of this knowledge  
# and programming this knowledge into the agent.  


==Intelligent agents==
This process of evaluating the needs of the agent to perform a task can lead to deeper understanding of the domain the agent needs to know about.
Intelligent agents possess qualities that make them autonomous entities that can exist in complex, dynamic, and open environments such as the Internet.  An agent can "sense, and act on,  its environment, and has a set of goals or motivations that it tries to achieve through these actions." (Maes, 1997).  
}} (Brophy et al. 1998).  


See also [[logo]], [[microworld]], [[simulation]].


Research in the area of intelligent agents focuses on questions like "How does an agent make an  appropriate decision; How does an agent learn; How does it adapt." In this context, learning focuses on algorithms designed to maintain an agents' autonomy and perform the desired goals of the person employing the agent.
==Designing effective teachable agents==


Previous work in learning by programming and the metaphor of computer agents provides insights toward creating opportunities that may yeild benefits similar to those of learning to those of learning by teaching.  Within the domain of computer technology, the idea of learning by teaching was emphasized by Papert (1980) in the context of helping students learn logo by teaching the "turtle" (see also Abelson & diSessa, 1980;  Mayer,  1988; Salomon, 1992). Extensions of this idea include programming lego toys and robots to interact with one another explicitly (e.g., Kafai & Resnick, 1996; Repenning & Sumner, 1995), programming computer agents to  collaboratively learn from one another (e.g., Dillenbourg, in press), creating micro worlds, and creating software to help others learn topics such as mathematics (e.g., Harel & Papert, 1991).
Swartz & Blair describe four features of teachable agents (TA) that enable students to take advantage of the inherent [[learning by teaching]] scenario:


==Use in ID==
# explicit well-structured shared visual representations of the TA's thinking and reasoning
A designer must evaluate how  an agent acts and how it  interacts with other agents (dependencies) in the simulation environment.  The designer must translate their hypothesis into an agent's actions/interactions using a visual programming environment to program sets of  propositions defining the behavior of the agents.  To create these propositions, users identify conditions when specific actions should be taken by agents, and they can specify actions using simple click and drag techniques.  Then the designer can evaluate the design by "running" the simulation to evaluate whether they correctly modelled the desired outcome. Used as a learning environment the designers are learners trying to  model the dynamics of a complex situations. This inquiry process of evaluation, implement and test provides an excellent learning opportunity that requires very little programming experience.
# independent performance of the agent so that the effects of the student's teaching provide feedback
# the agent’s ability to model productive learner behavior so as to provide guidance to the students about what needs to be taught or clarified.
# embedding the agent in environments that support teaching, including a larger context and external domain-specific resources.
 
Brophy et al. (1998) outline the major considerations and steps in designing an effective teaching agent within a simulation environment:
* evaluate how  an agent acts and how it  interacts with other agents (dependencies) in the simulation environment
* translate a hypothesis into an agent's actions/interactions  
** use a visual programming environment (e.g. [[AgentSheets]]) to program sets of  propositions defining the behavior of the agents
** identify conditions when specific actions should be taken by agents (instigated through ''click'' and ''drag'' options)
* evaluate the agent by running the simulation to see if desired outcome has been effectively modelled.
 
{{quotationbox |
Used as a learning environment, the designers are learners trying to  model the dynamics of complex situations. This inquiry process of evaluation, implement and test provides an excellent learning opportunity that requires very little programming experience}} ([http://www.teachableagents.org/papers/ta_workshop.html Brophy et al., 1998])
 
==Examples==
[http://www.teachableagents.org/betty.php BETTY]
 
==References==
* Blair, K., Schwartz, D., Biswas, G. & Leelawong, K. (2006). Pedagogical Agents for Learning by Teaching: Teachable Agents, Educational Technology & Society, Special Issue on Pedagogical Agents.[http://www.teachableagents.org/papers/Final-edtechTA.pdf PDF]
 
* Brophy S., Schwartz, D., Biswas, G., & Bransford, J. (1998, August). Learning Through Programmable Agents. Presented at Workshop on Pedagogical Agents, ITS '98, San Antonio, TX, August 1998.[http://www.teachableagents.org/papers/ta_workshop.html HTML]
[[Category:Artificial intelligence and education]]

Latest revision as of 17:26, 31 July 2009

Draft

Definition

Intelligent agents within computer-based environments are “autonomous entities that can exist in complex, dynamic, and open environments” (Maes, 1997). Agents can possess distinct behaviours and goals that can be defined and adapted by the system “as it learns appropriate “behavior” from the user and from other agents” (Maes, 1997).

Teachable agents in education are pedagogical agents and intelligent agents at once that through programming can be 'taught' to perform certain tasks within simulation-based environments to explore and solve problems. It is believed that by allowing students to program computer agents students will benefit from the effects of learning by teaching, discovery learning and more generally, project-oriented learning.


Programming intelligent agents requires several important processes:

  1. defining what the agent needs to know
  2. defining a representation of this knowledge
  3. and programming this knowledge into the agent.
This process of evaluating the needs of the agent to perform a task can lead to deeper understanding of the domain the agent needs to know about.

(Brophy et al. 1998).

See also logo, microworld, simulation.

Designing effective teachable agents

Swartz & Blair describe four features of teachable agents (TA) that enable students to take advantage of the inherent learning by teaching scenario:

  1. explicit well-structured shared visual representations of the TA's thinking and reasoning
  2. independent performance of the agent so that the effects of the student's teaching provide feedback
  3. the agent’s ability to model productive learner behavior so as to provide guidance to the students about what needs to be taught or clarified.
  4. embedding the agent in environments that support teaching, including a larger context and external domain-specific resources.

Brophy et al. (1998) outline the major considerations and steps in designing an effective teaching agent within a simulation environment:

  • evaluate how an agent acts and how it interacts with other agents (dependencies) in the simulation environment
  • translate a hypothesis into an agent's actions/interactions
    • use a visual programming environment (e.g. AgentSheets) to program sets of propositions defining the behavior of the agents
    • identify conditions when specific actions should be taken by agents (instigated through click and drag options)
  • evaluate the agent by running the simulation to see if desired outcome has been effectively modelled.


Used as a learning environment, the designers are learners trying to model the dynamics of complex situations. This inquiry process of evaluation, implement and test provides an excellent learning opportunity that requires very little programming experience

(Brophy et al., 1998)

Examples

BETTY

References

  • Blair, K., Schwartz, D., Biswas, G. & Leelawong, K. (2006). Pedagogical Agents for Learning by Teaching: Teachable Agents, Educational Technology & Society, Special Issue on Pedagogical Agents.PDF
  • Brophy S., Schwartz, D., Biswas, G., & Bransford, J. (1998, August). Learning Through Programmable Agents. Presented at Workshop on Pedagogical Agents, ITS '98, San Antonio, TX, August 1998.HTML