Elsevier

Computers & Education

Volume 53, Issue 3, November 2009, Pages 966-976
Computers & Education

Agent-customized training for human learning performance enhancement

https://doi.org/10.1016/j.compedu.2009.05.014Get rights and content

Abstract

Training individuals from diverse backgrounds and in changing environments requires customized training approaches that align with the individual learning styles and ever-evolving organizational needs. Scaffolding is a well-established instructional approach that facilitates learning by incrementally removing training aids as the learner progresses. By combining multiple training aids (i.e. multimodal interfaces), a trainer, either human or virtual, must make real-time decisions about which aids to remove throughout the training scenario. A significant problem occurs in implementing scaffolding techniques since the speed and choice of removing training aids must be strongly correlated to the individual traits of a specific trainee. We detail an agent-based infrastructure that supports the customization of scaffolding routines as triggered by the performance of the trainee. The motivation for this agent-based approach is for integration into a training environment that leverages augmented reality (AR) technologies. Initial experiments using the simulated environment have compared the proposed adaptive approach with traditional static training routines. Results show that the proposed approach increases the trainees’ task familiarity and speed with negligible introduction of errors.

Section snippets

Introduction to adaptive scaffolding

With new pervasive computing devices such as handheld computers, mobile phones, mp3 players and virtual reality headsets, there are numerous avenues to incorporate training into the daily societal routines. These types of devices allow users to receive voluminous amounts of information, textually, graphically, and audibly. Considering the popularity of these types of devices in the younger generation, information transfer and training considering multimodal interfaces is an important area for

Related work

Scaffolding is a more specific subset of the general notion of an experienced trainer (i.e. master) who nurtures and trains an apprentice (Collins, Brown, & Newman, 1989). The master must use multiple pedagogical approaches to train any variety of skills and tasks to the apprentice. Scaffolding is particularly effective because the master can identify and provide assistance only where it is needed. In fact, selectivity in this fashion benefits the trainee (Wood et al., 1976, Zhao and Orey, 1999

An intelligent scaffolding agent

A scaffolding agent can be defined as an intelligent software mechanism that uses the knowledge of its environment to reactively and proactively assist a human trainee with regards to incrementally learning a particular task or process. These aspects are similar to the traditional definitions of agents (Jennings et al., 1998). The focus of our work is, at least initially, to support the training of employees in an industry environment. In our initial studies, we have tailored our approach to

A proof of concept system scaffolding agent

The scaffolding agent was designed to be integrated with augmented reality headsets to provide advanced pedagogical support using multimodal instructions as delivered by the headsets. In future work, the agent-based techniques will be ported to our augment reality testbed (CAARS, 2008). However, in this study, the focus is on a game-oriented prototype. Using the Java programming language, we developed an application that teaches a trainee how to complete a jigsaw puzzle with the assistance of

Experimentation and evaluation

In order to show the effectiveness of the adaptive scaffolding approach, we performed an experiment with real human subjects. Georgia Institute of Technology offers a Women and Engineering (Tech camp) summer camp for middle-school females between the age of 11 and 14. This camp took place July 2007 on Georgia Tech’s campus. In 2007, forty-six women participated in the camp and our simulation software was administered to them. The experiment included two variations of the simulated augmented

Discussion

The experimental results show that the adaptive scaffolding approach is effective in encouraging the students to engage in meaningful training particularly for a memory-intensive task such as a jigsaw puzzle. The approach for selecting the number of scaffolds in this paper shows that even this initial level of customization is effective in improving training goals. This approach requires that the intelligent agent maintain a knowledge base of the training performance of past students. As a

Conclusion

This paper contributes an infrastructure that supports the integration of multimodal instructions and scaffold generation techniques using adaptive modules. This is one of few studies that combine this set of technologies for the advancement of human performance in learning. An agent-based prototype is created to manage the reduction of scaffolds during the training lifecycle. Experiments on a group of middle-school females demonstrated that the adaptive scaffolding approach trained the

Acknowledgements

The authors acknowledge conversations with Dr. Jayfus T. Doswell, Mr. Charles Griffin, Mr. Ojai Mallory, and Mr. Finny Mlemchkuwu. This work was funded by the National Science Foundation under award number 0512610 and partially under award number 0634302.

References (32)

  • B.J. Zimmerman

    Development of self-regulated learning: Which are the key sub processes?

    Contemporary Educational Psychology

    (1986)
  • Adaptable Learning Pathway Planning with Self-Organising Agents (2009)....
  • Aist, G., Kort, B., Reilly, R., Mostow, J., & Picard, R. (2002). Experimentally augmenting an intelligent tutoring...
  • Beal, C. R., & Cohen, P. (2005). Computational methods for evaluating student and group learning histories in...
  • B. Benson

    Scaffolding (coming to terms)

    English Journal

    (1997)
  • M.B. Blake

    Coordinating multiple agents for workflow-oriented process orchestration

    Information Systems and E-Business Management Journal

    (2003)
  • Blake, M. B., Butcher-Green, J. D., & Doswell, J. T. (2006). An agent-mediated multimodal scaffolding infrastructure....
  • Blake, M. B., Butcher-Green, J. D., & Doswell, J. T. (2007). Agent-mediated customized training for human learning...
  • Brusilovsky, P., & Peylo, C. (2003). Adaptive and intelligent web-based educational systems. In P. Brusilovsky, & C....
  • CAARS (2008)....
  • A. Collins et al.

    Cognitive apprenticeship: Teaching the craft of reading, writing, and mathematics

  • M.C.M. De Guerrero et al.

    Activating the ZPD: Mutual scaffolding in L2 peer revision

    Modern Language Journal

    (2000)
  • S.K. D’Mello et al.

    Towards an affect-sensitive autotutor

    IEEE Intelligent Systems, Special Issue on Intelligent Educational Systems

    (2007)
  • C.A. Ellis

    Workflow technology

  • A. Graesser et al.

    Teaching tactics and dialog in autotutor

    International Journal on Artificial Intelligence in Education

    (2001)
  • Cited by (18)

    • Augmented reality in science laboratories: The effects of augmented reality on university students' laboratory skills and attitudes toward science laboratories

      2016, Computers in Human Behavior
      Citation Excerpt :

      They reported that studies on AR in science education are few in number and that the field is in its infancy. The existing research focuses on issues such as development, usability, and initial implementation (Blake & Butcher-Green, 2009; El Sayed et al., 2011; Kaufmann & Schmalstieg, 2003). Students' laboratory skills and learning outcomes have been ignored to a great extent (Cheng & Tsai, 2013).

    • Clarifying the empirical connection of new entrants' e-learning systems use to their job adaptation and their use patterns under the collective-individual training environment

      2012, Computers and Education
      Citation Excerpt :

      Learners learn by sitting before computers or electronic devices. They can decide the pace, the style, and the order of the content in the learning process (e.g., Bryant, Campbell, & Kerr, 2003; Hamid, 2001). The learning control and responsibility then mostly transfer to the learners at e-learning environment, and thus learners play an important role in the learning process (e.g., Yanga & Liu, 2007).

    • Creating a performance-oriented e-learning environment: A design science approach

      2011, Information and Management
      Citation Excerpt :

      They may support learning processes as personal tutors, academic counselors, mentors, or peers. They play a major role in making e-learning suit the needs of learners by customizing the learning content and process based on the learners’ background, needs, and preferences [2]. An intelligent learning system typically has four parts: the domain expert, learner, pedagogical, and interface modules.

    • The effects of education compatibility and technological expectancy on e-learning acceptance

      2011, Computers and Education
      Citation Excerpt :

      Some learning approaches, such as instructional scaffolding (Brush & Saye, 2002; Yelland & Masters, 2007) and experiential learning theory (Rogers & Freiberg, 1994), provide useful suggestions on how to develop a learning environment that allows assist students to eliminate learning obstacles. In recent years, more studies concerning e-learning systems have stressed the importance of developing intelligent tools to facilitate students’ adaptations during learning (e.g., Barretto et al., 2003; Schiaffino et al., 2008; Blake & Butcher-Green, 2009). It is also argued that if a student perceives an e-learning system is able to provide compatible facilitation, he/she may believe that improved facilitations are available when using the system.

    • Linking employees' e-learning system use to their overall job outcomes: An empirical study based on the IS success model

      2010, Computers and Education
      Citation Excerpt :

      These facilitate employees’ renewal of knowledge and skills, and also help their better adaptation to changes. This importance of e-learning for self-study has been well documented (Blake & Butcher-Green, 2009; Kramarski & Gutman, 2006; Körndle, Narciss, & Proske, 2004; Newton & Doonga, 2007). It also facilitates enterprises’ implementation of employee training; through e-learning systems, organizations can quickly update training materials for employees.

    View all citing articles on Scopus
    View full text