Using pedagogical agents to increase engagement in eLearning

Written by Grazelde Langeveldt and Katherine Fourie
Nov 20, 2020

Many eLearning practitioners would agree that a successful digital learning experience relies on student engagement. However, facilitating engagement online can be challenging – particularly when it comes to sustaining students’ attention over time, in the absence of traditional face-to-face interaction. Fortunately, recommendations arising from numerous research studies have highlighted ways to address this issue. One such recommendation has been the use of pedagogical agents. Studies indicate that, when interactive pedagogical agents are used as instructional aids in the design of eLearning courses, student engagement increases – resulting in a more effective digital learning experience overall (Bickmore et al., 2011; Cook, 2017; Dinçer and Doğanay, 2015; Lane, 2016; and Martha and Santoso, 2019).

This article will provide an overview of pedagogical agents and their benefits, before making recommendations for how they can be used to increase student engagement in online courses.

 

What is a pedagogical agent?

 

A pedagogical agent can be described as an autonomous, computer-generated character that is designed to support students during learning – in particular, by reacting to changes in the environment, communicating with other agents, and engaging with students in the context of an interactive learning environment (Vallverdú and Casacuberta, 2009). The concept was originally borrowed from the fields of computer science, animation and artificial intelligence (AI), and applied to online learning environments for the purpose of simulating the social interaction that occurs in a traditional classroom setting (Phan, 2011). Pedagogical agents are embedded in the instructional application, acting in the background as part of the architecture of the educational system. They provide humanlike assistance, guiding students through the multimedia learning environment and contributing positively toward increasing student engagement – and, by implication, effective learning. Let’s take a brief look at how this is achieved.

Pedagogical agents can guide students’ learning processes by providing them with feedback while educational tasks are being performed. This can take the form of speech and/or nonverbal behaviours.

For example, after an activity has been completed successfully, the pedagogical agent can provide positive verbal and/or non-verbal feedback about the completed task. Alternatively, if the activity has not been completed successfully, the pedagogical agent can guide the student toward helpful material that will aid them in completing the task, as illustrated below. These social affordances have been found to promote a positive affect and aid in the transfer of learning (Kramer and Bente in Lane, 2016).

 

Furthermore, according to Johnson and Lester (2016), pedagogical agents can also support learning by:

  • demonstrating how tasks should be performed;
  • helping students to navigate from one point to another;
  • providing cues to guide students; and
  • eliciting emotions in students by expressing empathy toward them.

In doing so, pedagogical agents have a positive effect on aspects of learning such as motivation, satisfaction, appreciation, self-sufficiency and overall performance (Dinçer and Doğanay, 2015: 332). As such, their presence makes online courses more immersive and engaging, thereby increasing knowledge retention (Gettinger and Ball, 2007; Krause and Coates, 2008; and Kuh, 2009). The guidelines that follow explain how pedagogical agents can be used to increase student engagement in online courses.

 

Using pedagogical agents in eLearning

 

When using pedagogical agents in online courses, it is vital to ensure that they can guide, mentor, assess and engage students effectively during the learning process. The following points outline some recommendations for achieving this.

 

1. Research your target audience and the design of your pedagogical agent

 

Before creating a pedagogical agent for an online course, it is wise to research different designs, as well as how different target audiences might respond to these designs. Remember that agents often take on the role of virtual instructors (Pappas, 2014). As such, it is important that the chosen design appeals to, and motivates, your specific target audience (Shiban et al., 2015). Be mindful of selecting designs that may ‘steal the spotlight’. The role of a pedagogical agent is to enhance the learning experience, and to guide the student through the processes of knowledge acquisition and retention. As such, the design of your agent should improve the learning experience rather than disrupt it (Pappas, 2014).

 

2. Decide on the qualities that your pedagogical agent should possess

 

Based on the research you have conducted, you will need to decide on the qualities that your pedagogical agent should possess. Pedagogical agents should be relatable to your target audience, as this allows students to form a ‘connection’ with them – thereby providing further incentive for engagement with the learning content. Having said this, it may be surprising to learn that these agents don’t need to appear lifelike in order to be effective: cartoonlike agents can be used just as successfully as human-looking agents (Paulose, 2016). However, it is important for the pedagogical agent to display humanlike behaviour. Research indicates that students learn more effectively from agents that exhibit humanlike gestures, movements and facial expressions (Lusk and Atkinson in Paulose, 2016). Furthermore, Mayer (2014) has found that pedagogical agents are more effective when they sound conversational (personalisation principle) and humanlike (voice principle) in comparison to those with formal, machine-like speech patterns.

Adams (2017) and Pappas (2014) recommend that a conversational yet authoritative tone be used when guiding or conversing with students, to come across as helpful and friendly while still maintaining a sense of professionalism. If the tone is too authoritative, students might not find the agent relatable, motivating or interesting. Conversely, if the tone is too friendly, students may perceive the agent to lack credibility.

 

3. Ensure that the agent guides students through the course effectively

 

A key function of pedagogical agents is to guide students through the coursework effectively. This can be achieved in a number of ways. For example, the agent can assist students in navigating from one point to another, offer advice and helpful tips for completing tasks, or draw attention to (i.e. signal) important information. This can be achieved by means of non-verbal feedback, such as gestures or facial expressions, or through direct verbal feedback and conversation. This reduces the potential for confusion, while aiding performance and enabling students to attend to, acquire and retain essential information successfully (Johnson and Lester, 2016; O’Dowd et al., 2019; and Pappas, 2014).

 

4. Assist students in assessing their knowledge

 

Pedagogical agents can also assist students in assessing their knowledge throughout the course. This can be achieved through both questioning and feedback. For example, the agent may be used to ask the student direct questions about the course content in order to test their understanding. Additionally, they could pose probing questions to prompt the student to explore more complex ideas, thus uncovering the beliefs that inform their thinking (an approach that is referred to as Socratic questioning). Feedback (whether corrective or motivational) can then be provided based on the student’s performance. When used as part of formal assessments, pedagogical agents can also help to make the assessment process itself more interactive and enjoyable (Pappas, 2014).

 

Conclusion

 

Research indicates that educational systems aided by interactive pedagogical agents result in more effective eLearning courses (Bickmore et al., 2011). Pedagogical agents increase student engagement and enhance the overall learning experience by guiding, mentoring and interacting with students. In this way, they provide tangible, quantifiable benefits in stimulating learning and, ultimately, aiding academic success (Dinçer and Doğanay, 2015). By following appropriate guidelines and tailoring pedagogical agents to the target audience, educators can create an interactive learning experience in the absence of traditional face-to-face interaction.

________________________________________________________________________

 

References:

 

Adams, J. M. (2017), ‘For teachers, it’s not just what you say, it’s how you say it’. EdSource [website] <https://edsource.org/2017/for-teachers-its-not-just-what-you-say-its-how-you-say-it/574363> accessed 15 May 2020.

Bickmore, T., Pfeifer, L. and Schulman, D. (2011), ‘Relational Agents Improve Engagement and Learning in Science Museum Visitors’. Intelligent Virtual Agents: 11th International Conference. Reykjavik, Iceland. Conference paper, pp. 15–17.

Cook, C. (2017), ‘Avatars and Instruction: How Pedagogical Agents Can Improve Digital Learning’. Medium [website] <https://medium.com/inspired-ideas-prek-12/avatars-and-instruction-how-pedagogical-agents-can-improve-digital-learning-e7930f3b1e01> accessed 15 May 2020.

Dinçer, S. and Doğanay, A. (2015), ‘The Impact of Pedagogical Agent on Learners’ Motivation and Academic Success’. Practice and Theory in Systems of Education (10)4: 329–348

Gettinger, M. and Ball, C. (2007), ‘Best practices in increasing academic engaged time’. In Thomas, A. and Grimes, J. (Eds.) Best practices in school psychology V. Bethesda, MD: National Association of School Psychologists, pp. 1043–1075.

Higher Education Funding Council for England (2008), Tender for a Study into Student Engagement. Bristol: Higher Education Funding Council for England.

Johnson, W. L. and Lester, J. C. (2016), ‘Face-to-Face Interaction with Pedagogical Agents, Twenty Years Later’. International Artificial Intelligence in Education Society 26: 25–36.

Krause, K. L. and Coates, H. (2008), ‘Students’ engagement in first-year university’. Assessment and Evaluation in Higher Education 33(5): 493–505.

Kuh, G. D. (2009), ‘The National Survey of Student Engagement: Conceptual and Empirical Foundations’. In Umbach, P. D. (Ed.) New Directions For Institutional Research. Hoboken, NJ: Wiley InterScience, pp. 5–20.

Lane, H. C. (2016), ‘Pedagogical Agents and Affect: Molding Positive Learning Interactions’. In Tettegah, S. Y. and Gartmeier, M. (Eds.) Emotions, Technology, Design, and Learning. London: Academic Press, pp. 47–62.

Martha, A. S. D. and Santoso, H. B. (2019), ‘The Design and Impact of the Pedagogical Agent: A Systematic Literature Review’. Journal of Educators Online 16(1).

Mayer, R. E. (Ed.) (2014), The Cambridge Handbook of Multimedia Learning. 2nd edn. New York, NY: Cambridge University Press.

O’Dowd, R., Sauro, S. and Spector‐Cohen, E. (2019), ‘The Role of Pedagogical Mentoring in Virtual Exchange’. TESOL Quarterly (54)1: 146–172.

Pappas, C. (2014), ‘Top 10 Tips on How to Use Avatars in eLearning’. eLearning Industry [website] <https://elearningindustry.com/top-10-tips-use-avatars-in-elearning> accessed 15 May 2020.

Paulose, A. (2016), ‘How to Use Pedagogical Agents in Your eLearning’. Infopro Learning [website] <https://www.infoprolearning.com/blog/how-to-use-pedagogical-agents-elearning/> accessed 26 October 2020.

Person, N. K. and Graesser, A. C. ‘Instructional Design: Pedagogical Agents And Tutors’. Education Encyclopedia [website] <https://education.stateuniversity.com/pages/2095/Instructional-Design-PEDAGOGICAL-AGENTS-TUTORS.html> accessed 15 May 2020.

Phan, H. (2011), ‘A cognitive multimedia environment and its importance: A conceptual model for effective e-learning and development’. International Journal on E-Learning 10(2): 199–221.

Shiban, Y., Schelhorn, I., Jobst, V., Hörnlein, A., Puppe, F., Pauli, P. and Mühlberger, A. (2015), ‘The appearance effect: Influences of virtual agent features on performance and motivation’. Computers in Human Behavior 49: 5–11.

Vallverdú, J. and Casacuberta, D. (2009), Handbook of research on synthetic emotions and sociable robotics: New applications in affective computing and artificial intelligence. Hershey, PA: Information Science Reference.

 

0 Comments

Submit a Comment

Your email address will not be published. Required fields are marked *

Recent Journals

Three assumptions that may hamper online learner engagement

Three assumptions that may hamper online learner engagement

The use of online learning materials has many cited benefits in diverse pedagogical contexts (Appana, 2008). However, when designing online courses and materials, educators tend to hold certain presuppositions about learners’ attitudes and capabilities. This article...

read more
The future of learning will be community- and inquiry-based

The future of learning will be community- and inquiry-based

For thousands of years, humans have learned by making personal meaning and by confirming that meaning with those around them – there weren’t always dedicated classrooms, teachers or classmates as there are today. We learned within our community. We learned by...

read more
Five learning theories to consider in eLearning design

Five learning theories to consider in eLearning design

Understanding the principles of knowledge acquisition can help educators to create more effective learning experiences. This article explores five learning theories that are worth considering when designing eLearning courses, and discusses why practitioners may find...

read more
WordPress Image Lightbox