Academic journal article Educational Technology & Society

The Influence of a Pedagogical Agent on Learners' Cognitive Load

Academic journal article Educational Technology & Society

The Influence of a Pedagogical Agent on Learners' Cognitive Load

Article excerpt


Virtual characters continue to appear in a multitude of computer-based learning environments. One type of virtual character, the pedagogical agent, continues to be a focus of recent research. The term pedagogical agent refers to a wide range of virtual characters, but they always have a visual presence in the learning environment and their purpose is to help students learn the material (Moreno, 2005; Park, 2015; Yung & Paas, 2015).

While pedagogical agent research has been ongoing for approximately 20 years, researchers have continued to question the efficacy of a pedagogical agent's ability to facilitate the learning process. While Heidig and Clarebout's (2011) systematic review found that pedagogical agents often led to no significant differences between groups on learning outcomes, Schroeder, Adesope, and Gilbert's (2013) meta-analysis found that pedagogical agents can lead to small positive effects on learning outcomes. Despite these findings, the author concurs with Heidig and Clarebout (2011) in that "the question of whether pedagogical agents generally facilitate the learning process is too broad" (p. 30). However, it is important to note that this statement could apply to nearly any type of technology-based learning tool. As such, it continues to be important to examine what types of outcomes various pedagogical agents can influence in the learning environment.

Proponents of pedagogical agents have argued that agents can enhance the social aspects of the multimedia learning environment, thus causing the learner to be more cognitively engaged with the learning materials (Atkinson, Mayer, & Merrill, 2005; Mayer & DaPra, 2012; Mayer, Sobko, & Mautone, 2003). Similarly, researchers have found evidence that the pedagogical agent's design can influence learning outcomes and learners' perceptions (Domagk, 2010; Veletsianos, 2007). However, we still do not have definitive guidance as to what features pedagogical agents should have and there are many possible agent implementations yet to be thoroughly investigated.

As with many other technological innovations, the use of pedagogical agents is somewhat controversial. Arguments against their use are typically grounded in cognitive load theory. For instance, Clark and Choi (2007) argued that pedagogical agents may overload a learner's working memory capacity. This notion is echoed in Wouters, Paas, and van Merrienboer's (2008) review, which concluded that agents must be designed carefully to avoid overloading the working memory capacity. Similarly, van Mulken, Andre, and Muller (1998) argued that a pedagogical agent may cause distraction to the learner during the learning process.

Another issue that deserves further attention in the realm of multimedia learning is the pacing of instruction. Most pedagogical agent research has been conducted in learner-paced environments, where the instruction is broken into pre-determined instructional segments (Schroeder, Adesope, & Gilbert, 2013). However, developing this sort of program often requires computer programming skills and time that many instructors do not have. Accordingly, many instructors will host videos on services like YouTube. A YouTube video shares qualities of both learner-paced and system-paced instructional sequences. For instance, the learner can pause and fastforward -qualities that are similar to a learner-paced environment. However, the instructional sequence is not segmented and does not pause and wait for learners to click to continue - qualities that are similar to a systempaced instructional sequence. Since some principles of multimedia learning can have differential, or even inverse effects in differently paced environments (Ginns, 2005), it is important to clarify the type of pacing being used. Schroeder and Adesope (2013, 2015) have identified the pacing that occurs in these types of videos as learnerattenuated system-paced (LASP) learning environments. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.