The Effects of Social Cue Principles on Cognitive Load, Situational Interest, Motivation, and Achievement in Pedagogical Agent Multimedia Learning

Introduction The use of animated pedagogical agents in multimedia learning environments has increased as new technologies have made them more accessible (Gholson & Craig, 2002; Johnson, Rickel, & Lester, 2000). Pedagogical agents are animated life-like characters enabled with speech, gesture, movement, and human-like behaviors (Sweller, Ayres, & Kalyuga, 2011) and designed to facilitate learning in multimedia learning environments (Johnson et al., 2000). Pedagogical agents can embody different pedagogical roles to support learners by supplanting, scaffolding, coaching, testing, or demonstrating or modeling a procedure (Schroeder & Adesope, 2014). Previous studies have claimed the positive influences of pedagogical agents on student motivation and interest (Atkinson, 2002; Moreno, 2005) and also have indicated positive effects on student attitude toward learning and performance (Baylor, 2002a, 2002b; Baylor & Ryu, 2003; Moreno et al., 2001). However, other studies reported that pedagogical agents in multimedia learning could cause unnecessary cognitive load (Choi & Clark, 2006; Clark & Choi, 2005) called extraneous cognitive load. For example, the split-attention effect could occur when multiple sources of information are presented in split-attention without being integrated (Ayres & Sweller, 2014). The modality effect also can be caused when multiple sources of information are presented in single-modality not in dual modality (Low & Sweller, 2014). Both split-attention effect and modality effect are considered cognitive load effects caused by information presented through multiple information sources. According to cognitive load theory (CLT), our brain utilizes two primary types of memory, the working memory and the long term memory, to process, store, and access information (Kalyuga, 2011; Sweller, 2005; Sweller, van Merrienboer, & Paas, 1998). Due to the limited capacity of the working memory, learners must cope with a certain level of cognitive load to process newly presented information. In pedagogical agent multimedia learning, both auditory and visual channels of information can be engaged in working memory from two different sources that are pedagogical agent and on-screen multimedia material, thus influence learners' cognitive load as a whole. Cognitive load theorists (Pass, Renkl, & Sweller, 2003; Sweller, 1999, 2005) agree that three different types of cognitive load need to be considered in designing instruction: (a) intrinsic cognitive load, (b) extraneous cognitive load, and (c) germane cognitive load. Intrinsic cognitive load is imposed by the intrinsic nature of presented information or learning task itself and should be reduced (i.e., task difficulty). Extraneous cognitive load results from the ineffective instructional design and needs to be prevented (e.g., format of instructional materials). Germane cognitive load is also imposed by instructional design but is effective for learning (e.g., a learner's effortful process of understanding). The distinction between intrinsic cognitive load and germane cognitive load is not clearly made (see Sweller, Ayres, & Kalyuga, 2011) because germane cognitive load is not imposed by the nature and structure of the learning materials. However, germane cognitive load has been associated with various additional cognitive activities that are designed to foster schema acquisition (Kalyuga, 2010). Therefore, it would be reasonable to consider germane cognitive load as sources of auxiliary cognitive activities to enhance learning outcomes or to increase learner motivation (Kalyuga, 2010). Applying the types of cognitive load to designing multimedia learning environments, Mayer (2009, 2014b) suggested three kinds of cognitive processing demands in his cognitive theory of multimedia learning (CTML) that are extraneous processing, essential processing, and generative processing (also see Moreno & Mayer, 2010). …