| The purpose of this study was to examine the effectiveness of animated pedagogical agents designed to assist learners during their interaction with a multimedia learning environment involving proportional reasoning word problems. The animated agents included were designed to direct learner attention to appropriate problem states and consisted of: (a) a fully embodied agent, who assisted learners via both verbal and nonverbal modes of communication, (b) a minimally embodied agent, who provided only verbal instructions in the absence of any nonverbal communication cues, and (c) a no agent condition, in which learners received verbal instructions identical to the former two conditions simply without the presence of an on-screen animated agent. An additional goal of this study was to investigate specific types of worked examples incorporated into the computer-based learning environment. The proportional reasoning worked examples implemented in this study consisted of: (a) a simple learning environment wherein subgoals of the worked example gradually unfolded until the example was presented in its entirety (subgoal-oriented examples), or (b) a complex learning environment, which contained the simultaneous presentation of problem states at the onset of the worked example and lacked subgoal-orientation (simultaneous-oriented examples).; Results indicated that participants exposed to animated agent conditions outperformed students in the no-agent condition. Additionally, students receiving instruction in the simple learning environment produced higher performance scores on tests of transfer than those students in the complex learning environment. Results relative to the benefit of including animated agents into computer-based learning environments provide moderate support for the social agency theory. Findings related to the incorporation of subgoal-oriented examples into computer tutorials rather than simultaneous-oriented examples lend support to cognitive load theory.; This study contributes to current research by replicating an existing animated agent effect, which suggests incorporating an animated agent into a multimedia learning environment optimizes learning. Second, this study was the first to empirically investigate the difference between subgoal-oriented worked examples and simultaneous-oriented worked examples. Moreover, because the subgoal-oriented examples proved superior to simultaneous-oriented examples, the current study suggests that a sequential principle exists, in which sequentially presented subgoals are superior to simultaneous presented subgoals. |
