This study compared the use of an animated pedagogical agent (agent) with an electronic arrow and voice narration (arrow and voice) in a multimedia learning environment where 74 college level English as a Second Language (ESL) students learned English relative clauses. No significant differences in learning or performance were found between the group receiving the agent and the arrow and voice group. This result provides evidence for Clark's (2001) claim that learning is supported by appropriate instructional methods (in this study directing attention to and voicing key concepts), not a delivery medium (in this study an animated agent or the drawing of an arrow with voice). Clark has also argued against using expensive instructional media when lower cost alternatives achieve the same or similar learning and motivation goals. It was also found that the agent's visual appearances and social behaviors did not motivate, interest, or tutor learners better than the simple arrow with voice. Thus, this result does not support the Persona Effect (Lester, Converse, Kahler, Barlow, Stone, & Bhogal, 1997) which is derived from the hypothesis that an animated agent motivates human-computer interaction because it is perceived as more socially engaging and interesting than other attention directing and information dispensing devices, and thus leads learners to work harder and learn more. The study also did not support the cognitive efficiency hypothesis (Cobb, 1997) which refers to one medium requiring less effort from learners in achieving a specific learning criterion or leading to faster learning than another. No significant difference was found between the agent and arrow with voice groups in the measured levels of cognitive efficiency. Yet, as in past studies, the data suggested a significant learning benefit of animated pedagogical agents for learners with the lowest levels of prior knowledge but no differences at the intermediate or advanced prior knowledge levels. (Contains 5 figures and 4 tables.)