Hochschulschrift (Dissertation)
San Diego, USA, 2010
Growing interest of educational institutions in desktop 3D graphic virtual environments for hybrid and distance education prompts questions on the efficacy of such tools. Virtual worlds, such as Second Life?, enable computer-mediated immersion and interactions encompassing multimodal communication channels including audio, video, and text-. These are enriched by avatar-mediated body language and physical manipulation of the environment. In this para-physical world, instructors and students alike employ avatars to establish their social presence in a wide variety of curricular and extra-curricular contexts.
As a proxy for the human body in synthetic 3D environments, an avatar represents a 'real' human computer user and incorporates default behavior patterns (e.g., autonomous gestures such as changes in body orientation or movement of hands) as well as expressive movements directly controlled by the user through keyboard `shortcuts.' Use of headset microphones and various stereophonic effects allows users to project their speech directly from the apparent location of their avatar. In addition, personalized information displays allow users to share graphical information, including text messages and hypertext links. These 'channels' of information constituted an integrated and dynamic framework for projecting avatar `immediacy' behaviors (including gestures, intonation, and patterns of interaction with students), that may positively or negatively affect the degree to which other observers of the virtual world perceive the user represented by the avatar as 'socially present' in the virtual world.
This study contributes to the nascent research on educational implementations of Second Life in higher education. Although education researchers have investigated the impact of instructor immediacy behaviors on student perception of instructor social presence, students' satisfaction, motivation, and learning, few researchers have examined the effects of immediacy behaviors in a 3D virtual environment or the effects of immediacy behaviors manifested by avatars representing instructors.
The study employed a two-factor experimental design to investigate the relationship between instructor avatars' immediacy behaviors (high vs. low) and students' perception of instructor immediacy, instructor social presence, student avatars co-presence and learning outcomes in Second Life. The study replicates and extends aspects of an earlier study conducted by Maria Schutt, Brock S. Allen, and Mark Laumakis, including components of the experimental treatments that manipulated the frequency of various types of immediacy behaviors identified by other researchers as potentially related to perception of social presence in face-to-face and mediated instruction. Participants were 281 students enrolled in an introductory psychology course at San Diego State University who were randomly assigned to one of four groups. Each group viewed a different version of the 28-minute teaching session in Second Life on current perspective in psychology.
Data were gathered from student survey responses and tests on the lesson content. Analysis of variance revealed significant differences between the treatment groups (F (3,113) = 6.5, p = .000). Students who viewed the high immediacy machinimas (Group 1 HiHi and Group 2 HiLo) rated the immediacy behaviors of the instructor-avatar more highly than those who viewed the low-immediacy machinimas (Group 3 LoHi and Group 4 LoLo). Findings also demonstrate strong correlations between students' perception of instructor avatar immediacy and instructor social presence (r = .769). These outcomes in the context of a 3D virtual world are consistent with findings on instructor immediacy and social presence literature in traditional and online classes. Results relative to learning showed that all groups tested higher after viewing the treatment, with no significant differences between groups. Recommendations for current and future practice of using instructor-avatars include paralanguage behaviors such as voice quality, emotion and prosodic features and nonverbal behaviors such as proxemics and gestures, facial expression, lip synchronization and eye contact.
Additional information and a video you will find on http://sabinereljic.wordpress.com/dissertation/