The politeness effect: Pedagogical agents and learning outcomes

Wang, Ning; Johnson, W. Lewis; Mayer, Richard E.; Rizzo, Paola; Shaw, Erin; Collins, H. Thomas

doi:10.1016/j.ijhcs.2007.09.003

Cited by 241 publications

(145 citation statements)

References 36 publications

Supporting

Mentioning

141

Contrasting

Unclassified

Order By: Relevance

“…Although these features have been suggested by the psycholinguistic literature, reported correlations with personality ratings are generally weak: it was not obvious that they would improve accuracies of statistical models on unseen subjects. Computational work on modelling personality has primarily focused on methods for expressing personality in virtual agents and tutorial systems, and concepts related to personality such as politeness, emotion, or social intelligence (Walker, Cahn, & Whittaker, 1997;André, Klesen, Gebhard, Allen, & Rist, 1999;Lester, Towns, & FitzGerald, 1999;Wang, Johnson, Mayer, Rizzo, Shaw, & Collins, 2005) inter alia. Studies have shown that user evaluations of agent personality depend on the user's own personality (Reeves & Nass, 1996;Cassell & Bickmore, 2003), suggesting that an ability to model the user's personality is required.…”

Section: Discussion and Future Workmentioning

confidence: 99%

Using Linguistic Cues for the Automatic Recognition of Personality in Conversation and Text

Mairesse¹,

Walker²,

Mehl³

et al. 2007

jair

743

583

View full text Add to dashboard Cite

It is well known that utterances convey a great deal of information about the speaker in addition to their semantic content. One such type of information consists of cues to the speaker's personality traits, the most fundamental dimension of variation between humans. Recent work explores the automatic detection of other types of pragmatic variation in text and conversation, such as emotion, deception, speaker charisma, dominance, point of view, subjectivity, opinion and sentiment. Personality affects these other aspects of linguistic production, and thus personality recognition may be useful for these tasks, in addition to many other potential applications. However, to date, there is little work on the automatic recognition of personality traits. This article reports experimental results for recognition of all Big Five personality traits, in both conversation and text, utilising both self and observer ratings of personality. While other work reports classification results, we experiment with classification, regression and ranking models. For each model, we analyse the effect of different feature sets on accuracy. Results show that for some traits, any type of statistical model performs significantly better than the baseline, but ranking models perform best overall. We also present an experiment suggesting that ranking models are more accurate than multi-class classifiers for modelling personality. In addition, recognition models trained on observed personality perform better than models trained using selfreports, and the optimal feature set depends on the personality trait. A qualitative analysis of the learned models confirms previous findings linking language and personality, while revealing many new linguistic markers.

show abstract

Section: Discussion and Future Workmentioning

confidence: 99%

Using Linguistic Cues for the Automatic Recognition of Personality in Conversation and Text

Mairesse¹,

Walker²,

Mehl³

et al. 2007

jair

743

583

View full text Add to dashboard Cite

show abstract

“…The theoretical rationale is that personalization techniques-such as using conversational style or polite wording-creates a sense of social partnership with the narrator in which learners try harder to make sense of what their conversational partner is saying. In 11 of 11 experiments, involving computer-based lessons on lungs, lightning, botany, and industrial engineering, learners who received words in conversational style performed better on transfer tests than did learners who received words in formal style (Mayer, Fennell, Farmer, & Campbell, 2004, Experiments 1, 2, and 3; Moreno & Mayer, 2000b, Experiments 1, 2, 3, 4, and 5;Moreno & Mayer, 2004, Experiments 1a and 1b;Wang et al, 2008). The median effect size was 1.11, which is a large effect.…”

Section: Two Principles For Fostering Generative Processingmentioning

confidence: 99%

Applying the science of learning: Evidence-based principles for the design of multimedia instruction.

Mayer

2008

American Psychologist

Self Cite

659

601

View full text Add to dashboard Cite

Suppose you want to design (or select) (a) a computerbased training program to teach employees how to use a new database program, (b) a biology textbook for high school students, (c) an instructional video game aimed at promoting healthy eating, or even (d) a PowerPoint presentation on a topic in educational psychology. Each of these is an example of what I call a multimedia instructional message-a lesson containing words (e.g., printed words or spoken words) and pictures (e,g., illustrations, photos, animation, or video) that is intended to foster learning. How would you decide the best way to present your multimedia instruction? To help answer this question, you need some way of judging whether a proposed instructional method is consistent with research-based theories of how people learn (i.e., the science of learning) and evidence-based principles of how to design instruction (i.e., the science of instruction).Let's consider three approaches to the relation between theory and practice (Mayer, 1992). In the one-way street approach, psychologists develop a theory of how people learn that is based on research, and educators apply the theory in their lessons. The problem with the one-way street approach is that a clear specification of how people learn does not automatically yield a clear specification of effective instructional methods and materials. In the dead end street approach, psychologists busy themselves in building a theory of learning that is not closely related to authentic educational challenges-such as studying how laboratory animals run mazes or how adults memorize word lists-while educators seek research that determines the best method of instruction without regard to how or why it works. Finally, in the two-way street approach, there is a reciprocal relation between learning theory and educational practice in which the science of learning must be expanded to be able to explain how learning works in authentic learning situations, and the science of instruction must be expanded to consider the conditions for each instructional principle based on an understanding of how the human mind works. Figure 1 summarizes the relation between basic research-which has been attributed to the science of learning (SOL)-and applied research--which has been attributed to the science of instruction (SOI). A defining feature of the science of instruction is a focus on authentic learning situations rather than contrived learning situations, so Figure 1 shows "SOI" in the two quadrants on the right. A defining feature of the science of learning is a focus on testing learning theory, so "SOL" is shown in the two quadrants at the bottom of the figure. The four quadrants in Figure 1 are inspired by the work of Stokes (1997). The top left quadrant-conducting nontheoretical studies on contrived learning situations-is consistent with neither the SOL nor the SOI and is unlikely to yield information that is helpful for theory or practice. The top right quadrantconducting nontheoretical studies on authentic tasks-is consistent w...

show abstract

“…They consist of the aspects of the visual and auditory presence of the APA that are hypothesized to ''make the learning experience more interesting, believable, or natural'' (p. 508). Empirical studies of these properties have explored the influence of the APA's age, gender, ethnicity, politeness, responsiveness, dynamism, and visual appeal, among others (e.g., André et al 1996;Arroyo et al 2011Arroyo et al , 2009Baylor 2011;Baylor and Kim 2004;Baylor and Ryu 2003;Craig et al 2002;Domagk 2010;Kim et al 2007;Lester et al 1997;Moreno 2004;Moreno et al 2000;Plant et al 2009;Wang et al 2008;Woolf et al 2010). Given our special interest in gender and science education, we summarize below the empirical studies that have investigated gender and motivation in relation to APAs in virtual science learning environments.…”

Section: Animated Pedagogical Agents and Student Motivationmentioning

confidence: 99%

“…This is somewhat strange in view of the fact that motivation, ''the process whereby goal-directed activity is instigated and sustained'' (Pintrich and Schunk 2002, p. 5), can be considered a precondition of successful inquiry learning. Learning is at risk when students exert little effort because they lack confidence in their ability to complete inquiry activities successfully and fail to perceive the relevance of task engagement (Wang et al 2008). This study investigates whether student motivation in a science inquiry learning environment can be enhanced via motivational support from an animated pedagogical agent (APA).…”

Section: Introductionmentioning

confidence: 99%

Animated pedagogical agents effects on enhancing student motivation and learning in a science inquiry learning environment

Meij

Harmsen

2015

Education Tech Research Dev

View full text Add to dashboard Cite

This study focuses on the design and testing of a motivational animated pedagogical agent (APA) in an inquiry learning environment on kinematics. The aim of including the APA was to enhance students' perceptions of task relevance and selfefficacy. Given the under-representation of girls in science classrooms, special attention was given to designing an APA that would appeal to the female students. A review of the literature suggested that the best design solution would be an agent who was female, young, attractive, and ''cool''. An experiment compared three conditions: agent (image and voice), voice (no image), and control (no image and no voice). The research question was whether students' motivation and knowledge changed over time as they worked in the inquiry learning environment, and whether condition and gender affected such changes. Participants were 61 third-year students (mean age 14.7 years) from a secondary school. Gender was distributed evenly within and across conditions. A significant main effect of time on self-efficacy was found, with self-efficacy beliefs increasing significantly for both boys and girls. In addition, there was a significant interaction between time, condition, and gender for self-efficacy. About halfway during training, girls' self-efficacy beliefs significantly increased in both experimental conditions and decreased in the control condition. For boys the opposite pattern was found. Girls also gave higher appraisals for the agent. Students in all three conditions realized significant knowledge gains, which did not differ by gender. The discussion critically considers the need for, and design of motivational scaffolding in inquiry learning environments.

show abstract

The politeness effect: Pedagogical agents and learning outcomes

Cited by 241 publications

References 36 publications

Using Linguistic Cues for the Automatic Recognition of Personality in Conversation and Text

Using Linguistic Cues for the Automatic Recognition of Personality in Conversation and Text

Applying the science of learning: Evidence-based principles for the design of multimedia instruction.

Animated pedagogical agents effects on enhancing student motivation and learning in a science inquiry learning environment

Contact Info

Product

Resources

About