ABSTRACT. Learners have difficulty in decomposing conventionally designed animations to obtain raw material suitable for building high quality mental models. A composition approach to designing animations based on the Animation Processing Model was developed as a principled alternative to prevailing approaches. It provides learners with pre-decomposed material that is structured and sequenced to facilitate the relation building required for effective mental model construction. Study of a compositional animation that presented material in a contiguous fashion resulted in higher quality mental models of a piano mechanism than non-contiguous or control (conventional) versions. Eye fixation data indicated that the compositional animation led to superior mental models because it particularly fostered relational processing.

ABSTRACT. The present research investigates whether a drawing task improves monitoring accuracy and takes a closer look at the link between monitoring and control processes during learning. Accurate monitoring is essential for controlling and regulating one’s learning effectively. Monitoring accuracy can be improved by performing an active task after learning (e.g., drawing). In Study 1, we compared a condition with drawing task after each one of five text paragraphs with a control condition without drawing task. All participants made judgments of learning (Jols) for each paragraph and selected paragraphs for restudying. As expected, drawing improved monitoring accuracy (stronger link between Jols and performance) and increased the likelihood of restudying paragraphs that had not been understood (Jols more predictive for reselection). In Study 2 we investigated whether the drawing task would also lead to longer restudying of paragraphs that had not yet been understood, thereby fostering learning outcomes.

ABSTRACT. This paper presents part of a wider project to create, validate and apply a comprehensive dynamic spatial ability test suitable for use with learning from animation. Four types of test item were devised (two race tasks and two intercept tasks) and tested with forty undergraduate participants. Absolute and relative speed of the objects as well as trajectory directions were manipulated. Task performances and the effect of the manipulated factors are reported. Comparison with a conventional static spatial ability test revealed little correlation, suggesting that dynamic spatial ability can be regarded as a distinct capacity.

ABSTRACT. Does teacher enthusiasm benefit learning when students watch an online lecture? Although this intuitively seems to make sense, previous research with video material only found effects on perceived learning, and not on actual learning. In our study, 74 participants watched a short online lecture about the life of Ernest Hemingway in their home setting. In the Enthusiastic condition, the non-verbal expressiveness of the lecturer was high, whereas in the Unenthusiastic condition, gestures and vivid intonation were absent. The participants in the Enthusiastic condition rated the lecturer higher on enthusiasm, effectiveness and induced interest. And although their perceived learning was comparable to the Unenthusiastic condition, their performance on the actual achievement test was much higher. Thus, especially in naturalistic settings, teacher enthusiasm does seem to matter for learning from online lectures.

ABSTRACT. Supplementing multimedia learning materials with instructions in a conversational style rather than a formal style can facilitate learning. This is the so-called personalization principle. We investigated whether a specific language/cultural background could present a boundary condition for this principle. Across four experiments with a Czech sample (N = 278), we failed to find the superiority of instructions in a conversational style (d = 0.07) in the context where this superiority has been clearly demonstrated for US samples (i.e., short instructional animations; college and high school audiences). Twenty-nine percent of participants who received the conversational instructions expressed explicit reservations regarding the instructional style. The results demonstrate that language/cultural backgrounds indeed present a boundary condition for the personalization principle. The findings can be explained by the generally more formal approach to education in the Czech Republic compared to the US schooling system.

ABSTRACT. The redundancy principle states that presenting the same words in a narration and as on-screen text hinders learning. However, recent research shows that when redundant verbal information is presented to narrated multimedia presentations as on-screen labels, there is greater learning than when the complete verbally redundant text is presented. The present study extends this line of research by focusing on the usefulness of on-screen labels in an animation explaining a procedural task. The experiment had a 2x2x2 between-subject design (N = 129) with narration (yes vs. no), written text (yes vs. no), and on-screen labels (yes vs. no) as factors. Participants studied a first-aid procedure from an animation after which they performed and described the procedure. It is concluded that learning performance increased when the animation was presented with verbal information. On-screen labels improved learning, especially together with narration, but not more than when complete verbal redundant information was provided.

ABSTRACT. The expertise reversal effect in multimedia learning has been defined as a stimulus-design x learner knowledge interaction: the effect of design decisions about the multimedia stimulus differs depending on the knowledge level of the learner. In this literature review, I look closely into this research and find that some studies actually vary the task that learners complete, while holding the stimulus design constant, yielding task x knowledge interactions. Other studies actually vary the content over time, while holding the stimulus design constant, yielding a task x knowledge interaction. I conclude this synthetic review with ideas for how this perspective can enrich designers’ training and practices, theory, and future research

ABSTRACT. This round table will explore the potential of learner demonstrations to be a more effective alternative support strategy than self-generated drawing for improving learning from complex animations. Discussion will be stimulated by three presentations reporting recent work of the EARLI Center for Innovative Research ‘Drawing for Learning from Animation’ and key focus questions arising from that research. The presentations will cover the processing challenges posed to learners by requiring them to self-generated drawings from complex animations, how these challenges may be ameliorated by instead having learners perform demonstrations with manipulable models, and empirical investigations of the effectiveness of such demonstrations with respect to two different types of subject matter. Key themes for this round table are the central importance of extracting dynamic information to successful processing of animations and how learner demonstrations relate to the Animation Processing Model (Lowe & Boucheix, 2008).

ABSTRACT. This submission is a contribution to the round-table “Demonstrating dynamics to support learning from complex animations”. In a recent study we observed that drawing hindered understanding of broader spatial and temporal structures during learning from a complex animation. During drawing the students focused on visuospatial characteristics of the animation and neglected spatiotemporal aspects. In the present study, drawing is replaced by a task in which the studentsdemonstrate behaviors of the animated subject matter using a physical model. We hypothesize that demonstrations make the students better aware of the spatiotemporal characteristics of the animation. Further, in this study the animation builds up step by step. Thus, the visuospatial and spatiotemporal complexity is reduced, so that the identification of events as well as the identification of spatial and temporal relations between events might be facilitated. The effectiveness of demonstrating and incremental completion is investigated by means of a two-factorial experimental design.

ABSTRACT. The present experiment tested the effect of hand demonstration (with a plastic replica) on the mental model quality building from an animation of a scotch yoke mechanism. A hand demonstration condition is compared to a writing for understanding condition both followed by a post test-which assessed the mental model quality level of the scotch yoke mechanism for two independent groups of 20 participants each. The writing condition is more usual than the demonstration condition in conventional learning instructional design. However, on the basis of the A.P.M. model, we hypothesized that with the same controlled learning time, participants in the demonstration condition should outperformed participants in the writing condition. There could be an optimal "affordance" between the spatiotemporal properties of two hands demonstration and the mental model building steps requirements. The results of the experiment will be presented at the conference.

ABSTRACT. This submission is a contribution to the round-table “Demonstrating dynamics to support learning from complex animations”. Learner self-generation of static drawings from complex animations results in an emphasis on visuospatial rather than spatiotemporal information. Demonstration of the subject matter’s dynamics may be a more effective alternative for supporting learning. However, the potential of this approach is likely dependent on the specific design features of manipulable models used to make such demonstrations and how the quality of the resulting mental model is assessed.

ABSTRACT. Molecular animations are commonplace in today’s undergraduate biology or chemistry classroom, while studies examining their efficacy report mixed results. This study examines techniques for calling attention to the random movements of two proteins within a crowded molecular environment, while simultaneously depicting a binding event. Participants (n=148) watched one of five animations with or without additional visual cues highlighting the movements of two binding proteins. We hypothesized that one of the four cued treatments (linear path, velocity vectors, ghosting, and heat map) would provide students with important insights about the stochastic nature of molecular events without impeding their ability to follow the binding process. Multiple choice pre and post-test measures were collected using iClickers. While additional cueing strategies did moderate participants’ understanding of interactions they failed to correct prior conceptions relating to the stochastic nature of molecular environments.

ABSTRACT. Studies investigating the usability of bar vs. pie graphs show contrasting results and researchers disagree about which perceptual features are primarily responsible for their effect. In two studies, we offer evidence for the role and the effect of two crucial perceptual features in reading graphs: length and area. In an evaluation study, we examined which features are actually perceived by non-expert users as representing quantity. Results show that their judgments are less clear cut than the assumptions of researchers. For most graphs, more than one feature was perceived to play a considerable role, and overall, area was perceived to play a more crucial role than length in the majority of graphs. The study provided us with a classification of graph types based on the role of these features. In a second study, we asked respondents to make simple and more complex comparisons between quantities in a graph. Performance was more accurate and efficient with length than with area representing quantity, but only in complex comparisons.

ABSTRACT. The relationship between physiological responses during reading and text comprehensions has received very scant attention. The present study aimed to examine undergraduates’ heart rate reactivity while they read a science informational text in one of two reading goal conditions: to read for themselves or to read to answer to at least 50% of the post-reading questions to gain course credit. Gender, prior knowledge, topic interest and self-perception of knowledge, and final grade at the end of high school were also considered as control variables. Findings revealed that in the reading condition focused on good performance, students had larger heart rate reactivity reflecting physiological arousal, a component of engagement. They also comprehended the text better. Moreover, it emerged that the cardiovascular response to text reading partially mediated the effect of reading goal condition on text comprehension.

ABSTRACT. Decorated graphs are commonly used for communicating quantitative information in mass media and school textbooks. The present study investigated the possible effect of pictorial decorations on students’ self-generated titles for given graphs. Four groups of 8th graders participated in the study (total N=120 boys and girls). All groups received the same set of graphs comprising pie, bar and scatter graphs. Three groups received decorated graphs and one (control) received plain graphs. No differences in mean title scores were found between research and control groups. Nevertheless, differences in performance patterns between low-medium achievers and high achievers suggest a possible hindering effect of decorative elements on students graph comprehension. These findings bear direct implications for the design of learning materials.