Schedule SIG 2 meeting 2014 Rotterdam (Monday 25-Wednesday 27 August) - Abstracts below

Monday

25 Aug

08:30-09:15

Registration & coffee (CT & CB-hall)

09:15-09:30

Opening & Welcome - Huib Tabbers (CT-1)

09:30-10:30

Paper session A (CT-1) - chair: Jean-Michel Boucheix

1. Building bridges: Perspectives of researchers and teachers on visual representations in science education - Shaaron Ainsworth* and Len Newton

2. Amplifying multimedia effects by repeating the study-test cycle: Evidence for underconfidence with practice? - Alexander Eitel* and Katharina Scheiter

3. Bar (n)or pie - Annemarie Quispel* and Alfons Maes

10:30-11:00

Coffee (CB–hall)

11:00-12:30

Paper session B (CT-1) - chair: Ivar Braten

4. A comparison of paper-based and video tutorials for software learning - Jan van der Meij* and Hans van der Meij

5. Diagrams: Connecting text and image - Brad Jackel*

6. Contents and graphics in line: When is it beneficial to schematize pictures in expository prose? - Hans Westerbeek*, Marije van Amelsvoort, Alfons Maes and Marc Swerts

7. Decorative pictures in primary schools: Seductive or beneficial? - Maria Opfermann*, Annett Schmeck, Anne Wienand and Detlev Leutner

12:30-13:30

Lunch (CB-hall)

13:30-15:00

Parallel sessions (C2)

Parallel 1 (C2-3)

Roundtable: Animations & drawing-

chair: Ric Lowe

1. Ric Lowe
2. Benjamin Fillisch
3. Jean-Michel Boucheix

Parallel 2 (C2-4)

Roundtable:
Graphical representations -

chair: Marije van Amelsvoort

4. Marije van Amelsvoort
5. Sandra Nitz
6. Steffen Wagner

Parallel 3 (C2-5)

Interlinked papers:
Text comprehension and epistemic cognition -
chair: Panayiota Kendeou

7. Panayiota Kendeou
8. Marc Stadtler
9. Kalypso Iordanou
10. Gregory Trevors
11. Ivar Bråten

Parallel 4 (C2-6)

Interlinked papers:
A large-scale reading comprehension intervention -

chair: Bjorn de Koning

12. Bjorn de Koning
13. Lisanne Bos
14. Menno van der Schoot

15:00-16:00

Keynote I : Jan Willem Huisman (CT-1)

16:00-16:30

Coffee (CB-hall)

16:30-18:00

Paper session C (CT-1) - chair: Jean-Francois Rouet

8. Do versus think prompts: What is best for multimedia learning? - Irene T. Skuballa*, Marika Dobos, Panetta Sabina, Caroline Fortunski, Anke Dammert, Cornelia Hauser, Jasmin Leber and Alexander Renkl

9. Does animation in public transport information improve decision making? - Jonathan Groff*, Jean-Michel Boucheix and Laurence Paire-Ficout

10. Designing context aware instructions: Perceptual salience and task demands in the selection of natural landmarks - Adriana Alexandra Baltaretu*, Emiel Krahmer and Alfons Maes

11. Effects of building vs. studying a concept map on hypertext comprehension - Julie Lemarié*, Franck Amadieu, Ladislao Salmeron, Julien Cegarra, Pierre-Vincent Paubel and Aline Chevalier

18:00-20:00

Opening reception (Erasmus Paviljoen)

 

Tuesday

26 Aug

09:00-10:30

Paper session D (CT-1) - chair: Ladislao Salmeron

12. Fostering piano learning by dynamic mapping of notes - Melina Klepsch*, Bastian Könings, Michael Weber and Tina Seufert

13. Grounding website navigation: The effects of visuospatial cues on search behaviour - Martin Galilee*, Neil Schwartz and Erica de Vries

14. How do we integrate text and pictures? - Jana Arndt*, Anne Schüler and Katharina Scheiter

15. How graphs can mislead - Lisanne van Weelden*, Marije van Amelsvoort and Michael Bernhofer

10:30-11:00

Coffee (CB-hall)

11:00-12:30

Paper session E (CT-1) - chair: Katharina Scheiter

16. How to get better in performing mathematical tasks: A teacher-based training program - Michael Kraus*, Silja-Susann Taxis, Marion Geiger, Markus Vogel and Tina Seufert

17. ICT-related component skills and navigation in reading digital text - Carolin Hahnel*, Frank Goldhammer, Johannes Naumann and Ulf Kröhne

18. Learning through visual illustration and retrieval practice - Ann-Sofie Jägerskog*, Fredrik Jönsson, Staffan Selander and Bert Jonsson

19. Mind the gap: Multimedia instruction, pain, and cognition - Alexander Smith and Paul Ayres*

12:30-13:30

Lunch (CB-hall)

13:30-15:00

Poster session (CB-hall)

Theme 1 – Text

1. Fajardo Bravo
2. Franck Amadieu / Julie Lemarié*
3. Helge I. Strømsø
4. Jaan Mikk
5. Katharina Schiller
6. Monique Reichert
7. Sabine Klois
8. Sarah von der Muehlen
9. Sylvia Maria Savvidou
10. Yvonne Kammerer

Theme 2 – Graphics

11. Carolin Enzingmüller
12. Christine Tippett
13. Asuncion Lopez-Manjon / Cristina Marín*
14. Cristina Marín
15. Inga Ubben
16. Inouk Boerma
17. Jan Engelen
18. Muhammad Ashraf
19. Olga Kasatkina
20. Roxette van den Bosch

Theme 3 – Multimodal

21. Andrea Karpati
22. Carina Schubert
23. Jodie Jenkinson
24. Liesbeth Kester
25. Mona Holmqvist Olander
26/27. Richard Lowe (2x)
28. Ozge Yasar/ Janice Gobert*

 

Theme 4 – Instruction

31. Andrea Gauthier
32. Fang Zhao
33. Gerdien van Eersel/ Mario de Jonge*
34. Margot van Wermeskerken
35. Nora Anders
36. Pauline Reijners
37. Paulo Correia
38. Tina Seufert
39. Tinne Dewolf
40. Vincent Hoogerheide

15:00-16:00

Keynote II : Paul van den Broek (CT-1)

16:00-16:30

Coffee (CB-hall)

16:30-18:00

Paper session F (CB-2) - chair: Tina Seufert

20. Reducing transience of complex animation does not (always) lead to better learning performances - Jean-Michel Boucheix* and Claire Forestier

21. Success in solving mathematical word problems: A question of representational flexibility? - Katharina Hohn* and Wolfgang Schnotz

22. The effect of spatial text information and spatial ability on learning to tie knots with 2D- and 3D-visualizations - Tim Kühl*, Ulrike Kretschmar and Stefan Münzer

23. The role of visuo-spatial abilities in learning three-dimensional anatomical structures from animation - Sandra Berney* and Mireille Bétrancourt

18:00-18:45

Business meeting SIG 2 (CB-2)

19:30- ...

Conference dinner (restaurant ‘Prachtig’ [‘Beautiful’])

 

Wednesday

27 Aug

09:00-10:00

Paper session G (CT-1) - chair: Jan van der Meij

24. The multimedia effect and its stability over time - Judith Schweppe*, Alexander Eitel and Ralf Rummer

25. The quest for improving drawing quality to enhance learning outcomes is not over yet -Katharina Scheiter* and Katrin Schleinschok

26. Watching corresponding gestures facilitates learning about movements from spoken texts: An fNIRS study - Birgit Brucker, Ann-Christine Ehlis, Florian B. Häußinger and Peter Gerjets*

10:00-10:30

Coffee (CB-hall)

10:30-11:50

Paper session H (CT-1) - chair: Marije van Amelsvoort

27. Does personalization promote learners´ attention? An eye-tracking analysis compares neutral and aversive stimuli - Steffi Zander*, Maria Reichelt, Stefanie Wetzel and Sven Bertel

28. Multimedia effect within a single presentation - Manuela Glaser* and Stephan Schwan

29. Scanning and deep processing of information in hypertext: A cued retrospective think aloud study - Ladislao Salmeron*, Johannes Naumann, Victoria García and Inmaculada Fajardo

30. The role of popularity cues in students’ selection of scholarly literature online - Ole Skov*, Ludovic Le Bigot, Nicolas Vibert, Guillaume de Pereyra and Jean-François Rouet

12:00-13:00

Keynote III: Mary Hegarty (CT-1)

13:00-13:05

Closing SIG meeting (Huib Tabbers)

13:05-14:00

Lunch (CB-hall)

14:00- ..

start SIG 6/7 (CT-1)

14:30-16:00

Guided tour on Architecture Rotterdam (city center)

 

Paper session A

1. Building bridges: Perspectives of researchers and teachers on visual representations in science education - Shaaron Ainsworth and Len Newton

Abstract: Recently, many have worried about a gap between priorities of researchers and teachers. We set out to reveal the nature and extent of such a gap around visual representation in science education. To uncover research priorities, we identified 401 papers published in 2010-2012 that addressed visual representation in science education, which were coded for research questions, representations, methods and domains. Six teachers were interviewed about their use of representations, priorities and views about research. Findings revealed many commonalities: both researchers and teachers considered visual representation to be important, they shared interests in specific representations (e.g. animation), research questions and methodology. Nevertheless, it showed teachers to have different rationales for using representations, to utilize some representations far more frequently than their coverage by research and to treat as default that teachers mediated representations. We hope that in identifying where priorities converge and diverge, we can help build bridges of across communities.

2. Amplifying multimedia effects by repeating the study-test cycle: Evidence for underconfidence with practice? - Alexander Eitel and Katharina Scheiter

Abstract: When being faced with multimedia instructional materials, learners are prone to show overconfidence (judged learning better than actual learning). Reducing this overconfidence by repeating the study-test cycle should therefore foster multimedia learning even more than learning with text only (i.e., leading to stronger multimedia effect). These assumptions were investigated by presenting students with instructional materials (text only vs. text with pictures) about the toilet flushing system and the corresponding tests (retention and transfer), followed by the same instructional materials and the same tests. Results showed that overconfidence due to multimedia was not reduced from the first to the second study-test cycle. Nevertheless, stronger multimedia effects were found in the second than in the first study-test cycle supporting the idea that retrieval practice can be particularly beneficial for learning with multimedia. Investigating the exact mechanism underlying this effect is an interesting and important avenue for further research.

3. Bar (n)or pie - Annemarie Quispel and Alfons Maes

Abstract: We tested the usability of a series of graphs, each visualizing the same combination of nominal and quantitative data in a different way. Results show that performance with graph types that use length to represent quantity (e.g. bar graphs) is better than with types that use area to represent quantity (e.g. pie graphs). Further, performance with types that use length is better when parts are portrayed separately than when the total is portrayed. The reverse is true for types that use area to represent quantity. Further, performance with less familiar graph types appeared to be better than with bars and pies.

 

Paper session B

4. A comparison of paper-based and video tutorials for software learning - Jan van der Meij* and Hans van der Meij

Abstract: Software training has long been dominated by paper-based tutorials. This position is now challenged by video tutorials. Learning theories point to advantages and disadvantages and empirical research has yielded equivocal outcomes as to whether a paper-based or video tutorial is better. The present study compares the effectiveness of four tutorial configurations: Paper preview & Paper procedure, Paper preview & Video procedure, Video preview & Paper procedure, and Video preview & Video procedure. 111 participants from fifth or sixth grade of elementary school received instructions about Word’s formatting options. Participants in the video conditions performed tasks significantly better during training than those in the Paper preview & Paper procedure condition. A similar outcome was found for the post-test. The success of the video tutorials is ascribed to the dedicated set of design guidelines that combined the advantages of paper with those of video.

5. Diagrams: Connecting text and image - Brad Jackel*

Abstract: This paper will begin by presenting an educational approach to a general diagrammatic literacy that is based on conceptualizing diagrams as a limited set of overlapping functional categories. However the focus of the paper is on the manner in which diagrams rely on context for their definition and the ramifications of the observation that this necessary context is, in practice, usually constructed of words.

6. Contents and graphics in line: When is it beneficial to schematize pictures in expository prose? - Hans Westerbeek*, Marije van Amelsvoort, Alfons Maes and Marc Swerts

Abstract: Learners generally benefit from representational pictures that are added to expository text. But what determines whether it is better to design such pictures as schematized drawings or as detailed photographs? In some studies learning outcomes are positively affected by schematized pictures, but in other studies null effects are reported. We argue that learners' ability to identify key concepts in pictures is an important predictor of effectiveness of representational pictures, and that this ability can be facilitated by using schematized pictures. We present results of a pilot study (N=36), which indicate that the aforementioned ability correlates with learning outcomes. We are planning to test our hypotheses in a full-scale experiment, and to present results of this experiment at the conference.

7. Decorative pictures in primary schools: Seductive or beneficial? - Maria Opfermann*, Annett Schmeck, Anne Wienand and Detlev Leutner

Abstract: This study investigated, whether decorative pictures that were added to a text about the circulatory system, can have a positive effect on learning and whether they are as effective as instructional pictures and other forms of instructional support. 136 fourth grade children learned with either pure text or with text plus instructional or decorative pictures. Additionally, half of the children received metacognitive instructions during learning. Performance was assessed with a posttest including textual and pictorial items. Results showed that decorative pictures led to higher performance and caused less cognitive load for the text items than instructional or no pictures, especially when they were presented together with metacognitive instructions. For the picture items, a different pattern emerged – in this case, learning with instructional pictures appeared to be most beneficial. These results are discussed with regard to possible motivating effects of decorative pictures especially for younger children.

 

Paper session C

8. Do versus think prompts: What is best for multimedia learning? - Irene T. Skuballa*, Marika Dobos, Panetta Sabina, Caroline Fortunski, Anke Dammert, Cornelia Hauser, Jasmin Leber and Alexander Renkl

Abstract: We tested different prompt types designed to deepen novice learners' processing of multimedia representations, particularly the integration of pictorial and textual information. Two prompt types were tested: one prompt type encouraged the students in behavioral activity by highlighting corresponding information in text and picture (do prompts, behavioral processing hypothesis); another prompt type encouraged students to mentally elaborate the same information (think prompts, mental processing hypothesis). A pilot study (N = 44) and a main study (N = 94) were conducted with 8th graders to test the prompt effects on learning performance. Findings generally favored the mental processing hypothesis and accordingly prompts that encouraged mental processing proved to be superior. The think prompts fostered prompt-congruent and prompt-incongruent information processing. Moreover, the superiority of think prompts was reflected on participants’ knowledge about pictorial information. However, prompts were time-consuming and reduced learning efficiency in general.

9. Does animation in public transport information improve decision making? - Jonathan Groff*, Jean-Michel Boucheix and Laurence Paire-Ficout

Abstract: Decision making is a task during which comprehenders have to choose a course of action among several alternative possibilities over a period of time (Chanquoy, Tricot, Sweller, 2007). In public transport, travelers have several choices to reach their destination. They select the optimum solution. So, travelers behavior could be regarded as a decision making process. In this paper we examined whether in public transport, animated graphic information instead of spoken information, could facilitate decision making under time pressure. Ninety-eight participants were asked to visualize a series of animated train traffic disruption messages in two conditions of time pressure (with or without time pressure). They were immersed in a virtual train station and they have to explain what they should do to take their train to Paris. Decision score and response time were recorded. Results showed a positive effect of time pressure on response time.

10. Designing context aware instructions: Perceptual salience and task demands in the selection of natural landmarks - Adriana Alexandra Baltaretu*, Emiel Krahmer and Alfons Maes

Abstract: The current study examines a small but important ingredient of adequate route directions (RDs): the selection of landmarks ("go left at the red building"). The starting point is a collection of RDs elicited in visual environments systematically differing in two characteristics which are easy to be detected by navigation software (path complexity and visual clutter). In this paper we focus on the set of landmarks produced by respondents in this experiment and analyze to what extent they can be predicted on the basis of their perceptual salience (as a function of size and pixel salience). We conclude that perceptual salience influences landmark choice only to a limited extent, other factors related to the specific navigation task play an important role, such as the location and stability of landmark objects.

11. Effects of building vs. studying a concept map on hypertext comprehension - Julie Lemarié*, Franck Amadieu, Ladislao Salmeron, Julien Cegarra, Pierre-Vincent Paubel and Aline Chevalier

Abstract: The study examined how prior domain knowledge and concept mapping in hypertext affected students’ comprehension and online processing. Either participants made a concept map in a first step and then read the hypertext’s contents combined with concept mapping (mapping-before condition), or they read the hypertext’s contents first and then made a concept map and re-read the hypertext’s contents (reading-before condition). A post-pretest gain of knowledge was assessed for micro- and macrostructural explicit information as well as for implicit relations between concepts (i.e., inference questions). It was hypothesized that high prior knowledge learners benefit from the mapping-before condition while low prior knowledge learners benefit from the reading-before condition. Actually, the results revealed that the inference questions scores were higher in the reading-before for low and high prior knowledge students. However, the examination of online processes (eye movments, navigation and mapping data) confirmed that prior knowledge supported better retional processing between concepts in the mapping-before condition.


Paper session D

12. Fostering piano learning by dynamic mapping of notes - Melina Klepsch*, Bastian Könings, Michael Weber and Tina Seufert

Abstract: Learning to play piano is demanding because one has to map music notations from a sheet onto piano keys. Moreover learners have to look back and forth between the sheet and the piano. Overall learning performance will be low and cognitive load should be increased. To assist piano learners with these demands P.I.A.N.O., a piano learning system with interactive projection of music notation was developed. P.I.A.N.O directly projects visual hints upon the instrument and allows mapping of notes to piano keys without any sight-reading skills. The hint indicates which key has to be pressed and for how long. In an experimental study we analysed whether learners show better performance and less cognitive load after training with P.I.A.N.O compared to Finale, a software system with a dynamic marker running through the depicted notes in sheet notation. The P.I.A.N.O system in fact turned out to be highly effective.

13. Grounding website navigation: The effects of visuospatial cues on search behaviour - Martin Galilee*, Neil Schwartz and Erica de Vries

Abstract: In this study, principles from grounded cognition were applied in order to improve website navigation. Spatial consistency and animation of movement were used to visualize navigation in hyperspace and to enhance mental simulations of movement. Visuospatial cues improved navigation efficiency and allowed participants to navigate more freely through network linking.

14. How do we integrate text and pictures? - Jana Arndt*, Anne Schüler and Katharina Scheiter

Abstract: Integration of text and pictures into a coherent mental model is an important cognitive process in multimedia learning. The aim of the present experiments was to shed more light onto the integration process by testing whether specific details from a verbal or pictorial representation are integrated into a more general, mental representation that had been likewise constructed from processing either text or pictures. Results of four studies underline the assumption that integration of specific with more general information takes place. Moreover, Experiment 2 showed that it does not matter if sentences or pictures are processed first, speaking in favor of a bidirectional integration process. As evidenced in Experiment 3, the integrated model seems to remain stable over time. Using more complex semantic material weakened the integration process according to Experiment 4. Further research should investigate if these results are transferable to more complex multimedia materials.

15. How graphs can mislead - Lisanne van Weelden*, Marije van Amelsvoort and Michael Bernhofer

Abstract: Based on Kintsch’(1994) research on the interaction between prior knowledge and text coherence for learning from text, we investigate whether the same effect is present when learning from visual instruction. In our first experiment, participants had to tie a tie following visual instructions with low or high coherence. Background knowledge was manipulated by giving half of the participants two practice knots (the others did not practice and therefore had no background knowledge). Results show that the high background knowledge group conducted the task faster than the low background knowledge group, but there was no interaction between background knowledge and coherence on performance. In the round table session, we would like to discuss merits of our approach and possible future studies.


Paper session E

16. How to get better in performing mathematical tasks: A teacher-based training program - Michael Kraus*, Silja-Susann Taxis, Marion Geiger, Markus Vogel and Tina Seufert

Abstract: In maths students have to deal with different representations like terms, graphs or tables. They have to transform representations into another one or combine information of different given representations. But students have problems learning with multiple mathematical representations. Therefore we investigated the effects of a teacher based training in fostering transformation processes while dealing with multiple mathematical representations. 92 students of grade 8 and 9 were trained by their maths teachers in class in transformation processes and compared to 129 untrained students. Directly after the training, we found positive training effects for reception, production and integration, but no positive effect for translation. Positive long term effects of the training were found for all four transformation processes. Teachers trained their students effectively in processing multiple representations. In particular, students might need more opportunities to practice translation strategies in class.

17. ICT-related component skills and navigation in reading digital text - Carolin Hahnel*, Frank Goldhammer, Johannes Naumann and Ulf Kröhne

Abstract: The present study aims to investigate the relations between the ability to comprehend digital texts and both basic computer skills as well as skills in evaluating online sources as ICT-related components of 15-year-old students. We hypothesized that ICT skills explain performance in digital reading, additionally to reading comprehension of printed text, and that these linkages are mediated by students’ navigation behavior in digital environments. First results support the impact of ICT-related skills in digital reading suggesting that readers of digital texts also need basic experiences in dealing with computer interfaces and skills for anticipating the content of further hypertext nodes to comprehend hypertexts.

18. Learning through visual illustration and retrieval practice - Ann-Sofie Jägerskog*, Fredrik Jönsson, Staffan Selander and Bert Jonsson

Abstract: Previous research has shown that studying with (vs. without) visual illustrations as well as taking tests (vs. restudying) is beneficial for learning. Both are well-known learning strategies, but they have not previously been investigated in combination and rarely in the classroom. In this study, 133 upper secondary students were given a lecture presented only verbally or with the aid of a visual illustration. The students processed the information again either by retrieval practice or by restudying it. Recall and transfer tests were conducted after some few minutes, after a week and after 10 weeks. Visuoverbal presentation resulted in better learning than verbal presentation only. Although a modest testing effect was found, this effect was considerably weaker than the multimedia effect. Retrieval practice did not improve the participants’ memory performance beyond the beneficial effect of visuoverbal learning. Presentation format proved to be a more important factor for learning than study strategy.

19. Mind the gap: Multimedia instruction, pain, and cognition - Alexander Smith and Paul Ayres*

Abstract: Multimedia instruction, persistent pain, and cognitive performance were examined with 59 teachers from 2 schools in New South Wales (Australia). Participants were assessed according to their clinical pain levels on standardized tests, and then categorized as either pain or pain-free individuals. They were then randomly assigned to an instructional format (narration + visuals vs. narration). Participants received a 4-minute, system-paced instruction on lightning formation twice. Retention, transfer, and matching tests were completed under timed conditions. There were significant main effects for test score and test time, confirming the benefit of integrating visuals with narrated text (the multimedia principle). As predicted, pain participants performed significantly lower than pain-free participants on retention and transfer tests (the pain effect), even though they demonstrated clinically low levels of pain. Instructional design and persistent pain moderated the level to which new cause-and-effect information was successfully retained and applied.


Paper session F

20. Reducing transience of complex animation does not (always) lead to better learning performances - Jean-Michel Boucheix* and Claire Forestier

Abstract: When large amount of information is presented in long section animation, subsequent transient information effect has been shown to withdraw, or weaken, the superiority of dynamic visualizations presentation over static graphics in learning a hand-movement origami task (Wong, Leahy, Marcus & Sweller, 2012). In the experiment of these authors, children were allocated to one of four presentation conditions: long section animation, short section animation, long section static graphics and short section static graphics. In this paper, children learnt to tie complex nautical knots from hand movement’s video presentation, in a viewing and practicing task. In experiment 1, the study by Wong & al. was replicated with the conventional sequential presentation of the knots, in the four conditions. In experiment 2, transience was reduced using simultaneous presentations. Results showed long section animation did not lose their superiority over static graphics. Explanations in terms of inhibition processes and task affordance are suggested.

21. Success in solving mathematical word problems: A question of representational flexibility? - Katharina Hohn* and Wolfgang Schnotz

Abstract: Students are expected to think flexibly and adaptively when dealing with mathematical problems. Nevertheless, research has paid little attention to this topic yet. The present study examined students’ representational flexibility and task-based adaptivity when dealing with mathematical word problems and its association with successful solutions. Therefore, 122 students of different grade level solved four mathematical word problems independently. Their solution procedures were video-recorded and analyzed focusing on students’ self-generated representations and success in solution. Indicators for representational flexibility (variation) and task-based adaptivity were generated. Results revealed that there was no association between the mere variation of representations (flexibility) and success in solving mathematical word problems, but between task-based adaptivity and success. Thus, representational adaptivity in dependence of task characteristics seems to be an important tool for students. Anyway, more research is necessary to deepen the knowledge gained from the present study.

22. The effect of spatial text information and spatial ability on learning to tie knots with 2D- and 3D-visualizations - Tim Kühl*, Ulrike Kretschmar and Stefan Münzer

Abstract: It was investigated how spatial text information and spatial ability influenced learning with two-dimensional and three-dimensional visualizations. A 2x2-design with visualization format (2D vs. 3D) and spatial text information (high vs. low) as independent variables, and spatial ability as a continuous variable was chosen for the domain “tying knots”. The performance in tying knots, two picture recognition tests and two cloze tests served as dependent variables. Results revealed that learners performed better in tying knots when they learned with 3D-visualizations and also slightly benefited from spatial text information. For both picture recognition tests, an interaction between spatial ability and visualization format was observable: Only learners with higher spatial ability profited from 3D-visualizations. For a cloze test that asked for textual information that was given in all conditions, a redundancy effect occurred: Learners receiving 3D-visualizations performed worse when spatial information was given in the text, particularly those with lower spatial abilities.

23. The role of visuo-spatial abilities in learning three-dimensional anatomical structures from animation - Sandra Berney* and Mireille Bétrancourt

Abstract: The study investigated how displaying orientation references (internal axes, external character or none) in learning of two anatomical three-dimensional structures from animation affect performance on subsequent spatial judgment tasks. Learners’ visuo-spatial abilities in mental rotation and perspective-taking were also measured. The results showed no main effect of the orientation references on the accuracy of spatial judgments accuracy. However, participants who studied the material with the presence of orientation references, either internal or external, performed the spatial judgment tasks more quickly than did their control counterparts. There was no significant difference in the spatial ability measures for any of the orientation reference conditions. Nevertheless, visuo-spatial abilities affected the accuracy performance differently for the two tasks, suggesting different processing strategies.


Paper session G

24. The multimedia effect and its stability over time - Judith Schweppe*, Alexander Eitel and Ralf Rummer

Abstract: A central finding in learning with text and graphics is a benefit from learning with combined representations as compared to text only – the multimedia effect. Two experiments test for whether this advantage is restricted to short-term learning or whether it remains stable when learning is tested after a delay. Subjects learned about a pulley system at their own pace, either with only a written text or with text and a picture of the described pulley system presented simultaneously. We varied whether comprehension and retention were tested immediately after the learning phase or after a delay (1 week later in Experiment 1; 1 week or 2 weeks later in Experiment 2). In both experiments there was a multimedia effect that was independent of the time of testing. These findings suggest that the lack of a pictorial display hindered comprehension of the pulley system’s structure substantially and persistently.

25. The quest for improving drawing quality to enhance learning outcomes is not over yet -Katharina Scheiter* and Katrin Schleinschok

Abstract: The present study investigated whether the quality of drawings that learners generate while reading instructional texts can be improved by asking them to generate drawings on a tablet compared to paper. Ninety-three students were randomly assigned to a text-only control group, a paper-drawing group or two tablet-drawing groups, where students draw using either their finger or a stylus. All groups received an instructional text on the particle model of matters. The experimental groups were asked to generate drawings while reading. Subsequently, their learning outcomes were assessed. Unexpectedly, students in the two tablet groups produced lower-quality drawings and performed worse than either the control or the paper-drawing group. Results can be explained with the lack of familiarity students possessed in working with apps as well as with the didactical limitations of the app used to generate the drawings.

26. Watching corresponding gestures facilitates learning about movements from spoken texts: An fNIRS study - Birgit Brucker, Ann-Christine Ehlis, Florian B. Häußinger and Peter Gerjets*

Abstract: The present study examined the role of social-web specific epistemic beliefs in N = 124 vocational students' sourcing strategies with respect to social web contents. Results indicated that the stronger students' beliefs that knowledge claims encountered on the social web need to be checked against other information sources, reason, and prior knowledge, the more critically they evaluated the quality of a blog article and the blog author's credibility, and the more incomplete source references they identified. Furthermore, the stronger students' beliefs that the social web is a reliable knowledge resource that contains correct and detailed information, the more strategies they reported to find experts on a topic by using the social web.


Paper session H

27. Does personalization promote learners´ attention? An eye-tracking analysis compares neutral and aversive stimuli - Steffi Zander*, Maria Reichelt, Stefanie Wetzel and Sven Bertel

Abstract: People learn better from multimedia presentations when words are in conversational style rather than formal style (Mayer, 2009). Empirical results confirmed the personalization principle; however, possible moderator variables and underlying processes why personalization effects occur in multimedia learning are not clear. A second question was, whether the underlying processes remain the same when the personalization is realized in learning materials of emotionally aversive content. To gain further insights into the underlying processes, we conducted an eye-tracking experiment with two distinct learning materials (neutral vs. aversive). The participants received either a personalized or formal version of a computer-based material explaining weather phenomena or injuries of the brain. Results revealed significantly more fixations on main picture AOIs in the personalized version for the neutral material, whereas the results for the aversive material showed the opposite effect with significantly more fixations overall, longer average fixation durations on pictures and more transitions between pictures and text.

28. Multimedia effect within a single presentation - Manuela Glaser* and Stephan Schwan

Abstract: The present study examines in two experiments the knowledge acquisition with an archaeological reconstruction of a church. It compares knowledge acquisition of different elements of the reconstruction: elements presented only in the picture and elements presented audio-visually. Knowledge acquisition is measured by a free recall test, a multiple choice test, and a visual recognition test. The results of Experiment 1 show for all types of measurement that elements which are named in the audio text are remembered better than unnamed elements. Visual realism of the reconstruction has no significant effects on knowledge acquisition. Experiment 2 shows that the multimedia effect examined in Experiment 1 is due to remembering the named elements, and not due to the disadvantage of less remembering the unnamed elements.

29. Scanning and deep processing of information in hypertext: A cued retrospective think aloud study - Ladislao Salmeron*, Johannes Naumann, Victoria García and Inmaculada Fajardo

Abstract: A critical issue of information problem solving on the Internet is to find a trade-off between quickly scanning large sections of information in web pages and deeply processing those that are relevant for the task. We studied how high school students articulate scanning and deeper processing of information while answering questions using a Wikipedia document. By analyzing retrospective think aloud protocols and eye-tracking measures, we found that scanning of information was related to poor hypertext comprehension, while deep processing of information was related to better performance. This relationship between scanning, deep processing, and performance was qualified by reading comprehension skills in an unexpected way: scanning was linked to lower performance especially for good comprehenders, while the positive effect of deep processing was independent of reading skills. Finally, we discuss how these results relate to our current knowledge of Internet problem solving.

30. The role of popularity cues in students’ selection of scholarly literature online - Ole Skov*, Ludovic Le Bigot, Nicolas Vibert, Guillaume de Pereyra and Jean-François Rouet

Abstract: In two experiments, we asked undergraduate students to make binary choices of bibliographical references in a simulated Google Scholar environment. We manipulated the number of citations attributed to the references, and expected a higher rate of selection for highly cited items. In addition, we tested if self-reported topic knowledge and knowledge of what the number of citations means mediated the effect of popularity cues. Experiment 1 (n=161) found a strong main effect of popularity cues on selection; however, none of the other factors mediated the effect. Experiment 2 (n=11) further revealed that preference for highly cited referenced was related to a deeper processing of citation figures, perhaps triggered by a strong contrast across figures.


Parallel 1 - Animations & drawing

1. When may drawing help learning from animations? - Richard Lowe* and Jean-Michel Boucheix

Abstract: This theoretical presentation is part of a proposed round table on the utility of self-generated drawing for supporting more effective animation-based learning. The belief that animations are yet to fulfill their full potential as aids to learning is not uncommon amongst educational researchers. Empirical testing of a varied range of interventions designed to improve animation-based learning (mostly involving animation-text combinations) has produced mixed results. Could self-generated drawing, an approach found to improve learning from text, also lead to substantial and generalizable improvements in learning from animation in its own right? Types of processing activities likely to be involved in self generation of drawings from animations will be considered with a particular emphasis on how they may be able to enhance or impede learning.

2. Does sketching improve learning from an animation of a technical system? - Benjamin Fillisch*, Rolf Ploetzner, Judith Aurand, Onno Bahns and Fabian Moser

Abstract: This presentation is part of a proposed round table on the benefits of self-generated drawings for learning from animation. We describe an experimental study that was conducted to expand existing research on the educational effectiveness of drawing instructions while learning from animation. In the study, 39 students watched an animation of a car engine four consecutive times. Between two successive observations of the animation, approximately half of the students were instructed to reflect on the observed structures and processes whereas the other students were instructed to produce sketches of these characteristics. Before learning from the animation took place, both groups were shown an example of how the instructed learning technique could be applied. Subsequently, both groups had to write a summary of how the car engine works. On average, the students who produced sketches did not learn significantly more about the car engine than did the students who engaged in reflecting.

3. A process-oriented methodology for investigating drawing for learning from an animation - Jean-Michel Boucheix*, Richard.K. Lowe, Stéphane Argon, Zheng Gin and Jonathan Groff

Abstract: This methodological presentation is part of a proposed round table on the utility of self-generated drawing for supporting more effective animation-based learning. It focuses on a new on-line method for investigating the perceptual and cognitive processing involved in self-generated drawing when learning from an animation. The method involves synchronized recording of eye movements and drawing activity executed on a large screen graphics tablet. We identify the present lack of definitive process-related information as a major issue in this research field to justify the need for the methodology we have developed. The system used to collect coordinated data on visual and drawing activity is described and example results from its piloting are presented.


Parallel 2 - Graphical representations

4. Coherence and prior knowledge in visualizations - Marije van Amelsvoort*, Lisanne van Weelden, Nadine de Jong and Rick Thijssen

Abstract: Based on Kintsch’(1994) research on the interaction between prior knowledge and text coherence for learning from text, we investigate whether the same effect is present when learning from visual instruction. In our first experiment, participants had to tie a tie following visual instructions with low or high coherence. Background knowledge was manipulated by giving half of the participants two practice knots (the others did not practice and therefore had no background knowledge). Results show that the high background knowledge group conducted the task faster than the low background knowledge group, but there was no interaction between background knowledge and coherence on performance. In the round table session, we would like to discuss merits of our approach and possible future studies.

5. Bridging research to practice: Towards the assessment of representational competence in the classroom - Sandra Nitz* and Christine Tippett

Abstract: Interpreting, constructing, transforming, and evaluating scientific representations are crucial skills for students and an integral part of scientific literacy. Collectively, these skills contribute to representational competence (RC), a topic of scholarly interest within the research community in recent years. With regard to evaluating learning arrangements and diagnosing students’ skills systematically the question of how to assess students’ skills and understanding in the classroom is a crucial point. Drawing on our past research on measuring RC, we aim to develop a systematic approach to assessing students’ RC that is informed by research findings but also practicable for teachers in classroom contexts. To standardize our approach, we follow a model of test development as proposed by Terzer et al. (2013). We will create a RC measure systematically according to this approach and looking forward to discuss its items.

6. Representation of phenomena and models in physics - Steffen Wagner*, Franz Boczianowski and Burkhard Priemer

Abstract: In physics, a phenomenon is often presented in different pictorial representations. But students often struggle to understand, how these different representations are interconnected. To analyse these difficulties a questionnaire with different representations of the phenomenon of refraction of light was given to 1210 students. In addition, interviews were conducted in case studies.The results show that the concept of refraction can be pictorial represented in (i.e.) four different ways. Students were unable to see conncetions between these representations whilst experts could.. We assume that this happens because bridges that span the differences between different representations of one physical content are often not discussed in physics lessons. The talk will present these studies in detail and will discuss their general relevance.


Parallel 3 - Text comprehension and epistemic cognition

7. Epistemic cognition and reading comprehension - Panayiota Kendeou*, Kalypso Iordanou and Krista Muis

Abstract: In the present study we investigate graduate students’ cognitive processing during reading of a single document that includes conflicting accounts for a historical event and explore the extent to which these processes relate to students’ epistemic cognition in History. To explore the relations between epistemic cognition and reading comprehension processes, we used the think-aloud methodology because it allow for the consideration of the content of the actual students engage in. We also assessed text comprehension using a sentence verification task. Students’ epistemic cognition was assessed using the the Justification for Knowing Questionnaire (JFK-Q). Finally a control group also read the same text without performing a think-aloud. The findings showed that the three different forms of knowledge justification (justification by multiple sources; authority; personal opinion) can have a differential effect on student reading comprehension processes as they unfold during moment-by-moment reading.

8. Knowing when to defer: How text comprehensibility and beliefs about epistemic complexity affect laypeople’s evaluation of science-based claims - Marc Stadtler*, Lisa Scharrer and Rainer Bromme

Abstract: When laypeople read science-based texts to decide about the validity of related knowledge claims, they have been shown to more readily rely on their own evaluations rather than consulting an expert when the text information is easy compared to difficult to comprehend. The present study examined whether this comprehensibility effect is moderated by readers’ beliefs about the epistemic complexity of the topic. The results confirmed that laypeople’s readiness to decide based on their current knowledge increases to a lesser extent in response to reading comprehensible texts if they belief the topic knowledge to be highly complex.
 

9. Epistemological understanding and meta-level processing of evidence when reading a science text - Kalypso Iordanou*, Panayiota Kendeou and Krista Muis

Abstract: The relation between epistemological understanding and meta-level processing of evidence when reading a text was examined in a group of young adolescents and adults. Thirty-eight young adolescents and 25 graduate university students engaged in individual interviews where they read a science text, using a think aloud protocol. Participants also completed a prior-knowledge test and an instrument to assess their epistemological understanding. Results showed that participants who exhibited evaluativist epistemological understanding engaged more in high meta-level processing of evidence regarding the function of evidence in the context of an argument, compared to participants who held non-evaluativists epistemological understanding. Furthermore, a developmental difference was observed, with adult evaluativists engaging more in high meta-level processing of evidence than young adolescent evaluativists. These results have important theoretical and educational implications.

10. Epistemic beliefs and emotions predict the source of information recalled from multiple conflicting texts - Gregory Trevors*, Krista Muis, Reinhard Pekrun, Gale Sinatra and Eric Poitras

Abstract: The current study examined the effects of epistemic beliefs and epistemic emotions on written summaries of multiple conflicting documents about a controversial topic. University students reported their emotions after reading three conflicting texts on climate change and their epistemic beliefs about the topic. Written summaries were analyzed with text mining software for semantic similarity between individual texts. Results showed that emotions and beliefs differentially predicted subsequent recall of information from sources.

11.  Students’ epistemic beliefs and text comprehension: What we believe we know and what we believe we need to know - Ivar Bråten*, Helge I. Strømsø and Leila E. Ferguson

Abstract: This paper reviews theory and research on the role of student epistemic beliefs in text comprehension and discusses implications for educational research and practice. Based on current conceptualizations of epistemic beliefs as well as conceptualizations of the comprehension of single and multiple texts that address the role of epistemic beliefs in text comprehension, we review research linking individual differences in epistemic beliefs to the comprehension of single and multiple texts, including web-based texts. Finally, we discuss the implications of this body of work for future research and instructional practice.


Parallel 4 - A large-scale reading comprehension intervention

12. A large-scale reading comprehension intervention study aimed at the formation of an embodied situation model: Effects of the mental simulation training - Bjorn de Koning*, Lisanne Bos, Stephanie Wassenburg and Menno van der Schoot

Abstract: In this study we developed and tested a computer-based mental simulation training. The training was part of a large-scale reading comprehension intervention aimed at improving children’s ability to form an embodied situation model of text. The mental simulation training focused on stimulating (re)enactment of perceptual- and motor experiences. Results showed that the training increased reading motivation, but did not result in more effective use of trained mental simulation skills or higher general reading comprehension levels.

13. A large-scale reading comprehension intervention study aimed at the formation of an embodied situation model: Effects of the inference making training - Lisanne Bos*, Bjorn de Koning, Stephanie Wassenburg and Menno van der Schoot

Abstract: The study’s goal was to develop and test a computer-based inference making training. This training was part of a large-scale reading comprehension intervention aimed at improving children’s ability to form an embodied situation model of a text. The focus of the inference making training was on building coherent situation models. Results showed that the training resulted in more effective use of the trained inference making skills, increased general reading comprehension levels, and higher reading motivation.

14. A large-scale reading comprehension intervention study aimed at the formation of an embodied situation model: Effects of the monitoring training - Menno van der Schoot*, Lisanne Bos, Stephanie Wassenburg and Bjorn de Koning

Abstract: The goal of this study was to develop and test a computer-based comprehension monitoring training. This training was part of a large-scale reading comprehension intervention aimed at improving children’s ability to form an embodied situation model of a text. The focus of the monitoring training was on situation model updating. Results showed that the training resulted in more effective use of the trained monitoring skills, increased general reading comprehension levels, and higher reading motivation.


Poster session - Theme 1: Text

1. Type of anaphor effect’s in readers with Intellectual Disability: What eye movements say - Fajardo Bravo*, Vicenta Ávila, Laura Gil, Antonio Ferrer, Ladislao Salmeron, Marcos Gómez and Carlos Gelormini

Abstract: Syntactic simplification is a common solution used by text simplification tools which usually implies the conversion of a long sentence into two short ones. This technique involves the addition of an anaphor which paradoxically could hinder the reading process. The Discourse Prominence Theory (Gordon & Hendrick, 1998) predicts an interference of the repeated name anaphor compared to the pronoun anaphor in good readers when the anaphor refers back to a prominence element of the antecedent sentence. In this eye-tracking study, we tested the effect of anaphor form (repeated, pronoun and null anaphor) in Spanish for poor readers with intellectual disability. The results showed that readers made more regressions to antecedent sentences when texts contained null versus pronoun or repeated name anaphora. The duration of the first pass was shorter for sentences containing repeated name anaphor. The results are discussed in relation to the Explicitness principle (Gernsbacher, 1990).

2. Prior knowledge and coherence processing in concept mapping from hypertext - Franck Amadieu, Ladislao Salmerón, Julien Cegarra, Pierre-Vincent Paubel, Julie Lemarié* and Aline Chevalier

Abstract: The study examined how prior domain knowledge and concept mapping in hypertext affected students’ comprehension and online processing. Either participants made a concept map in a first step and then read the hypertext’s contents combined with concept mapping (mapping-before condition), or they read the hypertext’s contents first and then made a concept map and re-read the hypertext’s contents (reading-before condition). A post-pretest gain of knowledge was assessed for micro- and macrostructural explicit information as well as for implicit relations between concepts (i.e., inference questions). It was hypothesized that high prior knowledge learners benefit from the mapping-before condition while low prior knowledge learners benefit from the reading-before condition. Actually, the results revealed that the inference questions scores were higher in the reading-before for low and high prior knowledge students. However, the examination of online processes (eye movments, navigation and mapping data) confirmed that prior knowledge supported better retional processing between concepts in the mapping-before condition.

3. Students’ trust in conclusions from socio-scientific texts - Helge I. Strømsø* and Ivar Bråten

Abstract: We examined whether biased processing and use of critical reading strategies might predict students’ trust in conclusions from socio-scientific texts about potential risk topics. Results showed that an interaction between students’ prior beliefs and the nature of the texts’ conclusions (risk vs. no risk) predicted their trust in the texts’ messages, and that students’ self-reported use of critical reading strategies were related to their trust scores. Results indicated that students’ judgments of conclusions from socio-scientific texts on risk related topics were associated with both a “confirmation bias” and a “negativity bias”, and that the role of the “confirmation bias” varied somewhat according to the texts’ topics.

4. Readability formulae: Formulae of prior knowledge? - Jaan Mikk*

Abstract: Readability indexes of text are calculated depending on sentence length and word length which, in turn, is related to word familiarity. Therefore the familiarity of words and text content is one aspect of readability. The aim of the study was to investigate the relationship between prior knowledge of text content and the text characteristics. Questions were composed about the content of 48 texts. 124 students answered the questions without reading the texts. The percentage of correct answers correlated with word length, sentence length and other characteristics of the texts. Multiple correlation of the elaborated “prior knowledge formula” was about the same level as the correlation in some readability formulae.

5. How to improve text processing? Disfluency-effect and learning strategies - Katharina Schiller*, Martina Koch and Tina Seufert

Abstract: The effect that poorly readable learning material leads to better performance within tasks requiring keeping information in mind is called disfluency-effect. Additional there is the assumption, that the use of learning strategies can improve successful learning. To test the disfluency-effect, the effect of learning strategies and their combination on learning performance and cognitive load, a 2 x 2 - design was set up. The experimental study consisted of three parts requesting increasing levels of processing (recall, comprehension, transfer). The disfluency-effect was not evident for recall tasks and other performance levels but learning strategies turned out to be helpful for comprehension and transfer tasks. We also found an interaction effect for comprehension tasks: learners working with disfluent learning material and learning strategies showed a lower test performance than those working with fluent material and learning strategies. A higher germane cognitive load was discovered for learners using learning strategies in recall tasks.

6. Reading in German versus reading in French: Are there different attributes that determine the difficulty of reading comprehension items in both languages? - Monique Reichert*, Philipp Sonnleitner and Romain Martin

Abstract: The current study aims at identifying those cognitive and linguistic attributes that best describe and explain reading test performance in two languages, and of individuals with different language backgrounds. German and French language teachers from Luxembourg secondary schools were asked to rate a number of either German or French reading tasks with regard to a list of cognitive and linguistic attributes. The teachers’ item attribute ratings were then linked to the empirical data collected in a large-scale study among Luxembourg 9th graders. Based on the initial item-attribute assignments, ideal item-response patterns could be presumed and compared to real examinees’ response patterns by using a linear logistic test modeling approach. The results from the different modeling steps show (a) whether the theoretical assumptions underlying the difficulty in reading comprehension items hold, (b) whether they hold for both German and French, and (c) whether they hold for participants with different language backgrounds.

7. Strategy training facilitates children’s hypertext comprehension - Sabine Klois*, Eliane Segers, Linda de Leeuw and Ludo Verhoeven

Abstract: Primary school children have to read hypertext for comprehension. At school, children learn reading strategies for linear text, which do not automatically transfer to hypertext reading. In the present study we examined how strategy instruction combined with a mind-mapping task prompts children’s learning outcomes at three levels: (a) strategy use, (b) reading comprehension, (c) knowledge representation. Sixth-graders participated in a pre-/post-test design, 55 in the training group and 29 in the control group. Results showed that positive effects of the intervention for strategy use and reading comprehension. Children in both groups formed more sequential than expert-like knowledge representations. It can be concluded that strategy instruction tailored to hypertext is a first step in helping children to better comprehend hypertexts. A more elaborate training is probably necessary to further aid deeper text comprehension.

8. Judging the plausibility of argumentative statements in scientific texts: An expert-novice comparison - Sarah von der Muehlen*, Tobias Richter, Kirsten Berthold, Elisabeth-Marie Schmidt, Katherine Bruns and Sebastian Schmid

Abstract: Readers spontaneously validate presented information against their knowledge and beliefs but differ in their ability to strategically evaluate the soundness of arguments. This study compared the performance of university students and scientists in the field of psychology regarding plausibility ratings of statements and identification of argumentation fallacies. Results indicate that students base plausibility judgments on intuition and opinion rather than internal consistency of arguments. Moreover, they have deficits in detecting and classifying argumentation fallacies.

9. Multiple text comprehension and belief change: The influence of text characteristics and prior beliefs - Sylvia Maria Savvidou* and Irene-Anna Diakidoy

Abstract: In the present study, we explored the role of text characteristics and prior beliefs about vegan nutrition in memory and situation model of multiple texts, judgment of texts' trustworthiness, multiple text comprehension, and belief change. Text characteristics included different argument (concrete-emotional vs rational-objective) and claim (against vs in favor) types. After readings, participants changed their weak positive position to a more negative one. Findings showed that both argument and claim types influenced memory and situation model, judgment of texts' trustworthiness and belief change. The role of prior beliefs was only significant in belief change. Single text comprehension outcomes and prior beliefs did not significantly predict multiple text comprehension.

10. The role of social-web specific epistemic beliefs in sourcing strategies on the social web - Yvonne Kammerer*, Clara Oloff and Peter Gerjets

Abstract: The present study examined the role of social-web specific epistemic beliefs in N = 124 vocational students' sourcing strategies with respect to social web contents. Results indicated that the stronger students' beliefs that knowledge claims encountered on the social web need to be checked against other information sources, reason, and prior knowledge, the more critically they evaluated the quality of a blog article and the blog author's credibility, and the more incomplete source references they identified. Furthermore, the stronger students' beliefs that the social web is a reliable knowledge resource that contains correct and detailed information, the more strategies they reported to find experts on a topic by using the social web.


Poster session - Theme 2: Graphics

11. Constructing graphs in biology class: Does teacher support influence students’ self-efficacy and motivation? - Carolin Enzingmüller* and Helmut Prechtl

Abstract: There is a growing awareness that graphing is an essential part of the science curriculum. However, research indicates that students have significant difficulties in interpreting as well as constructing graphs and other inscriptions in science class. It has been argued that teacher support is crucial to students’ appropriate use and construction of inscriptions in scientific processes. This paper reports an analysis of the relations between teacher support and student outcomes in the area of graph construction. Teachers’ instructional behaviors and student outcomes were assessed by means of student ratings. The results of a multilevel regression analysis performed on perception data of 1266 secondary education students suggested that teacher support might be a strong predictor of students’ self-efficacy and motivation regarding graph construction.

12. A comparative content analysis: Visual representations of photosynthesis in biology textbooks - Christine Tippett* and Sandra Nitz

Abstract: This poster details a content analysis of high school biology textbooks, specifically the form and function of the visual representations that are used with the topic of photosynthesis. Textbooks from Canada and Germany are analyzed, both to provide a more diverse dataset and to allow an initial comparison of publisher/author choices in the two countries. To validate the inferences arising from the content analysis, results of a quantitative study examining students’ representational competence in photosynthesis are analyzed for patterns in student responses that relate to form and function of visual representations. Merging the content analysis results with the results of the quantitative study of representational competence allows us to make some statements about the match (or mismatch) between the types of visual representations used in textbooks and students’ ability to work with those representations.

13. Graphicacy on primary textbooks: Analysis of activities with images - Asuncion Lopez-Manjon, Yolanda Postigo and Cristina Marin*

Abstract: The aim of this research is to study how primary school textbooks promote graphicacy. We analyse the activities included with images of the human body in 3rd to 6th year textbooks from three publishers. The categories of analysis are related to the type of image, type of task, procedural content, cognitive processing level required and whether there are explicit proposals for teaching the images. The results show that illustrations and visual diagrams are the most frequent images. The main tasks involve image´ interpretation, while production tasks are unusual. There is not a balance among the different procedural contents. The processing level required is superficial which result in a learning process based on copy and repetition. We conclude that the activities in the textbooks foster a kind of graphicacy that is insufficient for the acquisition of the complex use of images.

14. Understanding of graphs in social science undergraduate students: Choosing the best graph - Cristina Marín*, Yolanda Postigo and Mª Puy Pérez-Echeverría

Abstract: This study analyzes which factors influence on interpretation of graphs of two groups of Social Science students. They had to choose the most suitable graph for the presentation of results from different research studies. Their achievement were analyzed according to their level of statistical training (different for each group), type of content (psychological-non psychological) and complexity of the problems (four levels of difficulty). Results showed no significant differences regarding the type of content for both student groups. Differences were identified regarding the difficulty of the problems and the kinds of misinterpretations of graphs. The students were capable of relating textual information to the adequate graph, but they also reveal significant shortcomings in their understanding of important syntactic aspects of graphs, like the nature of the variables represented and the relationship among them.

15. Climbing into the tree of life: An eye-tracking study of experts and novices interpreting phylogenetic trees - Inga Ubben*, Sandra Nitz and Annette Upmeier Zu Belzen

Abstract: Representational competence plays an important role in handling phylogenetic trees, which model evolutionary relatedness of organisms. These visual representations are used both in systematic biology to study phylogeny and in biology education to promote thinking in an evolutionary context. Nevertheless, students’ representational competence regarding phylogenetic trees usually remains on a low level. The present study aims at revealing patterns of scientists’ (experts) and students’ (novices) visual perception and reasoning by investigating how they analyze and compare phylogenetic trees. Therefore, eye-tracking technology is used in combination with verbal data and questionnaires. These insights into cognitive processes of experts and novices during reading and comparing phylogenetic tree representations will further serve as starting point for supporting students’ representational competence in biology education.

16. Reading pictures: The effect of pictures in narratives on primary school children’s reading comprehension and enjoyment - Inouk Boerma*, Suzanne Mol and Jelle Jolles

Abstract: In this study we examined the effect of pictures replacing parts of the text in a narrative on primary school children’s reading comprehension and reading enjoyment. Results showed that reading this experimental narrative was more difficult and less enjoyable than the text-only narrative, indicating that interpreting pictures adequately is an ability that children do not automatically possess. Furthermore, we found that children with high imagery skills comprehended the narrative better and enjoyed the story more than children with lower imagery skills. Within this ‘low-imagery group’ the text-only narrative appeared to be easier to comprehend than the experimental picture version. Within the ‘high-imagery group’, no differences were found between our three versions of the narrative. Finding effective ways to instruct children how to ‘read’ pictures might improve their reading comprehension and enjoyment. In addition, visualization trainings might help children improve their imagery skills, which can help them to understand texts better.

17. Does picture orientation constrain spatial situation models constructed from text? - Jan Engelen*, Samantha Bouwmeester, Anique de Bruin and Rolf Zwaan

Abstract: Does seeing a pictorial illustration prior to reading influence how readers mentally represent aspects of a situation other than those that were depicted? In four experiments, we investigate whether the orientation of a contextual picture of a character and the direction described in a sentence influence where readers focus their attention in the visual field. Contrary to our predictions, we found no evidence that response times in a picture verification task with objects presented left or right on a computer screen were affected by the interplay between visual and linguistic cues. In a meta-analysis, the effect of this interaction was close to zero. However, responses were reliably faster after sentences describing movement of the character toward an object, indicating that comprehenders may have adopted a character-internal perspective. These results suggest that if readers construct perceptual simulations of spatial relations, the perspective from which they do so is not affected by recent visual experiences.

18. Playing with cards: Constructing graphical genres - Muhammad Ashraf* and Erica De Vries

Abstract: This study investigates students’ classification of graphical representations by using card sorting. Maps and illustrations were two categories for which student and a priori groupings coincided.

19. Identifying dimensions of socio-technical tools for innovative design and engineering education - Olga Kasatkina*, Erica de Vries, Cedric Masclet and Jean-François Boujut

Abstract: In this paper, we identify a list of dimensions which form an analysis grid of socio-technical visualisation tools. The dimensions distinguish characteristics of 1) the design situation, 2) representational types, and 3) tool purpose. We elaborate on application of the analysis grid in engineering education.

20. Teachers' understanding and interpretation of progress-monitoring graphs - Roxette van den Bosch*, Siuman Chung and Christine Espin

Abstract: When teachers use graphed CBM progress-monitoring data for instructional decision-making their students’ performance improve, but teachers often do not use this type of data. As graph interpretation can be complex, teachers’ non-use of the data might be explained by graph-interpretation difficulties. In this study CBM progress-monitoring graphs will be presented to 20 teachers and they are asked to describe and interpret these graphs out loud, while their eye movements are tracked. Teachers’ general graph-reading ability will be taken into account as a possible mediating factor. Insight into teachers’ understanding and interpretation of graphs, as assessed with eye-tracking and think-aloud methodologies, may lead to interventions that stimulate the use of progress-monitoring data, which eventually may lead to improved student performance.


Poster session - Theme 3: Multimodal

21. Authentic, developmental assessment of spatial skills in a 3D virtual environment -Andrea Karpati*, Laszlo Budai and Bernadett Babaly

Abstract: Representing space in two-dimensional form has traditionally been one of the central tasks in art education because of its relevance for a wide range of professions and artistic expression. Representation and perception of space are valid indicators of visual skills development and often used for the detection of talent. However, traditional drawing tasks vaguely resemble real life experiences of space. As part of a national assessment of competences, a framework of visual skills were developed and tested through creative tasks and tests on a representative sample of 6-12-year-olds. In a follow-up study, the new GeoGebra 3D software and eDIA, an interactive virtual testing environment was used for developmental assessment of spatial skills. We compare performance in two-dimensional, on-line tasks and dynamic, virtual space. Our results indicate that the 3D environment is most appropriate for skills development, while 2D testing tools are optimal for evaluation.

22. Modeling eye movements to support multimedia learning - Carina Schubert*, Katharina Scheiter and Anne Schüler

Abstract: In two experiments, we investigated if presenting Eye Movement Modeling Examples (EMMEs) that illustrate how to process multimedia instruction adequately supports learning. In experiment 1a, students observed either a video of an expert’s eye movements on exemplary multimedia material, the same video using visual cues instead of EMMEs, or received no intervention. The interventions in the EMME and cueing conditions tapped different processes like selection, organization and integration of text and picture information. All groups learned with multimedia instruction on a novel topic and completed a posttest. There were no differences in learning outcome between the groups. Since this may have been due to lack of transfer of the processes to the new material, in study 1b students either saw EMMEs on the actual learning material or received no intervention Both groups learned with the same material and completed a posttest. Data from study 1b is currently under analysis.

23. Examining the effects of animated representations on students’ understanding of dynamic molecular events - Jodie Jenkinson*, Stuart Jantzen and Gaël McGill

Abstract: This research examines the effectiveness of visual complexity in animation for learning about molecular events. In three related studies, we examined the relative effectiveness of three-dimensional visualization techniques for learning about protein conformation and molecular motion in association with a ligand–receptor binding event. Undergraduate biology students (n=131; n=8; n=61) were assessed using short answer tests, verbal reports, and eye tracking data in order to characterize their understanding. Our findings suggest that students attend to the same narrative elements regardless of the complexity depicted in each animation and that in select learning contexts, increasingly complex representations may be more desirable for conveying the dynamic nature of cell binding events. However, our results also suggest that students have difficulty understanding the random nature of molecular events and anthropomorphize these processes.

24. Informal learning from educational television programs: A second screen approach - Liesbeth Kester*

Abstract: Second screens like tablets or smartphones allow for the enrichment of television programs with extra information or social interaction. Television producers become more and more aware of their use and invest in the development of second screen applications that can be consulted during a television program. This study investigates the effects of presenting extra information on a second screen while watching educational television on interest and retention. The extra information was presented as propositions or true/false propositions, either on a second screen during the program or after the program. Based on Hidi and Renninger’s Four Stage Model of Interest Development it was predicted that presenting true/false propositions increases interest and based on Mayer’s Cognitive Theory of Multimedia Learning it was predicted that presenting extra information during the program hampers retention. The first prediction was not confirmed. The second prediction was partially confirmed.

25. Individual differences in multi-modal viewing patterns - Mona Holmqvist Olander* and Eva Wennås Brante

Abstract: This study describes individual differences in viewing patterns between two groups of young adults with and without dyslexia. The participants (n=24; 14 non-dyslexic and 10 with dyslexia) met learning material in both text and picture while their eye movements were recorded. The self-paced experiment consisted of computer based material (six screens repeated in six sets) and questions with a mean experiment time of 13.49 minutes (9.39 – 20.24). The results show differences in viewing pattern regarding number of fixations at text and picture between the groups. The controls made 24% less fixations in total but +29% fixations more on the picture than the group with dyslexia. The controls spent 1.7% of their fixations at the picture (dyslexia 1%). The results indicate that information given in pictures is discerned to a lower degree by participants with dyslexia.

26. Learning functional relations in complex animations - Richard Lowe, Jodie Jenkinson and Gaël McGill

Abstract: Achieving major improvements in the quality of mental models that learners develop from complex animations has proven to be an elusive goal. Considerable research effort has been devoted to exploring the potential of visual cues (both static and dynamic) for helping learners to extract and internalize high relevance information from these animations. However, when the depicted information is already rich and multi-layered, there can be practical difficulties in cueing all relevant aspects within the animation. If cueing targets one particular level of information, it is possible that other levels will be neglected. Dynamic cues that signaled key overarching relational aspects were applied to an animation depicting the complex topic of ant locomotion. Most subsequent learner demonstrations using a replica ant included these cued high level aspects but not contingent lower level relations. Alternatives to conventional animation design approaches that better match information presentation to human visual processing will be considered.

27. Tactile graphics and vibrational search cues - Richard Lowe

Abstract: Information displays that rely on our sense of touch have been used both to compensate for vision impairment and to complement vision in situations with high visual processing demands. For example, tactile graphics are now an established option for equal access learning materials. However, touch also has the potential to function in an ancillary role to support more effective processing of visually-rich displays. The comprehension of complex animations imposes considerable demands on visual processing resulting in possible neglect of low salience, high relevance information due to misdirection of attention. Visually-based cueing can lack effectiveness in this already rich visual context. Touch-based cues have the potential to guide learners to key information within an animation. This paper reports a study comparing the relative efficiency of entity-based and path-based vibration for guiding search of tactile graphics.

28. Using eye tracking to measure students’ knowledge acquisition processes from text and graphics - Ozge Yasar, Janice Gobert* and Ermal Toto

Abstract: In this study, we provided students with scaffolding based on their eye-gaze movements about where to look/what to read via a pedagogical agent with the goal of improving their knowledge acquisition processes and content learning. 27 middle school students used Inq-ITS environment to read texts and view diagrams about Plate Tectonics. In addition to eye movements, learning gain scores were collected from pre and post-tests and open responses to explanation-type questions about the domain. In this work, we present the results of the analysis in which we investigate the relationship between integrative eye movements (text to diagram or diagram to text) and conceptual learning gains. The results of these analyses yielded that the students who engaged in more sophisticated knowledge acquisition processes, as evidenced by a greater proportion of integrative eye movements (text to diagram or diagram to text) had higher learning gains when compared to the students with less sophisticated knowledge acquisition processes who had lower conceptual learning gains.


Poster session - Theme 4: Instruction

31. Vascular invaders: Do game elements enhance learning? - Andrea Gauthier*, Michael Corrin and Jodie Jenkinson

Abstract: This research explores the costs and benefits of learning with an online gamified study aid for studying human vascular anatomy versus a similar non-game study aid and how it relates to usage over a period of 5 weeks. Preliminary data analyses show that undergraduate medical students are more likely to complete studying tasks with the game version in comparison with a non-game version (p=0.04). Though a significant positive correlation was found between game-usage and anatomy test improvement (r=0.41), it remains uncertain if students learn more from the added game elements in the experimental tool in comparison to the control tool (p=0.08). Further analysis is currently underway exploring the relationship between interaction with specific game elements, personal information such as game preferences, study habits, and gaming habits, and test improvement.

32. Item solving in text-picture comprehension - Fang Zhao* and Wolfgang Schnotz

Abstract: Our research aims at exploring question solving in text-picture comprehension, which has rarely been focused on before. We conducted an eye tracking experiment with 17 secondary school students, who comprehended our blended text and picture materials in two ways. (1) Unguided processing with text and picture and without question. (2) Information gathering to answer the question after the prior experience with text and picture. The results showed that readers have focuses differently on text and pictures in two conditions. Presumably text played the crucial role in guiding the meaning and picture is used to retrieve the perceived information. We also propose an Item Solving Model in Text-Picture Comprehension, which might provide pedagogical implication for learning in secondary school.

33. Test-enhanced learning: A replication attempt - Gerdien van Eersel, Mario de Jonge*, Huib Tabbers and Peter Verkoeijen

Abstract: Research has shown that testing during learning can enhance the long-term retention of information. In two experiments, Roediger and Karpicke (2006) investigated this so-called retrieval practice effect using educationally relevant materials (short prose passages) and found that free recall testing during initial learning enhanced long-term 1-week retention relative to a restudy (control) condition. In the present study, we attempted to replicate the results of their Experiment 1, using an identical research design, but different prose passages. Data collection is underway, preliminary result will be discussed.

34. Effects of seeing the model’s face on learning from video modeling examples - Margot van Wermeskerken* and Tamara van Gog

Abstract: Video modeling examples in which human models show learners how to perform a task are increasingly used in learning and instruction. However, little is known about design guidelines to optimize learning from such examples. Because faces automatically capture observers’ attention and because we automatically follow other people’s gaze, seeing the model’s face in a video modeling example might hinder (i.e., distract) or help learning (i.e., guide attention). Prior research using a puzzle problem showed a beneficial effect of seeing the model’s face after studying the example twice. This study aimed to replicate and extend those findings with a different task. Participants twice studied a video example in which it was demonstrated and explained how to build the molecule glutamine. After each view they attempted to build it themselves and answered questions related to the explanations. Preliminary results suggest that seeing the model’s face did not affect learning after one example.

35. Changing conceptions of nature of science through modified research articles: Effects of pre-training - Nora Anders*, Janna Müller and Isabel Braun

Abstract: Interventions targeting conceptions of nature of science should be explicit, authentic, and challenging. Reading research articles represents an authentic and challenging task to students. In a pilot study, we investigated the effects of providing explicit instruction before a reading activity involving a research article that had been modified for low-knowledge readers. Participants received explicit instruction on nature of science (beliefs pre-training condition) or genetic engineering (knowledge pre-training condition). Results obtained from interviews indicate that type of pre-training did not affect conceptions of nature of science. Knowledge pre-training resulted in significantly higher knowledge of key concepts. Results also suggest that participants under the knowledge pre-training condition achieved higher text comprehension. From participants’ responses to the interviews, we conclude that the reading activity should span several texts, the explicit instruction should be incorporated into the texts, and further modifications to the research articles should highlight their authentic portrayal of nature of science.

36. The 3R study strategy: When to offer instructions? - Pauline Reijners*, Liesbeth Kester, Gino Camp and Paul A. Kirschner

Abstract: The goal of the current study is to investigate at what moments within the 3R study strategy students in secondary education need instructions in order to profit from the strategy in terms of test results. In the first condition participants perform the traditional 3R study strategy with no manipulations during the read phase and free recall during the recite phase. In condition two participants answer questions after reading and in condition three participants receive these questions during the recite phase. In the last condition participants receive both during the read and recite phase questions. Furthermore, the relationship between normal study habits concerning studying prose texts and their test results is investigated.

37. Effect of prior knowledge on reading strategies of a concept map divided by level of difficulty - Paulo Correia* and Joana Aguiar

Abstract: The possibility of reading a concept map in different ways and the need to establish the reading sequence generates an additional cognitive demand (extrinsic load) that may compromise learning outcomes (intrinsic and generative loads). Learners with high prior knowledge (PK) have higher tolerance to deal with the ‘map shock’. This study aimed to explore the effects of PK on reading strategy of a digital Cmap about chemistry, divided by level of difficulty. Eye-tracking measurements were made during the Cmap reading with 21 participants. Count and length of fixation for the six areas of interest (AOI), defined according to the content level of difficulty, were recorded. The main results showed no effect of PK for cognitive mental effort reported by participants; however, a main effect was observed in reading strategy.

38. Matching text modality to learners’ modality preferences: Effects on learning and motivation - Tina Seufert*, Miriam Arnold, Julia Günzer, Felix Wagner and Georg Hauck

Abstract: Due to the increasing use of multimedia learning systems contents can be easily presented as visual or auditory text. Besides learners can have a preference for a particular sort of text presentation, the so called modality preference. It seems plausible that learners should learn better when the preferred modality is provided. We conducted two experiments, one in a scientific domain (n=43), the other in prose (n=53) each time conducting a 2x2 factorial design with the factor modality preferences (visual or auditive) and presented text modality (visual or auditive). We measured learning effects in both studies and in the second one we also analysed effects on learners motivation. While first we in fact found that learners show better learning performance when they learn in their preferred modality we could not confirm this in the second study. However, there we found higher motivation in the matching conditions.

39. The effect of visual aids in representational illustrations on realistic word problem solving - Tinne Dewolf*, Wim Van Dooren and Lieven Verschaffel

Abstract: The present study investigated the effect of two visual aids in representational illustrations on elementary school pupils’ realistic word problem solving. 288 pupils (age 10-12 years) participated in the study. In Part 1 they received a paper-and-pencil task with seven word problems with a realistic modelling challenge (P-items). These problems were presented together with representational line drawings, representational line drawings with an extra element to make the modelling complexity more apparent, or representational line drawings with this element being cued by means of highlighting. In Part 2, pupils received the same P-items together with a realistic reaction and a non-realistic reaction, and the request to choose the best answer. The findings showed no effect of the visual aids on the number of realistic reactions (Part 1), and, when reviewing possible answers, no positive effect of the visual aids on pupils’ appreciation for the realistic reaction either (Part 2).

40. A comparison of the effects of worked examples and modeling examples with and without a visible model on learning a problem-solving task - Vincent Hoogerheide*, Sofie Loyens and Tamara van Gog

Abstract: Example-based learning is an effective and efficient instructional strategy for novices to learn new procedural skills. Two experiments investigated whether worked examples and modeling examples with and without a visible model are equally effective in fostering learning outcomes and the acquisition of cognitive schemas. In Experiment 1 (N=76) and Experiment 2 (N=135), secondary education students were taught how to solve probability calculation problems. The students were randomly allocated one of three conditions: Worked Example Condition (written text, static visualizations), Modeling Example Condition With Model Visible (spoken text, dynamic visualization), or Modeling Example Condition Without Model Visible (spoken text, static visualizations). The results of both experiments taken together show that worked examples and modeling examples with and without a visible model are equally effective at fostering learning, and near and far transfer. Moreover, all three conditions invested an equal amount of effort in the acquisition and test phase.