Testing the additivity hypothesis of cognitive load theory [Elektronische Ressource] / vorgelegt von Babette Park

Mitarbeiter

Sujets

• Psychologie

Informations

• Publié par: universitat_des_saarlandes
• Publié le: 01 janvier 2010
• Nombre de lectures: 16
• Langue: English
• Poids de l'ouvrage: 2 Mo

Extrait

Testing the Additivity Hypothesis of
Cognitive Load Theory





Dissertation
zur Erlangung des akademischen Grades eines
Doktors der Philosophie
der Philosophischen Fakultäten
der Universität des Saarlandes


vorgelegt von

Babette Park

aus Freiburg im Breisgau




Saarbrücken, 2010




































Dekan:
Prof. Dr. Wolfgang Behringer

Berichterstatter:
Prof. Dr. Roland Brünken
Prof. Dr. Tina Seufert

Tag der Disputation: 5. Februar 2010
iiAbstract
The present work focuses on the additivity hypothesis of Cognitive Load Theory in the frame
of a research program supported by the Deutsche Forschungsgemeinschaft (German Research
Foundation; grant Br 2082/6-1). The additivity hypothesis predicts that effects of load-
inducing factors on learning are additive (Paas, Renkl, & Sweller, 2003a). Until now, this core
assumption of the widespread used theory has never been empirically tested. The present
work investigated different combinations of extraneous and germane load factors in self-
paced multimedia instruction to examine the additivity hypothesis. The instruction explained
a molecular process in the domain of Biology and contained 11 multimedia screens with static
pictures accompanied by verbal explanations. The highly complex learning issue is about
structural as well as procedural information of energy storage in cells. Modality of the verbal
explanation and seductive details served as extraneous load factors, while support for
coherence formation and mental animation tasks served as germane load factors. Four
preliminary single effect studies (n = 23-78 each) proved stable effects of these factors on
learning performance. Three main studies (n = 99-100 each) varied thereafter two extraneous
load factors (modality x seductive details), one extraneous and one germane load factor
(modality x support for coherence formation), as well as two germane load factors (support
for coherence formation x mental animation tasks), respectively, in a 2x2 factorial design. In
these three studies, overall 299 high school students (71.3 % female) of different German
schools with mean age of 17.05 years (SD = 1.39) participated. Results on learning success
and the subjectively rated cognitive load (Paas, 1994) served as dependent variables. Spatial
ability, prior knowledge and time-on-task were introduced as control variables. Analyses of
variance or covariance were conducted using experimental condition as between-subject
factor and post-test score and cognitive load ratings, respectively, to test the learning and total
cognitive load effects predicted by Cognitive Load Theory. Results showed none of the main
effects in learning success and cognitive load, predicted by the additivity hypothesis.
However, some interesting interaction effects turned out, which were generally not predicted
by the additivity hypothesis. In sum, results suggest that combined effects of different
cognitive load factors are not necessarily additive. They overall contradict the additivity
hypothesis. The last chapter discusses implications for cognitive load research and
implications of the found interesting boundary effects of the present work resulting in future
directions.


iiiZusammenfassung
Die vorliegende Arbeit untersucht die Additivitätshypothese der Cognitive Load Theory im
Rahmen eines von der Deutschen Forschungsgemeinschaft geförderten Forschungsprojekts
(Br 2082/6-1). Die Additivitätshypothese besagt, dass Effekte load-induzierender Faktoren
additiv wirken (Paas, Renkl, & Sweller, 2003a). Bislang wurde diese Kernannahme der weit
verbreitet genutzten Theorie jedoch empirisch nie überprüft. Um die Additivitätshypothese zu
testen, untersucht die vorliegende Arbeit verschiedene Kombinationen von extraneous und
germane load Faktoren in einer selbstgesteuerten multimedialen Instruktion. Diese erklärt den
Aufbau und Prozesse eines Moleküls aus der Biologie und umfasst 11 Bildschirmseiten mit
statischen Bildern und Begleittext. Das hoch komplexe Lernthema beinhaltet Struktur- und
Prozessinformationen zur Speicherung von Energie in den Zellen. Die Modalität des
Begleittexts und Seductive Details dienen als extraneous load Faktoren, wohingegen
Kohärenzbildungshilfen und mentale Animationsaufgaben als germane load Faktoren
fungieren. Vier Pilotstudien (jeweils n = 23-78) wiesen stabile Effekte dieser Faktoren auf den
Lernerfolg nach. Die drei daraufhin durchgeführten Hauptstudien (jeweils n = 99-100)
variierten jeweils zwei extraneous (Modalität x Seductive Details), einen extraneous und
einen germane (Modalität x Kohärenzbildungshilfen) bzw. zwei germane load Faktoren
(Kohärenzbildungshilfen x mentale Animationsaufgaben) in einem 2x2 faktoriellen Design.
Insgesamt nahmen an diesen drei Studien 299 Oberstufenschüler (71.3 % Mädchen) aus
verschiedenen deutschen Schulen mit einem durchschnittlichen Alter von 17.05 Jahren (SD =
1.39) teil. Der Lernerfolg und die subjektive Beurteilung der kognitiven Belastung (Paas,
1994) dienten als abhängige Variablen. Als Kontrollvariablen wurden das räumliche
Vorstellungsvermögen, das Vorwissen und die Lernzeit erfasst. In Varianz- bzw.
Kovarianzanalysen wurden die Experimentalbedingungen als Zwischensubjektfaktoren und
die Lernerfolgs- bzw. cognitive load-Werte genutzt, um die aus der Cognitive Load Theory
abgeleiteten Lern- und Load-Effekte zu testen. Die Ergebnisse zeigten keine der durch die
Addititvitätshypothese erwarteten Haupteffekte im Lernerfolg bzw. in der kognitiven
Belastung. Es ergaben sich jedoch einige interessante Interaktionseffekte, welche auf Basis
der Additivitätshypothese generell nicht zu erwarten waren. Zusammenfassend weisen die
Ergebnisse darauf hin, dass kombinierte Effekte verschiedener load-Faktoren nicht
notwendigerweise additiv wirken. Insgesamt widersprechen die Ergebnisse der
Additivitätshypothese. Welche Implikation diese Falsifikation und die gefundenen
Nebeneffekte für die cognitive load Forschung haben, wird im letzten Kapitel diskutiert und
mündet in einen Forschungsausblick.
ivAcknowledgement
By publishing this dissertation I would like to express my gratitude to all those who supported
me in this project. First of all, I would like to thank my supervisor Prof. Dr. Roland Brünken,
who always showed confidence, respect and kindness to me and shared his scientific
experience and ideas with me. This pleasant working environment allows constructive
criticism and guarantees moving forward with considerable latitude. One more mentor I
would like to thank is Prof. Dr. Tina Seufert (University of Ulm), who was willing to listen to
me whenever I looked for a professional dialog partner to discuss and rethink the strategy on
how to reach the goal. Tina, thank you for having been such a co-operative colleague in our
research group in Saarbrücken! Moreover, I would like to thank Prof. Dr. Robin Stark and
Prof. Dr. Norbert Gutenberg, who were mentoring me in an interdisciplinary research
program before I started my dissertation project. It was a pleasure to work with them and I
learned much about interdisciplinary networking, which builds a good basis for further
scientific working. In addition, I wish to thank the Deutsche Forschungsgemeinschaft
(German Research Foundation) for the generous support of the present research (Grant: Br
2082/6-1). Sincere thanks are given to all, who contributed to the positive development of this
research program: our technical support Dipl.-Ing. Dipl.-Päd. Tobias Gall, my new colleague
Dr. Stefan Münzer, and our co-operation partner Prof. Dr. Maria Bannert (TU Chemnitz) as
well as Prof. Dr. Jan Plass (New York University). I would also like to thank all my
colleagues at the Saarland University, who supported me in my daily work. Moreover, I am
very grateful to all my student assistants, Rebecca Baumann, Antje Zindler, Martin Klein,
Andreas Korbach and Laura Kielinger, who were very engaged. I would also like to thank all
teachers and students who participated in the studies. Furthermore, I would like to express my
deep gratitude to Prof. Roxana Moreno, Ph.D., J.D. for mentoring me during my doctoral
research trip at the University of New Mexico. This time I spent in Albuquerque was very
fruitful for my dissertation and this was also due to a lot of new friends I would like to thank
with these lines. ¡Gracias Fernando, Papa de Nueva México! And special thanks go to Kira
Carbonneau for correcting the present work as native speaker. Thanks to all other friends in
Germany and all over the world, who were keeping me in action off the job. Within this
context I would like to express my deep gratitude to my string quartet for giving me power,
warmth and a clear head by making music. The last lines are dedicated to my just expanded
family: