The Effect Of Oral Description In Complementing Animated Instruction 
In A Web-based Learning Environment On Undergraduate Students 
Achievement Of Different Educational Objectives

Jeff Swain
Wjs186@psu.edu
Mine Munyofu
Mxm939@psu.edu
Khusro Kidwai
Khk122@psu.edu
Huifen Lin
Huifen5612@yahoo.com.tw
Bradley Ausman
Bda135@psu.edu
Francis Dwyer
fmd@psu.edu
The Pennsylvania State University

Abstract
The study sought to test the principle of modality by using audio to 
deliver verbal information when that information is designed to 
support non-verbal information such as animations in a computer-based 
lesson.  This was done by comparing the effect of two types of audio 
support mechanisms- a simple support mechanism consisting of 
declarative statements explaining the animated sequence and a complex 
support mechanism consisting of questions and answers explaining the 
animated sequence- on undergraduate student achievement. Learning was 
measured through drawing, terminology, and comprehension tests. The 
results indicate student achievement was not enhanced by the addition 
of auditory support.
Introduction
With computers coming equipped with multi-media features such as the 
ability to play sound and animation, teachers and instructional 
designers are able to develop and deliver lessons in novel ways using 
both the visual and auditory channels of students. But, is student 
performance improved by the use of both channels as opposed to the 
traditional delivery method relying solely on the visual channel? 
And, at what learning level (the factual, conceptual, rule & 
procedure level) is there improvement in performance when both 
channels are used?
According to cognitive load theory (Kalyuga, Chandler and Sweller 
1988, 1989; Sweller 1988; Baddeley 1986, 1992) methods of instruction 
reducing working memory load in order facilitate the encoding and 
storing of the information in long-term memory are effective.  One 
such method is dual coding theory (Sadoski and Paivio 2001; Clark and 
Paivio 1991; Paivio 1971, 1986, 1990). Dual coding theory assumes we 
have two information processing systems: a verbal system, comprised 
of words, whose strength lies in its sequentially ordered hierarchy, 
each bit of information paves the way for the next, and a non-verbal 
system whose strength lies in its synchronous (holistic) hierarchy.
Using audio to deliver verbal information when that information is 
designed to support non-verbal information such as graphics, 
pictures, and animations can enhance the effect of using both the 
verbal and visual systems. This is known as the modality effect 
(Clark & Mayer 2003; Penney 1989; Paivio 1986). The modality effect 
(Clark & Mayer, 2003, pp. 93) states "people learn more deeply from 
multimedia lessons when words explaining concurrent animations or 
graphics are presented as speech rather than as onscreen text."
The studies reviewed indicate that the modality effect works well at 
improving student's verbal recall of factual information (Mayer 1991, 
Mayer 1992, Barron 1993, Mousavi 1995, Mann 1995, Mayer 1996, Mayer 
1998, Moreno 1999, Mayer 2001, and Moreno 2002). There is also 
indication that the modality effect works at improving student's 
ability to solve problems (Mayer 1991, Mayer 1992, Barron 1993, Mayer 
1994, Mousavi 1995, Mayer 1996, Mayer 1998, Moreno 1999, Chuang 1999, 
Mayer 2001, Moreno 2002). However, there is limited information 
regarding the effect of modality of student achievement of learning 
concepts, rules and procedures. The studies conducted by Mayer (1998) 
and Moreno (1999, 2002) found evidence to support the positive effect 
of modality on student's ability to learn conceptual information but 
these learning levels were not isolated and studied on their own. In 
these studies, student's ability to recall facts, identify concepts, 
and solve problems were tested together. This leaves open the 
question of how effective is the use of modality if the goal of the 
lesson is to facilitate achievement of conceptual and rule/procedure 
knowledge? This study seeks to begin filling in the gap in the literature 
by isolating these two intellectual skill learning-levels.
Literature Review
Studies Exploring The Effect Of Dual-Coding: Using The Visual And 
Verbal Channels
Mayer, in his study on how computer based animations can be used to 
promote scientific understanding (1991) and in his study aimed at 
identifying the role of student's spatial ability in learning from 
words and pictures (1994), found that undergraduate college students, 
given a lesson in an area of non-expertise, performed better on 
recall and problem solving tests when both the verbal and visual 
systems were utilized. Mayer also found that the effect of dual 
coding was enhanced when the verbal and visual information was 
presented concurrently, at the same time as the animation rather than 
before or after it (Mayer 1991, 1992, 1994). This finding was 
duplicated by Moreno (1999) who tested undergraduate college students 
with low prior knowledge of meteorology, ability to recall 
information about the process of lightning. Chuang (1999) found 
similar results working with seventh grade students in Taipei, Taiwan 
when studying the role of gender and field dependence/independence on 
the ability to solve math problems.
Placing supporting text near the animation it is meant to support in 
known as the contiguity principle (Clark and Mayer 2000) or the 
split-attention effect (Chandler and Sweller 19992; Sweller, 
Chandler, Tierney, and Cooper 1990; Tarmizi and Sweller 1988). The 
split-attention affect happens when students must divide their 
attention between multiple sources of information (Mousavi 1995). The 
split-attention effect can be reduced by placing printed words next 
to the animation they are supporting (Clark and Mayer 2000).
The positive effect of dual coding in reducing cognitive load also 
was evident in studies exploring the impact of reducing cognitive 
load in the lesson summary.  In his study to see if reducing 
cognitive load in lesson summaries would help increase student's 
retention, Mayer (1996) found that undergraduate students performed 
better on tests of recall and problem solving when summaries included 
both illustrations and text.
The results indicate the effectiveness of verbal, in the form of 
text, support of animation in reducing cognitive load. Animation 
complemented with a textual explanation enabled students to take 
greater advantage of their capability to process information on two 
levels by stimulating the visual system and by reducing the load 
placed on the verbal processing system. This re-shuffling of 
information in working memory increased their ability to make meaning 
out of the information in preparation for storage in long-term 
memory. The placement of the supporting textual explanation next to 
the animation further reduced cognitive load and enhanced performance.
Studies Exploring The Effect Of Modality: Using The Spoken Word In 
Place Of The Written Word To Support The Visual Channel
While animation helped to reduce cognitive load it was not reduced as 
much as it could be because both text and animation have to pass 
through the same (the visual) sensory channel (Mousavi 1995; Chandler 
and Sweller 1992). This meant that students were forced to shift 
their attention between the text and the animation while going 
through the pattern recognition and selective perception processes. 
Miller (1956, pp. 85) referred to the limitation of the sensory 
register as our "channel capacity." Channel capacity is the maximum 
amount of information we can hold in our sensory memory at any given 
point in time.
When animation is supported by a spoken explanation, as opposed to a 
textual explanation, cognitive load is further reduced. This time the 
reduction comes through the way that information passes from the 
environment through sensory memory and into working memory (Chandler 
and Sweller 1992; Paivio 1986; Penney 1989).
Mann (1995), in as study testing student's ability to construct a 
solution to an educational problem, used temporal sound, spoken 
information that highlights or details static or moving visuals, in a 
computer-based lesson to more effectively use student's channel 
capacity in sensory memory by using sound with text to support 
concurrent animation. Mann found that students were able to recall a 
greater amount of critical detail when temporal sound was used.
While Mann used both written text and temporal sound simultaneously, 
Mayer (1998), in a study that tested student's ability to recall how 
lightening works, compared the effect of using either temporal sound 
or written text to support concurrent animation and found that 
students were able to recall more information and perform better on 
problem solving tests if they received the lesson using temporal 
sound.
Lai (2000), in her study testing the various types of visual 
illustrations on concept learning, also found that students receiving 
lessons using temporal sound to support static graphics performed 
better on matching tests than students receiving lessons that used 
text to perform the same function.
Similar to spatial contiguity, the placement of text near the 
animation it supports, the contiguity affect also applies when 
temporal sound is used to support animation. Temporal contiguity 
occurs when visual and spoken materials are presented simultaneously 
rather than successively (Moreno 1999). Moreno (1999), using a lesson 
on the process of lightning formation, found that learning was 
negatively impacted if the temporal sound was not placed concurrent 
with the animation it supported matching Mayer's (1994) findings with 
written text.
Using written text or temporal sound that closely matches the 
animation it supports without a lot of extraneous details is also an 
important factor for success. In a study on reducing the cognitive 
load in lesson summaries, Mayer (1996), using a lesson on the process 
of lightning formation, found that students receiving concise lesson 
summaries, that included both visual and verbal information, 
performed best on verbal recall and problem solving tests. Similarly, 
Moreno (2002) found that extraneous details hurt student performance 
when using the lesson on the process of lightning formation.
Issues Of Instructional Consistency
Instructional consistency (Canelos 1983; Gagne and Medsker 1996)) 
states that intellectual skills are hierarchical in nature and that 
lower order skills are prerequisite for learning higher order skills. 
Verbal skills comprise the base of the hierarchy and are the ability 
to recall factual information. Verbal skills are a prerequisite for 
learning discriminations. Discriminations, the ability to distinguish 
between things, come next and are a prerequisite for concept 
formation.
Concepts are the ability to classify information based on its 
critical attributes and are a prerequisite for the learning of rules, 
which are the ability to specify the relationship between concepts. 
At the top of the hierarchy sits higher order rules, the ability to 
use multiple rules in order to perform a task or solve a problem 
(Gagne and Medsker 1996, pp. 32-33).
Therefore if the objective is to generate solutions, the 
instructional unit must contain the rules/procedures, concepts, and 
facts that represent the prerequisite knowledge needed to solve the 
problem. In the studies reviewed animation supported by narration 
increased students ability to recall factual information and to 
generate solutions to problems. The majority of studies reviewed 
explored student's ability to recall factual information and solve 
problems (Mayer 1991, 1992 1996; Mousavi 1995), to solve problems 
(Barron 1993; Mayer 19994, 2001), or to recall information (Mann 
1995). These studies do not give us an indication of where or how 
animation supported by narration works in the instructional 
hierarchy. Is animation supported by concurrent narration effective 
at teaching concepts or rules/principles? Or is there something about 
the animation with narration that enables students to build 
connections among intellectual skills?
This study aims to build on the existing knowledge base and begin to 
fill in the gaps in the literature by testing the hypothesis that 
animation supported by concurrent temporal sound is better at 
teaching concepts and rules/principles than animation alone.
Purpose Of This Study
The purpose of this study was to test the effect of the modality 
principle when applied within the instructional consistency paradigm 
by using audio to deliver verbal information when that information is 
designed to support non-verbal information such as animations in a 
computer-based lesson.  This was done by comparing the effect of two 
types of audio support mechanisms- a simple support mechanism 
consisting of declarative statements explaining the animated sequence 
and a complex support mechanism consisting of questions and answers 
explaining the animated sequence- on undergraduate student 
achievement of conceptual, rule and procedure knowledge. The 
questions the current study seeks to provide insight into include:
1. Do students receiving the treatment consisting of the simple audio 
support (declarative statements explaining the animation) perform 
better on tests of conceptual and rule & procedural knowledge than 
students receiving the treatment with animation alone?
2. Do students receiving the treatment consisting of complex audio 
support (questions followed up with declarative statements explaining 
the animation) perform better on tests of conceptual and rule & 
procedural knowledge than students receiving the simple audio 
treatment and the treatment consisting of animation alone?
Design And Methodology
A posttest only design method was used. The placement and use of 
animation and temporal sound was derived based on the results of two 
pilot studies. In the first study a computer-based lesson using 
Dwyer's (1977) lesson on the human heart, titled "The Heart And Its 
Functions", was used to determine where to place the animation. Since 
the goal of the primary study was to test student's ability to 
perform on tests of conceptual and rule/procedure knowledge, the 
lesson was designed using programmed instruction to ensure that 
adequate factual knowledge was achieved. An item analysis was 
completed to determine where to place the animation in the lesson for 
the subsequent study. A difficulty level of .60 was used as the 
cutoff meaning that any item with a difficulty level below sixty 
percent was targeted for animated support.
A series of four tests consisting of 20 questions each were used to 
assess student achievement. An identification test was used to assess 
factual knowledge. A drawing test and a terminology test were used to 
measure conceptual knowledge. A comprehension test was used to 
measure rule/procedure knowledge. Dwyer and Moore (1978) established 
the validity and reliability of the tests.  Based on the outcomes of 
the pilot study, 18 animations were developed and placed in the 
lesson adjacent to the textual material they were designed to support.
A second pilot study was conducted in order to determine which of the 
animated sequences required the use of temporal support. Again, item 
difficulty was set at .60 with any items supported by animation 
scoring below sixty percent targeted for support using temporal 
sound. The same four tests were used to assess student achievement. 
Based on the results of this analysis two treatments, one using 
simple audio support and another using complex audio support, were 
developed to support the animations.
For the primary study, eighty-eight undergraduate students were 
recruited from a management class, an educational psychology class, 
and an information systems class. These students were randomly 
assigned to one of three experimental groups: A control group that 
received the lesson with animation but no audio (NA) support. A 
treatment group assigned a lesson that used simple audio (SA) 
explanations, in the form of declarative sentences, in support of the 
animation. And, a treatment group assigned a lesson that used complex 
audio (CA) explanations, in the form of Questions and answers, in 
support of the animation. Twenty-nine students received the treatment 
with no audio support. Thirty students received the treatment with 
simple audio support. Twenty-nine students received the treatment 
with the complex audio support. All students received extra credit 
towards their final grade in the class for participating in the study.
All three experimental groups (NA, SA, CA) received a treatment where 
the beginning of the lesson was comprised of programmed instruction 
to ensure the prerequisite factual knowledge was gained. However, due 
to the nature of the content it was impossible to deliver the lesson 
with each learning level isolated. Therefore, the programmed 
instruction section also contained conceptual information along with 
the factual information. The programmed instruction consisted of a 
web page containing one or two pieces of factual knowledge. This 
meant that these pages also contained animation or animation with 
temporal support if the item analysis indicated it was needed.

Figure 1 Sample Programmed Instruction Page

After a series of three to four pages like this, students were asked 
to answer a series of practice questions based on the material just 
presented. If the student's score was satisfactory they were able to 
move on to the next part of the lesson. If the score was 
unsatisfactory they student was brought back to the beginning of that 
series of content. There was no limit placed on the amount of time or 
the number of times the student could spend on one section. Once the 
student satisfactorily completed the programmed instruction segment 
they were given a pencil and paper drawing test in which they were 
asked to draw and label the main sections of the human heart. Once 
that was completed the students were asked to complete an online 
identification test. In this test a picture of a heart was presented 
with an arrow pointing to the section to be identified. Students were 
asked to select the name of the section from a list of four choices.
The second part of the lesson was primarily focused on rule/procedure 
information although there was also some conceptual information 
presented.  Students went through a series of web pages describing 
the flow of blood as it passes through the heart. Some pages 
contained animated support of the content and some pages contained 
animation along with temporal support. The control group (NA) lesson 
contained only animation. Where the SA group received temporal 
support for animation it was in the form of simple declarative 
sentences. For example, if the animation were designed to show that 
the ventricles are the thickest walled chambers of the heart, the 
student, when he/she selected the play button would see the animation 
and simultaneously hear a statement that said, "The ventricles are 
the thickest walled chambers of the heart."
The CA group received the same treatment as the SA group except that 
the temporal support was delivered in a question and answer format. 
Continuing the example above regarding the ventricles, a student in 
the CA group would hear while the animation was playing, "What are 
the thickest walled chambers of the heart? The ventricles are the 
thickest walled chambers of the heart."

Figure 2 Sample Page With Animation
The same voice was used in both treatment groups and the speed at 
which the animation was played was adjusted to fit the length of the 
temporal support. In all three experimental groups, students were 
allowed to replay the animation, as many times as they felt was 
necessary to understand the material.
At the end of the lesson students in each group were asked to 
complete a terminology test where they were asked to complete a 
sentence by selecting the appropriate word or phrase from the choices 
provided, and a comprehension test where they were asked to answer a 
question by selecting the appropriate answer from a list of choices. 
The identification, terminology, and comprehension tests were built 
into the computer-based lesson and completed on-line.
Results And Implications
A one-way ANOVA was conducted to compare the differences between the 
three experimental groups on scores on the four criterion tests using 
the standard version of SPSS 11.0. The alpha level was set at the 
.05. Comparisons were made at two levels: using all the test items 
and using only the test items identified for improvement through the 
item analysis done after each pilot study.
Results Using All Items
The table below lists the mean scores and the standard deviations for 
each treatment group counting all test items:
Identification Test
Group
M
SD
No Audio (Control)
17.62
2.470
Simple Audio
17.40
2.094
Complex Audio
17.88
2.215
Total for All Groups
17.62
2.247

Drawing Test
Group
M
SD
No Audio (Control)
15.97
3.257
Simple Audio
16.00
3.280
Complex Audio
17.31
2.714
Total for All Groups
16.42
3.125

Terminology Test
Group
M
SD
No Audio (Control)
11.93
5.028
Simple Audio
12.70
4.170
Complex Audio
12.58
5.077
Total for All Groups
12.40
4.174

Comprehension Test
Group
M
SD
No Audio (Control)
11.00
3.464
Simple Audio
11.27
3.600
Complex Audio
12.08
4.069
Total for All Groups
11.42
3.688



The analysis of variance did not indicate a significant difference 
between the mean scores of each group. This was true both for the 
individual tests and the total test [Drawing Test: F (2, 87) =1.787, 
p > .05; Identification Test: F (2, 84) = .319, p > .05; Terminology 
Test: F (2, 84) = .218, p > .05; Comprehension Test: F (2, 84) = 
.621, p > .05; Total Test: F (2, 84) = .659, p > .05.
Results Using Only Items Identified Through Item Analysis
The tables below details the mean and standard deviation for each 
treatment group for only the items identified as deficient through 
item analysis. The reliability coefficient was .7778. There is no 
table for the Identification Test because none of the questions had 
an item difficulty of below .60.
Drawing Test
Group
M
SD
No Audio (Control)
3.28
1.601
Simple Audio
2.97
1.450
Complex Audio
3.93
1.163
Total for All Groups
3.39
1.458

Terminology Test
Group
M
SD
No Audio (Control)
4.14
2.489
Simple Audio
4.93
2.243
Complex Audio
4.73
2.539
Total for All Groups
4.60
2.416

Comprehension Test
Group
M
SD
No Audio (Control)
4.38
1.635
Simple Audio
4.60
2.010
Complex Audio
4.69
2.074
Total for All Groups
4.55
1.893

Conflicting results were found between the analysis of variance of 
the drawing and terminology tests. Both tests were designed to 
measure conceptual knowledge yet while the drawing test [F (2,87) = 
3.547, p = .033] indicated significant differences between groups the 
terminology test [F (2, 84) = .851, p > .05] did not. A possible 
explanation for this discrepancy could lie in the type of conceptual 
knowledge each test was designed to measure. The drawing test was an 
associative style test measuring low-level conceptual knowledge while 
the terminology test required the linking together of these 
associations to create higher-level concepts. Much of the lower-level 
conceptual knowledge could have been gained through the 
programmed-instruction portion of the lesson. The analysis of 
variance indicated no significant differences between mean scores of 
the groups for the comprehension test indicating that neither simple 
nor complex audio were better at improving student performance in the 
acquisition of rule/procedural knowledge (Comprehension Test: F (2, 
84) = .198, p >.05). Also, no significant differences were found when 
all the items were combined to create a total test score (Total 
Items: F (2, 84) = .835, p > .05).
Implications Of The Results
The lack of significance in performance between the experimental 
groups indicates that using either animation or animation along with 
temporal support may be problematic when it comes to teaching 
concepts and rules/principles. These results are interesting in lieu 
of prior studies whose results indicated animation supported with 
temporal sound was effective at teaching facts and problem-solving 
skills. With factual knowledge being a prerequisite for learning 
concepts, rules/principles and rules/principle knowledge being a 
prerequisite for problem solving learning it was thought that 
animation supported by temporal sound would have been effective at 
teaching concepts and rules/principles. However, the results of this 
study do not support this hypothesis.
These results do, however, suggest possibilities for further 
research. Some possible avenues for exploration include: What is it, 
of there is anything, about the nature of concepts and 
rules/procedures that may not make them amenable to animation based 
learning? Does temporal sound increase cognitive load when the lesson 
is aimed at conceptual or rule/principle information? Does the kind 
of content being presented limit the effectiveness of modality? What 
impact do learner traits, such as attention, attitude and motivation 
toward the content, play? Are methods other than simple declarative 
sentences or questions followed by answers better suited for teaching 
concepts and rules/procedures?
A potential limitation of the study lies in the similarity between 
the content to be read and the audio used to support the animation. 
Clark and Mayer (2003) report that redundancy of information could 
negatively impact learner performance. While the audio was not a 
verbatim speaking of the text, it may have been similar enough that 
its impact was negated.
Summary
Using temporal sound to support animation in computer-based lessons 
has been effective when the goal is to teach factual knowledge. There 
is also indication that it is effective for teaching problem-solving 
skills. This study, however, did not find evidence to support that 
using animation with temporal sound to teach concepts, 
rules/principles is effective. The results indicate that there is a 
problem of instructional consistency when applying the modality 
effect to the learning of intellectual skills.

References

Baddeley, J. R. (1986). Working Memory. Oxford, England: Oxford 
University Press.
Baddeley, J. R. (1992). Working Memory. Science(255), 556-559.
Barron, A. E. (1993). The Effectiveness of Digital Audio in 
Computer-Based Training. Journal of Research on Computing in 
Education, 25(3), 277-290.
Chandler, P., & Sweller, J. (1992). The split-attention effect as a 
factor in the design of instruction. British Journal of Educational 
Psychology, 62, 233-246.
Chuang, Y.-R. (1999). Teaching in a Multimedia Computer Environment: 
A Study of the Effects of Learning Style, Gender, and Math 
Achievement. Interactive Multimedia Electronic Journal of 
Computer-Enhanced Learning, 1(10).
Clark, J. M., & Paivio, A. (1991). Dual Coding Theory In Education. 
Educational Psychology Review, 3, 149-210.
Clark, R. C., & Mayer, R. E. (2003). e-Learning and the Science of 
Instruction. San Francisco, CA: John Wiley & Sons.
Driscoll, M. (2000). Psychology of Learning for Instruction (2nd 
ed.). Needham Heights, MA: Allyn and Bacon.
Gagne, R., & Medsker, K. (1996). The Conditions of Learning: Training 
Applications. Orlando, FL: Harcourt Brace.
Lai, S.-L. (2000). Influence of Audio-Visual Presentations on 
Learning Abstract Concepts. International Journal of  Instructional 
Media, 27(2), 199-206.
Mann, B. L. (1995). Focusing Attention with Temporal Sound. Journal 
of Research on Computing in Education, 27(4), 402-425.
Mayer, R. E. (1992). The Instructive Animation: Helping Students 
Build Connections Between Words and Pictures in Multimedia Learning. 
Journal of Educational Psychology, 84(4), 444-452.
Mayer, R. E. (1993). Cognitive Constraints on Multimedia Learning. 
Journal of Educational Psychology, 93(1), 187-198.
Mayer, R. E. (1994). For Whom is a Picture Worth A Thousand Words?: 
Extensions of a Dual-Coding Theory of Multimedia Learning. Journal of 
Educational Psychology, 86(3), 389-401.
Mayer, R. E. (1996). When Less is More: Meaningful Learning From 
Visual and Verbal Summaries of Science Textbook Lessons. Journal of 
Educational Psychology, 88(1), 64-73.
Mayer, R. E. (1998). A Split-Attention Effect in Multimedia Learning: 
Evidence for Dual Processing Systems in Working Memory. Journal of 
Educational Psychology, 90(2), 312-320.
Mayer, R. E., & Anderson, R. (1991). Animations Need Narrations: An 
Experimental Test of a Dual-Coding Hypothesis. Journal of Educational 
Psychology, 83(4), 484-490.
Mayer, R. E., & Heiser, J. (2001). Cognitive Constraints On 
Multimedia Learning: When Presenting More Material Results In Less 
Understanding. Journal of Educational Psychology, 93(1), 187-198.
Mayer, R. E., & Moreno, R. (2002). Aids to Computer-based Multimedia 
Learning. Learning and Instruction, 12(1), 107-119.
Miller, G. (1956). The Magical Number Seven, Plus or Minus Two: Some 
Limits on Our Capacity for Processing Information. The Psychological 
Review, 63, 81-97.
Moreno, J. (1999). Cognitive Principles of Multimedia Learning. 
Journal of Educational Psychology, 91(2), 358-368.
Moreno, J., & Mayer, R. E. (2002). Verbal Redundancy in MultiMedia 
Learning. Journal of Educational Psychology, 94(1), 156-163.
Mousavi, J. (1995). Reducing Cognitive Load by Mixing Auditory and 
Visual Presentation Modes. Journal of Educational Psychology, 87(2), 
319-334.
Paivio, A. (1979). Imagery and Verbal Processes. Hillsdale, NJ: 
Lawrence Erlbaum Associates.
Paivio, A. (1986). Mental Representation: A Dual Coding Approach. 
Oxford, England: Oxford University Press.
Penney, C. G. (1989). Modality Effects and the Structure of 
Short-term Verbal Memory. Memory and Cognition, 17, 398-422.
Sadowski, M., & Paivio, A. (2001). Imagery And Text: A Dual Coding 
Theory Of Reading And Writing. Mahwah, NJ: Lawrence Erlbaum 
Associates.
Sweller, J. (1988). Cognitive Load During Problem Solving: Effects on 
Learning. Cognitive Science, 12, 257-285.
Sweller, J., Chandler, P., Tierney, P., & Cooper, M. (1990). 
Cognitive load and selective attention as factors in the structuring 
of technical material. Journal of Experimental Psychology: General, 
119, 176-192.
Tarmizi, R., & Sweller, J. (1988). Guidance During Mathematical 
Problem Solving. Journal of Educational Psychology, 80, 424-436.


ITT Proceedings	17	Swain et. al.