Content uploaded by Wolfgang Schnotz
Author content
All content in this area was uploaded by Wolfgang Schnotz on Jun 09, 2018
Content may be subject to copyright.
Running head: ENABLING, FACILITATING, AND INHIBITING EFFECTS
Enabling, Facilitating, and Inhibiting Effects of Animations in Multimedia Learning:
Why Reduction of Cognitive Load Can Have Negative Results on Learning
Wolfgang Schnotz & Thorsten Rasch
University of Koblenz-Landau
Enabling, Facilitating, and Inhibiting Effects 2
Abstract
New technologies allow the display of text, static visuals, and of animations.
Although animations are inherently attractive, they are not always beneficial
for learning. Problems may arise especially when animations modify the
learner’s cognitive load in an unintended way. In two learning experiments
with 40 and 26 university students, the effects of animated pictures on
knowledge acquisition were investigated. Some pictures displayed visual
simulations of changes over time, whereas other pictures could be manipulated
by learners to represent different states in time. Results showed that
manipulation pictures had an enabling function for individuals with high
learning prerequisites, whereas simulation pictures had a facilitating function
for individuals with low learning prerequisites. However, the facilitating
function was not beneficial for learning, because learners were prevented from
performing relevant cognitive processes on their own. A careful analysis of the
interrelation between different kinds of cognitive load and the process of
learning is therefore required.
Enabling, Facilitating, and Inhibiting Effects 3
Enabling, Facilitating, and Inhibiting Effects of Animations in Multimedia Learning:
Why Reduction of Cognitive Load Can Have Negative Results on Learning
Computer-based multimedia learning environments can provide rapid access to
information. They can display multiple representations, and they can enhance active learning
through interactivity and exploration. However, these learning environments can also
introduce new demands for learners. In many environments learners have to orient themselves
and to navigate within complex information spaces. They have to search for and evaluate
information, and they have to understand and integrate multiple representations to build
coherent knowledge structures.
One of the frequently used features in computer-based multimedia learning
environments is animation. Any element on a computer screen can be animated, but the most
frequent use of animation concerns animated pictures. Animated pictures can be used to
support 3D perception by showing an object from varying perspectives. They can be used to
direct the observer’s attention to important (and unimportant) aspects of a display, convey
procedural knowledge (as e.g. in software training), demonstrate the dynamics of a subject
matter, and allow exploratory learning through manipulating a displayed object. Furthermore,
they can have a supplantative effect (Salomon, 1994), when they help learners to perform a
cognitive process that they could not otherwise perform without this external support. Despite
a widespread belief that animation is a powerful instructional device, however, it is still an
open question under which conditions animated pictures really enhance comprehension and
learning (Tversky, Morrison, & Betrancourt, 2002).
Functions of Animations
Animations can have two basic functions based on a reduction of cognitive load. If
they reduce the cognitive load of tasks in order to allow cognitive processing that would
otherwise be impossible, then animations have an enabling function. If they reduce the
cognitive load of tasks that could otherwise be solved only with high mental effort, then
animations have a facilitating function (cf. Mayer, 2001; Sweller & Chandler, 1994; Sweller,
van Merriënboer, & Paas, 1998). For example, when students learn about time phenomena
Enabling, Facilitating, and Inhibiting Effects 4
related to the earth’s rotation, animated pictures like those in Figures 1 and 2 can be useful. In
these figures, the earth is depicted as a sphere viewed from the North Pole that rotates in a
space where different locations are associated with different states of time. The picture shown
in Figure 1 can be manipulated by the learner who can define specific days-times for specific
cities. After clicking on the OK-button, the earth moves into the corresponding time state. We
will call this a manipulation picture. Because a manipulation picture enables learners to
investigate a high number of different time states, which would not be possible on the basis of
a static picture, such a picture is assumed to have an enabling function.
-----------------------------------------------------
INSERT FIGURE 1 ABOUT HERE
-----------------------------------------------------
The picture shown in Figure 2 can be used to simulate the earth‘s rotation. The learner
can choose different ways that a traveler can circumnavigate around the earth (symbolized by
a black dot moving in Western or Eastern direction with different traveling speed depending
on the learner’s choice). After pressing the SIMULATION-button, the earth starts rotating
and the traveler’s dot starts moving on the rotating earth. We will call this a simulation
picture. It might be much easier for a student to observe the rotation of the earth and the
movement of an object in a simulation picture than to perform the corresponding mental
simulations on his/her own with only a static picture (Lowe, 1999; Sims & Hegarty, 1997).
Thus, such a picture is assumed to have a facilitation function. The study was aimed at
analyzing how the assumed functions of animations affect cognitive processing and learning
results.
-----------------------------------------------------
INSERT FIGURE 2 ABOUT HERE
-----------------------------------------------------
Study 1
This study focused on a comparison between learning from animated (manipulation
and simulation) pictures and learning from static pictures. If animated pictures enable the
learner to perform additional cognitive processing, the learner’s total amount of processing
should increase. As additional processing needs additional time, the enabling function of
Enabling, Facilitating, and Inhibiting Effects 5
animations should lead to an increase of learning time compared to corresponding static
pictures. The enabling function is expected to be more pronounced when individuals have
high learning prerequisites (high cognitive ability and high prior knowledge) because these
learners will be able to use the possibilities of animations more extensively than individuals
with low learning prerequisites. If animated pictures facilitate cognitive processing, the
learner needs less effort with animated pictures than with static ones, because the animation
reduces cognitive load to a degree that is easier to cope with. Thus, if the facilitating function
of animations applies, learners will invest less learning time into animated pictures than into
corresponding static pictures. The facilitating function is expected to be more pronounced
when learners have low prerequisites because these individuals need more external support
than learners with high prerequisites.
If animated pictures enable individuals with high learning prerequisites to do
additional cognitive processing, these learners will spend more time observing animated
pictures than static pictures. If animated pictures facilitate processing for individuals with low
learning prerequisites, these learners will spend less time observing animated pictures than
static pictures. Following this line of reasoning it seems plausible to assume that there is an
interaction between learning prerequisites (high/low) and type of pictures (animated/static) on
learning time. This leads to the following hypothesis:
(H1) Students with high learning prerequisites will spend more time studying animated
pictures than static pictures, whereas students with low learning prerequisites will
spend less time studying animated pictures than static pictures.
If animated pictures enable learners with high prerequisites to do more cognitive
processing, this additional processing should also lead to better learning. Thus, one can
assume the following hypothesis regarding the enabling function of animated pictures for
students with high learning prerequisites:
(H2a) Students with high learning prerequisites learn more from animated pictures than from
static pictures.
If animated pictures facilitate cognitive processing for learners with low prerequisites
and allow them to process information more successfully, this should also lead to better
learning. Thus, one can assume the following hypothesis regarding the facilitating function of
animated pictures for students with low learning prerequisites:
Enabling, Facilitating, and Inhibiting Effects 6
(H2b) Students with low learning prerequisites learn more from animated pictures than from
static pictures.
The two hypotheses H2a and H2b could be integrated into an overall hypothesis,
which assumes that independent of individual learning prerequisites students learn more from
animated pictures than from static pictures. It should be noted, however, that there are
different reasons behind this overall prediction.
Method
Learners and Learning Material.
40 university students were randomly assigned to 2 groups of 20 participants.
Learning material was a computer-based hypertext that consisted of 22 cards (paragraphs)
with 2750 words about time and date differences on the earth and about the results of
circumnavigations around the earth. One group received the text with animated pictures and
the other group with static pictures. The pictures showed the earth as a sphere rotating in a
space in which different locations were associated with different time states. In the animation
group, 5 pictures allowed manipulations by defining specific days-times for specific
geographical locations as shown in Figure 1, and 5 pictures allowed to choose among different
ways of circumnavigating around the earth with a visual simulation of the earth‘s rotation and
a visualization of different circumnavigation as shown in Figure 2. The static pictures were
identical but did not include buttons for manipulation or simulation. In both groups, the
learners had free access to the text paragraphs and pictures via a hierarchically organised
menu.
Procedure and Scoring.
In the pre-test phase, participants were given a paper and pencil test for prior
knowledge, in which they had to explain a series of concepts referring to time phenomena on
the earth, and an intelligence test (Intelligenz-Struktur-Test 70 of Amthauer, 1973). They
were given prior knowledge scores on the basis of their written protocols and intelligence
scores based on their test results. In the subsequent practice phase, learners made themselves
familiar with the hypertext system referring to another subject matter unrelated to that used in
the experiment. The practice phase served to avoid any extraneous cognitive load caused by
an unfamiliar learning environment. In the following learning phase, all students received the
Enabling, Facilitating, and Inhibiting Effects 7
hypertext about time phenomena on the earth with either animated pictures or static pictures.
In order to provide an orientation for learning, participants received a sequence of 10
questions. Five of these questions were related to time differences between different places on
the earth while the remaining questions addressed time and date changes related to
circumnavigations of the earth.
Participants had free access to the available text and picture information. They could
take notes on a sheet of paper, and they had unlimited learning time. The students were
informed that they would subsequently be tested for their comprehension with similar
questions but without further access to the learning material. In order to avoid a too strong
task-oriented limitation of their exploratory activities, participants did not receive feedback
whether their answers were correct or incorrect. Picture observation times were automatically
recorded for each learner by the hypertext system.
In the final posttest phase, participants were required to apply the acquired knowledge
in a comprehension test without further access to the learning material or their notes. The test
consisted of 24 multiple-choice items and there was no time limit imposed for answering.
Twelve items referred to time differences between different places on the earth (e.g., What is
the time in Anchorage, if it is Thursday 9 o‘clock p.m. in Tokyo?). These questions required
knowledge about the subdivision of the earth‘s surface into time zones and about the time co-
ordinates of different cities and are referred to as time difference questions. Twelve items
dealt with time phenomena related to circumnavigations of the world (e.g., Why did
Magellan’s companions think, upon their arrival after sailing around the world, that it was
Wednesday when it was actually already Thursday?) These questions required participants to
perform internal simulations based on a mental model of the earth including time zones, the
date line as well as date zones, and are referred to as circumnavigation questions. For each
participant the number of correctly answered time-difference questions was determined as
his/her time-difference score and the number of correctly answered circumnavigation
questions as his/her circumnavigation score.
Results
In order to dierentiate between participants with high and low
learning prerequisites, regression analyses were performed with prior
knowledge and intelligence as predictors and learning results (sum of
Enabling, Facilitating, and Inhibiting Effects 8
correctly answered time dierence questions and sum of correctly
answered circumnavigation questions) as the dependent variables. The
resulting linear combination of prior knowledge and intelligence served for
determining the individual value of learning prerequisites for each
participant. Subsequently the sample was divided through a median split
into a group of 20 learners with high learning prerequisites and a group of
20 learners with low learning prerequisites. The means and standard deviations of
the picture observation times, of the time-difference scores, and of the circumnavigation
scores of learners with static pictures and of learners with animated pictures are presented in
Table 1, which further differentiates between high and low learning prerequisites.
-------------------------------------
TABLE 1 ABOUT HERE
-------------------------------------
A 2 x 2 ANOVA (Picture Type x Learning Prerequisites) of the picture observation
times yielded neither a significant main effect of picture type (F(1, 36) = 0.195, n.s.) nor a
significant main effect of learning prerequisites (F(1, 36) = 1.627, n.s.), but a significant
interaction Picture Type x Learning Prerequisites (F(1,36) = 3.171, MSE = 1 453 313, p
= .042,
2=.081). When students had high learning prerequisites, they spent more time for
animated pictures than for static pictures, which was marginally significant (t(18) = 1.306, p =
.104, d = 0.58). When students had low learning prerequisites, they spent less time for
animated pictures than for static pictures, which was also marginally significant (t(18) =
1.294, p = .106, d = 0.58). Although the observed a priori contrasts failed to become
significant, the significant interaction Picture Type x Learning Prerequisites can be considered
as preliminary support of hypothesis H1, which implies that students with high learning
prerequisites are more likely to spend more time studying animated pictures than static
pictures compared to students with low learning prerequisites, who are more likely to spend
more time studying static pictures than animated pictures. The results correspond to the
assumption that the enabling function of animations applies to students with higher learning
prerequisites, whereas the facilitating function applies to students with lower learning
prerequisites.
A corresponding 2 x 2 ANOVA of the time-difference scores showed a significant
effect of learning prerequisites (F(1, 36) = 5.528, MSE = 7.603, p = .012,
2 = .133) and a
Enabling, Facilitating, and Inhibiting Effects 9
highly significant effect of picture type (F(1, 36) = 8.553, MSE = 7.603, p = .003,
2 = .192).
The interaction Picture Type x Learning Prerequisites was not significant (F(1, 36) = 0.082,
n.s.). Students with high learning prerequisites outperformed students with low learning
prerequisites, and students with animated pictures outperformed students with static pictures
in answering time difference questions. Students with high learning prerequisites performed
significantly better in answering time-difference questions after learning from animated
pictures than after learning from static pictures (t(18) = 2.316, p = .017, d = 1.04), which
corresponds to hypothesis H2a. Students with low learning prerequisites also answered these
questions significantly better after learning from animated pictures rather than static ones
(t(18) = 1.830, p = .042, d = 0.82), which corresponds to hypothesis H2b.
Contrary to the analysis of time-difference scores, the corresponding 2 x 2 ANOVA of
the circumnavigation scores showed neither a significant effect of picture type (F(1, 36) =
2.380, MSE = 6.564, n.s.) nor a significant effect of learning prerequisites (F(1, 36) = 0.187,
n.s.), and no significant interaction Picture Type x Learning Prerequisites (F(1, 36) = 2.380,
n.s.). With regard to hypothesis H2a, students with high learning prerequisites answered
circumnavigation questions equally well after learning with animated pictures and after
learning with static pictures (t(18) = 0.0, n.s.). With regard to hypothesis H2b, surprisingly
students with low learning prerequisites answered circumnavigation questions better after
learning with static pictures than after animated ones (t(13.5) = 2.380, p = .033, d = 1.07). In
other words: Hypothesis H2a did not receive any support from the circumnavigation
questions. There was also no evidence for hypothesis H2b, but rather for the opposite
prediction: Animated pictures did not have positive effects on answering these questions, but
were harmful when students had lower learning prerequisites.
Discussion
It was assumed that animations can have both an enabling function and a facilitating
function in the process of learning. The picture observation times indicate that the individual’s
learning prerequisites decide which function dominates. For individuals with high
prerequisites animations seem to have an enabling rather than a facilitating function. For
individuals with low learning prerequisites, animations seem to have a facilitating rather than
an enabling function. The findings concerning answering time-difference questions supported
the assumption that animations result in better learning due to their enabling or facilitating
Enabling, Facilitating, and Inhibiting Effects 10
function. The findings concerning answering circumnavigation questions, however, did not
give any evidence for this assumption: Learners with high learning prerequisites did not profit
from the animations, and learners with low learning prerequisites surprisingly even performed
better with static pictures than with animated pictures.
In order to understand this unexpected divergence between time-difference and
circumnavigation scores, it might be helpful to analyse the cognitive processes required by the
corresponding questions more closely. Answering time difference questions requires
knowledge about time coordinates of various cities in the world and the time differences
between them. Manipulation pictures like that shown in Figure 1 can be used to display a high
number of different time states, which should be a good basis to extract information about
time differences. Thus, the high performance of the animation group in answering time
difference questions might correspond to the enabling function of such animations.
Answering circumnavigation questions requires mental simulations. Simulation
pictures like that in Figure 2 provide external support for such simulations. It may be well
possible that under specific conditions this function will be beneficial for learning, namely if
the individual has too low abilities to perform a mental simulation on his/her own (Salomon,
1994; Sweller & Chandler, 1994; van Gog, Ericsson, Rikers & Paas, this issue). Our study
indicates, however, that facilitation can also have a negative effect on learning. If individuals
are capable of performing such mental simulations by themselves, then external support can
make processing unnecessarily easy and, thus, students invest less cognitive effort in learning
from animation than in learning from static pictures. From the perspective of cognitive load
theory, animation can unnecessarily reduce germane load associated with deeper meaningful
cognitive processing (Sweller, 1999; Sweller, van Merriënboer, & Paas, 1998; van
Merriënboer, 1997). Most learners had obviously sufficient skills for mental simulations
without external support, but students with lower cognitive prerequisites were apt to accept
unneeded external support.
This negative effect of the facilitating function of animations has similarities with the
effects of redundancy in cognitive load theory. Sometimes, information from one information
source is self-contained, when it provides all the required information for knowledge
construction. If the same information is provided again in a different form, this creates
redundancy. Such redundancy usually increases cognitive load instead of reducing it, because
processing of the unneeded information means a waste of cognitive capacity (Kalyuga,
Enabling, Facilitating, and Inhibiting Effects 11
Chandler, & Sweller, 1998). However, whereas the redundancy effect is explained by an
increase of extraneous cognitive load due to the processing of additional (unneeded)
information, which reduces the remaining mental capacity, the assumed negative effect of the
facilitating function of animation is interpreted as a result of an (unintended) decrease of
germane cognitive load, because available mental capacity is left unused for the process of
learning.
Furthermore, our study indicates that animations can have different effects on different
tasks. As the manipulation pictures seem to allow deeper analysis of time differences, their
enabling function results in better performance with time difference questions. Simulation
pictures seem to make mental simulations easier, but this facilitating function can be harmful
for learning, when individuals who were able to perform these mental simulations also on
their own are indirectly hindered in doing so by unnecessary external support. In this case,
animation has an inhibiting effect on learning due to an inadequate reduction of germane
cognitive load. The simulation pictures stimulate behavioral interactivity, but do not stimulate
mental activity – a result that corresponds to the findings of Moreno and Valdez (this issue).
As the animation treatment in Study 1 included manipulation pictures as well as
simulation pictures, it was not possible to distinguish between different effects of different
kinds of animation. We assumed that manipulation pictures have primarily an enabling
function, which is especially important for time-difference questions, whereas simulation
pictures have primarily a facilitating function, which is especially important for
circumnavigation questions. In order to analyse in more detail whether the different kinds of
animation have different cognitive functions for different kinds of learners a second study was
conducted.
Study 2
Study 2 compared different kinds of animation: manipulation pictures and simulation
pictures. The manipulation pictures allowed a high number of different time states to be
generated for explorative purposes as shown in Figure 1. The simulation pictures allowed
continuous external simulation of the earth’s rotation with a fixed rotation speed of 4.8 rpm
that could not be controlled by the learner, combined with a circumnavigation around the
earth at different speeds as shown in Figure 2. We assumed that the manipulation pictures
have an enabling function that is especially helpful for answering time difference questions
Enabling, Facilitating, and Inhibiting Effects 12
and which is more pronounced (according to Study 1) if learners have high rather than low
learning prerequisites. The following hypothesis for students with high learning prerequisites
was posited:
(H3a) Manipulation pictures lead to better performance in answering time-difference
questions than simulation pictures, if learners have high learning prerequisites.
We further predicted an interaction between picture type and learning prerequisites according
to the following hypothesis:
(H3b) Manipulation pictures are more beneficial for answering time-difference questions
(compared to simulation pictures) for learners with high learning prerequisites than for
learners with low learning prerequisites.
Furthermore, we assumed that the simulation pictures have a facilitating function that affects
primarily circumnavigation questions and which is more pronounced, if learners have low
rather than high learning prerequisites. In Study 1, we found that this facilitating function had
negative effects on learning, because the external support had made processing unnecessarily
easy for our students. We therefore expected that simulation pictures would result in lower
performance with circumnavigation questions than manipulation pictures and that this effect
will be more pronounced when students have low rather than high learning prerequisites. The
following hypothesis was therefore derived for students with low learning prerequisites:
(H4a) Simulation pictures lead to lower performance in answering circumnavigation
questions than manipulation pictures, if learners have low learning prerequisites.
We furthermore predicted an interaction between picture type and learning prerequisites
according to the following hypothesis:
(H4b) Simulation pictures are more harmful for answering circumnavigation questions
(compared to manipulation pictures) for learners with low learning prerequisites than
for learners with high learning prerequisites.
Method
Participants of Study 2 were 26 university students who were randomly assigned to 2
groups. 13 students were assigned to the manipulation group, and 13 were assigned to the
simulation group. The learning material was the same as in Study 1 except that the
Enabling, Facilitating, and Inhibiting Effects 13
manipulation group received a text that included only 5 manipulation pictures, whereas the
simulation group received a text that included only 5 simulation pictures. The procedure of
Study 2 was exactly the same as in Study 1. In the final post-test phase, participants were
again asked to apply the acquired knowledge in a comprehension test without further access
to the learning material or to notes. The test included 12 time difference questions and 12
circumnavigation questions. For each participant the number of correctly answered time
difference questions and the number of correctly answered circumnavigation questions were
determined.
Results
In order to dierentiate between participants with high and
participants with low learning prerequisites, regression analyses were
performed with prior knowledge and intelligence as predictors and
learning results (sum of correctly answered time dierence questions and
sum of correctly answered circumnavigation questions) as the dependent
variables. Based on the resulting linear combination of prior knowledge
and intelligence, participants were assigned through a median split to a
group of 14 learners with high learning prerequisites and a group of 12
learners with low learning prerequisites. Table 2 shows the means and standard
deviations of the time difference scores and the circumnavigation scores of the manipulation
group and the simulation group. It further differentiates between high and low learning
prerequisites.
-------------------------------------
TABLE 2 ABOUT HERE
-------------------------------------
A 2 x 2 ANOVA of the time-difference scores with the factors animation type and
learning prerequisites showed a marginally significant effect of animation type (F(1, 22) =
1.743, MSE = 2.501, p = .10), a highly significant effect of learning prerequisites (F(1, 22) =
10.211, p = .002,
2 = .317) and a significant interaction Animation Type x Learning
Prerequisites (F(1, 22) = 4.511, p = .023,
2 = .170). Students with manipulation pictures
outperformed students with simulation pictures, and students with high learning prerequisites
performed better than students with low learning prerequisites. When learners had high
Enabling, Facilitating, and Inhibiting Effects 14
learning prerequisites, they had significantly higher time-difference scores after learning from
manipulation pictures than after learning from simulation pictures (t(12) = 2.287, p = .021, d
= 1.22), whereas learners with low learning prerequisites had lower scores with manipulation
pictures than with simulation pictures. Thus, the results support hypotheses H3a and H3b.
Accordingly, manipulation pictures have an enabling function that is helpful for answering
time difference questions, but only if learners have sufficiently high learning prerequisites.
A 2 x 2 ANOVA of the circumnavigation scores with the factors animation type and learning
prerequisites yield a significant effect of animation type (F(1, 22) = 5.020, MSE = 2.896, p
= .018,
2 = .186) and a significant effect of learning prerequisites (F(1, 22) = 4.109, p = .028,
2 = .157), but no significant interaction Animation Type x Learning Prerequisites (F(1, 22) =
0.558, n.s.). Learners with simulation pictures showed lower performance than learners with
manipulation pictures, and students with high learning prerequisites outperformed students
with low learning prerequisites. Due to the non-significant interaction Animation Type x
Learning Prerequisites, there was no support for hypothesis H4a. However, students with low
learning prerequisites had lower performance in answering circumnavigation questions after
learning from simulation pictures than after learning from manipulation pictures (t(10) =
2.928, p = .008, d = 1.70). This corresponds to hypothesis H4b, which assumed that
simulation pictures result in lower performance with circumnavigation questions than
manipulation pictures, especially if learners have low learning prerequisites. Thus, simulation
pictures seem to have a facilitating function especially for students with low learning
prerequisites, which affect the answering of circumnavigation questions. However, this
function turned out again to be harmful for these learners, because the external support had
made processing unnecessarily easy.
Discussion
The findings should be interpreted with caution due to the relatively small number of
participants.The results indicate that the different kinds of animations have indeed different
functions in the process of learning. Whereas the manipulation pictures seem to have
primarily an enabling function, the simulation pictures seem to have primarily a facilitating
function. Manipulation pictures seem to be primarily beneficial for answering time-difference
questions. Learners can use such pictures to generate various time states of the earth in order
to extract information about time differences, which was obviously helpful later for answering
Enabling, Facilitating, and Inhibiting Effects 15
time-difference questions. This function seems to be especially pronounced when students
have higher learning prerequisites, because these learners have sufficient resources available
to use these possibilities (cf., Clarke, Ayres, & Sweller, this issue). Simulation pictures seem
to affect primarily the answering of circumnavigation questions. They have a facilitating
function insofar as they allow following an external simulation process that makes the
corresponding mental simulation much less demanding. This function might be beneficial for
learners who would not be able to perform this mental simulation at all without external
support (cf., Mayer, 1997, 2001; Salomon, 1994; Schnotz, Boeckheler, & Grzondziel, 1999).
However, if learners are able to perform the mental simulation on their own, the external
support prevents students from performing learning-relevant cognitive processes on their own.
In this case, the facilitating function is beneficial for processing, but not for learning.
General Discussion
Compared to static pictures, animated pictures provide additional information that
seems to have different functions for learning. On the one hand, animations can enlarge the set
of possible cognitive processes and, thus, allow the learner to perform more processing than
s/he would be able to perform with static pictures. This is the enabling function of animations.
On the other hand, animations can trigger dynamic cognitive schemata that make specific
cognitive processes easier. This is the facilitating function of animations.
Different kinds of animated pictures seem to fulfil different functions for learning.
Manipulation pictures which allow to generate and display a large number of static pictures,
showing different states or showing a subject matter from different perspectives, seem to have
primarily an enabling function. They enable a learner to perform more cognitive processing
than s/he would be able to do with static pictures. Simulation pictures, on the contrary, which
allow displaying dynamic processes, seem to have primarily a facilitating function. They
provide external support for corresponding mental simulations and, thus, make these mental
processes easier to perform. Individuals with high learning prerequisites seem to benefit
primarily from the enabling function, whereas individuals with low learning prerequisites
seem to be affected primarily by the facilitating function of animations.
Both the enabling function and the facilitating function of animation can be considered
as a reduction of cognitive load (Sweller, van Merriënboer, & Paas, 1998). The facilitating
function of animations can be helpful for learners with very low ability or prior knowledge
Enabling, Facilitating, and Inhibiting Effects 16
who would not be able at all to perform the corresponding mental simulations without
external support (cf., Wallen, Plass, & Brünken, this issue). The two studies presented above,
however, have shown that the facilitating function of animations can also be harmful, namely
if learners, who would be already able to perform the mental simulations on their own,
nevertheless make use of the unneeded external support. Animation can keep learners from
doing relevant cognitive processing not due to increased task difficulty, but due to an
inappropriate facilitation of the task. In this case, the animation reduces cognitive load, but
unfortunately it reduces germane load that is necessary for learning instead of the extraneous
load. The use of animation in multimedia learning environments seems to be beneficial only
under some circumstances, whereas it can have negative effects under other circumstances.
When generalizing these results, one should keep in mind, however, that the
distinction between high and low learning prerequisites is always relative. Only those learners
who try to perform their own mental simulations can be hindered by unneeded help or suffer
from interference with the external animation. Learners with low cognitive abilities may not
even try to perform a mental simulation without external support by an animation. If no
learning occurs from such mental simulations with static pictures or manipulation pictures,
then simulation pictures cannot be harmful simply because there cannot be less learning than
no learning.
As animation provides additional and transient information, one could also argue that
animation does not decrease, but rather increases cognitive load as an effect of redundancy. It
is well known that presenting redundant information can be considered as an increase of
extraneous cognitive load, because learners have to process additional, but unneeded
information. Such redundancy can result in an expertise reversal effect, when individuals with
higher learning prerequisites perform better without, rather than with, additional information
(Kalyuga, Chandler, & Sweller, 1998; Kalyuga & Sweller, this issue).
In the studies presented above, the negative effects of animation derived from the
facilitating function were found primarily when students had low learning prerequisites rather
than high learning prerequisites. Consequently, this pattern of results is not consistent with an
expertise reversal effect Nevertheless, it is a weakness of the studies that they did not directly
measure cognitive load by a rating scale (Paas & van Merriënboer, 1994), but used cognitive
load theory only as an interpretative framework. A direct measure of cognitive load in further
studies could help to decide whether the negative effects of animation can be attributed to an
Enabling, Facilitating, and Inhibiting Effects 17
increase of extraneous cognitive load due to redundancy or to an unintended decrease of
germane cognitive load due to an inappropriate facilitation of the task. An important aspect
that has been ignored in this study is how a learner’s willingness to invest mental effort into a
learning task is affected by the cognitive load associated with the task (cf., Paas, Tuovinen,
van Merriënboer, & Darabi, this issue). Another topic of interest would be the different roles
played by prior knowledge (which could be considered as the learner’s inventory of cognitive
schemata) and intelligence (which could be considered as the quality of cognitive processing
based on these schemata) in learning from animation. Further research will be required to
clarify these questions.
Enabling, Facilitating, and Inhibiting Effects 18
References
Amthauer, R. (1973). Intelligenz-Struktur-Test 70 (IST 70). Göttingen, Germany: Hogrefe.
Clarke, T., Ayres, P., & Sweller, J. (this issue). The impact of sequencing and prior
knowledge on learning mathematics through spreadsheet applications. Educational
Technology, Research & Development.
Kalyuga, S., Chandler, P., & Sweller, J. (1998). Levels of expertise and instructional design.
Human Factors, 40, 1-17.
Kalyuga, S., & Sweller, J. (this issue). Rapid dynamic assessment of expertise to improve the
efficiency of adaptive e-learning. Educational Technology, Research & Development.
Lowe, R. K. (1999). Extracting information from an animation during complex visual
learning. European Journal of Psychology of Education, 14, 225-244.
Mayer, R. E. (1997). Multimedia learning: Are we asking the right questions? Educational
Psychologist, 32, 1-19.
Mayer, R. E. (2001). Multimedia learning. New York: Cambridge University Press.
Moreno, R., & Valdez, F. (this issue). Cognitive load and learning effects of having students
organize pictures and words in multimedia environments: The role of student interactivity
and feedback. Educational Technology, Research & Development.
Paas, F., & van Merriënboer, J. J. G. (1994). Instructional control of cognitive load in the
training of complex cognitive tasks. Educational Psychology Review, 6, 357-371.
Paas, F., Tuovinen, J. E., van Merriënboer, J. J. G., & Darabi, A. (this issue). A motivational
perspective on the relation between mental effort and performance: Optimizing learners’
Enabling, Facilitating, and Inhibiting Effects 19
involvement in instructional conditions. Educational Technology, Research &
Development.
Salomon, G. (1994). Interaction of media, cognition, and learning. Hillsdale, NJ: Erlbaum.
Schnotz, W., Boeckheler, J., & Grzondziel, H. (1999). Individual and co-operative learning
with interactive animated pictures. European Journal of Psychology of Education, 14,
245-265.
Sims, V. K., & Hegarty, M. (1997). Mental animation in the visuospatial sketchpad: Evidence
from dual-task studies. Memory & Cognition, 25, 321-332.
Sweller, J. (1999). Instructional design in technical areas. Camberwell, Australia: ACER
Press.
Sweller, J., & Chandler, P. (1994). Why some material is difficult to learn. Cognition and
Instruction, 12, 185-223.
Sweller, J., van Merriënboer, J. J. G., & Paas, F. (1998). Cognitive architecture and
instructional design. Educational Psychological Review, 10, 251-296.
Tversky, B., Morrison, J. B., & Betrancourt, M. (2002). Animation: Can it facilitate?
International Journal of Human-Computer Studies, 57, 247-262.
Van Gog, T., Ericsson, K. A., Rikers, R. M. J. P., & Paas, F. (this issue). Instructional design
for advanced learners: Establishing connections between the theoretical frameworks of
cognitive load and deliberate practice. Educational Technology, Research & Development.
Van Merriënboer, J. J. G. (1997). Training complex cognitive skills. Englewood Cliffs, NJ:
Educational Technology Publications.
Enabling, Facilitating, and Inhibiting Effects 20
Wallen, E., Plass, J. L., & Brünken, R. (this issue). The function of annotations in the
comprehension of scientific texts – Cognitive load effects and the impact of verbal ability.
Educational Technology, Research & Development.
Address for correspondence:
Wolfgang Schnotz
Dept. of General and Educational Psychology
Thomas-Nast-Str. 44
D-76829 Landau, Gemany
schnotz@uni-landau.de
Enabling, Facilitating, and Inhibiting Effects 21
Table 1
Means and Standard Deviations of Picture Observation Times and Learning Results in Study 1.
Mean Standard Deviation Sample Size
Learning Static Anim. Total Static Anim. Total Static Anim.
prerequisites Pictures Pictures Pictures Pictures Pictures Pictures
Picture Observation Times (seconds)
Whole sample 1728 1559 1643 1638 661 1236 20 20
Low 2310 1463 1887 1998 542 1489 10 10
High 1145 1656 1400 960 780 891 10 10
Learning Results: Time Difference Questions
Whole sample 4.10 6.65 5.38 2.38 3.31 3.13 20 20
Low 3.20 5.50 4.35 2.20 3.31 2.98 10 10
High 5.00 7.80 6.40 2.31 3.05 3.00 10 10
Learning Results: Circumnavigation Questions
Whole sample 8.95 7.70 8.33 2.86 2.27 2.63 20 20
Low 9.40 6.90 8.15 2.95 1.52 2.62 10 10
High 8.50 8.50 8.50 2.84 2.68 2.69 10 10
Enabling, Facilitating, and Inhibiting Effects 22
Table 2
Means and Standard Deviations of Learning Results in Study 2.
Mean Standard Deviation Sample Size
Learning Manipul. Simul. Total Manipul. Simul. Total Manipul. Simul.
prerequisites Pictures Pictures Pictures Pictures Pictures Pictures
Learning Results: Time Difference Questions
Whole sample 3.62 2.69 3.15 2.06 1.84 1.97 13 13
Low 1.83 2.33 2.08 1.17 1.51 1.31 6 6
High 5.14 3.00 4.07 1.21 2.16 2.02 7 7
Learning Results: Circumnavigation Questions
Whole sample 5.46 4.00 4.73 1.90 1.68 1.91 13 13
Low 5.00 3.00 4.00 1.41 0.89 1.54 6 6
High 5.86 4.86 5.36 2.27 1.77 2.02 7 7
Enabling, Facilitating, and Inhibiting Effects 23
Figure Captions
Figure 1. Example of a manipulation picture.
Figure 2. Example of a simulation picture.
Enabling Effects
Figure 1
Enabling Effects
Figure 2