ArticlePDF Available

Abstract and Figures

In 7 free recall experiments, the benefit of creating drawings of to-be-remembered information relative to writing, was examined as a mnemonic strategy. In Experiments 1 and 2, participants were presented with a list of words and asked to either draw or write out each. Drawn words were better recalled than written. Experiments 3-5 showed that the memory boost provided by drawing could not be explained by elaborative encoding (deep level of processing (LoP)), visual imagery, or picture superiority, respectively. In Experiment 6, we explored potential limitations of the drawing effect, by reducing encoding time, and increasing list length. Drawing, relative to writing, still benefited memory despite these constraints. In Experiment 7, the drawing effect was significant even when encoding trial types were compared in pure-lists between-participants, inconsistent with a distinctiveness account. Together these experiments indicate that drawing enhances memory relative to writing, across settings, instructions, and alternate encoding strategies, both within- and between-participants, and that a deep LoP, visual imagery, or picture superiority, alone or collectively, are not sufficient to explain the observed effect. We propose that drawing improves memory by encouraging a seamless integration of semantic, visual, and motor aspects of a memory trace.
Content may be subject to copyright.
The drawing effect: Evidence for reliable and robust
memory benets in free recall
Jeffrey D. Wammes, Melissa E. Meade, and Myra A. Fernandes
Department of Psychology, University of Waterloo, Waterloo, ON, Canada
(Received 23 December 2014; accepted 31 August 2015)
In 7 free-recall experiments, the benet of creating drawings of to-be-remembered information relative
to writing was examined as a mnemonic strategy. In Experiments 1 and 2, participants were presented
with a list of words and were asked to either draw or write out each. Drawn words were better recalled
than written. Experiments 35 showed that the memory boost provided by drawing could not be
explained by elaborative encoding (deep level of processing, LoP), visual imagery, or picture superiority,
respectively. In Experiment 6, we explored potential limitations of the drawing effect, by reducing
encoding time and increasing list length. Drawing, relative to writing, still beneted memory despite
these constraints. In Experiment 7, the drawing effect was signicant even when encoding trial types
were compared in pure lists between participants, inconsistent with a distinctiveness account.
Together these experiments indicate that drawing enhances memory relative to writing, across settings,
instructions, and alternate encoding strategies, both within- and between-participants, and that a deep
LoP, visual imagery, or picture superiority, alone or collectively, are not sufcient to explain the
observed effect. We propose that drawing improves memory by encouraging a seamless integration
of semantic, visual, and motor aspects of a memory trace.
Keywords: Subject-performed tasks; Memory; Drawing; Imagery; Levels of processing.
In the current study, we rst sought to determine
whether drawing was an efcacious strategy for
boosting later retention and memory performance.
Previous work has indicated that many subject-per-
formed tasks, such as production (MacLeod,
Gopie, Hourihan, Neary, & Ozbuko, 2010), gener-
ation (Slamecka & Graf, 1978), and enactment
(Guttentag & Hunt, 1988), carried out during
encoding can provide a memorial benet relative
to more passive encoding strategies such as silent
reading. Though useful, these strategies may not
be practical in a typical learning environment such
as a classroom or lecture hall due to their disruptive
nature. For this reason, there is a need to nd prac-
tical unobtrusive techniques that people can apply
in their everyday lives to remember important
information, or that students can apply to
enhance retention. One traditional study approach
in the aforementioned circumstances would be to
write detailed notes based on the professors
chosen lecture topic, or writing down a list of to-
be-remembered information.
Correspondence should beaddressed toJeffrey Wammes, Department of Psychology, University of Waterloo, 200 University Ave W,
Waterloo, ON, N2L 3G1, Canada. E-mail: jwammes@uwaterloo.ca
The authors would thank Grace Sim, Liat Koer, and Terri Middleton for their dedication in completing this project, as well as
insightful comments and suggestions for our work.
This work was funded by a scholarship from the Natural Science and Engineering Research Council (NSERC) awarded to author
J.W.; and an NSERC Discovery grant awarded to author M.F.
© 2016 The Experimental Psychology Society 1
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
http://dx.doi.org/10.1080/17470218.2015.1094494
Downloaded by [University of Waterloo] at 06:13 22 February 2016
Drawing and retention
Given that many studies show memorial benets
for information presented as pictures compared to
words (e.g., Paivio, Rogers, & Smythe, 1968), it
stands to reason that creating a drawing of to-be-
learned information may lead to superior long-
term retention relative to a more commonplace
note-taking strategy. Pulling on a similar thread
to that of previous researchers (Guttentag &
Hunt, 1988; MacLeod et al., 2010; Slamecka &
Graf, 1978) who explored the anecdotal notion
that there was an advantage to learning by doing,
we sought to determine whether drawing provided
a measurable advantage over passive note-taking.
Surprisingly, research corroborating or even
testing this intuition is sparse. Some educational
studies have touched on the matter, but often
using educational paradigms in which drawing is
confounded with a number of other additional
learning tools, such as the provision of visual aids
such as backgrounds and cutouts to assemble
(Lesgold, De Good, & Levin, 1977;
Schwamborn, Mayer, Thillmann, Leopold, &
Leutner, 2010), existing images upon which to
model their drawings, or additional time on task
for the drawing relative to control conditions (see
Van Meter & Garner, 2005, for a review). The
latter example makes it difcult to ascertain
whether the benet of drawing was simply a func-
tion of the total time spent studying each item (Van
Meter, 2001), a factor known to bolster memory
(Cooper & Pantle, 1967). The goal of the current
work was to rst approach this question in its
most basic form and then build to a comparison
with other common learning techniques (visual
imagery, elaborative encoding, and picture super-
iority). The reasoning behind this was twofold:
rst, to demonstrate that drawing may be a better
alternative to more extensively researched mnemo-
nic strategies, and second, because each could be
argued as one potential mechanism through
which drawing inuences memory. Lastly, we
examined potential boundary conditions, by
testing whether any benet of drawing was
reduced by dramatically shortening the encoding
duration, by increasing the length of the study
list, or when the encoding strategy was manipulated
between participants rather than intermixed and
within participants. The latter case directly tests a
distinctiveness account as an explanatory mechan-
ism for any benet conferred by drawing at
encoding.
The rst allusion to a potential benetof
drawing on later memory was suggested by the
dual-coding hypothesis by Paivio and Csapo
(1973). They showed a memorial advantage on a
free-recall test for words that were drawn rather
than written out during encoding. However, they
gave participants under 5 s to create a rough
sketch, which was probably too brief for partici-
pants to produce a complete drawing, and thus
may have underestimated the extent to which
drawing benets memory performance. Further,
because participants were asked to draw pictures,
an unusual task, one cannot rule out a distinctive-
ness account to explain the advantage. This
account, invoked as a partial explanation for pro-
duction effects (MacLeod et al., 2010), holds that
performing an encoding task for a select subset of
items adds an extra layer of encoding that dis-
tinguishes these from others, therefore making
them more memorable (Conway & Gathercole,
1987). This distinctiveness explanation is bolstered
by the fact that participants were only asked to write
the word once as their control condition (Paivio &
Csapo, 1973), which in addition to being less dis-
tinct, can be completed quicker, is arguably easier
to carry out, and is less engaging than drawing,
leaving additional time for the mind to stray from
the encoding task at hand. Accordingly, their task
manufactured an inherent disadvantage to their
writing condition in that the total time spent study-
ing each word may not have been equated across
their writing and drawing conditions.
Later work (Peynirciog˘lu, 1989) reincarnated
the study of drawing as a facilitator of memory
for scenes, though in lieu of more typical retrieval
tasks, participants were required to later draw the
studied scenes from memory. Scenes were graded
as correct retrievals only if they were drawn
exactly as initially presented. They found a benet
of drawing at encoding; however, it is difcult to
determine whether the observed drawing benet
2THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
was due to the act of drawing, or simply attributable
to transfer-appropriate processing (Morris,
Bransford, & Franks, 1977). In addition, neither
of the previously discussed studies compared
drawing directly to any other competing encoding
strategies.
Evidence from subject-performed tasks and
picture superiority
Many subject-performed tasks, which involve some
additional activity at encoding, have been shown to
boost memory performance. As such, it is worth
considering whether drawing would lead to the
same memorial benets. Encoding tasks requiring
a deep level of processing (LoP), for example,
were some of the rst to be explored in detail.
Craik and Lockhart (1972) suggested that encod-
ing that required deep semantic processing (e.g.,
deciding whether a word ts in a sentence) would
lead to superior later memory performance relative
to shallower perceptual-based processing (e.g.,
tracking whether letters in word are in upper or
lower case). In the rst description of another
subject-performed task, generation, Slamecka and
Graf (1978) found better memory for the words
generated from a cue, which was a synonym of
the target word, and the target wordsrst letter
(e.g., rapid-f___), relative to words that were
simply read (see Bertsch, Pesta, Wiscott, &
McDaniel, 2007, for a review). Previous work has
also explored the benets of reading a word aloud
during encoding, relative to silently, and showed a
robust production effectthat manifests in both
within- (MacLeod et al., 2010) and between-par-
ticipants designs (though to a much lesser extent;
Fawcett, 2013) and even extends to writing the
word (Forrin, MacLeod, & Ozubko, 2012).
Finally, one of the strongest subject-performed
tasks known to greatly enhance memory is enact-
ment. In this manipulation, participants are asked
to perform a movement associated with the
studied word, and this has been shown to provide
a boost to subsequent memory performance
(Guttentag & Hunt, 1988). Given the extensive
research documenting these methods of enhancing
memory, it is surprising that few have considered a
potentially easy-to-use means to do so, such as
drawing. Currently, however, there are no compre-
hensive explorations of the effectiveness of drawing
as a viable encoding manipulation. Accordingly,
our research aims to ll this void by empirically
testing the hypothesis that drawing produces sig-
nicant benets to memory performance.
Beyond the efcacy of subject-performed tasks,
there are other theoretical streams that provide
encouraging evidence in support of our hypothesis.
The nding that images are generally better
remembered than words, termed the picture super-
iority effect, has been well supported and replicated
in the literature, consistent across various method-
ologies, paradigms, and demographic groups
(Hockley, 2008; Kinjo & Snodgrass, 2000;
Maisto & Queen, 1992; Mintzer & Snodgrass,
1999; Paivio & Foth, 1970; Paivio et al., 1968;
Rowe, 1972; Snodgrass & McClure, 1975; but
see Boldini, Russo, Punia, & Avons, 2007;
Vaidya & Gabrieli, 2000; Weldon & Roediger,
1987). Paivios(1971,1991) dual-code theory
suggests that pictures are better remembered than
words because they are represented both visually
and verbally. In our paradigm, while an image is
not directly provided to our participants, their
encoding task is to create one. As such, this
manipulation ought to lead to benets to memory
similar to that observed in the picture superiority
effect, by encouraging visualization of the word.
Why would drawing boost memory?
The available evidence, though conicted, foresha-
dows that the current work may demonstrate a
memory benet as a result of drawing. If the
effect is reliable, the unsolved problem remains, of
determining why this would be the case. As men-
tioned, many subject-performed tasks that result in
improved item-specic memory (e.g., production,
generation, enactment) are presumed to be driven
by a distinctiveness mechanism (MacLeod et al.,
2010;Mulligan,2002), wherein the items that
benet from various such mnemonic strategies do
so relatively, and at the expense of theother, less dis-
tinct items (e.g., read silently, passive viewing).
This account was explored in Experiment 7. Prior
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 3
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
to this, however, we examined the extent to which
the benet of drawing may be driven by one of
the following individual components. Specically,
we argue that the process of drawing a verbal item
engages at least four components: (a) elaboration,
(b) visual imagery, (c) motor action, (d) pictorial
representation. We hypothesize that in order to
transfer a verbal item into a drawn visual represen-
tation, participants must rst generate some physical
characteristics of an item (elaboration), create a
visual image of the item (visual imagery), engage
in the actual hand movements required of drawing
(motor action), and then are left with the picture
as a memory cue for later retrieval.
The current work delved rst into whether
drawing words at encoding would lead to improved
later recall relative to writing, which has been
described as a distinctiveness-based strategy in the
production effect family (Forrin et al., 2012). The
writing trial type was chosen in part to temper
any distinctiveness-based explanations, and in
part to control for the need for motor action
(Component 3) involved in drawing. In
Experiment 1, we measured the relative benetof
drawing using two different sets of drawing instruc-
tions. As our study was motivated in part by an
interest in typical learning environments, in
Experiment 2 we investigated whether any benet
of drawing would apply when the study phase was
completed in a large lecture hall, where actions
taken at encoding may be disruptive to others in
the room. In Experiments 3 through 5, while main-
taining our draw and write trial types to control for
motor action (Component 3 described above), we
directly contrasted these with three different
encoding trial types. Each was presumed to play a
fundamental role in the drawing process, and thus
a likely candidate to explain any benecial effects
of drawing. In Experiment 3, we contrasted the
drawing manipulation with a task requiring more
elaborative semantic processing (Component 1),
in order to rule out a deep level of processing as a
potential explanatory factor. It is possible that
drawing affords a participant memorial benets
because it encourages visualization of each word,
and it is this imagery, a well-established mnemonic
technique in its own right (e.g., Elliott, 1973), that
is driving the effect rather than the physical act of
drawing. To rule out this explanation, in
Experiment 4, we compared drawing directly to a
visual imagery encoding task (Component 2). In
Experiment 5 we ruled out a dual-code or picture
superiority mechanism for the drawing effect, by
comparing drawing directly to viewing a picture
of the word (Component 4) during encoding.
Next we probed for potential limitations of the
drawing effect by increasing the number of to-be-
remembered items and decreasing the encoding
time for each in Experiment 6. Lastly, in
Experiment 7, we contrasted drawing and writing
as between-participants conditions, to determine
directly whether distinctiveness, invoked to
explain the production effect (MacLeod et al.,
2010), was also the mechanism driving the
drawing effect.
We propose that in addition to encouraging
deeper semantic processing, visual imagery, and
picture superiority, the act of drawing provides
some mechanical information, akin to when
words are enacted. Accordingly, in all cases these
motor contributions (Component 3) are controlled
for by including repetitive writing as a control. Our
proposed mechanism for this drawing effect, which
is addressed in more detail in the General
Discussion, is that by encouraging the cohesion of
multiple modes of representation of target infor-
mation, we effectively create a more resilient
trace, which is robust in the face of changes in
setting, instructions, or competing encoding
strategies.
EXPERIMENT 1: INITIAL
DEMONSTRATION
The aim of Experiment 1 was to determine
whether, in a controlled and relatively simple para-
digm, drawing would provide a benet to later
memory performance. The basic framework of the
paradigm remains consistent across all experiments
in this study, with only minor deviations in task
instructions. In this initial experiment, recall of
incidentally encoded words was tested after a brief
retention interval. In Experiment 1A, during
4THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
incidental encoding, words were either drawn in
detail or written out multiple times. Based on pre-
vious work (Paivio, 1971; Paivio & Csapo, 1973),
we predicted that despite the inherent advantage
that production would afford written words
(Forrin et al., 2012), drawn words would be
better recalled than written ones.
We were concerned that instructing participants
to add detail to their drawing, but not their writing,
was implicitly biasing participants toward favouring
retention of the drawn items. Further, it is possible
that semantic satiation, or the loss of meaning that
occurs with repeated exposure, would occur with
repeated writing (Balota & Black, 1997). To
control for this confound, the instructions were
crossed over in Experiment 1B, such that the draw-
ings were to be done multiple times, and the
writing done once with the instruction to add
detail to the writing. We predicted that the advan-
tage of drawing would still manifest under these
instructions.
Method
Participants
Participants for Experiment 1A were 30 under-
graduate students (19 female), and for 1B were
25 undergraduate students (9 female) at the
University of Waterloo, who completed the exper-
iment for course credit or monetary remuneration.
Participants ranged in age from 18 to 47 years
(M= 20.67, SD = 4.15), with between 14 and 27
years of education (M= 16.72, SD = 2.22). All par-
ticipants had normal or corrected-to-normal vision,
and learned English before the age of seven.
Materials
Target items. An 80-item word list (see Appendix
A) was created from a selection of the verbal
labels for Snodgrass images (Snodgrass &
Vanderwart, 1980), to ensure that all words could
be easily drawn. Complex drawings were avoided
(e.g., clown) in favour of simpler items (e.g.,
apple). This measure was taken to reduce the
time it would take participants to create each of
the drawings; every word could be drawn in the
time provided, based on a pilot study, and no
item required excessive visual detail to be discern-
able. Words ranged in frequency between 1 and
25 (M= 8.23, SD = 6.44), in length between 3
and 11 letters (M= 5.56, SD = 1.79), and in
number of syllables from 1 to 4 (M= 1.63, SD =
0.72).
Filler task. A continuous reaction time task (CRT)
was created by making sound les representing
low-, medium-, and high-pitched tones. This was
done using Audacity software (Mazzoni &
Dannenberg, 2000), such that each sine wave
tone was exactly 500 ms long, at frequencies of
350, 500, and 650 Hz, respectively.
Questionnaires. We also asked participants to com-
plete the Vividness of Visual Imagery
Questionnaire (VVIQ), created by Marks (1973),
and three questions regarding participantshistory
of drawing. The VVIQ is a short questionnaire
that assesses individual differences in ability to
create a mental image of an item or scene.
Individuals are provided with four scenarios and,
for each, are asked to rate how clearly they can visu-
alize nuanced aspects of each of the scenarios on a
5-point scale. There were no reliable correlations
with either questionnaire throughout the course
of study. Accordingly, more detailed methodologi-
cal information and data regarding these corre-
lations are presented in the Supplemental Material.
Procedure
Participants completed the experiment individually
in a testing room. Stimulus presentation and
response recording were controlled using E-prime
v2.0 software (Psychology Software Tools Inc.,
Pittsburgh, PA) via an IBM computer with 17-
inch monitor. Instructions were presented in
English on screen and were also read aloud by the
experimenter. Participants were told that, depend-
ing on the promptword they saw, they were to
either drawor writethe subsequent word on
the pad of paper (14 cm ×21 cm) provided. In
Experiment 1A, a prompt of drawmeant the par-
ticipant was to draw a picture illustrating the word
on the screen and to continue adding detail until
their allotted time was exhausted. A prompt of
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 5
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
writemeant the participants were to clearly and
carefully write out the word multiple times. In
Experiment 1B, the instructions were crossed
over, such that drawmeant they were to repeat-
edly draw the item presented, while writemeant
they were to continue adding detail to a single iter-
ation of writing the word. This instruction was
unorthodox, but was included to potentially shift
emphasis to the write condition. Participants
response to this instruction (see Appendix B)
included using block letters, adding decorative
ourishes to their letters, shading, or incorporating
some elements of the concept into the writing (e.g.,
music notes on the word harp). They were
informed of time constraints for each item and
that they would hear a tone to warn them that
the next item would appear. Participants were not
told that their memory would be tested.
Encoding. Participants underwent a brief practice
phase in order to familiarize them with the encod-
ing phase, after which the experiment began.
Participants were not informed that they would
be required to complete a later memory test. This
incidental encoding paradigm was selected to
reduce the possibility that participants would
develop a strategy of preferentially focusing on
drawn items in anticipation of later testing. From
the list of 80 words, 30 were randomly selected to
be studied, a list unique for each participant. Of
these 30, 15 were randomly selected to be drawn,
and 15 written (see Appendix B for samples of
the outcome of each trial type). This set of words
was then presented in a randomized order, such
that drawn and written items were randomly inter-
mixed. On each trial, the prompt appeared in the
centre of the screen for 750 ms, followed by a
500-ms xation, after which the word to be
encoded appeared for 750 ms. Participants then
had 40 s to perform the encoding task, either
draw or write. A 500-ms tone alerted them that
the next item was forthcoming, after which they
had 3 s to ip their pad of paper to the next page
in preparation for the next prompt.
Retention. Following the encoding trials, partici-
pants were asked to perform the CRT as a ller
task. Tones were to be classied as low, medium,
or high, by pressing the 1, 2, or 3 key on a small
response pad. After hearing samples of each kind
of tone, participants proceeded to classify 60
tones, selected at random. For each trial, the tone
was played for 500 ms, after which participants
had 1500 ms to make their response, for a total of
2000 ms per trial. Thus the retention interval was
two minutes.
Retrieval. In the next phase of the experiment, par-
ticipants were asked to freely recall as many words
as they could, in any order, either written or
drawn, from earlier in the experiment. They were
given 60 s to complete their recall, which was
spoken aloud by the participant and recorded.
Questionnaires. Immediately following the retrieval
phase, participants completed a version of the
VVIQ (Marks, 1973) and the three questions per-
taining to drawing experience and ability (see
Supplemental Material).
Results and discussion
Participantsrecall output was sorted into the
number of recalled words that were drawn at
encoding and the number that were written. Each
of these values was divided by the total number of
words studied within each encoding trial type
(15), to create a proportional recall score. Data
were analysed in a 2 ×2 mixed measures analysis
of variance (ANOVA), with Experiment (1A,
1B) as a between-participants variable and encod-
ing (draw, write) as a within-participants variable.
Analyses showed a signicant main effect of encod-
ing trial type, F(1, 53) = 82.83, MSE = .02,
p,.001, η
2
= .61, such that drawn words were
recalled better than written words (see Figure 1).
The average number of times participants wrote
out the word in Experiment 1A (M= 16.71, SD
= 5.19) or drew the word in Experiment 1B (M=
2.96, SD = 3.85) was not correlated with the
number of recalled drawn items, recalled written
items, total recalled items, or the magnitude of
the benet of drawing, ps..21. Twenty-six of
the 30 participants in 1A showed the pattern of
6THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
improved recall for drawn relative to written items,
t(29) = 8.29, SE = .03, p,.001, d= 1.51, while 20
of the 25 did in 1B, t(24) = 4.81, SE = .04,
p,.001, d= 0.96. Hereafter, we use the term
drawing effectto refer to this distinct advantage
of drawing words relative to writing them out in
terms of later memory performance.
The main effect of experiment and the inter-
action were not signicant, allowing us to conclude
that rst, there was no overall difference in memory
performance depending on whether participants
were instructed to add detail to one trial type or
the other and, second, that the benet of drawing
did not differ depending on the instruction to
add detail or not. Such a nding rules out the con-
found that the instruction to add detail on drawn
trial types (Experiment 1A) could account for the
benecial effect of drawing on subsequent
memory, since the instruction to add detail to the
written trial types did not confer the same benet
(Experiment 1B).
First and most importantly, our results indicated
that there was a signicant recall advantage for
words that were drawn during incidental encoding
as compared to those that were written.
Participants recalled more than two times as
many drawn than written words, and most partici-
pants recalled more drawn than written words. This
nding expands upon previous work suggesting
some special advantage of drawing (Paivio &
Csapo, 1973; Peynirciog˘lu, 1989). Contrary to
previous work, however, we carefully controlled
the time our participants were given to complete
their drawings, thereby allowing sufcient time
for the creation of said drawing, and also equated
it with the time given for writing trial types.
Unlike in previous work, in which participants
wrote the word once then sat idle, we created trial
types wherein the entirety of the time allotted for
each trial type was most likely to be used. Our
design thus increased the likelihood that partici-
pants were processing the word throughout the
entire encoding trial, rather than engaging in
task-unrelated thoughts.
Second, because the drawinstruction in
Experiment 1A was to add detail to drawings to
ll the study time for each trial, while the write
instruction was to simply rewrite each word on a
given trial, we wanted to ensure that the drawing
benet was not just a result of this difference in
instruction. For example, it is possible that the
instruction to repeatedly write out the word on
each instance of the writetrial type was reducing
access to semantic information about the words,
thereby reducing subsequent memory (Balota &
Black, 1997). In Experiment 1B, despite removing
repetition in the writetrial types, we still found a
relative memorial benet for drawn words. When
participants were told to add detail to their
writing, presumably shifting the emphasis to the
writetrial type, and removing repetition, the
drawing effect was still evident, and the advantage
was shown in the majority of participants.
EXPERIMENT 2: GROUP TESTING
In Experiment 1 we showed that drawing, com-
pared to copying words during study, afforded par-
ticipants a signicant benet in their later recall. In
the next experiment we wanted to explore whether
this was a useful strategy that students could use in
a group classroom setting. Other mnemonic strat-
egies, such as production and enactment, could be
quite distracting in a group setting, making them
less suited for such public environments.
Accordingly, while the previous experiment was
completed individually in a testing room devoid
of distractions, we aimed to determine whether
Figure 1. Proportion of words recalled out of 15 in Experiments 1A,
1B, and 2, from the draw, list, and write trial types. Error bars
represent the standard error of each mean.
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 7
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
drawing could generalize to a group lecture hall
setting in Experiment 2.
It is possible that in this context, the drawing
effect would still manifest, but that it might be
smaller due to distraction from the activity or
movement of other participants in a classroom
setting. As well, due to the group setting, recall in
this experiment was written rather than aloud.
The transfer-appropriate processing theory
(Morris et al., 1977) would dictate that since
words were to be output using the same format as
that at study, a written memory test should lead
to a more robust reexperiencing of the encoding
experience for target words that were written than
for those that were drawn, providing them with a
memory boost. The end result of the change in
test format might therefore be a reduction in our
reported drawing effect. In addition to looking at
whether the drawing effect was maintained even
in the classroom setting, we analysed the combined
data from Experiment 1 and 2 in order to deter-
mine whether there were any possible interactions
based on test setting.
Method
Participants
Participants in Experiment 2 were 49 undergradu-
ate students (38 female) at the University of
Waterloo, who completed the experiment in
return for course credit or monetary remuneration.
Participants ranged in age from 17 to 24 years (M=
19.10, SD = 1.26), with between 13 and 18 years of
education (M= 14.78, SD = 1.12). All participants
had normal or corrected-to-normal vision and
learned English before the age of seven.
Materials
Study words and tones used for the ller task were
identical to those in Experiment 1.
Procedure
Participants were tested in two different group ses-
sions in a large lecture hall with stadium-style
seating for up to 126 students. One session con-
sisted of 15 and another of 34 undergraduate stu-
dents. Stimulus presentation and response
recording was controlled using E-prime v2.0 soft-
ware (Psychology Software Tools Inc., Pittsburgh,
PA) and displayed via projection screen at the
front of the lecture hall. Given the group setting,
some additional preliminary instructions were
given, to ensure participants did not interact with
one another or look at one anothers responses:
Participants were always separated by at least one
empty seat and were instructed to treat the exper-
iment as they would an examination.
The instructions from Experiment 1A were
used here, as the data from Experiment 1B ruled
out any differential effects due to instruction. Due
to the group setting, some obvious changes were
necessary to avoid participantsresponses interfer-
ing with othersfocus. Specically, responses in
the tone classication task were done using pen
and paper, as was their recall output and response
to the questionnaires. Apart from this, the pro-
cedure was identical to that of Experiment 1A.
Results and discussion
Data were analysed in a paired-samples ttest, with
encoding trial type (draw, write) as a within-partici-
pants variable. Analyses showed a signicant main
effect of encoding trial type, t(48) = 9.11, SE = .03,
p,.001, d= 1.30, such that drawn words were
better recalled than written words (see Figure 1).
Forty-three of the 49 participants showed this
pattern. Our results indicate that the drawing
effect reported in Experiment 1 is resilient to
changes in environment and is thus generalizable
to a group lecture setting. What they do not tell
us, however, is whether the change in setting
altered the size of the drawing effect. For this
reason, data from Experiments 1 and 2 were com-
pared for the following analyses.
Comparison with Experiment 1
Given that the group lecture hall setting is a much
more efcient way of testing participants and is
probably more analogous to an actual learning
environment in the real world, additional analyses
were conducted to ensure that our effect of
drawing was not interacting with the experimental
setting. Data from Experiments 1A and 1B, in
8THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
which participants were tested individually, and
Experiment 2, in which participants were tested
in groups, were combined into a pooled analysis.
A2×2 mixed measures ANOVA was conducted
with setting (individual, group) as a between-par-
ticipants variable, and encoding trial type (draw,
write), as a within-participants variable. There
was a signicant main effect of encoding trial
type, F(1, 102) = 165.06, MSE = .02, p,.001, η
2
= .62, such that drawn words were better recalled
than written words. Interestingly, the main of
setting was also signicant, F(1, 102) = 16.24,
MSE = .02, p,.001, η
2
= .14, such that those par-
ticipants who were tested in groups performed
much better than those tested individually, recalling
nearly two more words than their individually
tested counterparts. The Encoding Trial
Type ×Setting interaction, however, was not sig-
nicant, F(1, 102) = 0.001, MSE = .02, p..05,
η
2
= .00. In other words, the memory boost
afforded by drawing an item as compared to
writing it out was stable when the task was
moved from an individual to a group lecture
setting. This nding should serve to assuage any
concerns about the venue of testing. Accordingly,
our remaining empirical questions were often
addressed in a group setting.
The aim of the subsequent three experiments
was to directly contrast drawing with what we
believe to be primary components of the pro-
gression from a verbal item to a participant-
created drawing of that item, its visual referent.
As a reminder, these components include (a) elab-
oration, (b) visual imagery, (c) motor action, (d)
pictorial representation. In all cases, writing was
included as a baseline partially to control for dis-
tinctiveness, but primarily to control for the
motor action (Component 3) that drawing
necessitates.
EXPERIMENT 3: DEEP LEVEL OF
PROCESSING
Having ruled out the alternative account that our
instruction to add detail was producing our
drawing effect (in Experiment 1B), the aim of
Experiment 3 was to rule out another possible
mechanism driving the drawing effect. We sought
to determine whether the observed benet resulting
from drawing relative to writing words during a
study phaseoccurred because drawing encourages
a deeper, more elaborative level of processing. As
outlined by Craik and Lockhart (1972) in their
LoP framework, a stimulus that is encoded at a
deeperlevel is one that evokes a greater degree
of semantic analysis. This framework suggests
that items that undergo deeper encoding promote
enrichment or elaboration that leads to greater
success during later retrieval than for items that
were encoded shallowly. It follows that creating a
drawing of an item would probably require recog-
nition of the stimulus, then some elaboration in
order to recreate ones verbal representation in a
sketch. It has been suggested that this elaboration,
even in expert artists, draws rst from denotative
semantic information, including knowledge of
physical characteristics of the to-be-drawn item
rather than visual imagery (McMahon, 2002).
To rule out LoP as an explanation for the
drawing advantage, we considered the inuence
of a listtrial type during encoding, which required
participants to write a list of semantic character-
istics of the presented target word, thereby encoura-
ging more elaborative processing. It is important to
note that while most previous work investigating
the effect of a deep LoP had a companion
shallowLoP condition (e.g., indicating whether
a word has the letter eor g; Walsh & Jenkins,
1973), our paradigm included only the deep LoP
listtrial type, along with our original draw
and writetrial types. Notably, the writetrial
type is far from a shallow LoP condition, as even
writing a word once resulted in 13% more hits in
a recognition task than passive reading (Forrin
et al., 2012).
In a sense, we were also stacking the odds
against our drawing trial types in this experiment.
The words in the listtrial type were to be
encoded at a deep level, and in line with the
levels of processing framework, they should enjoy
memorial benets. Second, as in previous exper-
iments, transfer-appropriate processing (Morris
et al., 1977) and distinctiveness (Forrin et al.,
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 9
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
2012) should provide some benet to written items.
Lastly, previous work showed that a semantic
encoding task led to words being better remem-
bered than pictures (Vaidya & Gabrieli, 2000). It
follows from this result that because our encoding
task in the listtrial type required semantic proces-
sing, those words would show a stronger memorial
advantage than those in the drawtrial types at
encoding. If the drawing effect still occurs, which
we hypothesize it will based on our prior exper-
iments, it would indicate rmly that drawing is an
extremely robust and reliable encoding manipu-
lation, and one that cannot be explained by elabora-
tive encoding alone.
Method
Participants
Participants were 47 undergraduate students (40
female) at the University of Waterloo, who com-
pleted the experiment in return for course credit
or monetary remuneration. Participants ranged in
age from 17 to 51 years (M= 20.17, SD = 5.07),
with between 13 and 18 years of education (M=
15.26, SD = 1.37). All participants had normal or
corrected-to-normal vision and learned English
before the age of seven.
Materials
Word stimuli and tones were the same as those in
previous experiments.
Procedure
The procedure was identical to that used in
Experiment 2, except that a third trial type was
added during encoding. Participants were
instructed to draw,list,orwritethe word
being presented to them. Instructions for draw
and writewere identical to those in previous
experiments. For the listprompt, participants
were instructed to write out a list of physical charac-
teristics of the word presented until the time for
that trial type ended. Because there were now
three different prompts instead of two, the ran-
domly selected list of 30 words was divided into
three lists of 10 words each (10 to be drawn, 10
to be visualized, 10 to be written) instead of two
lists of 15 words of each trial type as in the prior
experiments. This set of words was then presented
in a randomized order, such that drawn, visualized,
and written items were randomly intermixed. Apart
from these modications, the experimental proto-
col was identical to that of Experiment 2.
Results and discussion
Data from one participant were excluded as they
performed the incorrect encoding instruction for
9 of the 30 words. Data from the remaining partici-
pants were analysed using repeated measures
ANOVA, with encoding trial type (draw, list,
write) as a within-participants variable. Results
indicated a signicant main effect of encoding
trial type, F(2, 92) = 21.20, MSE = .02, p,.001,
η
2
= .32. Paired-samples ttests using Bonferroni
adjusted alpha levels of .0167 per test (.05/3)
revealed that this was driven by signicantly
better recall for words in the draw than in both
the list, t(46) = 5.16, SE = .03, p,.001, d= 0.75,
and the write trial types, t(46) = 5.89, SE = .03,
p,.001, d= 0.86. The difference in recall
between words from the list and write trial types
was not signicant, t(46) = 0.39, SE = .03, p= .70,
d= 0.06 (see Figure 2). In the listtrial types, par-
ticipants listed a mean of 6.37 characteristics (SD =
1.65) per word, but the number of characteristics
listed was not correlated with the number of
Figure 2. Proportion of words recalled out of 10 in Experiment 3
from the draw, list, and write trial types. Error bars represent the
standard error of each mean.
10 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
words recalled overall, or in any of the three encod-
ing trial types, ps..11.
Results indicated again that drawing was superior
to both of the other encoding instructions (listing
physical characteristics or writing out the word).
Such a nding suggests that the effect of drawing
cannot be dismissed as another iteration of a classic
deep LoP manipulation. What is more, our deep
LoP manipulation, which required listingsemantic
characteristics describing the study word, showed no
signicant advantage over a writinginstruction.
Some might argue that our deep LoP trial type
ought to have improved memory performance relative
to writing out the studied word repeatedly. As men-
tioned previously, deep LoP manipulations typically
are compared to shallow LoP counterparts (Walsh
&Jenkins,1973), something that was absent in our
paradigm. The aforementioned research showed
that deep processing is superior to shallow, but our
writetrial type was not shallow in the traditional
sense, and this may explain why listwas not superior
to write.Toclarify,writingawordoutmultiple
times over the span of 40 s may have allowed partici-
pants to think more deeplyabout the word or visualize
the item, as the task of writing a word repeatedly is
probably not as cognitively demanding as the other
trialtypes.Accordingly,ourwritetrial type cannot
be categorically considered to be a shallow LoP trial
type, which may explain the absence of a signicant
difference in recall of words from the listrelative
to writetrial types. That the drawing effect was
larger than the benet conferred to words receiving
a deep LoP orientation at encoding, known to reliably
boost memory (Walsh & Jenkins, 1973), is a testa-
ment to the strength of the effect. Further, listing a
greater number of characteristics for any particular
word did not lead to better memory for that word,
suggesting that regardless of how deeply words were
encoded in the listing task, the benetofdrawingas
an encoding orientation remained superior.
EXPERIMENT 4: VISUAL IMAGERY
It was worth considering that the benet from
drawing at encoding occurred as a result of
adding visual imagery to the memory trace, as
dual-code theory suggests is the case when creating
a memory for a picture (Paivio, 1971). As dis-
cussed, drawing requires the translation of material
from a verbal into a visual code. While retrieving
physical characteristics, as in Experiment 3, is a
likely contributor to this translation process, it
seems intuitive that participants may also engage
visual imagery as a subsequent or even concurrent
step. One must create a mental image of their pro-
totypical visual representation of that word, so that
they can attempt to reproduce this prototype
through drawing. Visual imagery alone has been
shown to improve memory performance relative
to rote repetition (Elliott, 1973; Kieras, 1978;
Lupiani, 1977; Winnick & Brody, 1984), so it is
very possible that visual imagery is a contributor
to, and potential explanatory mechanism for, any
benet documented as a result of drawing.
As discussed, Paivio (1971) posited that
memory for pictures was superior to that of words
due to the fact that they invoke both the readily
apparent visual representation and also the rep-
resentation of the pictures verbal referent. In
other words, upon seeing a picture of a horse, one
readily retrieves the word horseas well, providing
two codes for memory. Thus, this theory suggests
that disparate memory performance for target pic-
tures and words is driven by differences in visual
imagery. By encouraging participants to draw a
studied word, we are coaxing a detailed visual rep-
resentation in addition to their existing verbal code,
thereby asking participants to create a (parallel)
dually coded memory trace.
In order to rule out the possibility that the
drawing effect is simply a residual outcome of
differential invocation of visual imagery, we intro-
duced an additional competing encoding instruc-
tion to visualizeor engage in mental imagery of
a study word, to compare to the trial types in
which drawor writeinstructions were given.
The dual-code theory (Paivio, 1971) led us to
hypothesize that the instruction to create a mental
image would boost subsequent memory, relative
to the instruction to simply write out the word.
Importantly, though we speculated that creating
a mental image would boost memory, we predicted
that this boost would be smaller than that from our
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 11
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
drawing manipulation. This is because with the
drawing instruction, participants must not only
create a mental image of the word they are pre-
sented with and process the meaning of this
word, but also undergo the mechanistic process of
moving the pencil to create the image, which may
be akin to a muted enactment effect. In other
words, whereas the visualizetrial type encourages
mental imagery, the drawtrial type not only
encourages imagery, but also a deeper level of pro-
cessing and potentially a form of enactment.
Additionally, the participant is left with an actual
picture to use as a cue once they are nished
drawing. By requiring integration of these aspects
of the memory trace, we expect that drawing will
have an advantage over typical writing or visual
imagery alone.
Method
Participants
Participants were 28 undergraduate students (21
female) at the University of Waterloo, who com-
pleted the experiment in return for course credit
or monetary remuneration. Participants ranged in
age from 19 to 23 years (M= 20.29, SD = 1.01),
with between 14 to 18 years of education (M=
15.86, SD = 1.04). All participants had normal or
corrected-to-normal vision and learned English
before the age of seven.
Materials
Word stimuli and tones were the same as those in
previous experiments.
Procedure
The procedure was the same as that of Experiment
3, except for one substitution. The prompt list
was replaced with visualize, such that the three
prompts were draw,visualize, and write.A
prompt of visualizemeant the participant was to
create a mental image of what the word represents.
They were instructed to maintain their focus on the
image, adding detail to it until that trial type time
had elapsed.
Results and discussion
Data were analysed using repeated measures
ANOVA, with encoding trial type (draw, visualize,
write) as a within-participants factor. Results indi-
cated a signicant main effect of encoding trial
type, F(2, 54) = 8.23, MSE = .04, p,.005, η
2
= .23. Paired-samples ttests using Bonferroni
adjusted alpha levels of .0167 per test (.05/3)
revealed that this was driven by signicantly
better recall for drawn trial types than for write
trial types, t(27) = 4.62, SE = .04, p,.001, d=
0.87, and marginally better recall for drawn than
for visualized trial types, t(27) = 2.20, SE = .06, p
= .037, d= 0.42. The difference in recall between
visualize and write trial types was not signicant,
t(27) = 1.58, SE = .05, p= .126, d= 0.30 (see
Figure 3).
Results from Experiment 4 suggest that instruc-
tions to create a mental image of a word during
encoding do not confer a memorial benet relative
to writing or drawing the word. We acknowledge,
however, that visual imagery is a manipulation
that presents difculties in verifying participant
compliance. As a reviewer pointed out, this could
result in lazy imaging or writing, akin to the lazy
reading hypothesis presented to explain the pro-
duction effect. This is a limitation inherent in the
duration provided for encoding, but a confound
we address, at least for the general drawing effect,
Figure 3. Proportion of words recalled out of 10 in Experiment 4
from the draw, visualize, and write trial types. Error bars
represent the standard error of each mean.
12 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
in Experiment 6. Importantly though, assuming
participants were compliant with our instructions,
drawing led to much better memory performance
than visualizing, Thus, we can rule out visual
imagery alone as a potential mechanism by which
drawing exerts its memorial benet and move
forward in exploring picture superiority as a poten-
tial mechanism.
EXPERIMENT 5: PICTURE
SUPERIORITY
Picture superiority, the reliable effect wherein pic-
tures are better remembered than words (Paivio
et al., 1968), was detailed earlier as a theoretical
reason for pursuing drawing as an encoding strat-
egy. It is perhaps the most obvious alternative
explanation for our reported drawing benet.
After having retrieved a set of physical character-
istics describing the to-be-drawn item, and gener-
ating a visual image of that item, participants
must produce the action necessary to produce the
drawing. Participants are then left with an image
to associate with its encoded verbal referent.
Accordingly, one could argue that any benet
observed as a result of drawing is simply due to
the creation of pictures, which we know are better
remembered than words. If this were the case,
one would expect that memory for drawn words
would be no different from memory for pictures
of those words.
In order to more directly assess whether picture
superiority could be the mechanism through which
drawing exerts its inuence, we replaced the visu-
alizecondition in Experiment 4 with a viewcon-
dition, wherein participants were instructed to
simply view an image of the studied word.
Method
Participants
Participants were 37 undergraduate students (29
female) at the University of Waterloo, who com-
pleted the experiment in return for course credit
or monetary remuneration. Participants ranged in
age from 18 to 23 years (M= 19.14, SD = 1.40),
with between 14 and 19 years of education (M=
14.78, SD = 1.16). All participants had normal or
corrected-to-normal vision and learned English
before the age of seven.
Materials
Word stimuli and tones were the same as those in
previous experiments. A picture was chosen for
each word stimulus from a picture set containing
thousands of unique images of objects (Brady,
Konkle, Alvarez, & Oliva, 2008). If the object
was not found in this stimulus set, one was retrieved
using a Google image search. All images were
cropped to the same dimensions (256 ×256
pixels) and were converted to greyscale using
ImageMagick software (ImageMagick Studio
LLC, 19992013).
Procedure
The procedure was the same as that of Exper-
iment 4, except for one minor change. The
prompt visualizewas replaced with view, such
that the three prompts presented during encoding
were draw,view, and write. For the view
prompt, participants were instructed to view a
picture representative of the to-be-remembered
word. In this condition, after the prompt and
word disappeared, a picture of what the word rep-
resents appeared on screen, and remained on
screen for the remainder of the encoding time.
Results and discussion
Data were analysed using repeated measures
ANOVA, with encoding trial type (draw, view,
write) as a within-participants variable. Results
indicated a signicant main effect of encoding
trial type, F(2, 72) = 7.24, MSE = .03, p,.005,
η
2
= .17. Paired-samples ttests using Bonferroni
adjusted alpha levels of .0167 per test (.05/3)
revealed that this was driven by signicantly
better recall for words in the draw than in both
the view, t(36) = 2.56, SE = .04, p,.0167, d=
0.42, and the write trial types, t(36) = 4.08, SE
= .04, p,.001, d= 0.67. The difference in recall
between words from the view and write trial types
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 13
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
was not signicant, t(36) = 1.07, SE = .04, p= .293,
d= 0.18 (see Figure 4).
Results indicate that drawing led to better recall
than viewing images of the to-be-remembered
words, and, therefore, the effects of picture super-
iority alone cannot be invoked as an explanation
for the observed benet of drawing on memory.
EXPERIMENT 6: BRIEF ENCODING
TIME
All experiments thus far have required that partici-
pants study 30 words, with 40 s to perform the
required encoding strategy for each word. Within
these 40 s, it is possible, as a reviewer pointed
out, that participants may not be engaging in the
required encoding strategy for the entirety of the
allotted encoding time. This has been described
as the problem of lazy reading in the production
effect, though when this concern was controlled
for, the effect still remained (MacLeod et al.,
2010). This is certainly a concern with our prior
experiments, especially in our less directly observa-
ble manipulations (visualize, view), but also argu-
ably in the writing encoding trial types as well.
Given this, the goal of Experiment 6 was to
address some of these concerns experimentally, as
well as to search for potential boundary conditions
of the drawing effect. Some might argue that the
40 s allotted for drawing, and the various encoding
trial types in the rst ve experiments, was excessive
and thus not as practical as other mnemonic strat-
egies that can be completed in periods of time
under 5 s. Further, we used relatively short lists in
the foregoing experiments as a direct result of our
lengthy encoding duration. To address these con-
cerns, in Experiment 6, the number of studied
items was more than doubled, and the encoding
duration for each item was reduced from 40 to
4 s. If the drawing effect was due to this long
encoding duration, or due to lazy writingover
the course of 40 s, then the drawing effect should
not be present when encoding duration is drasti-
cally reduced. With this methodological change,
the drawing effect might still occur; although we
predicted that the effect may be smaller, due to
the fact that the drawing created within a 4-s
time period would probably contain fewer details
than a drawing created in 40 s.
Method
Participants
Participants were 28 undergraduate students
(22 female) at the University of Waterloo, who
completed the experiment in return for course
credit or monetary remuneration. Participants
ranged in age from 18 to 25 years (M= 20.64,
SD = 1.90), with between 13 and 22 years of edu-
cation (M= 16.41, SD = 1.96). All participants
had normal or corrected-to-normal vision and
learned English before the age of seven.
Materials
Word stimuli were the same as those in previous
experiments. Using a notepad and ipping the
pages between each stimulus at encoding was not
conducive to this new rapid presentation of
stimuli. Accordingly, we switched to using a
Fisher-Price Doodle Pro®. This small drawing
pad has roughly the same drawable surface as did
our drawing pads (11 cm ×16 cm), but uses mag-
netic drawing technology similar to that of an
Etch-a-Sketch®. Unlike an Etch-a-Sketch®, the
Doodle Pro® comes with a small stylus to draw,
Figure 4. Proportion of words recalled out of 10 in Experiment 5
from the draw, view, and write trial types. Error bars represent
the standard error of each mean.
14 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
rather than knobs, making it more analogous to the
paper and pencil technique used in the rst ve
experiments. This product features a sliding knob
on the top, which quickly wipes clean the
drawing pad in preparation for the next trial.
Procedure
The procedure was very similar to Experiment 1A,
with the exception of some timing and presentation
changes to support the more rapid encoding time.
Additionally, in order to facilitate comparison
with investigations of other encoding strategies
thought to depend on distinctiveness (McDaniel
& Bugg, 2008), this experiment employed inten-
tional, rather than incidental, encoding. Like
Experiment 1A, participants were tested individu-
ally in a testing room. Participants studied 66
words instead of 30, split equally between the
write and draw encoding trial types. These 66
words were then presented with drawn and
written items randomly intermixed. During
encoding, rather than the typical protocol of
showing a participant a prompt (750 ms), a xation
(500 ms), and the word (750 ms), and then asking
them to complete the task (40 s), participants were
simply shown the word. Words were presented on a
black background in either white or red font, and
participants were instructed that the font color
indicated whether they were to draw or write the
item (with colors counterbalanced across partici-
pants). Each word was presented for 4 s, during
which time the participant was to complete the
task indicated by the font color of the word.
Between trials, a blank screen was shown for 1
second to give participants time to reset their
drawing pads.
The retention task from the previous experiment
was replaced with a 5-min tone classication and a
5-min visual CRT task, in which participants
identied within which three numbered boxes
(left, middle, or right) an asterisk was presented,
using the keys corresponding to the numbers pre-
sented above the boxes (1, 2, 3).
Participants were then given two minutes to
recall as many words as they could from the encod-
ing phase, by typing them in an input eld.
Participants heard a tone to notify them when
there were 20 s remaining in the recall test.
Participants did not complete the VVIQ or the
drawing-related questions after this experiment.
Results and discussion
First, the data were analysed with cue colour (red or
white) as a between-participants factor. There were
no main effects or interactions with cue colour, so
the data were collapsed across this variable. Data
were analysed in a paired-samples ttest, with
encoding trial type (draw, write) as a within-partici-
pants variable. Analyses showed a signicant main
effect of encoding trial type, t(27) = 12.01, SE
= .02, p,.001, d= 2.27, such that drawn words
were better recalled than written words (see
Figure 5).
These results indicate that even with a longer
study list, a much shorter encoding duration, and
a slightly longer retention interval between study
and test, drawn words were remembered much
better than written ones. Further, the effect was
substantially more pronounced (see Figure 5) than
in previous experiments (see Figures 14), indicat-
ing that the drawing encoding strategy is poten-
tially even more potent at shorter encoding
durations. It is worth noting, however, that it is
also possible that writing was simply a less potent
strategy at shorter encoding strategies, or that this
larger effect size was an artefact of the change
from incidental to intentional encoding.
Figure 5. Mean proportion of words recalled out of 33 for each trial
type in Experiment 6 (mixed lists) and out of 66 for each trial type in
Experiment 7 (pure lists). Error bars represent the standard error of
each mean.
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 15
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
Comparisons between the results of Experiment 6
and those of the prior experiments may be proble-
matic because the paradigms differ in more ways
(e.g., list length, intentional learning) than simply
encoding duration.
EXPERIMENT 7: BETWEEN
PARTICIPANTS
Having ruled out emphasis on adding detail to a
particular encoding trial type (the instruction to
continue adding detail to drawing, or writing in
Experiment 1), and each of the three components
proposed to be integral to the drawing process
(elaboration, visual imagery, and picture superiority
in Experiments 35), we turned our focus to exam-
ining whether the drawing effect can be accounted
for, like many other item-specic encoding strat-
egies (McDaniel & Bugg, 2008; Nairne, Riegler,
& Serra, 1991), by a distinctiveness account. We
maintain our proposal that drawing exerts its mem-
orial benets through integration of a number of
different types of memory codes into one cohesive
trace. Alternatively, though, drawing could be
exerting its effects by making some items more dis-
tinct than others. The current work draws obvious
parallels with the production effect, wherein item
memory for words read aloud is superior to that
for words read silently (MacLeod et al., 2010).
The mechanism commonly proposed to explain
this, and other item-specic mnemonic strategies,
is a distinctiveness account (Bodner & Taikh,
2012; MacLeod et al., 2010; McDaniel & Bugg,
2008; Ozubko & MacLeod, 2010). In other
words, in a mixed-list paradigm, some items are
read aloud (or generated from a cue, or enacted),
while others are simply read silently (or enacted
by experimenters while the participant passively
views; Engelkamp & Zimmer, 1997; Slamecka &
Graf, 1978). Accordingly, the trials that are more
unique, in which a participant completes a novel
or bizarre encoding task, have some distinctive
information that can later be used as an extra retrie-
val cue. The empirical support for this account
comes from a commonly observed absence or rever-
sal of most of these effects when trial types are
completed between participants, in pure lists, as
opposed to within participants. This is true of pro-
duction (MacLeod et al., 2010; but see Bodner,
Taikh, & Fawcett, 2014), generation (Slamecka
& Katsaiti, 1987), enactment (Engelkamp &
Dehn, 2000), and many other commonly cited
mnemonic strategies (McDaniel & Bugg, 2008;
Mulligan, 2002). While most of the foregoing nd-
ings used recognition, the experiments in the
current work used free recall. Some previous work
has explored how the production effect is modu-
lated depending on whether trial types are com-
pared within or between subjects. In contrast to
recognition, where some previous work has found
a signicant between-participants production
effect (e.g., Bodner et al., 2014), when employing
free recall, there is no evidence for the production
effect in pure lists (Jones & Pyc, 2014; Jonker,
Levene, & MacLeod, 2014). What this tells us is
that the benet of most of these effects manifests
as a relative enhancement, rather than an absolute
one. Specically, distinctive items are better
encoded at the expense of the less distinctive
ones. In an attempt to counter this potential
account of the observed drawing effect, throughout
all of our experiments thus far, we employed writing
as a baseline condition, with which to compare
drawing. This baseline was chosen as writing has
been previously described as an iteration of the pro-
duction effect, and as a task that is distinctive in its
own right (Forrin et al., 2012). Further, we com-
pared drawing with alternate encoding trial types
of visual imagery, elaborative encoding, and
picture superiority, which could also be categorized
under the distinctive encoding umbrella.
However, in order to conduct the most direct
experimental exploration of a distinctiveness
account, in Experiment 7 we compared drawing
and writing encoding strategies in pure lists, a
manipulation that typically eliminates the effects
of other encoding strategies that have been
described as relying on distinctiveness (McDaniel
& Bugg, 2008; Mulligan, 2002). With this, if the
benet of drawing is driven exclusively by distinc-
tiveness of drawn relative to written items, we
would not observe a benet of drawing when com-
pared between participants. If, however,
16 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
distinctiveness can only partially explain our nd-
ings, which we expect will be the case given the
large effect size we observed thus far, then results
should indicate an extant, albeit smaller, boost as
a result of drawing items, relative to writing them.
Method
Participants
Participants were 47 undergraduate students (37
female) at the University of Waterloo, who com-
pleted the experiment in return for course credit
or monetary remuneration. Participants ranged in
age from 18 to 24 years (M= 19.51, SD = 1.51),
with between 13 and 21 years of education (M=
15.55, SD = 1.58). All participants had normal or
corrected-to-normal vision and learned English
before the age of seven. Participants were randomly
assigned to be in the pure draw (n=24) or pure
write (n= 23) conditions.
Materials
All materials were the same as those in Experiment 6.
Procedure
The procedure was identical to that in Experiment 6,
including the intentional encoding, except that par-
ticipants were randomly assigned to one of two con-
ditions, instructing them to either draw all words
(pure draw) or write all words (pure write) presented
during study. Accordingly, there was no need for
colour coding of stimuli, and all study words were
presented in white font on a black background.
Participants still studied 66 total words, but rather
than 33 falling under each encoding trial type, all
66 were either drawn or written.
Results and discussion
Data were analysed using an independent-samples
ttest, with encoding trial type (draw, write) as a
between-participants variable. Analyses showed a
signicant main effect of encoding trial type,
t(45) = 3.63, SE = 0.02, p,.005, d= 1.08, such
that those in the pure draw condition recalled sig-
nicantly more words than those in the pure
write condition (Figure 5). It is worth noting,
however, that the effect is actually likely to be
larger than indicated, as one participant in the
pure write condition recalled a striking 33 words,
which is over 3.5 standard deviations above the
mean. When analysed without this outlier, the
effect was substantially larger, t(44) = 5.54, SE
= .02, p,.001, d= 1.63. Overall, these results
suggest that a distinctiveness account is not suf-
cient to explain the benet that drawing affords
recall. Other effects (e.g., production, generation)
depend on a relative boost at the expense of the
less distinct words, resulting in a lack of any mem-
orial effect when encoding strategy is presented in
pure lists, between participants (Mulligan, 2002).
Our data indicate that even when compared
between subjects, drawing allows for a signicant
boost in memory performance.
It should be noted that by inspection of the
means, we see that the proportion of written
words that were recalled is substantially lower in
mixed list presentation than in pure list presen-
tation; however, the increase in proportion of
drawn words recalled in mixed lists relative to
pure lists was only marginal. An analysis designed
to contrast within- and between-participant com-
parisons of the same conditions was performed, as
outlined by Erlebacher (1977). This analysis is
designed to explore the interaction of design type
with a dependent variable of interest.
1
Results indi-
cated that there was a signicant Design
Type ×Encoding Trial Type interaction,
F(1, 44) = 19.63, p,.001, η
2
= .09. Such a result
indicates that the effect of drawing is slightly aug-
mented in mixed lists, but this is due to lower
recall of written words, rather than a massive
boost in recall of drawn words. Because of well-
documented issues with output interference in
free-recall paradigms (Roediger, 1974), it is also
unclear whether the poorer memory for written
1
The analysis detailed in Erlebacher (1977) requires that sample sizes in all three groups be equal. Our within-subjects design had
28 participants, and the between-subjects groups had 23 and 24 participants, respectively. In order to complete the analysis, 23 par-
ticipants were randomly selected from the larger groups, and the analysis was conducted on this truncated data set.
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 17
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
words is due to slight distinctiveness advantages of
drawn words, or due to the fact that output interfer-
ence, especially with long lists, prevents output of
written words later in the allotted recall time. Our
results may be comparable to the ndings of
Bodner et al. (2014) in their between-participants
production effect experiments. Similar to our
work, these researchers found that memory for
words encoded aloud was not substantially
improved in a within-participants relative to a
between-participants design, as would be expected
given the distinctiveness account (MacLeod et al.,
2010), indicating that a distinctiveness account
might not be applicable. It is worth noting,
though, that these researchers used a recognition
paradigm, while we used free recall. In previous
explorations of the benets of production on free
recall, there is no evidence to support a between-
participants production effect (Jones & Pyc, 2014;
Jonker at al., 2014). Thus, it is clear that the
drawing effect cannot be fully explained by a rela-
tive distinctiveness account and can be differen-
tiated from other item-specic encoding strategies
thought to rely on distinctiveness (McDaniel &
Bugg, 2008). In order to comprehensively explore
the applicability of a distinctiveness account, and
because of the discussed issues of output interfer-
ence, future work should test the pure- versus
mixed-list paradigms using a recognition task.
Most importantly, though, there remains a 9%
benet of drawing in our pure-lists design, such
that 23% of drawn words, but only 14% of
written words, were later recalled. This shows
that even if distinctiveness can account for some
of the advantage provided by drawing, it is far
from an exhaustive mechanism.
GENERAL DISCUSSION
Our results showed unequivocally that drawing pic-
tures of words presented during an incidental study
phase provides a measurable boost to later memory
performance relative to simply writing out the
words, once or repeatedly. Across all seven exper-
iments, memory for drawn words was superior to
all other instructional manipulations during encod-
ing (see Table 1 for effect sizes). Specically,
drawing led to better later memory performance
than adding detail to written words, listing physical
characteristics of words, creating mental images of
words, and viewing pictures of the words. We also
demonstrated that the drawing effect was a feasible
mnemonic strategy, even at much shorter (4 relative
to 40 s) encoding durations, with much longer (66
relative to 30 words) lists, and that the benet
Table 1. Summary of results from Experiments 17
Experiment Description Results
Cohensd
(draw vs.
write)
1A Studied items are drawn (draw) in detail or written (write) repeatedly; later
free recall
Draw .write 1.51
1B Same as 1A, except drawings are repeated, and writing is in detail Draw .write 0.96
2 Same as 1A, except participants were tested in large groups in lecture halls Draw .write 1.30
3 Drawing compared to a deep LoP condition (list), in addition to writing Draw .list = write 0.86
4 Drawing compared to a visual imagery condition (visualize), in addition to
writing
Draw .visualize
a
=
write
0.87
5 Drawing compared to a picture superiority condition (view), in addition to
writing
Draw .view = write 0.67
6 Same as 1A, but with shortened encoding time, and longer list of items Draw .write 2.27
7 Same as 6, except between subjects; one group only drew items, other group
only wrote
Draw .write 1.63
Note: LoP = level of processing.
a
The difference between draw and visualize was marginal.
18 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
afforded by drawing items was robust in a pure-list,
between-participants design. The drawing effect
was stable despite both subtle (emphasis on detail)
and drastic (addition of a third encoding trial
type) changes to our paradigm. Our results show
that drawing should be considered among the
ranks of production (MacLeod et al., 2010), gener-
ation (Slamecka & Graf, 1978), and enactment
(Engelkamp & Zimmer, 1997) as a robust encod-
ing manipulation that can, and does, improve
memory performance dramatically.
Experiments 1 and 2 showed that drawing is far
superior to simply writing out a word, which has
been documented as an effective iteration of the
production effect (Forrin et al., 2012). We next
turned our attention to the exploration of a
number of potential mechanisms by which
drawing could exert its effect. We reasoned that
the act of creating a drawing based on a presented
word would invoke (a) elaboration, (b) visual
imagery, (c) motor action, and (d) a picture
memory. Accordingly, it was prudent to rule out
each of these individually as mechanisms driving
the drawing effect.
Components of the drawing process
In Experiment 3, participants were asked to list
physical characteristics of the studied words, as an
elaborative, deep LoP instruction manipulation
during encoding (Craik & Lockhart, 1972). In
this, as well as in Experiment 4, results showed
that drawing resulted in enhanced memory, and
was even was superior to visual imagery, which
probably contributes to the picture superiority
effect (Paivio, 2014). Such results suggest that the
act of drawing a word during study engages a par-
ticipant more than does a simple visual imagery
encoding orientation, and accordingly drawing pro-
duces a more powerful memorial benet. In
Experiment 5, participantsmemory for drawn
items was better than that for words presented
with their associated pictures during study; this
suggests that the benet of drawing was beyond
that provided by the picture superiority effect
alone (Paivio et al., 1968). This nding is especially
compelling given the recent nding that
participantsdiscrimination of 2500 studied
images was extraordinary (almost 90%; Brady
et al., 2008). As participants were asked to choose
between image pairs with as little as small perspec-
tive changes to differentiate them, these results
suggest that people have a massive capacity for
remembering the detail in images. The list
(Experiment 3), visualize(Experiment 4), and
view(Experiment 5) encoding orientations led
to roughly similar recall performance as did the
writing encoding orientation in our various exper-
iments. Such a nding suggests that the differences
we have observed, in memory for words drawn
compared to written at encoding, were not due to
a cost in memory from writing, but rather due to
a boost in memory for the drawn words.
Experiments 3 through 5 also served the added
purpose of ruling out these three individual com-
ponents of the drawing effect (elaboration, visual
imagery, picture memory), as explanations for the
effect of drawing on memory. Our results indicate
that not only do LoP, imagery, and picture super-
iority fail to fully account for our data, but that
the drawing effect is signicantly more effective
than these reliable alternative encoding strategies.
Distinctiveness accounts
The question remained: What is driving the
drawing effect? Most effects (e.g., production,
generation) from item-specic encoding strategies
tend to be conspicuously absent in between-partici-
pants designs, leading to a compelling distinctive-
ness account to explain them. In other words,
when produced and silently read words are com-
pared in a mixed list, the produced words are
more salient than read words, leading to better
recall of produced words (MacLeod et al., 2010).
This distinctiveness results in a reliable production
effect in within-participant designs, but often
no effect in between-participants designs.
Accordingly, we tested the drawing effect in pure
lists to directly explore the notion of a distinctive-
ness account. We found a substantial (9%) boost
in recall (23% of drawn, relative to 14% of written
words were later recalled), even in a between-par-
ticipants pure list paradigm, suggesting that
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 19
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
distinctiveness fails to account for the observed
benet, though we acknowledge it may contribute,
since the effect was reduced in our pure compared
to mixed design. It is important to note that
some studies have shown that reliable, albeit
smaller, production effects can be observed with
recognition scores compared in between-partici-
pants designs (Bodner et al., 2014), and meta-ana-
lyses of between-participants production effect
experiments do show a reliable benet (Bodner
et al., 2014; Fawcett, 2013). This slightly
weakens the case for distinctiveness as an expla-
nation for production. In comparison with the
current work, however, it should be noted that
the observed between-participants production
effects are only roughly 4% differences, while the
between-participants drawing effect showed a
much larger 9% benet, indicating that distinctive-
ness is probably not a mechanism that could fully
explain our foregoing ndings.
When looking at the recall proportions from our
between-participants data (Experiment 7), com-
pared with the same conditions within participants
(Experiment 6), it is clear that the effect is smaller,
due in part to an increase in proportion of drawn
items recalled in mixed relative to pure lists, but
mainly driven by a decrease in written items
output in mixed relative to pure lists. Though this
appears to point toward a small role of distinctive-
ness in driving our effect, it is difcult to interpret
this reduction. It could be due to distinctiveness, in
which case this account can partially, but far from
fully, explain our data. Alternatively, written items
could be suppressed due to output interference on
recall later in the allotted retrieval time. This type
of output interference has been well documented
in the literature (e.g., Roediger, 1974). If this is
the reason for the suppression, the implication is
an even smaller or nonexistent role of distinctive-
ness, in favour of an alternative account, peripher-
ally related to strength accounts (Ratcliff, Clark,
& Shiffrin, 1990).
Synergistic interaction
There is evidence to suggest that the combination
of various encoding strategies can provide a
benet that together is better than the sum of
each individually. For example, Fawcett, Quinlan,
and Taylor (2012) showed that the production
effect was larger when naming pictures than
words. Similarly, studies have shown that the
well-replicated generation effect is larger when
the generated word is read aloud, in a form of the
production effect (MacLeod et al., 2010,2012),
suggesting that integrating two of these effects
can lead to additive performance. Research has
also shown a generation effect for pictures (Kinjo
& Snodgrass, 2000), indicating that even though
memory for images is clearly very detailed, there
are still encoding manipulations that can further
drive performance upward. We propose that
drawing, through the seamless integration of its
constituent parts, produces a synergistic effect,
whereby the whole benet is greater than the sum
of the benet of each component. To reiterate,
we reasoned that drawing relies on the integration
of (a) elaboration, captured by the benet of the
listtrial type in Experiment 3, (b) visual
imagery, captured by the benet of the visualize
trial type in Experiment 4, (c) motor action, con-
trolled for inherently in the writetrial type, and
(d) creation of a pictorial representation, captured
by the benet of the viewtrial type in
Experiment 5. The mechanism driving our
drawing effect may be one that integrates these
traces into a more cohesive unit.
The benet of list (.01), visualize (.08), and view
(.05) conditions (computed as proportion recall in
these trial types minus proportion recall in their
respective write conditions), added on top of the
average proportion of written recall across all exper-
iments (.310) falls short of the performance levels
achieved, on average, from the draw encoding
trial types (.48). This unexplained difference,
however small, provides some very preliminary
support for a synergistic effect of each of the con-
stituent components of drawing that leads to
greater memory than from the sum of the individ-
ual contributions. The consistent superiority of
drawing relative to other encoding strategies
throughout the experiments in the current work is
also consistent with this account (see Table 1).
Specically, we believe that because drawing
20 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
results in more interconnected memory cues to
draw upon at recall, the memory trace for drawn
words is much more likely to be effectively retrieved
than when it was simply written, listed, visualized,
or viewed at encoding.
The evidence cited above in favour of our
synergistic account is indirect and certainly not
conclusive. Accordingly, before this account can
be invoked to explain the drawing effect, further
experimental work is required. Future research
could examine this proposed synergistic account
more directly in one of two ways. First, memory
for drawn versus written items could be compared
using functional magnetic resonance imaging
(fMRI), and functional connectivity analyses
could help determine whether drawn items acti-
vate more diverse brain regions associated with
visual imagery, motor cortex, and prefrontal areas
associated with greater depth of processing.
Functional connectivity analyses would allow the
exploration of whether there is greater connectivity
between the aforementioned regions and the hip-
pocampus (known to be active for rich, detailed
retrieval; Wiltgen et al., 2010) when drawn relative
to written items are retrieved. Second, one could
address the feasibility of the synergistic account
more directly in a carefully controlled behavioural
paradigm. This could be achieved by systematically
varying the presence or absence of various pro-
posed components of the drawing process (e.g., a
condition where participants are unable to see
their drawing, or where they must trace an existing
image). Thus, direct comparisons of conditions in
which the only difference is the presence or
absence of one of the components would give an
estimate of the contribution to recall that that
component alone provides. Doing similar subtrac-
tions for each component (e.g., isolating the ela-
borative component by subtracting recall in a
trace the picturecondition, from recall in a
draw the picturecondition) would provide
some indication of roughly how much of a
benet each component provides to memory on
its own. If the total benet of drawing surpassed
the sum of these components, this would provide
a compelling case for our synergistic account of
drawing.
In our Experiments 15, participants were pro-
vided with one minute for free recall. Previous work
exploring cumulative recall functions have shown
that relational tasks reach asymptote faster than
item-specic tasks, which reach asymptote later,
when more time is given for recall (Burns &
Hebert, 2005; Burns & Schoff, 1998). While all
of the tasks used in the current work were item-
specic, it is still a possibility that if provided
with more time for recall, the number of recalled
written words would approach the number of
recalled drawn words. Given the large effect size,
and that the effect was even larger in Experiment
6, when more time (2 minutes) was provided for
recall, it seems unlikely that this would be the
case. Given this concern, it is important for future
work to explore a number of methodological vari-
ations to rule out the inuence of the amount of
time allotted for recall, as well as any potential
effects of output interference. To address the
former issue, further explorations of the drawing
effect could allow 10 minutes for free recall, to
determine whether the effect might decrease in
magnitude as a result of slower recall of written
words. A possible iteration to rule out the latter
output interference issue (Roediger, 1974) could
instruct participants to recall exclusively written
words rst, followed by drawn words.
While we did show that the drawing effect is
reliable in group testing in our experiments, the
content was still only single words and hardly
representative of an academic setting. As men-
tioned, there is a body of work that has examined
the efcacy of drawing for retaining information
in an academic setting, though manipulations
were often complex and frequently involved con-
founds or additional tools beyond basic drawing
(e.g., Schwamborn et al., 2010; Van Meter, 2001;
Van Meter & Garner, 2005). With this, follow-
up work from the present study could address
whether the advantage of drawn relative to
written words will apply to memory for more
complex conceptual concepts across a cross-
section of educational domains, as has been
suggested by previous work discussing the inuence
of drawing on science learning (Ainsworth, Prain,
& Tytler, 2011). Future work could explore
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 21
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
drawing as a strategy for learning information from
video lectures, presented to groups of participants
in large lecture halls. Such research would highlight
whether drawing is a viable technique for the
improvement of recall of more representative scho-
larly material. In educational settings, students
often doodle, which is essentially drawing, but of
information unrelated to the presented material.
Andrade (2009) asked participants to listen to
phone messages and assigned half of the partici-
pants to shade in shapes while listening to the
message. The doodlinggroup outperformed the
other group on a later memory test. The authors
suggest that this could be due to a reduction in
mind wandering. Future work could explore
whether drawing or doodling leads to better later
performance by reducing the mind-wandering
rates of participants, thereby improving later
retention.
Conclusion
In the rst systematic exploration of the efcacy of
drawing as an encoding strategy, we showed a large
and reliable advantage in memory performance for
items that were previously drawn relative to those
that were written, which we label the drawing
effect. Drawing a to-be-remembered stimulus
was superior to writing it out (Experiment 1),
regardless of whether the encoding orientation
instructed the participant to emphasize either detail
or repetition. Further, this effect was replicated in
an ecologically valid large group setting
(Experiment 2). Next, we determined that the
drawing effect was superior to a deep level of pro-
cessing (Experiment 3), visual
imagery (Experiment 4), and viewing pictures
(Experiment 5) as encoding orientations, thereby
ruling these out as possible alternative explanations
for the benet of drawing on subsequent memory.
Experiment 6 showed that drawing effects were
still observable with long lists and short encoding
durations, while Experiment 7 effectively ruled
out a distinctiveness account for the drawing
effect. We argue that the mechanism driving the
effect is that engaging in drawing promotes the
seamless integration of many types of memory
codes (elaboration, visual imagery, motor action,
and picture memory) into one cohesive memory
trace, and it is this that facilitates later retrieval of
the studied words.
Supplemental material
Supplemental content is available via the Supple-
mentaltab on the articlesonlinepage(http://dx.
doi.org/10.1080/17470218.2015.1094494).
REFERENCES
Ainsworth, S., Prain, V., & Tytler, R. (2011). Drawing
to learn in science. Science,333, 10961097.
Andrade, J. (2009). What does doodling do? Applied
Cognitive Psychology,24, 100106.
Balota, D. A., & Black, S. (1997). Semantic satiation in
healthy young and older adults. Memory & Cognition,
25, 190202.
Bertsch, S., Pesta, J. P., Wiscott, R., McDaniel, A. M.
(2007). The generation effect: A meta-analytic
review. Memory & Cognition,35, 201210.
Bodner, G. E., & Taikh, A. (2012). Reassessing the basis
of the production effect in memory. Journal of
Experimental Psychology: Learning, Memory, and
Cognition,38(6), 17111719.
Bodner, G. E., Taikh, A., & Fawcett, J. M. (2014).
Assessing the costs and benets of production in rec-
ognition. Psychonomic Bulletin & Review,21(1), 149
154.
Boldini, A., Russo, R., Punia, S., & Avons, S. E. (2007).
Reversing the picture superiority effect: A speed-
accuracy trade-off study of recognition memory.
Memory and Cognition,35(1), 113123.
Brady, T. F., Konkle, T., Alvarez, G. A., & Oliva, A.
(2008). Visual long-term memory has a massive
storage capacity for object details. Proceedings of the
National Academy of Sciences of the United States of
America,105(38), 1432514329.
Burns, D. J. & Hebert, T. (2005). Using cumulative-
recall curves to assess the extent of relational and
item-specic processing. Journal of Experimental
Psychology: Learning, Memory, and Cognition,13,
189199.
Burns, D. J., & Schoff, K. (1998). Slow and steady often
ties the race: Effects of item-specic and relational
22 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
processing on cumulative recall. Journal of
Experimental Psychology: Learning, Memory, and
Cognition,24, 10411051.
Conway, M. A., & Gathercole, S. E. (1987). Modality
and long-term memory. Journal of Memory and
Language,26(3), 341361.
Cooper, E. H., & Pantle, A. J. (1967). The total-time
hypothesis in verbal learning. Psychological Bulletin,
68(4), 221234.
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of pro-
cessing: A framework for memory research. Journal of
Verbal Learning and Verbal Behavior,11(6), 671684.
Elliott, L. (1973). Imagery versus repetition encoding in
short-and long-term memory. Journal of Experimental
Psychology,100(2), 270276.
Engelkamp, J., & Dehn, D. M. (2000). Item and order
information in subject-performed tasks and exper-
imenter-performed tasks. Journal of Experimental
Psychology: Learning, Memory, and Cognition,26(3),
671682.
Engelkamp, J., & Zimmer, H. D. (1997). Sensory factors
in memory for subject-performed tasks. Acta
Psychologica,96(12), 4360.
Erlebacher, A. (1977). Design and analysis of exper-
iments contrasting the within-and between-subjects
manipulation of the independent variable.
Psychological Bulletin,84(2), 212219.
Fawcett, J. M. (2013). The production effect benets
performance in between-subject designs: A meta-
analysis. Acta Psychologica,142(1), 15.
Fawcett, J. M., Quinlan, C. K., & Taylor, T. L. (2012).
Interplay of the production and picture superiority effects:
A signal detection analysis. Memory,20(7), 655666.
Forrin, N. D., MacLeod, C. M., & Ozubko, J. D.
(2012). Widening the boundaries of the production
effect. Memory and Cognition,40(7), 10461055.
Guttentag, R. E., & Hunt, R. R. (1988). Adult age
differences in memory for imagined and performed
actions. Journals of Gerontology,43(4), P107P108.
Hockley, W. E. (2008). The picture superiority effect in
associative recognition. Memory and Cognition,36(7),
13511359.
Jones, A. C., & Pyc, M. A. (2014). The production
effect: Costs and benets in free recall. Journal of
Experimental Psychology: Learning, Memory, and
Cognition,40(1), 300305.
Jonker, T. R., Levene, M., & MacLeod, C. M. (2014).
Testing the item-order account of design effects
using the production effect. Journal of Experimental
Psychology: Learning, Memory, and Cognition,40(2),
441448.
Kieras, D. (1978). Beyond pictures and words:
Alternative information-processing models for
imagery effect in verbal memory. Psychological
Bulletin,85(3), 532554.
Kinjo, H., & Snodgrass, J. G. (2000). Is there a picture
superiority effect in perceptual implicit tasks?
European Journal of Cognitive Psychology,12(2),
145164.
Lesgold, A. M., De Good, H. & Levin, J. R. (1977).
Pictures and young childrens prose learning: A sup-
plementary report. Journal of Reading Behavior,9(4),
353360.
Lupiani, D. A. (1977). The facilitative effects of
imagery instructions and stimulus characteristics
on immediate and long term free recall and clus-
tering. The Journal of General Psychology,97(1),
7387.
MacLeod, C. M., Gopie, N., Hourihan, K. L., Neary, K.
R., & Ozbuko, J. D. (2010). The production effect:
Delineation of a phenomenon. Journal of
Experimental Psychology: Learning, Memory, and
Cognition,36, 671685.
MacLeod, C. M., Pottruff, M. M., Forrin, N. D., &
Masson, M. E. (2012). The next generation: The
value of reminding. Memory & cognition,40(5),
693702.
Maisto, A. A., & Queen, D. E. (1992). Memory for pic-
torial information and the picture superiority effect.
Educational Gerontology,18, 213223.
Marks, D. F. (1973). Visual imagery in the recall of pic-
tures. British Journal of Psychology,64,1724.
Mazzoni, D., & R. Dannenberg. 2000.Audacity.
Retrieved from http://audacity.sourceforge.net/
McDaniel, M., & Bugg, J. (2008). Instability in memory
phenomena: A common puzzle and a unifying expla-
nation. Psychonomic Bulletin & Review,15(2), 237
255.
McMahon, J. A. (2002). An explanation for normal and
anomalous drawing ability and some implications for
research on perception and imagery. Visual Arts
Research,28(1), 3852.
Mintzer, M. Z., & Snodgrass, J. G. (1999). The picture
superiority effect: Support for the distinctiveness
model. American Journal of Psychology,112(1), 113
146.
Morris, C. D., Bransford, J. D., & Franks, J. J. (1977).
Levels of processing versus transfer appropriate pro-
cessing. Journal of Verbal Learning and Verbal
Behavior,16(5), 519533.
Mulligan, N. (2002). The emergence of item-specic
encoding effects in between-subjects designs:
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 23
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
Perceptual interference and multiple recall tests.
Psychonomic Bulletin & Review,9(2), 375382.
Nairne, J. S., Riegler, G. L., & Serra, M. (1991).
Dissociative effects of generation on item and order
retention. Journal of Experimental Psychology:
Learning, Memory, and Cognition,17(4), 702709.
Ozubko, J. D., & MacLeod, C. M. (2010). The pro-
duction effect in memory: Evidence that distinctive-
ness underlies the benet. Journal of Experimental
Psychology: Learning, Memory, and Cognition,36(6),
1543.
Paivio, A (1971). Imagery and verbal processes. New York:
Holt, Rinehart, and Winston.
Paivio, A. (1991). Dual coding theory: Retrospect and
current status. Canadian Journal of Psychology,45(3),
255287.
Paivio, A. (2014). Intelligence, dual coding theory, and
the brain. Intelligence,47, 141158.
Paivio, A. & Csapo, K. (1973). Picture superiority in free
recall: Imagery or dual coding? Cognitive Psychology,
5, 176206.
Paivio, A., & Foth, D. (1970). Imaginal and verbal
mediators and noun concreteness in paired-associate
learning: The elusive interaction. Journal of Verbal
Learning and Verbal Behavior,9(4), 384390.
Paivio, A., Rogers, T. B., & Smythe, P. C. (1968). Why
are pictures easier to recall than words? Psychonomic
Science,11(4), 137138.
Peynircioğlu, Z. F. (1989). The generation effect with
pictures and nonsense gures. Acta Psychologica,70
(2), 153160.
Ratcliff, R., Clark, S. E., & Shiffrin, R. M. (1990). List-
strength effect: I. Data and discussion. Journal of
Experimental Psychology: Learning, Memory, and
Cognition,16, 163178.
Roediger, H. L. (1974). Inhibiting effects of recall.
Memory & Cognition,2, 261269.
Rowe, E. J. (1972). Discrimination learning of pictures
and words: A replication of picture superiority.
Journal of Experimental Child Psychology,14(2), 323
328.
Schwamborn, A., Mayer, R. E., Thillmann, H.,
Leopold, C., & Leutner, D. (2010). Drawing as a
generative activity and drawing as a prognostic
activity. Journal of Educational Psychology,102(4),
872879.
Slamecka, N., & Graf, P. (1978). The generation effect:
Delineation of a phenomenon. Journal of
Experimental Psychology: Human Learning and
Memory,4, 592604.
Slamecka, N. J., & Katsaiti, L. T. (1987). The generation
effect as an artifact of selective displaced rehearsal.
Journal of Memory and Language,26(6), 589607.
Snodgrass, J. G., & McClure, P. (1975). Storage and
retrieval properties of dual codes for pictures and words
in recognition memory. Journal of Experimental
Psychology: Human Learning and Memory,1,521529.
Snodgrass, J. G., & Vanderwart, M. (1980). A standar-
dized set of 260 pictures: Norms for name agreement,
image agreement, familiarity, and visual complexity.
Journal of Experimental Psychology: Human Learning
& Memory,6(2), 174215.
Vaidya, C. J., & Gabrieli, J. D. E. (2000). Picture super-
iority in conceptual memory: Dissociative effects of
encoding and retrieval tasks. Memory and Cognition,
28(7), 11651172.
Van Meter, P. (2001). Drawing construction as a strategy
for learning from text. Journal of Educational
Psychology,93(1), 129140.
Van Meter, P., & Garner, J. (2005). The promise and
practice of learner-generated drawing: Literature
review and synthesis. Educational Psychology Review,
17(4), 285325.
Walsh, D. A., & Jenkins, J. J. (1973). Effects of orienting
tasks on free recall in incidental learning: Difculty,
effort,and processexplanations. Journal of Verbal
Learning and Verbal Behavior,12(5), 481488.
Weldon, M. S., & Roediger, H. L. (1987). Altering
retrieval demands reverses the picture superiority
effect. Memory & Cognition,15(4), 269280.
Wiltgen, B. J., Zhou, M., Cai, Y., Balaji, J., Karlsson, M.
G., Parivash, S. N., Silva, A. J. (2010). The hippo-
campus plays a selective role in the retrieval of detailed
contextual memories. Current Biology,20(15), 1336
1344.
Winnick, W. A., & Brody, N. (1984). Auditory and
visual imagery in free recall. The Journal of psychology,
118(1), 1729.
24 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016
WAMMES, MEADE, FERNANDES
Downloaded by [University of Waterloo] at 06:13 22 February 2016
APPENDIX A
List of target items used in Experiments 1 through 7
APPENDIX B
Samples from participants
airplane couch kite ruler
ant cow knife sailboat
axe desk ladder scissors
balloon doll lamp screwdriver
banana door lemon sheep
bee drum lion shoe
beetle duck lips skirt
blouse ear monkey spider
boot elephant mushroom spoon
broom ute owl stool
buttery fork pants stove
camel frog peanut strawberry
cannon giraffe pear sweater
carrot glove penguin toaster
cat grapes pepper trumpet
caterpillar guitar pig turtle
cherry hammer pineapple violin
clock harp pumpkin wagon
coat jacket rabbit whistle
corn kettle rooster wrench
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2016 25
THE DRAWING EFFECT
Downloaded by [University of Waterloo] at 06:13 22 February 2016
... In our more recent work (Wammes et al., 2016) we demonstrated just how effective drawing can be as an encoding tool. During encoding of a to-be-remembered set of words, we asked participants to either draw a picture of what the word represents, or to write out the word repeatedly, on different trial types. ...
... Participants were later asked to recall as many words as they could remember from the encoding phase. We demonstrated, repeatedly and overwhelmingly, that drawing reliably improved later memory compared to simply writing (Wammes et al., 2016). The benefit from drawing at encoding was also greater than creating lists of elaborative details about a to-be-remembered word, was found in both within-and between-subjects designs , and replicated across varied settings such as within classrooms or individual testing rooms (Wammes et al., 2016). ...
... We demonstrated, repeatedly and overwhelmingly, that drawing reliably improved later memory compared to simply writing (Wammes et al., 2016). The benefit from drawing at encoding was also greater than creating lists of elaborative details about a to-be-remembered word, was found in both within-and between-subjects designs , and replicated across varied settings such as within classrooms or individual testing rooms (Wammes et al., 2016). Our work suggested that drawing a picture of a to-be-remembered stimulus is a potent, yet simple, strategy to improve later memory. ...
Article
Drawing at encoding has been shown to improve later recall of to-be-remembered words, pictures, and academic terms, compared to when one simply writes out the target information. Here we examined whether drawing in a diary, compared to writing in it, differentially improved later memorability of personal autobiographical events, and whether aging influenced the magnitude of this effect. Thirty younger and thirty older adults were given a diary booklet, and instructed to use it to reminisce about a random daily event of their choosing, on 12 days within a two-week period. They recorded the event on each day in one of two ways, counterbalanced within-subject: by writing about or by drawing a picture of the event. Participants also generated a keyword descriptor for each event. After the two-week period, participants were cued using their keyword descriptors to recall each autobiographical memory by writing it down. Self-reported match accuracy, between reminisced and recalled events was tabulated. Across age groups, match accuracy was significantly higher for those drawn than written during the reminiscing phase. In addition, we compared the word count, level of detail, visual imagery, and point of view of the recalled memories. Self-reports of the level of detail in the recalled memory were better maintained for events drawn compared to written during the reminiscing stage, and better maintained in older than younger adults. Our findings suggest the use of drawing, while reminiscing within a diary format, enhances accuracy and quality of later recollections.
... Approaches commonly target specific tasks that are generally classified as "deep" processing tasks according to the levels-of-processing framework (Craik, 2002;Craik & Lockhart, 1972). Examples include pleasantness ratings (Hunt & Einstein, 1981), generation (Bertsch et al., 2007;Slamecka & Graf, 1978), production (Conway & Gathercole, 1987;MacLeod & Bodner, 2017), survival processing (Nairne et al., 2007), and more recently, drawing an image of a word's referent (Wammes et al., 2016(Wammes et al., , 2017. Although the benefits of these encoding tasks on memory for studied information are well supported using a variety of study materials (e.g., Fernandes et al., 2018;Ozubko et al., 2012), it is equally important to gauge task effectiveness on memory errors when considering overall memory accuracy. ...
... Drawing images of DRM list items increased correct recognition relative to a read-control group-a pattern consistent with other drawing studies (Meade et al., 2020;Wammes et al., 2016). This benefit was equivalent for both the standard black-pencil group and the coloured-pencil group, although correct recognition was at ceiling which may have masked potential differences between the two drawing groups. ...
... In Experiment 1, we compared between-groups that were tasked with drawing referents of DRM list items with either a standard black pencil or using coloured pencils relative to a read-only control group. Consistent with the well documented benefits of drawing on correct memory (Fernandes et al., 2018;Wammes et al., 2016), we similarly found that both drawing tasks improved correct recognition over reading. Importantly, false recognition of critical lures also reflected drawing benefits in which both drawing groups reduced false recognition relative to the read-only control. ...
Article
Full-text available
We examined the effects of drawing on correct and false recognition within the Deese/Roediger-McDermott (DRM) false memory paradigm. In Experiment 1, we compared drawing of a word’s referent using either a standard black pencil or colored pencils relative to a read-only control group. Relative to reading, drawing in either black or colored pencil similarly boosted correct recognition and reduced false recognition. Signal-detection analyses indicated that drawing reduced the amount of encoded memory information for critical lures and increased monitoring, indicating that both processes contributed to the false recognition reduction. Experiment 2 compared drawing of individual images of DRM list items relative to drawing integrated images using sets of DRM list items. False recognition was lower for drawing of individual images relative to integrated images—a pattern that reflected a decrease in encoded memory information but not monitoring. Therefore, drawing individual images improves memory accuracy in the DRM paradigm relative to a standard read-control task and an integrated drawing task, which we argue is due to the recruitment of item-specific processing.
... The act of taking notes may be more beneficial than the notes themselves. Note-taking is cognitively demanding because it requires listening intently to the lecture while reading the PowerPoint and writing organized and comprehensible notes [9]. Passively listening to a lecture does not require this effort. ...
... Tablets are unique due to the ability to draw, add color, type, handwrite, highlight text, and add images and links to study resources. In a study by Wammes et al. [9], drawing compared to writing was found to better enhance memory by integrating visual, motor, and semantic processes. Revising notes is quick and easy on a tablet, and studies have shown that students who edit notes will retain more information [13]. ...
Article
Purpose: Technological advances are changing how students approach learning. The traditional note-taking methods of longhand writing have been supplemented and replaced by tablets, smartphones, and laptop note-taking. It has been theorized that writing notes by hand requires more complex cognitive processes and may lead to better retention. However, few studies have investigated the use of tablet-based note-taking, which allows the incorporation of typing, drawing, highlights, and media. We therefore sought to confirm the hypothesis that tablet-based note-taking would lead to equivalent or better recall as compared to written note-taking. Methods: We allocated 68 students into longhand, laptop, or tablet note-taking groups, and they watched and took notes on a presentation on which they were assessed for factual and conceptual recall. A second short distractor video was shown, followed by a 30-minute assessment at the University of California, Irvine campus, over a single day period in August 2018. Notes were analyzed for content, supplemental drawings, and other media sources. Results: No significant difference was found in the factual or conceptual recall scores for tablet, laptop, and handwritten note-taking (P=0.61). The median word count was 131.5 for tablets, 121.0 for handwriting, and 297.0 for laptops (P=0.01). The tablet group had the highest presence of drawing, highlighting, and other media/tools. Conclusion: In light of conflicting research regarding the best note-taking method, our study showed that longhand note-taking is not superior to tablet or laptop note-taking. This suggests students should be encouraged to pick the note-taking method that appeals most to them. In the future, traditional note-taking may be replaced or supplemented with digital technologies that provide similar efficacy with more convenience.
... This level of performance reached much beyond the prediction of nine items at maximum that would be made by verbal recall memory studies (Murdock, 1962). In fact, drawing as a mnemonic strategy has been shown to outperform verbally based, imagery-based, and semantic elaborative strategies (Wammes et al., 2016;Wammes et al., 2019), even for memory of highly abstract concepts (Roberts & Wammes, 2020). Furthermore, learning to draw may boost one's ability to efficiently perceive and encode visual information (Perdreau & Cavanagh, 2014;Perdreau & Cavanagh, 2016;Vogt & Magnussen, 2007). ...
... Artists, who are highly trained at drawing, show different fixation patterns from non-artists when viewing an image, and recall more details (Vogt & Magnussen, 2007). Even for non-artists, drawing when learning information shows larger benefits over verbally rehearsing information in adults (Perdreau & Cavanagh, 2016;Wammes et al., 2016) and children (Gross & Hayne, 1998). However, drawing can also inflate the existence of falsely recalled information (Bruck et al., 2000;Otgaar et al., 2016). ...
Article
Full-text available
When we draw, we are depicting a rich mental representation reflecting a memory, percept, schema, imagination, or feeling. In spite of the abundance of data created by drawings, drawings are rarely used as an output measure in the field of psychology, due to concerns about their large variance and their difficulty of quantification. However, recent work leveraging pen-tracking, computer vision, and online crowd-sourcing has revealed new ways to capture and objectively quantify drawings, to answer a wide range of questions across fields of psychology. Here, I present a tutorial on modern methods for drawing experiments, ranging from how to quantify pen-and-paper type studies, up to how to administer a fully closed-loop online experiment. I go through the concrete steps of designing a drawing experiment, recording drawings, and objectively quantifying them through online crowd-sourcing and computer vision methods. Included with this tutorial are code examples at different levels of complexity and tutorials designed to teach basic lessons about web architecture and be useful regardless of skill level. I also discuss key methodological points of consideration, and provide a series of potential jumping points for drawing studies across fields in psychology. I hope this tutorial will arm more researchers with the skills to capture these naturalistic snapshots of a mental image.
... Enfin, le dessin renforce la mémoire (Shepard, 1967;Wammes et al., 2016) et contribue au traitement des émotions (Ottarsdottir, 2018), tout comme le rêve semble également le faire (Malinowski et Horton, 2015). Il est donc pertinent d' explorer l'apport du dessin du rêve pour reconsolider des mémoires émotionnelles. ...
Article
RÉSUMÉ La représentation visuelle du rêve est utilisée en complément au récit du rêve dans le but d’explorer son apport pour reconsolider des mémoires émotionnelles. La méthode proposée a permis d’identifier et d’activer des mémoires émotionnelles ainsi que d’activer et de créer des expériences dissonantes pendant la fenêtre temporelle de labilité des mémoires activées. Les images de rêves et le processus créatif ont contribué de différentes façons au processus de reconsolidation. L’utilisation de la représentation visuelle du rêve semble pertinente pour reconsolider des mémoires émotionnelles. Les résultats présentés ont des implications pour la pratique clinique de la psychothérapie. D’autres recherches sont requises.
... These researchers also report that children and adults who gesture during learning recall more information than those who do not (Cutica & Bucciarelli, 2013;Cutica et al., 2014;Ianì et al., 2017; see also Stevanoni & Salmon, 2005). A related point is the 'drawing effect': for example, Wammes et al. (2016) found in seven experiments that, in comparison to writing, drawing while learning resulted in between two and five times as many words being recalled. Since both gestures and drawing involve the coordination of semantic, spatial, perceptual and motor cognitive functions, these findings would appear to be consistent with those of Knauff et al.'s (2002) fMRI study of cortical areas activated in deductive reasoning. ...
Article
Full-text available
Previous researchers (e.g., Vosniadou and Brewer, Cognitive Psychology 24:535–585, 1992) have claimed that children have naïve, but coherent, mental models of the earth, such as the flat earth and the hollow sphere. Recent studies have challenged this view, focusing on the original researchers’ methods of testing (drawing and open questions) and coding. In this study we tested the construct validity of these methods by asking expert astronomers to complete the same test originally used with 5-year-olds. Many astronomers gave responses that, if given by young children, would have been considered non-scientific and as evidence of naïve mental models. Many gave two or more seemingly contradictory answers to the same questions and, even when only their ‘most scientific’ responses were considered, fewer than 50% of these expert scientists were classified as having coherent scientific mental models. Comparison with children’s responses to a rephrased and disambiguated task indicated that even 6–7-year-olds gave more scientific answers. The astronomers’ comments revealed the main reasons why the original task lacks validity: they found many questions confusing and ambiguous. This may well explain many children’s responses, too. Since the task incorrectly indicates that experts in the field have fundamental misconceptions, it is likely also to have led to substantial underestimates of children’s scientific understanding.
Article
Full-text available
We introduce biology to the artist’s design tool, the storyboard. This versatile organizing and visualizing artistic platform is introduced into the biology classroom to aid in an inventive and focused discovery process. Almost all biological concepts are dynamic, and storyboards offer biology, lecture, wet and computational labs, flexibility, inventiveness, and an opportunity for students to slow down the so-called steps of biological processes and moderate their observations. Storyboarding is a thoughtful and reflective discovery device with enormous potential to break with traditional biology classroom experiences and return to the root of the educational process: storytelling. It will encourage teachers to embark on the remodeling of the biological curriculum with specific technical skills that students and teachers should consider developing to make the STEAM experience tailored to the uniqueness of biological systems. Storyboards offer hands-on, illustrative, and interactive conversations about biology concepts. They are an “unplugged” and contemplative experiences, organizing frameworks for personal expression focused on biological wonders.
Article
This paper reviews and discusses research on notetaking during academic listening conducted in both first (L1) and second language (L2) contexts and is organized into two main categories: research that is beginning to impact English for academic purposes (EAP) classrooms and that which has yet to make an impact on EAP, but should. Overall, I assert that, while some relevant research on notetaking is reaching EAP classrooms, there is an abundance of knowledge from L1 contexts as well as a number of unexplored areas that have potential to improve instruction and student ability. Throughout the paper, I hypothesize why certain research findings are not being applied in classrooms. In concluding, I provide suggestions for how researchers and teachers might further support colleagues in applying research findings.
Article
The widespread offering of online engineering classes has not been without its share of controversy [1] – [7] . There are those who question the pedagogic quality of courses made freely available through video-sharing platforms like YouTube, which feature largely crowdsourced and crowd-vetted content, despite such technologies figuring prominently in the evolving skill sets and preferences of engineering students. The same group typically considers the more traditional in-person format to be a critical element in learning engineering fundamentals, serving the secondary purpose of fostering much-needed discipline, patience, imagination, and visualization. Most established engineering institutions have recognized the value of both approaches (consider these popular examples: the Massachusetts Institute of Technology [8] and NPTel [9] ) and supplemented a majority of their in-person classes with a smattering of online offerings, though reservations may persist [10] – [12] .
Article
Medical students tend to have difficulties in developing a holistic view of metabolic pathway and hormone regulation. To address this issue, an interactive activity was implemented for first‐year medical students at the School of Medicine, University of Minho, Portugal. Students' previous knowledge on metabolic pathways was evaluated by a pre‐test followed by an interactive activity. In the supervised activity, students were challenged to elaborate a diagrammatic representation regarding enzymes, co‐factors, and hormonal metabolic regulation in early fasting during the night, as well as in well‐fed conditions. The activity was concluded with a post‐test to determine the students' learning gains and a few days later students were evaluated by a final exam. Afterwards, students evaluated the activity by filling a questionnaire. Results from four different cohorts showed that the activity resulted in significant learning gains, particularly favoring students who have less prior knowledge. The comparison between the pre‐test and the final exam also revealed significant learning gains for low achievers students. On the questionnaires, the majority of the students rated the activity as good or very good. Students agreed that this activity promotes: (a) reactivation of previous knowledge; (b) a better understanding of the interconnections between the metabolic pathways; (c) the application of learned concepts in real scenarios; and (d) sharing knowledge with peers. This study describes an active, unpretentious, and easily implemented activity available for early medical and biochemical curricula.
Article
Full-text available
Should science learners be challenged to draw more? Certainly making visualizations is integral to scientific thinking. Scientists do not use words only but rely on diagrams, graphs, videos, photographs, and other images to make discoveries, explain findings, and excite public interest. From the notebooks of Faraday and Maxwell (1) to current professional practices of chemists (2), scientists imagine new relations, test ideas, and elaborate knowledge through visual representations (3–5).
Article
Full-text available
A number of memory phenomena evident in recall in within-subject, mixed-lists designs are reduced or eliminated in between-subject, pure-list designs. The item-order account (McDaniel & Bugg, 2008) proposes that differential retention of order information might underlie this pattern. According to this account, order information may be encoded when a common form of processing is used alone in a list (e.g., reading), but not when an unusual form of processing is used (e.g., generation) or when a common form and an unusual form are mixed within a list. The production effect-better memory for words said aloud than for words read silently-shows this same design-contingent pattern. In 2 experiments, we investigated whether differential order retention might underlie the production effect. Consistent with the item-order account, we found that retention of order information was better in pure silent lists than in either pure aloud lists or mixed lists, as measured using an order reconstruction test. Moreover, in Experiment 2, order was better preserved in free recall of pure silent lists than of either pure aloud or mixed lists. Thus, production joins the set of tasks identified by McDaniel and Bugg (2008), and our findings suggest a role for order processing in explaining the production effect. (PsycINFO Database Record (c) 2013 APA, all rights reserved).
Article
Full-text available
Evidence is reviewed indicating that output interference—the deleterious effects of recall of some information on information recalled later—occurs both in primary and secondary memory. It appears that output interference provides at least a partial account for the disparity between information available in memory and its accessibility at recall. It is argued that consideration of output interference may provide a helpful perspective in resolving problems in the study of episodic and semantic memory, including the negative effects of part-list cueing and the tip-of-the-tongue phenomenon.
Article
Full-text available
The production effect is a memory advantage for items studied aloud over items studied silently. Although it typically is found within subjects, here we also obtained it between subjects in a recognition task-providing new evidence that production can be an effective study strategy. Our experiment, and a set of meta-analyses, also evaluated whether the within effect reflects costs to silent items and/or benefits to aloud items. Contrary to a strong distinctiveness account, we found little evidence that aloud items show an additional within-subjects benefit. Instead, silent items suffered an additional within-subjects cost. Blocking silent and aloud items eliminated this cost, suggesting that the cost was due to mixing silent and aloud items. Our discussion focuses on implications for distinctiveness and strength accounts of the production effect and on how to implement production as an encoding strategy depending on the learner's goals.
Article
Full-text available
Facilitative effects of illustration activity on prose learning have been previously established for young (first grade) children in a listening task. The present study tested for illustration effects with passages of varying complexity and length. Positive effects were found for long and short passages (100 vs. 50 words) that were either simple (one locational setting) or complex (two settings). The pattern of results for cued and free recall suggested that illustration facilitates individual sentence comprehension more than passage integration for these children. Analysis of subject-constructed illustrations showed a clear relationship between illustration content and passage recall.
Five experiments are reported comparing memory for words that were generated by the subjects themselves with the same words when they were simply presented to be read. In all cases, performance in the generate condition was superior to that in the read condition. This held for measures of cued and uncued recognition, free and cued recall, and confidence ratings. The phenomenon persisted across variations in encoding rules, timed or selfpaced presentation, presence or absence of test information, and between- or within-subjects designs. The effect was specific to the response items under recognition testing but not under cued recall. A number of potential explanatory principles are considered, and their difficulties enumerated. It is concluded that the generation effect is real and that it poses an interesting interpretative problem. This is an empirically oriented article whose purpose is to report a set of simple experiments that establish the existence of a robust and interesting phenomenon of memory. This phenomenon, called the generation effect, is robust in that it manifests itself across a variety of testing procedures, encoding rules, and other situational changes. It is interesting in that it does not seem to be easily or satisfactoril y accommodated by any of the currently familiar explanatory notions. We expect that once the phenomenon is described in its initial form, it will be the subject of wider experimental analysis and will eventually become better understood. In contrast to the usual objective reasons for embarking upon a line of research, the present work was neither initiated by any extant theoretical issue nor inspired by any previously published findings. It was carried out with the sole purpose of arriving at a
Presents a standardized set of 260 pictures for use in experiments investigating differences and similarities in the processing of pictures and words. The pictures are black-and-white line drawings executed according to a set of rules that provide consistency of pictorial representation. They have been standardized on 4 variables of central relevance to memory and cognitive processing: name agreement, image agreement, familiarity, and visual complexity. The intercorrelations among the 4 measures were low, suggesting that they are indices of different attributes of the pictures. The concepts were selected to provide exemplars from several widely studied semantic categories. Sources of naming variance, and mean familiarity and complexity of the exemplars, differed significantly across the set of categories investigated. The potential significance of each of the normative variables to a number of semantic and episodic memory tasks is discussed. (34 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).
Article
The distinction between verbal and nonverbal cognitive abilities has been a defining feature of psychometric theories of intelligence for the past century. Despite their popularity, however, these theories have not included functional connections between verbal and nonverbal systems that are necessary if they are to explain performance in intellectual tasks involving interactions between language and nonverbal knowledge. This functional gap limits the capacity of psychometric theories to explain and predict fundamental aspects of individual differences in cognitive abilities that have long been studied experimentally. This article summarizes the history, nature, and possible causes of the problem, and then concludes with a neuroscientifically-enhanced, multimodal dual coding approach to intelligence that focuses on the synergistic interactivity of verbal and nonverbal systems.
Article
Pictures are remembered better than their names. This picture superiority effect in episodic memory has been attributed either to the greater sensory distinctiveness of pictures or to their greater conceptual distinctiveness. Weldon and Coyote (1996) tested the conceptual distinctiveness hypothesis by comparing how well pictures as opposed to words primed in two conceptual implicit memory tasks (category production and word association). They found no picture superiority in priming and concluded that the basis of the picture superiority effect must then be pictures' greater sensory distinctiveness. Using the same logic, we compared how well pictures as opposed to words primed in a perceptual implicit memory task (picture and word fragment identification). The sensory distinctiveness theory would predict that pictures should prime picture fragment identification better than words prime word fragment identification, a result we call the picture superiority in within-form priming. Across three experiments which manipulated the encoding task at study, only one showed picture superiority in within-form priming. In contrast, in all three experiments there was robust picture superiority in recall, and exposure to pictures and words at study and test produced independent effects in which both study and test exposure to pictures was more effective for recall than exposure to words. We consider how these results might be reconciled by differences in retrieval demands between recall and fragment identification.