Appetite 57 (2011) 321–328

Contents lists available at ScienceDirect

Appetite
journal homepage: www.elsevier.com/locate/appet

Research report

Cognitive advantages of chewing gum. Now you see them, now you don’t
Serge V. Onyper *, Timothy L. Carr, John S. Farrar, Brittney R. Floyd
Department of Psychology, St. Lawrence University, Canton, NY 13617, USA

A R T I C L E I N F O A B S T R A C T

Article history: The current series of experiments investigated the effects of the timing of gum chewing on cognitive
Received 2 March 2011 function, by administering a battery of cognitive tasks to participants who chewed gum either prior to or
Received in revised form 25 April 2011 throughout testing, and comparing their performance to that of controls who did not chew gum.
Accepted 19 May 2011
Chewing gum was associated with performance advantages on multiple measures when gum was
Available online 27 May 2011
chewed for 5 min before, but not during, cognitive testing. The benefits, however, persisted only for the
first 15–20 min of the testing session, and did not extend to all cognitive domains. To explain this pattern
Keywords:
of results, it is proposed that the time-limited nature of performance benefits can be attributed to
Chewing gum
Time
mastication-induced arousal. Furthermore, the lack of improvement in cognitive function when gum is
Cognition chewed throughout testing may be because of interference effects due to a sharing of resources by
Memory cognitive and masticatory processes. This dual-process mechanism is not only consistent with the
Processing speed outcome of present experiments but can potentially account for a wide range of findings reported in the
Executive function literature.
ß 2011 Elsevier Ltd. All rights reserved.

Introduction 1908) to suggest that the facilitative effects of gum chewing on

cognitive performance were due to the accompanying increases in
The study of the cognitive benefits of chewing gum has received alertness and arousal. For instance, numerous studies have shown
increased attention from researchers in the past few years. In one of that gum chewing elevates heart rate and blood pressure (Farella,
the first studies on the subject, Wilkinson, Scholey, and Wesnes Bakke, Michelotti, Marotta, & Martina, 1999; Hasegawa et al., 2009;
(2002) demonstrated that chewing a piece of sugar-free gum Smith, 2010; Wilkinson et al., 2002), cortisol levels (Smith, 2010) and
improved immediate and delayed recall as well as working memory, cerebral blood flow (Hasegawa, Ono, Hori, & Nokubi, 2007; Onozuka
compared to sham chewing (i.e., mimicking chewing motions) or et al., 2002; Sesay, Tanaka, Ueno, Lecaroz, & de Beaufort, 2000), as
quiet control conditions. Similar results were reported by Stephens well as EEG and fMRI markers of cortical arousal (Hirano et al., 2008;
and Tunney (2004). Furthermore, the chewing of gum was Morinushi, Masumoto, Kawasaki, & Takigawa, 2000; Takada &
associated with a small overall increase in performance on a battery Miyamoto, 2004; for a review, see Weijenberg, Scherder, &
of cognitive tests (Scholey et al., 2009). In other studies, chewing Lobbezoo, 2011). Improved cerebral blood flow during mastication
gum was found to benefit verbal working memory (Hirano et al., in particular is thought to be associated with improved cognitive
2008; Zoladz & Raudenbush, 2005), free recall (Baker, Bezance, function (Weijenberg et al., 2011).
Zellaby, & Aggleton, 2004; Johnson & Miles, 2008), attention (Smith, If chewing gum is indeed associated with increases in arousal
2010; Tucha, Mecklinger, Maier, Hammerl, & Lange, 2004; Tucha & and a corresponding shift in cognitive function, it is unclear why
Simpson, 2011), as well as performance on reaction time measures many studies find little or no performance advantages of chewing.
(Sakamoto, Nakata, & Kakigi, 2009; Smith, 2010). For instance, Tucha et al. (2004) found that chewing gum was
Several mechanisms have been proposed to account for the associated with improvements on only one measure (sustained
facilitation in performance observed when gum is chewed during (as attention) out of 25 or so administered as part of their experiment.
well as prior to) cognitive testing. For instance, Stephens and Tunney Other studies likewise failed to find any effects of gum on memory
(2004) argued that the improved performance of those who chewed (Smith, 2009a, 2010), attention (Kohler, Pavy, & Van den Heuvel,
gum was due to an increased availability of glucose in the brain 2006; Smith, 2009b; see also Tucha et al., 2010), and learning of
associated with increased metabolic activity. Others have relied on lecture material (Allen, Norman, & Katz, 2008). Gum chewing also
the optimum arousal theory (e.g., Sanders, 1986; Yerkes & Dodson, did not affect performance on a mental rotation task (Nader,
Gittler, Waldherr, & Pietschnig, 2010) or the ability to solve
anagrams (Torney, Johnson, & Miles, 2009).
* Corresponding author. It is possible that methodological differences between studies
E-mail address: [email protected] (S.V. Onyper). (e.g., differences in study design and task demands, the timing of

0195-6663/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.appet.2011.05.313
322 S.V. Onyper et al. / Appetite 57 (2011) 321–328

measures and the duration of the testing session, as well as the half of whom chewed gum during the entire testing session and
type and brand of chewing gum) can explain some of the half served as controls.
discrepancies in findings (e.g., Scholey, 2004a,b). It is also possible
that chewing might usurp cognitive resources needed to maintain Experiments 1a and 1b
adequate performance on an attentionally demanding task, as
recently suggested by Tucha et al. (2010). Tucha and colleagues The aims of the following two experiments are to test the
found that chewing impaired performance on a task of sustained hypothesis that chewing gum prior to testing will enhance
attention in children with ADHD (who are known to suffer from an cognitive functioning, and to examine the claim that the
inability to remain attentive for prolonged periods of time) and, to performance benefits will be more pronounced earlier in
a lesser extent, healthy controls, concluding that chewing might testing—that is, immediately after the gum has been chewed
act as a distractor task (see also Tänzer, von Fintel, & Eikermann, and discarded. Participants in both experiments discarded gum
2009 and Tucha & Simpson, 2011, for a similar argument). after chewing and immediately before starting the test battery,
Notably, studies that provide the strongest support for the making our measures of performance free from the potential of any
claim that chewing gum facilitates cognitive function (and, in interference/distractor effects due to masticatory processes dis-
particular, episodic and working memory) required that partici- cussed by Tucha and colleagues (2010), and Tucha & Simpson
pants began chewing 3 min (Wilkinson et al., 2002) or 15 min (2011).
(Stephens & Tunney, 2004) prior to testing. Similarly, Sakamoto
and colleagues (2009) demonstrated that those who chewed gum Methods
for 5 min before measurements took place performed faster on a
measure of simple reaction time. The evidence provided by these Participants and design
studies offers an intriguing possibility—that the chewing of gum Eighty St. Lawrence University undergraduates participated in
prior to testing ‘‘fortifies’’ one against the effects of interference Experiment 1a and 79 different students participated in Experi-
that might arise when gum is chewed during testing, and in order ment 1b in exchange for course credit. The design of each
to observe facilitative effects of gum on performance it must be experiment was between-groups: Approximately half of the
chewed for a period of time before engaging in cognitively participants were assigned to the chewing gum condition and
demanding activity. To the best of our knowledge, there is no half served as controls. Participants were tested in sessions that
research that explored this possibility directly; therefore, one of lasted approximately 35 min. Up to five people were tested in the
the claims examined in the current study is whether the benefits of same session. The study was approved by an ethics review board
gum are more likely to emerge when it is chewed prior to, rather and all participants provided written informed consent prior to
than during, cognitive testing. completing the study.
We are also interested in exploring the time course of cognitive
facilitation due to the chewing of gum. Empirical evidence Materials and procedure
indicates that heart rate, blood pressure, and cerebral blood flow Each participant completed a battery of five tasks that
increase during gum chewing and remain elevated for 15–20 min measured performance in several cognitive domains. Two tasks
afterwards (Farella et al., 1999; Hasegawa et al., 2007; Momose provided a measure of episodic memory. In each, participants were
et al., 1997; Shinagawa et al., 2004). While these increases are shown a different list of 30 words displayed one at a time for 1.5 s
statistically significant, they are moderate in magnitude (e.g., an with a .25-s break in between. The words were 6–7 letter nouns
increase in heart rate of 9–10 heart beats per minute), and low-to- generated from the MRC Psycholinguistic Database (Coltheart,
moderate levels of arousal typically benefit memory and cognitive 1981) with concreteness, familiarity, and image ability ratings of
functioning (Revelle & Loftus, 1992). Therefore, if improved 400–650. The stimuli were projected onto a large screen at the
performance of participants who chew gum prior to testing is front of a laboratory equipped with personal computers. In the full
indeed attributable to changes in alertness and arousal, such attention task, the participants were instructed to try to remember
improvements would coincide with a window of optimal arousal the words for a later test. In the divided attention task, participants
that might reasonably be expected to last from the time chewing encoded words while also pressing random keys on the computer
has ceased until baseline levels of arousal are re-established, i.e., no keypad. They were required to press a key to each beat of a
more than 15–20 min. On the other hand, if gum is chewed metronome set at 52 beats per minute, and instructed to make
throughout testing, it may result in no improvement in perfor- digit sequences as random as possible (i.e., not to enter 1–1–1 or 1–
mance, possibly due to interference arising when the attentional 2–3, etc.). A short practice session that familiarized participants
demands of completing a cognitive task must be shared with the with the key entry procedure preceded the encoding phase. After a
demands of masticatory processes, particularly early in the testing delay of approximately 3 min during which a different task was
session (Tänzer et al., 2009; Tucha & Simpson, 2011; see also Tucha performed (see below), the participants were given 2 min to recall
et al., 2010). the words by writing them down in any order in the booklet
To summarize, the current experiments investigated whether provided. The dependent measure was the number of words
the cognitive benefits of chewing gum would be greater when gum recalled correctly.
is chewed prior to testing, and weaken or disappear altogether To measure working memory, we used a dot-matrix task
when gum is chewed during testing. We also examined the time involving a visuospatial storage component with a concurrent
course of the effect of gum chewing on cognitive function, visuospatial processing load adapted from Miyake and colleagues
anticipating that the strongest effects would manifest immediately (2001). A visuospatial working memory task was chosen to
after the cessation of chewing. Participants completed a battery of minimize interference from other, primarily verbal, tasks included
five tasks representing broad domains of cognition that included in the battery. Participants viewed sets containing 2–4 matrix
tests of episodic and working memory, processing speed, and equations (see Fig. 1) and verified whether they were true or false.
executive functioning. In Experiments 1a and 1b, participants Equations were displayed for 4.5 s during which a response (true or
performed the tasks in two different orders. In either experiment, false) was to be made using the ‘z’ and the forward slash keys of the
one group of participants chewed gum prior to the battery, while keyboard. Each equation was succeeded by a 5  5 grid shown for
another group did not chew gum at all. In Experiment 2, the order 1.5 s with one randomly placed dot inside it, and the grid was
of tasks was counterbalanced across participants, approximately followed by a 1-s mask (a black square equal in size to the grid).
[(Fig._1)TD$IG] S.V. Onyper et al. / Appetite 57 (2011) 321–328 323

Following the 5-min time estimation interval, participants in

both experiments completed the cognitive battery. The tasks in
Experiment 1a were given as a single block in the following order
(duration of each task, including time for instructions/practice, is
given in parentheses): Episodic memory encoding of a word list
under full attention conditions (2 min), followed by the symbol
Fig. 1. A sample dot-matrix equation. Participants verified whether each equation
digit modalities test (3 min), followed by the recall of words
was true or false (performing addition or subtraction operations), and then
attempted to remember the location of a dot randomly placed inside a 5  5 grid. (2 min). Next, the participants completed the working memory
Dot placements were to be marked on an answer grid after viewing a series of 2, 3, test (12 min), episodic memory encoding of a different list of words
or 4 equations/grids. In the figure, the correct response to the equation is true. under conditions of divided attention (2 min), followed by the
animal naming task (3 min), followed by recall of the second word
Participants were instructed to remember the location of the dot in list (2 min). Participants in Experiment 1b completed the same
the grid, and indicated the position of all dots in a given set on a measures in reverse order (i.e., starting with the episodic memory
blank answer grid after viewing each complete set of equations/ task under divided attention condition). After completing the
grids (the placement of the dots was not duplicated within a given performance tasks, participants indicated the frequency with
set). Fifteen such sets were completed by each participant: Five which they typically chewed gum on a scale from 1 (never) to 5
sets of 2 equations each were shown first, followed by five sets of 3 (most of the time).
equations, and finally five sets of 4 equations. The score was the
number of sets completed correctly (max. 15). Two practice sets Results and discussion
were completed prior to the start of the main task.
Participants also completed the Symbol Digit Modalities Test One participant in Experiment 1a did not complete the animal
(Smith, 1991), by matching digits 1 through 9 to symbols (simple naming task as instructed, and a participant in Experiment 1b
geometric figures) using a reference key. Symbol digit measures discontinued participation prior to completing the two final tasks.
perceptual speed of processing, motor speed, as well as visual Thus, their data from those tasks were excluded from analyses.
scanning and tracking ability (Strauss, Sherman, & Spreen, 2006). The Hostelling’s Trace multivariate test of the overall effect of
Following some brief practice, participants were presented with the chewing gum on performance was significant in both, Experiment
randomly ordered symbols and instructed to copy down the correct 1a, F(5, 73) = 8.04, p < .001, Hostelling’s trace = .55, h2 = .36, and
digit under each symbol according to the reference key as quickly Experiment 1b, F(5, 72) = 3.49, p = .007, Hostelling’s trace = .24,
and accurately as possible, without skipping any. The dependent h2 = .20. Therefore, t-tests were used to examine group differences
measure was the number of digits copied correctly in 90 s. on each performance measure. Table 1 displays mean scores for the
The fifth task was an animal naming task, a measure of verbal chewing gum and control participants, t values, and corresponding
fluency and executive functioning (Gladsjo et al., 1999) during measures of effect size. Effect sizes are reported as a standardized
which participants were given 90 s to write down the names of as difference between means (Cohen’s d). Cohen (1988) characterizes
many animals as possible. effect sizes of 0.2, 0.5, and 0.8 and above as small, medium, and
On arrival at the laboratory participants in both experiments large, respectively.
were assigned to either a chewing gum or a control condition. In Participants who chewed gum prior to completing the battery
Experiment 1a, half of the participants in the gum condition chewed performed significantly better (p < .05) than controls on measures
Wrigley’s brand spearmint flavored gum with sugar (approximately of recall (for words encoded under full attention conditions),
2 g), and half chewed Wrigley’s spearmint flavored sugar-free gum.1 working memory, and perceptual speed of processing in Experi-
Experiment 1b used Wrigley’s doublemint chewing gum. In neither ment 1a, and on measures of recall (for words encoded under
experiment did control participants receive any gum. divided attention) and working memory in Experiment 1b.
At the start of testing those in the chewing gum condition were Furthermore, there were no differences between the two groups
told that they would complete a battery of cognitive tests, and that in the frequency of habitual gum chewing, either in Experiment 1a
one of the goals of the study was to determine whether chewing (M = 3.08, SE = .15 for the group that chewed gum and M = 2.83,
gum affected the ability to estimate passage of time (this was done SE = .14 for controls, t(798) = 1.21, p = .23) or Experiment 1b (the
as a cover story to minimize demand characteristics). The chewing gum group: M = 3.20, SE = .16; controls: M = 2.89,
participants were given a stick of gum and instructed to bite it SE = .16; t(74) = 1.36, p = .18), and entering the frequency of
once every second while keeping track of the time elapsed, using a chewing as a covariate into an analysis of variance for group
grid of 600 numbered cells projected onto the screen at the front of differences on each of the five measures did not alter the statistical
the testing room as a visual cue. Participants were timed to a conclusions displayed in the table.
metronome set at 60 beats per minute for the first 10 s of the time An examination of the data shown in the table also reveals that
estimation task, and then continued in silence until asked to stop, the advantage of chewing gum was particularly evident in the
at which point they wrote their estimate in a booklet and discarded earlier portion of the test battery in both experiments. This would
the gum. The time interval was always set at 5 min. Those in the suggest that the benefits of chewing gum on performance may be
control condition were given the same instructions as participants time-limited. To examine this possibility further, we plotted effect
in the chewing gum condition, except that they received no gum. sizes representing standardized differences in mean scores of the
Since estimates of the time interval were part of the cover story chewing gum and control groups as a function of when in the session
rather than of any theoretical interest to the study, they are not the task was administered (see Fig. 2). The results indicate that tasks
discussed further. that were closest in time to when gum was chewed manifested a
stronger advantage for the gum group than those performed later in
1
The decision to include a group that chewed gum with sugar was motivated in the testing session. Across both experiments, all but one task (the
part by Stephens and Tunney’s (2004) observation that administering a large dose animal naming test in Experiment 1b) administered in the first
of glucose before chewing gum may augment cognitive effects due to the gum 20 min of testing resulted in enhanced performance of the group
chewing alone. However, no differences between the sugar-free and sugar-added
gum groups were found on any of the performance measures (perhaps because the
that chewed gum prior to completing cognitive measures. Perfor-
amount of sugar contained in gum was very small), therefore their results were mance of the two groups did not differ significantly on any of the
collapsed for all analyses. tasks completed at the end of the testing session.
324 [(Fig._2)TD$IG]
S.V. Onyper et al. / Appetite 57 (2011) 321–328

Note: Positive values of Cohen’s d indicate better performance for the group that chewed gum. The order of tasks in Experiment 1a corresponds to the order in which they are given in the table. In Experiment 1b, the order of task
Cohen’s d

0.16
0.17
0.14
0.02

0.04

administration was reversed. The values for Experiment 2 are collapsed across the two orders of administration which were counterbalanced across participants (no order effects were present for any of the tasks).
t-value

0.66
0.67
0.58

0.18
0.08
4.63(2.98)
58.31(7.37)
6.09(2.52)

3.45(2.03)

23.50(3.90)
Control

32
Experiment 2

3.41 (2.23)
56.85(9.98)

23.73(6.12)
5.70(2.34)

5.06(3.11)
Gum

33

Fig. 2. Association between the time of task administration and the magnitude of
Cohen’s d

the effect of chewing gum on cognitive performance (Cohen’s d) for Experiments 1a

(solid circles) and 1b (open circles). The horizontal axis provides a timeline of
0.11

0.59
0.45

0.36
0.05

administration of cognitive tasks (given as a difference between task start and task
completion times), from the time the session commenced (for recall tasks, the plot
shows the encoding phase). The solid line represents the best-fit linear regression to
effect size estimates from both experiments. Note: Participants in the gum group
t-value

chewed gum during the first 5 min of the testing session. AN = animal naming;
2.61*
2.01*
0.49
0.22

1.36

RF = recall of list encoded under full attention; RD = recall of list encoded under
divided attention; SD = symbol digit substitution task; WM = working memory.
57.08(11.28)
6.44(1.87)

3.94(3.28)
3.14(1.72)

24.25(6.07)

The effects of temporal factors can also be examined by

Control

considering differences in recall for words studied under full and

36

divided attention conditions either early or late in the course of

testing. There is little reason to expect that the size of group
Experiment 1b

differences in recall on either one of those tasks will vary across

57.71(13.24)

22.56(4.96)
6.17(2.95)

5.95(3.51)
4.05(2.20)

the two orders of task administration unless the timing of the test
has an impact on the magnitude of the chewing gum advantage.
Gum

The number of words recalled was examined in a three-way

43
Means (and standard deviations) of cognitive variables for the group that chewed gum and controls.

ANOVA with Condition (chewing gum or control), Time of testing

(early or late in the session), and Attentional load at encoding
Cohen’s d

(full or divided) as factors. In Experiment 1a, the full-attention

0.53

0.28
1.07

0.80

0.01

memory retrieval took place early in the testing session, and the
divided-attention retrieval took place late in testing. In Experi-
ment 1b, the reverse was true. As expected, there were main
t-value

effects of experimental condition (the group that chewed gum

4.78**

3.56**
2.37*

1.25

0.07

recalled more words, M = 5.49, than controls, M = 4.44),

F(1,154) = 13.19, p < .001, h2 = .08, and attentional load (words
studied under full attention were more likely to be remembered,
54.62(9.24)

2.98(1.79)

25.23(4.51)
4.70(2.66)
5.20(2.30)

M = 6.50 for the full attention and M = 3.43 for divided attention
Control

encoding conditions), F(1,154) = 148.50, p < .001, h2 = .49, and a

time of testing by condition interaction, F(1,154) = 6.68, p = .01,
40

h2 = .04. More importantly, there was also a three-way interac-

Experiment 1a

tion between condition, attentional load, and time of testing,

60.33(12.06)

F(1,154) = 12.21, p = .001, h2 = .07. As shown in Fig. 3, this

25.15(6.38)
8.18(3.19)

7.10(3.33)
3.60(2.61)

interaction indicates that the magnitude of the effect of chewing

Gum

gum on recall varied with the timing of the test and also
40

depended on the resource availability at encoding—while early

testing led to a greater advantage of the chewing gum group
Recall (full attention encoding)

under conditions of divided attention (d = 0.45), this advantage

was more than twice as large when full attention was given to the
Symbol digit substitution

Recall (divided attention

encoding task (d = 1.07). These results also suggest that the

benefits of gum chewing may be less pronounced with increased
Working memory

task load.
Animal naming
encoding)

Experiment 2
p < .01.
p < .05.
Measure
Table 1

As indicated earlier, a number of research studies have failed to

**
n

reveal a cognitive advantage to gum chewing. The vast majority of

[(Fig._3)TD$IG] S.V. Onyper et al. / Appetite 57 (2011) 321–328 325

Fig. 3. The effects of full and divided attention at encoding on free recall performance early and late in the testing session. The magnitude of the effect of chewing gum on recall
varied with the timing of the test but also depended on resource availability at encoding. Completing the task early in the session produced a chewing gum advantage for lists
encoded under full and divided attention conditions. The effect was stronger, however, in the full attention condition. Performance of the chewing gum and control groups did
not differ when the task was administered late in the testing session regardless of resource availability at encoding.

these studies used a procedure in which gum was chewed testing as factors revealed no main effects or interactions for any of
throughout testing with the rate of chewing under participant the individual measures, indicating that there are no order effects
control. The reasons for a lack of an effect are not clear, but may on task performance. In other words, our results are consistent
include dual-task interference resulting from concurrent mastica- with previous research that fails to find cognitive benefits of
tion (e.g., Tucha et al., 2010; Tucha & Simpson, 2011). The current chewing when gum is chewed concurrently with testing.
experiment adopts a procedure similar to those studies, with the
expectation that few (if any) of the cognitive advantages of gum General discussion
chewing seen in the preceding two experiments will be manifested
here. As an additional goal, we examine whether the testing order The current series of experiments evaluated two hypotheses
effects seen in Experiments 1a and 1b and observed for measures of regarding the relationship between chewing gum and cognitive
attention by Tänzer and colleagues (2009) and Tucha and Simpson function. First, we hypothesized that participants that chewed gum
(2011) occur for tasks included in the current experiment. for 5 min prior to engaging in cognitive testing would outperform
those that did not chew gum. This prediction was guided by an
Methods examination of literature that revealed that chewing gum was
more likely to benefit cognitive function when chewing com-
The participants were 65 St. Lawrence University under- menced for at least several minutes prior to engaging in mental
graduates who did not participate in the previous two experi- activity, but that little or no improvement would be seen if gum
ments. They were tested in sessions lasting approximately 30 min were chewed solely during cognitive tests. Second, we predicted
in groups of up to five in the same session. Upon arriving at the that if cognitive advantages of chewing gum were obtained, they
laboratory and signing informed consent, the participants com- would be most evident soon after chewing. This hypothesis was
pleted the same cognitive battery as in Experiments 1a and 1b based on the assumption that heightened arousal associated with
(minus the time estimation task). Approximately half of the masticatory processes may be responsible for enhanced cognitive
participants (N = 33) were assigned to the condition in which performance, and that this arousal would gradually diminish after
Wrigley’s doublemint gum was chewed throughout the entire cessation of chewing.
testing session, and the rest were controls that did not receive The results provide strong evidence that chewing gum is
chewing gum during the experiment. To make this experiment associated with an overall increase in cognitive functioning,
comparable to previous studies, the group that chewed gum did so particularly working memory, episodic memory, and perceptual
naturally—that is, without a metronome-imposed rhythm. The speed of processing, but only when chewing takes place prior to
order of tasks was counterbalanced within each condition; the cognitive testing. The benefits, furthermore, appear to be of limited
duration of each task was the same as in the earlier experiments. duration. In both Experiments 1a and 1b, five of the six tasks
initiated in the first half of the 30-min battery resulted in
Results and discussion statistically significant increases in performance of participants
who chewed gum (see Fig. 2). Performance on tasks that were
Table 1 presents descriptive statistics, t-values, and effect sizes completed toward the end of the testing session, however, did not
for the five cognitive tasks. The multivariate test of the overall differ between the gum and the control groups in either
effect of gum chewing on cognitive performance was not experiment. Notably, the advantages associated with the chewing
significant, F < 1. In addition, none of the t-tests between the of gum disappeared altogether when it was chewed throughout
chewing gum and the control groups were statistically significant. the entire 30-min period of testing in Experiment 2. This outcome
Furthermore, analyses of variance with condition and order of contrasts with the findings of Tänzer et al. (2009) and Tucha and
326 S.V. Onyper et al. / Appetite 57 (2011) 321–328

Simpson (2011) who found that performance increases on a task of recovery coupled with declining – yet still elevated compared to
sustained attention associated with chewing gum emerge later in baseline – levels of arousal results in facilitation of cognitive
testing, although none of our tasks were measures of attention. performance. Interestingly, forms of exercise that are less
The precise mechanisms associated with increases in perfor- physically and attentionally demanding (such as pedaling a
mance due to gum chewing are not well understood. Stephens and stationary bike versus running on a treadmill) result in little
Tunney (2004), for instance, suggested that enhanced delivery of cognitive impairment during physical activity as well as greater
glucose to the brain as a result of insulin secretion during cognitive enhancement afterwards (Lambourne & Tomporowski,
mastication was responsible for the improved performance of 2010). If we consider that chewing gum leads to cardiovascular and
participants who chewed gum. They found that participants who neurophysiological changes similar to those that occur during mild
started to chew prior to testing performed as well as those that did to moderate exercise (Hasegawa et al., 2009; Weijenberg et al.,
not chew gum but ingested a glucose-enriched drink on several 2011), then the results of our experiments would be directly
tests of episodic and working memory, and a measure of attention comparable to those reported by Lambourne and Tomporowski
and processing speed. It is also possible, however, that the outcome (2010).
observed by Stephens and Tunney can be explained by a more The possibility of shared resources is supported by the findings
generalized increase in metabolic demands as a result of that mastication activates many of the same areas of the brain as
mastication, leading to elevated arousal. Arousal theory (Sanders, mental activity (Weijenberg et al., 2011). It is also consistent with
1986; Yerkes & Dodson, 1908) holds that moderate arousal would our own observation of a reduced chewing gum advantage in recall
typically be optimal for performing a variety of cognitive tasks, when information is encoded under increased task load conditions
although task complexity mediates this effect (e.g., Revelle & (Fig. 3), although we must note that this reduction may also be due
Loftus, 1992). Mastication leads to low-to-moderate increases in to a possible floor effect in the recall scores of words in the divided
arousal and alertness as shown by physiological and behavioral attention condition. Thus, we propose that the arousal-mediated
measures (Weijenberg et al., 2011), and there is evidence that explanation of the chewing gum advantage is incomplete without
elevated levels of arousal persist for a period of time after the also considering the potential detrimental effects of the processes
cessation of chewing. For instance, Shinagawa et al. (2004) involved in the chewing of gum on maintaining optimal cognitive
observed that the chewing of gum for 5 min led to a 25% increase functionality. This dual-mechanism theory has the potential to unite
in blood flow in several cortical regions that peaked 10 min after the discrepant findings observed among studies that use different
gum was discarded and returned to baseline levels after 20 min. experimental procedure and stimulus materials, although it may
Others observed that heart rate (Farella et al., 1999; Hasegawa not readily account for all of existing findings, such as the late
et al., 2007) and blood oxygenation levels (Kamiya et al., 2010; emergence of the attentional benefits observed in the Tänzer et al.
Momose et al., 1997) would peak during chewing but also remain (2009) and Tucha and Simpson (2011) studies.
elevated afterwards. Thus, if arousal reaches levels conducive to One notable limitation of the current study is the absence of
enhanced cognitive function while gum is chewed and remains measures of physiological arousal, which makes our conclusions
within the ‘‘optimal window’’ for 15–20 min following chewing, regarding the role of arousal in modulating the cognitive
then improved performance on tasks susceptible to arousal- advantages of gum chewing speculative. Direct measures of
related benefits should emerge. Even though no arousal measures arousal coupled with careful controls will be necessary to establish
were collected in the current study, the results of the first two whether performance increases for participants who chew gum are
experiments are consistent with this possibility. due to arousal or some other factor, especially because most of the
Conversely, if levels of arousal fall outside of the optimal studies to date have provided evidence that is correlational in
window, no performance increments would be expected, which nature. For instance, while chewing could result in increased
may explain the null effect observed for participants that chewed cerebral blood flow thus enhancing performance, elevated blood
gum during cognitive testing in the final experiment. We would flow may also be the outcome of greater resource demand
argue, however, that such an explanation is not sufficient by itself necessitated by the need to control chewing and perform cognitive
to account for Experiment 2 results. One reason is that several operations concurrently. Degree of interference or dual-tasking is
recent studies demonstrated that chewing gum can interfere with another factor that must be operationalized and controlled in
performance of attentionally demanding tasks (Tänzer et al., 2009; subsequent studies. Future research incorporating controls that
Tucha et al., 2010; Tucha & Simpson, 2011; see also Miles & attain levels of arousal comparable to those induced by the
Johnson, 2007). In addition, the naturalistic pattern of chewing in chewing of gum (yet achieved via other means, such as exercise or
studies that find positive effects of gum on cognitive performance a cold pressor test) would provide a more direct validation of the
does not appear very different from the studies that find no such arousal mediation hypothesis and yield a clearer picture of the
effects, although more reliable benefits seem to emerge mainly mechanisms responsible for performance benefits associated with
when participants start chewing gum prior to testing (e.g., the chewing of gum.
Stephens & Tunney, 2004; Wilkinson et al., 2002). One of the differences between the current study and previous
It is also possible that in situations where gum chewing occurs research concerns the rate of gum chewing. Participants in
during cognitive testing potential performance benefits due to Experiments 1a and 1b were instructed to chew rhythmically, in
chewing-induced arousal are masked by the distracting nature of time to a metronome, whereas past studies have typically allowed
the chewing task. One source of insight into this possibility comes participants to set their own pace of chewing, which was also the
from studies of exercise and cognitive function. For instance, case in Experiment 2. While it is possible that at least some of the
exercise is known to heighten sympathetic nervous activity, observed differences in outcomes of the first two experiments and
increasing physical arousal. Nonetheless, a recent meta-analysis the final experiment may have been due to a discrepancy in
concluded that cognitive performance is typically reduced during chewing instructions, it should be emphasized that in Experiments
short (i.e., 20 min or less) periods of exercise but improves 1a and 1b participants assigned to the gum condition did not chew
immediately thereafter (Lambourne & Tomporowski, 2010). gum during cognitive testing (but rather prior to testing, which
Cognitive function, which relies on a limited pool of metabolic should eliminate any confounding influences of the act of chewing
resources, is thought to be compromised when those resources are itself on cognitive measures), while those in Experiment 2 did. It is
directed toward meeting the more immediate demands of also unlikely that the rate of chewing is more important than the
exercise. Once exercise is concluded, however, gradual metabolic act of chewing itself, since in previous studies that observed a
 S.V. Onyper et al. / Appetite 57 (2011) 321–328 327

cognitive advantage of gum that was chewed prior to testing the Farella, M., Bakke, M., Michelotti, A., Marotta, G., & Martina, R. (1999). Cardiovascular
responses in humans to experimental chewing of gums of different consistencies.
rate of chewing was uncontrolled (e.g., Sakamoto et al., 2009; Archives of Oral Biology, 44, 835–842.
Stephens & Tunney, 2004; Wilkinson et al., 2002). Nonetheless, a Gladsjo, J. A., Schuman, C. C., Evans, J. D., Peavy, G. M., Miller, S. W., & Heaton, R. K.
direct examination of the effects of the manner in which gum is (1999). Norms for letter and category fluency. Demographic corrections for age,
education, and ethnicity. Assessment, 6, 147.
chewed (rhythmically or naturally) is necessary to fully resolve Johnson, A. J., & Miles, C. (2008). Chewing gum and context dependent memory. The
this question. independent roles of chewing gum and mint flavour. British Journal of Psychology,
It is also possible that the cognitive effects of gum chewing 99, 293–306.
Hasegawa, Y., Ono, T., Hori, K., & Nokubi, T. (2007). Influence of human jaw movement
observed in our study are confounded with flavor—for instance, on cerebral blood flow. Journal of Dental Research, 86, 64–68.
there is evidence that some flavors lead to an amplification of Hasegawa, Y., Sakagami, J., Ono, T., Hori, K., Zhang, M., & Maeda, Y. (2009). Circulatory
cognitive effects, even in the absence of chewing (Zoladz & response and autonomic nervous activity during gum chewing. European Journal of
Oral Sciences, 117, 470–473.
Raudenbush, 2005; see also Johnson & Miles, 2008). On the other
Hirano, Y., Obata, T., Kashikura, K., Nonaka, H., Tachibana, A., Ikehira, H., et al. (2008).
hand, studies (e.g., Sakamoto et al., 2009) also reveal performance Effects of chewing in working memory processing. Neuroscience Letters, 436, 189–
improvements after participants chew flavorless gumbase, sug- 192.
gesting that the presence of gum itself is sufficient to induce Kamiya, K., Fumoto, M., Kikuchi, H., Sekiyama, T., Mohri-Iluzawa, Y., Umino, M., et al.
(2010). Prolonged gum chewing evokes activation of the ventral part of prefrontal
changes in cognitive functioning. cortex and suppression of nociceptive responses. Involvement of the serotonergic
Finally, even though the performance benefits seen in our system. Journal of Medical and Dental Sciences, 57, 35–43.
results apply to several types of cognitive processing – in Kohler, M., Pavy, A., & Van den Heuvel, C. (2006). The effects of chewing versus caffeine
on alertness, cognitive performance and cardiac autonomic activity during sleep
particular, working memory, episodic memory, and processing deprivation. Journal of Sleep Research, 15, 358–368.
speed – we stop short of making a claim that the benefits of gum Lambourne, K., & Tomporowski, P. (2010). The effect of exercise-induced arousal on
chewing are domain-general. For instance, those who chewed gum cognitive task performance. A meta-regression analysis. Brain Research, 1341, 12–
24.
prior to the test battery did not show a performance advantage on Miles, C., & Johnson, A. J. (2007). Chewing gum and context dependent memory. A
the animal naming task, a measure of verbal fluency typically reexamination. Appetite, 48, 154–158.
associated with executive functions. Others (e.g., Stephens & Miyake, A., Friedman, N. P., Rettinger, D. A., Shah, P., & Hegarty, M. (2001). How are
visuospatial working memory, executive functioning, and spatial abilities related?
Tunney, 2004) also failed to demonstrate the effects of chewing on A latent-variable analysis. Journal of Experimental Psychology: General, 130(4), 621–
verbal fluency. Measures of fluency are often thought to reflect 640.
prefrontal functioning, specifically in the dorsolateral region Momose, I., Nishikawa, J., Watanabe, T., Sasaki, Y., Senda, M., Kubota, K., et al. (1997).
Effect of mastication on regional cerebral blood flow in humans examined by
(Troyer, Moscovitch, Winocur, Alexander, & Stuss, 1998). While
positron-emission tomography with 15O-labelled water and magnetic resonance
we know relatively little about the causes of performance variation imaging. Archives of Oral Biology, 42, 57–61.
among different cognitive measures due to gum chewing, there is Morinushi, T., Masumoto, Y., Kawasaki, H., & Takigawa, M. (2000). Effect on electro-
evidence that the effects of chewing on the brain are selective, encephalogram of chewing flavored gum. Psychiatry and Clinical Neurosciences, 54,
645–651.
rather than general, and that they do not typically lead to Nader, I. W., Gittler, G., Waldherr, K., & Pietschnig, J. (2010). Chew on this. No support
differential activation of dorsolateral prefrontal cortex (Kamiya for facilitating effects of gum on spatial task performance. Archives of Oral Biology,
et al., 2010; Momose et al., 1997; Onozuka et al., 2002; but see 55, 712–717.
Onozuka, M., Fujita, M., Watanabe, K., Hirano, Y., Niwa, M., Nishiyama, K., et al. (2002).
Takada & Miyamoto, 2004). Mapping brain region activity during chewing. A functional magnetic resonance
In summary, the current study demonstrates that the dis- imaging study. Journal of Dental Research, 81, 743–746.
crepancies in research findings of the burgeoning literature on the Revelle, W., & Loftus, D. A. (1992). The implications of arousal effects for the study
of affect and memory. In S. Christianson (Ed.), The handbook of emotion and
effect of gum chewing on cognitive function can be attributed to memory. Research and theory (pp. 113–145). Hillsdale, NJ: Lawrence Erlbaum
the timing of chewing. Clear performance advantages emerge Associates .
when gum is chewed prior to (but not during) cognitive testing, Sakamoto, K., Nakata, H., & Kakigi, R. (2009). The effect of mastication on human
cognitive processing. A study using event-related potentials. Clinical Neurophysi-
although the benefits persist only for the first 15–20 min of the
ology, 120, 41–50.
testing session. Our findings also suggest that the benefits of gum Sanders, A. F. (1986). Towards a model of stress and human performance. Acta
chewing extend only to some cognitive domains. A dual-process Psychologica, 53, 61–97.
Scholey, A. (2004a). Chewing gum and cognition performance. A case of a functional
mechanism is proposed to account for the presence of the chewing
food with function but no food? Appetite, 43, 215–216.
gum advantage when gum is chewed prior to engagement in Scholey, A. (2004b). Further issues regarding the possible modulation of cognitive
demanding cognitive processing, and its elimination when gum is function by the chewing of gum. Response to Stephens and Tunney (2004) and
chewed during mental activity. Mastication-induced arousal is Tucha et al. (2004). Appetite, 43, 221–223.
Scholey, A., Haskell, C., Robertson, B., Kennedy, D., Milner, A., & Wetherell, M. (2009).
postulated as the potential mechanism to explain the time-limited Chewing gum alleviates negative mood and reduces cortisol during acute labora-
nature of performance enhancement, and interference due to a tory psychological stress. Physiology and Behavior, 97, 304–312.
sharing of metabolic resources by cognitive and masticatory Sesay, M., Tanaka, A., Ueno, Y., Lecaroz, P., & de Beaufort, D. G. (2000). Assessment
of regional cerebral blood flow by xenon-enhanced computed tomography
processes is thought to explain why performance may not increase during mastication in humans. Keio Journal of Medicine, 49(Suppl. 1), A125–
or even decline. Additional tests of the dual-process theory of the A128.
chewing gum advantage are necessary to delineate both its Shinagawa, H., Ono, T., Honda, E., Sasaki, T., Taira, M., Iriki, A., et al. (2004). Chewing-
side preference is involved in differential cortical activation patterns during tongue
generality and limitations, and further studies are needed to movements after bilateral gum-chewing. A functional magnetic resonance imag-
provide a more complete picture of the relationship between ing study. Journal of Dental Research, 83, 762–766.
physiological changes and cognitive functioning due to the Smith, A. (1991). Symbol digit modalities test. Los Angeles, CA: Western Psychological
Services.
chewing of gum.
Smith, A. (2009a). Effects of chewing gum on mood, learning, memory and perfor-
mance of an intelligence test. Nutritional Neuroscience, 12(2), 81–89.
Smith, A. (2009b). Effects of caffeine in chewing gum on mood and attention. Human
References Psychopharmacology, 24, 239–247.
Smith, A. (2010). Effects of chewing gum on cognitive function, mood and physiology in
Allen, K. L., Norman, R. G., & Katz, R. V. (2008). The effect of chewing gum on learning as stressed and non-stressed volunteers. Nutritional Neuroscience, 13(1), 7–16.
measured by test performance. Nutrition Bulletin, 33, 102–107. Stephens, R., & Tunney, R. J. (2004). Role of glucose in chewing gum-related facilitation
Baker, J. R., Bezance, J. B., Zellaby, E., & Aggleton, J. P. (2004). Chewing gum can produce of cognitive function. Appetite, 43, 211–213.
context-dependent effects upon memory. Appetite, 43, 207–210. Strauss, E., Sherman, E., & Spreen, O. (2006). A compendium of neuropsychological tests.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Administration, norms, and commentary (3rd edition). New York: Oxford University
Lawrence Erlbaum Associates. Press.
Coltheart, M. (1981). The MRC psycholinguistic database. Quarterly Journal of Experi- Tänzer, U., von Fintel, A., & Eikermann, T. (2009). Chewing gum and concentration
mental Psychology, 33A, 497–505. performance. Psychological Reports, 105, 372–374.
328 S.V. Onyper et al. / Appetite 57 (2011) 321–328

Takada, T., & Miyamoto, T. (2004). A fronto-parietal network for chewing of gum. A Tucha, L., Simpson, W., Evans, L., Birrel, L., Sontag, T. A., Lange, K. W., et al. (2010).
study on human subjects with functional magnetic resonance imaging. Neurosci- Detrimental effects of gum chewing on vigilance in children with attention deficit
ence Letters, 360, 137–140. hyperactivity disorder. Appetite, 55, 679–684.
Torney, L. K., Johnson, A. J., & Miles, C. (2009). Chewing gum and impasse-induced self- Weijenberg, R. A. F., Scherder, E. J. A., & Lobbezoo, F. (2011). Masstication for the mind.
reported stress. Appetite, 53, 414–417. The relationship between mastication and cognition in ageing and dementia.
Troyer, A. K., Moscovitch, M., Winocur, G., Alexander, M. P., & Stuss, D. (1998). Neuroscience and Biobehavioral Reviews, 35, 483–497.
Clustering and switching on verbal fluency. The effects of focal frontal- and Wilkinson, L., Scholey, A., & Wesnes, K. (2002). Chewing gum selectively improves
temporal lobe lesions. Neuropsychologia, 36, 499–504. aspects of memory in healthy volunteers. Appetite, 38, 235–236.
Tucha, O., Mecklinger, L., Maier, K., Hammerl, M., & Lange, K. W. (2004). Chewing gum Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of
differentially affects aspects of attention in healthy subjects. Appetite, 42, 327–329. habit-formation. Journal of the Comparative Neurology and Psychology, 18, 459–482.
Tucha, L., & Simpson, W. (2011). The role of time on task performance in modifying the Zoladz, P. R., & Raudenbush, B. (2005). Cognitive enhancement through stimulation of
effects of gum chewing on attention. Appetite doi:10.1016/j.appet.2010.12.021. the chemical senses. North American Journal of Psychology, 7, 125–138.