This article was downloaded by: [37.188.236.47]
On: 15 February 2013, At: 08:31
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Cognitive Psychology
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/pecp21
Caffeine increases false memory in nonhabitual
consumers
Caroline R. Mahoney
a b
, Tad T. Brunyé
a b
, Grace E. Giles
a b
, Tali Ditman
a c
,
Harris R. Lieberman
d
& Holly A. Taylor
a
a
Department of Psychology, Tufts University, Medford, MA, USA
b
US Army Natick Soldier Research, Development, and Engineering Center, Natick,
MA, USA
c
Massachusetts General Hospital/Harvard Medical School, Athinoula A. Martinos
Center for Biomedical Imaging, Charlestown, MA, USA
d
US Army Research Institute for Environmental Medicine, Natick, MA, USA
Version of record first published: 07 Mar 2012.
To cite this article: Caroline R. Mahoney , Tad T. Brunyé , Grace E. Giles , Tali Ditman , Harris R. Lieberman & Holly
A. Taylor (2012): Caffeine increases false memory in nonhabitual consumers, Journal of Cognitive Psychology, 24:4,
420-427
To link to this article: http://dx.doi.org/10.1080/20445911.2011.647905
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial or
systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution
in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the
contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and
drug doses should be independently verified with primary sources. The publisher shall not be liable for
any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused
arising directly or indirectly in connection with or arising out of the use of this material.
Caffeine increases false memory in nonhabitual
consumers
Caroline R. Mahoney1,2
, Tad T. Brunye´ 1,2
, Grace E. Giles1,2
, Tali Ditman1,3
,
Harris R. Lieberman4
, and Holly A. Taylor1
1
Department of Psychology, Tufts University, Medford, MA, USA
2
US Army Natick Soldier Research, Development, and Engineering Center, Natick, MA,
USA
3
Massachusetts General Hospital/Harvard Medical School, Athinoula A. Martinos Center
for Biomedical Imaging, Charlestown, MA, USA
4
US Army Research Institute for Environmental Medicine, Natick, MA, USA
Insight into caffeine’s equivocal effects on memory can be derived from work suggesting both emotional
arousal and psychosocial stress increase false memory rates without increasing veridical memory. This
study investigated how a range of caffeine doses affect veridical and false memory formation in
nonhabitual consumers. A double-blind, repeated-measures design with caffeine (0 mg, 100 mg, 200 mg,
400 mg caffeine) was used to examine memory using the Deese-Roediger-McDermott (DRM) paradigm.
Results showed that caffeine modulated arousal levels, peaking at 200 mg and returning to near baseline
levels at 400 mg. Main effects of caffeine demonstrated higher critical lure recall and recognition ratings
(i.e., false memory) as a function of dose, again peaking at 200 mg. Those who showed the highest arousal
increases as a function of caffeine also tended to produce the highest false recall and recognition rates.
Veridical memory was not affected. Results demonstrate that consumption of as little as 100 mg of
caffeine elicits reliable inverted-U shape changes in arousal and, in turn, false memories in individuals
who do not habitually consume caffeine.
Keywords: Arousal; Caffeine; False memory; Veridical memory.
Caffeine (1,3,7-trimethylxanthine) is found naturally
in food and beverages, such as coffee, tea,
and chocolate, and has recently become popular
as a supplement in commercially available energy
drinks and food bars. Population surveys of
consumption in the United States indicate that
over 80% of adults habitually consume caffeine
(average 280 mg/day; Barone & Roberts, 1996).
Caffeine has well known effects on the central
nervous system that include increased alertness,
wakefulness, motivation, and motor activity, as
well as increased neuronal activity (for a review,
see Lieberman, 2003). It is often cited for its
positive effects on attention and basic psychomotor
tasks, such as visual vigilance, simple reaction
time, and choice reaction time (for example,
Lieberman, Tharion, Shukitt-Hale, Speckman, &
Tulley, 2002). The beneficial effects of caffeine on
performance are commonly attributed to caffeine’s
antagonistic role at adenosine A1 and
A2A receptors in dopamine-rich brain areas
(Garrett & Griffiths, 1997).
However, arousal does not always have positive
effects on cognitive processes. Data regarding
Correspondence should be addressed to Caroline R. Mahoney, US Army Soldier Center, Kansas Street, Natick, MA 01760-5020,
USA. E-mail: caroline.mahoney@us.army.mil
The opinions expressed herein are those of the authors and do not reflect those of the United States Army.
JOURNAL OF COGNITIVE PSYCHOLOGY, 2012, 24 (4), 420Á427
# 2012 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business
http://www.psypress.com/ecp http://dx.doi.org/10.1080/20445911.2011.647905
Downloadedby[37.188.236.47]at08:3115February2013
caffeine’s effects on memory are equivocal, with
some studies reporting positive effects and others
reporting no effects or negative effects (for
example, Childs & de Wit, 2006). Some insights
into caffeine’s effects on memory, however, can
be derived from recent work suggesting both
emotional arousal, such as that accompanying
anger or fear, and arousal following psychosocial
stress, increased false memory recall, and recognition
rates without necessarily increasing veridical
(true) memory (Corson & Verrier, 2007;
Payne, Nadel, Allen, Thomas, & Jacobs, 2002).
The effects of emotional arousal on false recall
and recognition may be due to increased relational
processing (Brunye´, Mahoney, Augustyn, &
Taylor, 2009; Corson & Verrier, 2007), possibly
resulting from arousal induced changes in cortisol
in the PFC and hippocampal systems (Mayer &
Gaschke, 1988). In general, false memories for
word lists are thought to reflect the associative
nature of verbal memory (Hutchinson & Balota,
2005; Roediger & McDermott, 1995). The extent
to which participants recall or recognise previously
presented words provides an indication
of veridical verbal memory, whereas spontaneously
recalling or giving high recognition ratings
to highly associated, but never presented,
words provides a measure of false memory. Given
the ubiquity of caffeine consumption and the
applied and legal implications of both veridical
and false memory formation, it is important to
understand if the physiological arousal resulting
from caffeine consumption also affects false recall
and recognition.
To understand how arousal affects both veridical
and false memory formation, a double-blind,
within-participants repeated-measures design
tested the effects of a range of caffeine doses
(0Á400 mg) on recall and recognition in nonhabitual
caffeine consumers using the Deese-Roediger-McDermott
(DRM) paradigm. This design
allowed for a targeted (i.e., isolated effects of
arousal) and comprehensive (i.e., a parametric
range of doses) evaluation of arousal effects on
veridical and false memory. The DRM paradigm
involves learning lists of words that are each
highly associated with one nonpresented word,
the critical lure (Roediger & McDermott, 1995).
After encoding, participants complete recall and
recognition tests. Veridical memory rates are
determined by measuring correctly recalled and
recognised words, and false memory rates are
determined by measuring falsely recalled or recognised
critical lures. Given earlier work suggesting
emotional arousal may increase rates of false
memory without affecting veridical memory in the
DRM paradigm (Corson & Verrier, 2007; Payne et
al., 2002), it can be predicted that if caffeine
consumption leads to increased arousal and, in
turn, increased relational processing (i.e., associative
lexical activation; Hutchinson & Balota,
2005), then rates of false recall and recognition
should follow a positive doseÁresponse relationship
and veridical memory should remain relatively
unaffected.
MATERIALS AND METHODS
Design
The present study used a double-blind, repeatedmeasures
design with four levels of treatment
(0 mg, 100 mg, 200 mg, 400 mg caffeine). The
highest dose of caffeine matches that found in a
20 oz coffee portion served at a major franchise
coffee house (i.e., 415 mg; www.starbucks.com).
Treatment order was counterbalanced across
participants.
Participants
Thirty six male (16) and female (20) volunteers
between the ages of 18 and 35 were recruited
from the Tufts University student population.
They were 19.0 (SD01.26; range 18Á22) years
old, 171.5 (SD08.6; range 62Á75) cm in height,
and weighed 68.3 (SD011.5; range 110Á208) kg.
All participants were low caffeine consumers
(self-report of M041.3, SD028.8, range 0Á102
mg/day), nonsmokers, in good health, did not use
prescription medication other than oral contraceptives,
and did not use nicotine in any form.
Written informed consent was obtained, and all
procedures were jointly approved by the Tufts
University Institutional Review Board and the
Human Use Review Committee of the US Army
Research Institute for Environmental Medicine.
Manipulation check
Participants used the Brief Mood Introspection
Scale (BMIS) to rate their current mood and
arousal state in accordance with 16 adjectives
(eight positive and eight negative) on a series of
CAFFEINE AND FALSE MEMORY 421
Downloadedby[37.188.236.47]at08:3115February2013
Likert scales anchored at 1 (‘‘definitely do not
feel’’) to 4 (‘‘definitely feel’’).
Cognitive tests
DRM paradigm. Fifty word lists were used.
Each list contained 15 words and was randomly
assigned to one of five sets. Thus, each participant
saw 10 new word lists on each of the five test
sessions. The sets of word lists were counterbalanced
across test sessions. Lists were presented
via headphones, at a rate of one word every 2 s.
Lists were presented in random order. Immediately
after each list was presented, participants
engaged in a 1-minute recall test. After the 10th
list was presented and recall completed, participants
completed a recognition test consisting of
60 randomly ordered words including 30 ‘‘old’’ or
presented words (the first, eighth, and 10th item
of each list), the 10 critical lures and 20 ‘‘new’’
unrelated words. Participants rated each word on
a scale ranging from 1 (‘‘sure it was new’’) to 4
(‘‘sure that the item was studied’’).
Caffeine or placebo administration
In order to control for taste, caffeine or placebo
was administered in capsule form. All treatment
doses were administered in an identical colour,
size, weight, and shape capsule. Capsules contained
either 0 mg, 100 mg, 200 mg, or 400 mg of
caffeine. Placebo capsules were filled with physiologically
inert microcrystalline cellulose powder,
which was also used as filler material in the
two lower-dose caffeine capsules. The caffeine
was 99.8% pure anhydrous USP-grade powder.
Procedure
Participants completed all four treatment conditions
and a normal consumption day on separate
days, resulting in five test sessions. There was a
minimum 3-day wash out period between test
sessions. Participants were instructed not to eat or
drink anything (with the exception of water) after
9:00 p.m. the night before a test session and not to
use any over-the-counter medications or herbal
supplements 24 hours prior to testing. During the
normal consumption day, participants were
allowed to consume their normal amount of
caffeine prior to arrival for testing. Test sessions
began between 8:00 and 9:30 a.m.
When participants arrived in the morning, they
completed a baseline BMIS and then consumed a
capsule containing varying doses of caffeine or
placebo along with a cup of water. Sixty minutes
after consuming the capsule, participants completed
another BMIS and then immediately
began the DRM task. Timing of testing was based
on literature showing peak plasma concentrations
of caffeine occur approximately 45Á60 minutes
after ingestion. Participants completed the
same test sequence for each of the treatment
conditions.
RESULTS
Statistical analyses
DRM data were analysed with repeated measures
analysis of variance (ANOVA) with treatment
(0 mg, 100 mg, 200 mg, and 400 mg) as a withinsubject
variable. Baseline BMIS data were analysed
by ANOVA with treatment (0 mg, 100 mg,
200 mg, and 400 mg) and adjective (16 separate
adjectives) as within-subjects variables. Posttreatment
BMIS difference score data [(60 minutes
post-treatment) Á (baseline)] were analysed
by ANOVA with treatment (0 mg, 100 mg, 200
mg, and 400 mg) and adjective (16 separate
adjectives) as within-subject variables. An effect
was deemed statistically significant if the likelihood
of its occurrence by chance was p5005. If
an ANOVA yielded a significant main effect,
planned comparisons in the form of t-tests were
performed (results of planned comparisons are
provided in Figure 1). All statistical analyses
were performed using SPSS 12.0. Two participants
were excluded for failing to perform one or more
recall tests.
Manipulation check: BMIS
As expected, baseline affect measures did not
vary as a function of treatment condition, across
any of the 16 adjectives in a 4 (treatment
condition: 0, 100, 200, 400 mg))16 (adjectives)
ANOVA (Fs B1).
An ANOVA on BMIS difference scores
revealed an effect of adjective, F(15,
495)010.12, p B.01, h2
0.09; more importantly,
422 MAHONEY ET AL.
Downloadedby[37.188.236.47]at08:3115February2013
adjective interacted with treatment condition,
F(45, 1485)02.79, p B.01, h2
0.05. To specifically
examine treatment condition effects on
individual adjective ratings, we conducted a
series of ANOVAs. Eight adjectives (caring,
content, gloomy, jittery, grouchy, loving, fed up,
active) did not vary as a function of treatment
condition. The other eight adjectives showed
effects of treatment condition (all ANOVA
psB.05); these are detailed in Table 1, along
with results from paired t-tests comparing each
treatment condition to the 0 mg condition. Note
that in no case did any individual adjective rating
differ between the 200 mg and 400 mg conditions.
Exploratory analysis included gender in the
earlier ANOVA; no main or interactive effects
were identified (pmin0.14).
Recall
A main effect of caffeine demonstrated higher
critical lure recall as a function of treatment
condition, peaking at 200 mg, F(3, 99)08.83,
p B.01, h2
0.20 (see Figure 1). Exploratory
analysis included gender in the previous ANOVA;
no main or interactive effects were identified
(pmin0.07). Veridical recall of old items did not
vary as a function of treatment condition, F(3,
99)01.39, p0.25.
Figure 1. Results from the recall and recognition tests, with standard error bars, for each of the four caffeine doses (0, 100, 200,
400). Asterisks indicate results from paired t-tests comparing to 0 mg condition: **p B.01,*p B.05, m
p B.10. Within critical recall,
200 mg 100 mgm
, and within critical recognition, 200 mg 100 mg*.
TABLE 1
Mean and standard error BMIS difference scores (pre- to post-capsule administration) for each treatment condition
NCD 0 mg 100 mg 200 mg 400 mg
Adjective M SE M SE M SE M SE M SE
Lively 0.13 0.13 (0.09 0.17 0.24 0.14 0.29* 0.14 0.29* 0.16
Happy (0.22 0.14 (0.56 0.12 0.02** 0.12 0.00** 0.09 (0.21* 0.13
Sad (0.10 0.13 (0.09 0.11 (0.21* 10 0.18 0.11 (0.15 0.06
Tired (0.57 0.13 (0.03 0.18 (0.50* 0.13 (0.59** 0.12 (0.45* 0.06
Drowsy (0.42 0.15 0.12 0.18 (0.41** 0.13 (0.53** 0.15 (0.59** 0.16
Peppy (0.01 0.12 (0.12 0.15 0.15 0.19 0.24* 0.13 0.30* 0.14
Nervous (0.22 0.12 0.06 0.12 (0.12 0.08 0.18 0.19 0.41 0.15
Calm (0.13 0.11 (0.12 0.16 (0.09 0.11 (0.12 0.16 (0.33 0.15
NCD: Normal Consumption Day. Asterisks indicate significance in t-tests comparing each treatment level to the 0 mg condition
(*pB.05, **pB.01). Nervous and calm showed only nonsignificant differences. All other BMIS adjectives p.05.
CAFFEINE AND FALSE MEMORY 423
Downloadedby[37.188.236.47]at08:3115February2013
Recognition
As with recall data, critical lure recognition
ratings increased as a function of caffeine dose,
again peaking at 200 mg, F(3, 99)07.53, pB.01,
h2
0.19 (see Figure 1). Exploratory analysis
included gender in the earlier ANOVA; no main
or interactive effects were identified (pmin0.55).
Veridical ratings of old items decreased marginally
across caffeine doses, F(3, 99)02.67, p0.052,
h2
0.07, and correct rejection of unrelated items
did not change (FB1).
We also assessed false recognition using a gist
memory d? measure, which compares critical lure
ratings to ratings of unstudied unrelated lures
(with a 1 or 2 rating recoded as ‘‘no’’, and a 3 or 4
rating recoded as ‘‘yes’’). This method treats
‘‘yes’’ responses to critical lures as hits and
‘‘yes’’ responses to unrelated lures as false alarms.
As such, higher d? reveals higher global associative
processing (i.e., gist memory; Koutstaal &
Schacter, 1997). An ANOVA comparing d? across
treatment conditions revealed a main effect, F(3,
99)011.87, pB.01, h2
0.26. Overall, relative to
placebo (M01.55, SE00.18) there were higher
d? levels in the 100 mg (M01.91, SE00.19),
t(33)02.63, p0.01, and the 200 mg (M02.16,
SE00.23), t(33)03.48, p B.01, conditions
(400 mg, M01.27, SE00.19), p .05.
A more traditional d? measure, calculated by
comparing hits to old words versus false alarms to
critical lures (using recoded ratings, as before),
also showed a main effect, F(3, 99)06.52, pB.01,
h2
0.16. Overall, relative to placebo (M01.17,
SE00.21) there were lower d? levels in the
100 mg (M00.64, SE00.21), t(33)03.44,
p B.01, and the 200 mg (M00.51, SE00.22),
t(33)03.41, p B.01, conditions (400 mg, M01.08,
SE00.20), p .05. We also calculated response
criterion, which demonstrated a main effect of
treatment, F(3, 99)08.28, pB.01, h2
0.20, with a
more lenient criterion at 200 mg (c0(0.70),
relative to placebo (c0(0.40), t(33)03.29,
p B.01 (all other ps .05). This type of criterion
shift often accompanies higher critical lure false
alarms (e.g., Koutstaal & Schacter, 1997).
Withdrawal effects
To confirm that our results are not attributable to
caffeine withdrawal effects, we conducted t-tests
to compare the normal consumption day recall
and recognition data to performance on the 0 mg
day. No significant differences were found (all
ps .05). Normal consumption day data are
detailed in Table 2, for both the BMIS and
memory tasks.
Using arousal to predict false memory
To test whether caffeine-induced BMIS arousal
differences between the 200 mg and 0 mg
conditions predict recall and recognition results,
we conducted two linear regressions. First, for
each participant we calculated a composite
arousal measure for each treatment dose that
averaged the four arousal-related (Mayer &
Gaschke, 1988) adjectives that showed differences
between 200 and 0 mg (i.e., Table 1):
lively, tired (reverse-scored), drowsy (reversescored),
and peppy. We then used this composite
measure to calculate difference scores between
the 200 mg and 0 mg conditions (200Á0 mg). We
used these BMIS arousal difference scores to
predict, in separate regressions, recall and recognition
critical lure false alarm difference scores
(200Á0 mg). For recall results, those with greater
arousal differences as a function of caffeine dose
also tended to show the greatest increases in
critical lure recall, b00.44, t(33)02.77, p B.01;
this was not true for veridical recall, b00.15,
t(35)00.84, p .05. For recognition results, the
same effect was found for critical lure recognition,
b00.43, t(35)02.69, p B.05, and inversely
for old item recognition, b0(0.38, t(35)02.32,
p B.05.
TABLE 2
Mean and standard error recall and recognition ratings for the
normal consumption day
M SE
Recall rates
Critical lure recall 0.12 0.02
Veridical recall 0.68 0.01
Recognition rates
Critical lure recall 2.49 0.09
Veridical recall 3.51 0.04
CR of unrelated items 1.09 0.02
d? traditional sensitivity 1.09 0.14
d? gist sensitivity 1.75 0.12
c response criterion (0.50 0.07
424 MAHONEY ET AL.
Downloadedby[37.188.236.47]at08:3115February2013
DISCUSSION
The present study examined effects of a range of
caffeine doses (0Á400 mg) on veridical verbal
memory and false recall and recognition using the
Deese-Roediger-McDermott (DRM) paradigm.
Converging evidence from our manipulation
check and memory tasks demonstrates that caffeine
intake approximating a 12Á16 oz coffee at
major franchise coffee houses elicits reliable and
parametric increases in arousal and, in turn, false
memories, without substantially altering veridical
memory. Specifically, we found higher phenomenological
experiences of arousal and higher false
recall and recognition critical lures after participants
consumed as little as 100 mg of caffeine.
Our results are consistent with previous work
demonstrating that increases in arousal lead to
greater relational, sometimes at the cost of itemspecific,
processing (Brunye´ et al., 2009) and are
in agreement with research showing that arousal
enhances elaborative encoding and increases
suggestibility (for example, Porter, Spencer, &
Birt, 2003). They also extend previous work
examining caffeine influences on false memory.
Although false memory did not continue to
increase at the 400 mg dose, as was hypothesised,
this finding is consistent with other work that
suggests performance advantages diminish with
high caffeine doses. In fact, cognitive processes
such as vigilance, executive control, and mood
(e.g., Lieberman et al., 2002) commonly show an
inverted-U pattern of caffeine effects, particularly
in low consumers.
To date, only two studies have examined the
influence of caffeine-related arousal on false
memory formation and both used only a single
moderate dose (200 mg). The first study assessed
whether caffeine administration would enhance
recall memory for word lists as well as increase
false recall for semantically related words (Capek
& Guenther, 2009). The results are consistent
with ours in that a 200 mg caffeine dose
increased false recall, but differ from ours in
that this dose improved veridical recall. The
second study examined how sleep deprivation
affects false recall and if caffeine administration
prior to retrieval would attenuate this effect
(Diekelmann, Landolt, Lahl, Born, & Wagner,
2008). Contrary to our findings, the results of this
study suggest that sleep-deprived participants
had more false memories and caffeine consumption
led to fewer false memories. These findings
are contrary to ours, but it is difficult to compare
data from well-rested individuals with those who
are 9 hours sleep deprived. Further, it could be
the case that the caffeine administered in Diekelmann
et al. (2008) returned sleep-deprived
participants to a baseline level of arousal at
which, as in the present study, there were
relatively moderate false alarm rates. In addition,
both Capek and Guenther (2009) and Diekelmann
et al. used between-participant designs and
did not account for normal consumption habits.
Therefore, it should be noted that not only does
our study further existing work by providing a
parametric assessment of arousal effects on false
recall and recognition, but it also better controlled
for individual differences in caffeine
metabolism and possible effects of subjective
experiences of arousal by using a within-subjects
design and controlling for normal consumption
habits. In addition, by testing only in the morning
after an overnight fast, our study limited potential
confounding effects associated with circadian
rhythms and differences in macro- and micronutrient
consumption prior to testing. Finally, testing
only low consumers and including a normal
consumption test session reduced the confound
of changes in glucocorticoid release associated
with habitual caffeine use (e.g., nonconsumer vs.,
regular consumer) and demonstrated that effects
we observed were due to increased arousal
associated with caffeine consumption, rather
than symptoms of caffeine withdrawal.
The mechanism by which caffeine may influence
false memory formation is unknown. One
possibility is that caffeine influences memory via
an increase in glucocorticoid production. Glucocorticoids
are produced by the hypothalamicpituitary-adrenal
(HPA) axis in response to
physical and psychological stress. In humans,
cortisol is the primary glucocorticoid. Cortisol
is known to influence neuronal metabolism in
the brain, particularly the hippocampus, a structure
linked to memory function with dense
concentrations of glucocorticoid receptors. For
example, cortisol can reduce glucose uptake in
the hippocampus as shown with positron emission
tomography (PET) and reduce activation in
the hippocampus during memory retrieval as
shown with fMRI (Oei et al., 2007). Furthermore,
evidence suggests that emotional arousal,
psychosocial stress, and/or high levels of glucocorticoids
that accompany stress can impair
performance on memory tasks in normal participants
and those with cognitive impairments
CAFFEINE AND FALSE MEMORY 425
Downloadedby[37.188.236.47]at08:3115February2013
that can affect memory function, such as Cushing
syndrome, Alzheimer’s, schizophrenia, and depression
(Corson & Verrier, 2007; Kirschbaum,
Wolf, May, Wippich, & Hellhammer, 1996; Payne
et al., 2002; Wolf, 2009).
Consumption of caffeine typically results in
increased cortisol release, especially in nonhabitual
consumers given moderate or high doses of
caffeine (Lovallo et al., 2005). This modulation of
cortisol may disrupt the PFC and hippocampal
systems contributions to contextual remembering.
It is generally agreed the hippocampus and PFC
are important sites for contextual remembering
either by binding together elements of the contextual
situation or through source monitoring
(Mitchell, Johnson, Raye, & D’Esposito, 2000).
According to the theory of spreading activation,
the ability to distinguish between presented words
and nonpresented words that were associatively
activated depends in part on the ability to apply
context, such as the order of a particular word on
a list or what preceded it. Stress, or glucocorticoid
release following stress, may disrupt formation of
contextual representations via the modulation of
the PFC and hippocampal systems, and therefore
increase the difficulty in distinguishing words that
were actually presented (i.e., ‘‘old’’ words) and
semantically related words that were associatively
activated (i.e., critical lures) (Payne et al., 2002).
Of course, the present design does not allow us to
disentangle the effect of caffeine on the different
stages of memory formation; given that both
encoding and retrieval processes were done under
the influence of caffeine. This limitation presents
a possibility for future research regarding the
possible differential influence of caffeine administration
on encoding and retrieval processes.
The present results might also be attributed to
caffeine-modulated changes in the distinctiveness
between studied and critical items. The distinctiveness
heuristic proposes that word memory
accuracy is determined by both verifying the
correctness of list items and rejecting critical
lures, and distinctive processing serves both functions
(Dodson & Schacter, 2001). Distinctive
processing allows people to recognise differences
between items and lures within the context of
high similarity. Predictions offered by the distinctiveness
heuristic provide a potential explanation
for the relatively lenient response criterion to
critical lures at 200 mg: Caffeine-related arousal
might promote nondistinctive processing of old
items and critical lures. Given that caffeinerelated
memory effects were specific to critical
lures, this pattern suggests an effect on event- but
not item-based distinctiveness (cf., Hunt, 2003).
Future research might more specifically examine
whether caffeine might interact with the application
of a distinctiveness heuristic, for instance by
manipulating the salience of encoded words.
CONCLUSION
In conclusion, the present study extends previous
work by demonstrating that pharmaceutically
induced physiological arousal can increase false
memories in the DRM paradigm in a dosedependent
manner, without affecting veridical
memory. This is most likely due to greater
relational processing, perhaps as a result of
glucocorticoid-mediated dysregulation of contextual
binding and source monitoring within the
PFC and hippocampus. It might also be partially
attributed to shifted response criteria, perhaps
indicating a differential application of the distinctiveness
heuristic under high arousal conditions.
Given the ubiquity of caffeine consumption in the
form of coffee, tea, energy drinks, and foods such
as chocolate and energy bars, the present results
hold everyday implications for false memory
formation, at least for individuals who do not
regularly consume caffeine. One caveat to this is
that the translation of DRM results to real-world
behaviours is not known, suggesting a need for
future work using more naturalistic situations.
Future work could also attempt to determine the
mechanism by which caffeine influences false
memory by examining the effects of other stimulants
that do not primarily affect adenosine
receptors (e.g., amphetamine or modafinil). Studies
to examine the relationship between caffeine
consumption, cortisol release, and false memory
in a doseÁresponse fashion in habitual and nonhabitual
consumers should also be conducted, as
the present effects, are likely to be less pronounced
in habitual caffeine consumers.
Original manuscript received March 2011
Revised manuscript received September 2011
First published online March 2012
REFERENCES
Barone, J. J., & Roberts, H. R. (1996). Caffeine
consumption. Food and Chemical Toxicology, 34,
119Á129.
426 MAHONEY ET AL.
Downloadedby[37.188.236.47]at08:3115February2013
Brunye´, T. T., Mahoney, C. R., Augustyn, J. S., &
Taylor, H. A. (2009). Emotional state and local
versus global spatial memory. Acta Psychologica,
130, 138Á146.
Capek, S., & Guenther, R. (2009). Caffeine’s effect on
true and false memory. Psychological Reports, 104,
787Á789.
Childs, E., & de Wit, H. (2006). Subjective, behavioral,
and physiological effects of acute caffeine in light,
nondependent caffeine users. Psychopharmacology,
185, 514Á523.
Corson, Y., & Verrier, N. (2007). Emotions and false
memories: Valence or arousal? Psychological
Science, 18(3), 208Á211.
Diekelmann, S., Landolt, H., Lahl, O., Born, J., &
Wagner, U. (2008). Sleep loss produces false memories.
PLosone. Retrieved 12 December 2009 from
http://www.plosone.org/article/info:doi%2F10.
1371%2Fjournal.pone.0003512
Dodson, C. S., & Schacter, D. L. (2001). ‘‘If I had said it
I would have remembered it’’: Reducing false
memories with a distinctiveness heuristic. Psychonomic
Bulletin and Review, 8(1), 155Á161.
Garrett, B. E., & Griffiths, R. R. (1997). The role of
dopamine in the behavioral effects of caffeine in
animals and humans. Pharmacology, Biochemistry
and Behavior, 57, 533Á541.
Hunt, R. R. (2003). Two contributions of distinctive
processing to accurate memory. Journal of Memory
and Language, 48, 811Á825.
Hutchinson, K. A., & Balota, D. A. (2005). Decoupling
semantic and associative information in false memories:
Explorations with semantically ambiguous
and unambiguous critical lures. Journal of Memory
and Language, 52, 1Á28.
Kirschbaum, C., Wolf, O. T., May, M., Wippich, W., &
Hellhammer, D. H. (1996). Stress- and treatmentinduced
elevations of cortisol levels associated with
impaired declarative memory in healthy adults. Life
Sciences, 58(17), 1475Á1483.
Koutstaal, W., & Schacter, D. L. (1997). Gist-based
false recognition of pictures in older and younger
adults. Journal of Memory and Language, 37,
555Á583.
Lieberman, H. R. (2003). Nutrition, brain function and
cognitive performance. Appetite, 40(3), 245Á254.
Lieberman, H. R., Tharion, W. J., Shukitt-Hale, B.,
Speckman, K. L., & Tulley, R. (2002). Effects of
caffeine, sleep loss, and stress on cognitive performance
and mood during US Navy SEAL training.
Psychopharmacology, 164, 250Á261.
Lovallo, W. R., Whitsett, T. L., al’Absi, M., Sung, B. H.,
Vincent, A. S., & Wilson, M. F. (2005). Caffeine
stimulation of cortisol secretion across the waking
hours in relation to caffeine intake levels. Psychosomatic
Medicine, 67, 734Á739.
Mayer, J. D., & Gaschke, Y. N. (1988). The experience
and meta-experience of mood. Journal of Personality
and Social Psychology, 55, 102Á111.
Mitchell, K. J., Johnson, M. K., Raye, C. L., &
D’Esposito, M. (2000). fMRI evidence of age-related
hippocampal dysfunction in feature binding in working
memory. Cognitive Brain Research, 10(1Á2),
197Á206.
Oei, N. Y., Elzinga, B. M., Wolf, O. T., de Ruiter, M. B.,
Damoiseaux, J. S., Kuijer, J. P., et al. (2007).
Glucocorticoids decrease hippocampal and prefrontal
activation during declarative memory retrieval in
young men. Brain Imaging and Behavior, 1(1Á2),
31Á41.
Payne, J. D., Nadel, L., Allen, J. J. B., Thomas, K. G. F.,
& Jacobs, W. J. (2002). The effects of experimentally
induced stress on false recognition. Memory, 10(1),
1Á6.
Porter, S., Spencer, L., & Birt, A. R. (2003). Blinded by
emotion? Effect of the emotionality of a scene on
susceptibility to false memories. Canadian Journal
of Behavioural Science, 35, 165Á175.
Roediger, H., & McDermott, K. (1995). Creating false
memories: Remembering words not presented in
lists. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 21, 803Á814.
Wolf, O. T. (2009). Stress and memory in humans:
Twelve years of progress. Brain Research, 1293,
142Á154.
CAFFEINE AND FALSE MEMORY 427
Downloadedby[37.188.236.47]at08:3115February2013