ARTICLE IN PRESS

Psychoneuroendocrinology (2008) 33, 328–339

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/psyneuen

Olfactory influences on mood and autonomic,

endocrine, and immune function
Janice K. Kiecolt-Glasera,b,, Jennifer E. Grahamc, William B. Malarkeyb,d,e,
Kyle Porterf, Stanley Lemeshowb,f,g, Ronald Glaserb,d,e

a
Department of Psychiatry, The Ohio State University, 1670 Upham Drive, Columbus, OH 43210, USA
b
The Ohio State Institute for Behavioral Medicine Research, 2175 Graves Hall,
333 West 10th Avenue, Columbus, OH 43210, USA
c
Department of Biobehavioral Health, The Pennsylvania State University,
315 East Health and Human Development Building, University Park, PA 16802, USA
d
Department of Internal Medicine, Davis Medical Clinic, The Ohio State University,
480 Medical Center Drive, Columbus, OH 43210, USA
e
Molecular Virology, Immunology, and Medical Genetics, 2078 Graves Hall, 333 West 10th Avenue, Columbus, OH 43210, USA
f
The Ohio State University Center for Biostatistics, M200 Starling Loving, 320 West 10th Avenue, Columbus, OH 43210, USA
g
The Ohio State University College of Public Health, M006 Starling Loving, 320 West 10th Avenue, Columbus, OH 43210, USA

Received 3 August 2007; received in revised form 8 October 2007; accepted 29 November 2007

KEYWORDS Summary
Psychoneuro- Despite aromatherapy’s popularity, efficacy data are scant, and potential mechanisms are
immunology; controversial. This randomized controlled trial examined the psychological, autonomic,
Aromatherapy; endocrine, and immune consequences of one purported relaxant odor (lavender), one
Odor; stimulant odor (lemon), and a no-odor control (water), before and after a stressor (cold
Complementary pressor); 56 healthy men and women were exposed to each of the odors during three
medicine separate visits. To assess the effects of expectancies, participants randomized to the
‘‘blind’’ condition were given no information about the odors they would smell; ‘‘primed’’
individuals were told what odors they would smell during the session, and what changes to
expect. Experimenters were blind.
Self-report and unobtrusive mood measures provided robust evidence that lemon oil
reliably enhances positive mood compared to water and lavender regardless of
expectancies or previous use of aromatherapy. Moreover, norepinephrine levels following
the cold pressor remained elevated when subjects smelled lemon, compared to water or
lavender. DTH responses to Candida were larger following inhalation of water than lemon

Corresponding author at: Department of Psychiatry, The Ohio State University, 1670 Upham Drive, Columbus, OH 43210, USA.
Tel.: +1 614 292 0033; fax: +1 614 292 0038.
E-mail address: Kiecolt-Glaser.1@osu.edu (J.K. Kiecolt-Glaser).

0306-4530/$ - see front matter & 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.psyneuen.2007.11.015
 ARTICLE IN PRESS
Olfactory influences on mood and physiology 329

or lavender. Odors did not reliably alter IL-6 and IL-10 production, salivary cortisol, heart
rate or blood pressure, skin barrier repair following tape stripping, or pain ratings
following the cold pressor.
& 2007 Elsevier Ltd. All rights reserved.

1. Introduction fewer physical symptoms than individuals exposed to the

unpleasant smell of dimethyl sulfide (Knasko, 1992).
Widely utilized, aromatherapy is employed for relief of pain, Although some essential oils like lavender and lemon have
relaxation and anxiety reduction, and enhanced energy; been classified as either sedative or stimulating and these
essential oils have been used to help women cope with labor categories are broadly related to their putative CNS actions,
pain, to relieve chemotherapy side effects, to enhance the there does not seem to be general agreement on mechan-
rehabilitation of cardiac patients, to promote restful sleep, isms of action, and the health benefits are unclear;
and to reduce post-surgical discomfort (Price and Price, autonomic and self-report data have typically been used
1999). However, efficacy data are scant, and potential as a surrogate for health outcomes. In fact, despite the clear
mechanisms of action are controversial. presumption that aromatherapy has immune consequences,
The ‘‘lock and key’’ or systemic effect theory posits that we found only one human study that included immunological
essential oils act like a drug or enzyme, such that particular measures (Komori et al., 1995). Indeed, more broadly, most
odors should have very specific effects (Hirsch, 2001), a studies lack physiological or objectively assessed health
popular view among aromatherapy practitioners who endpoints.
prescribe certain odors for distinct health problems Placebo and expectancy effects are central problems in
(Hirsch, 2001). For example, short-term inhalation of human olfactory research, and interpretation of a number of
lavender oil (typically over the space of a few hours) is studies is difficult for this reason; few studies are double or
described as therapeutic for insomnia, influenza, head- even single blind, and many assessed only a single odor
aches, migraines, anxiety, nervousness, and melancholy without any control conditions (Martin, 2006). The paucity
(Price and Price, 1999). Others have argued that lavender of these key controls is important because of what Jellinek
enhances immune function, as well as treating lung and terms the ‘‘placebo mechanism’’ (Jellinek, 1997); unlike the
sinus infections, laryngitis, and asthma (Keville and Green, systemic effects theory described above, this theory holds
1995). These and other authors have emphasized lavender’s that the characteristics of the odorant are irrelevant, and
relaxant properties (Grace, 1999; Tisserand and Balacs, individuals’ expectancies determine the pattern of re-
1995). sponses. In support of this perspective, participants to
Following a long history in folklore, lavender has been whom it was suggested that an odor would affect perfor-
used as a sleep aid, and one small study suggested that the mance showed an improvement in math calculations, even
ambient odor of lavender can significantly enhance the when they were in fact exposed to no odor (Knasko et al.,
amount of time asleep after withdrawal of medication for 1990). The general affective theory or reflectorial effect
insomnia (Hardy et al., 1995). In another sleep study, theory (Hirsch, 2001) provides yet another conceptual
lavender oil presented the first 2 min of every 10 min period framework; it suggests that odors perceived as positive
for 40 min increased deep or slow-wave sleep compared to a may induce positive moods, and these mood changes may
control (distilled water) stimulus (Goel et al., 2005). enhance both physical and psychological well being.
Contingent negative variation (CNV), an EEG shift that To compare and contrast the diverse perspectives about
occurs when individuals are expecting an event (e.g., a light whether and how odors affect health, we examined the
that will signal a tone), is diminished by sedatives and autonomic, endocrine, and immune consequences of one
enhanced by stimulants; lavender decreased CNV, but it did purported sedating or relaxant odor, lavender, one activat-
not affect reaction time or heart rate as sedatives do (Torii ing or stimulant odor, lemon, and distilled water as a no-
et al., 1988). Further, a comparison of rosemary and odor control during both resting and ‘‘challenge’’ or stress
lavender oils using EEG and math computations showed that conditions in a mixed or between-within repeated measures
lavender increased patterns consistent with drowsiness, and design; each subject served as his or her own control during
subjects reported greater relaxation, while rosemary three separate 6 h visits. Depending on their random
produced EEG patterns interpreted as increased alertness, assignment, participants were either given no information
consistent with faster and more accurate math test results about what odors they would be smelling or what to expect
(Diego et al., 1998). (the ‘‘blind’’ group), or they were told what odors they
In contrast to lavender’s sedative characterization, lemon would smell and what changes to expect from the relaxant,
oil is described as activating, immunomodulatory, and mood stimulant, or no odor exposures (the ‘‘primed’’ group).
enhancing (Buchbauer et al., 1993; Keville and Green, 1995; Our protocol for each session included a cold pressor, a
Lis-Balchin and Hart, 2002; Price and Price, 1999); it has also laboratory stressor that elevates stress-related hormones,
been touted as an inhalation remedy for respiratory tract heart rate, and blood pressure (Blandini et al., 1995; Hirsch
infections (Grace, 1999). Lemon oil has been associated with and Liebert, 1998). Both before and after the cold pressor
increased heart rate and enhanced mental and physical task we performed tape stripping, a common dermatological
performance in human studies (Jellinek, 1997). In addition, paradigm for studying restoration of the skin barrier, a
in one study participants exposed to lemon oil reported process mediated by both endocrine and immune systems
 ARTICLE IN PRESS
330 J.K. Kiecolt-Glaser et al.

(Choi et al., 2005). Our design thus provided a way to We also conducted a structured interview to assess prior
examine the ability of lemon and lavender odors to experiences with aromatherapy. Questions elicited partici-
modulate stress and pain responses to the cold pressor, as pants’ evaluation of and previous experience with aro-
well as wound healing via the speed of skin barrier repair. matherapy, and their expectancies about the extent to
Specific predictions can be derived from the various which their own psychological and physiological responses
theories posited to explain the effects of essential oils. For would be influenced by odors.
example, if the systemic effect theory is correct, even Finally, in order to assess affective responses, cotton balls
relatively short-term exposure to lavender would be with each of the three stimuli were taped under each
expected to produce larger declines in the production of subject’s nose in a randomly determined order. During
cortisol and catecholamines, faster skin barrier repair, lower exposure to each odor, participants rated their responses to
pain ratings in response to the cold pressor, and smaller a standardized series of positive and neutral slides (Lang et
stress-related immunological changes compared to lemon al., 1999). On completion of each series, subjects rated the
and the no-odor control; short-term exposure to lemon oil odor’s pleasantness, familiarity, and intensity using a 1–10
should produce greater transient increases in positive scale (Doty, 1986).
affect, heart rate, blood pressure, and catecholamines than
either lavender oil or the no-odor control. If expectancies 2.3. General Clinical Research Center (GCRC) visit
determine the pattern of responses (Jellinek, 1997), then
timeline
the primed group’s mood and physiological responses to
lemon and lavender odors would be greater than the blind
After the screening session, subjects scheduled three visits
group; similarly, those with positive expectancies about
to the GCRC, a hospital research unit. At least 2 weeks were
aromatherapy in advance of participation would be ex-
scheduled between visits, and the average time to complete
pected to show greater changes. By assessing olfactory
all three visits was 64.46 days (SD ¼ 48.40). All visits
influences on mood and autonomic, endocrine, and immune
followed the same timeline, illustrated in Figure 1, differing
function, our design allowed us to contrast these diverse
only in the odor condition.
conceptual perspectives, clarify mechanisms, and assess
On arrival a heparin well was placed in one arm, and
possible clinical efficacy.
participants then received a standard breakfast (after
fasting since midnight). Subjects next completed a battery
2. Method of questionnaires including a baseline PANAS to assess mood
and the first of three emotional Stroop tests, both of which
were repeated at the times shown in Figure 1. Following a
2.1. Participants
10-min baseline, we began automated blood pressure and
heart rate assessments using a Dynamap/Critikon 1846SX/P.
Participants recruited through ads were told that the study Baseline blood samples were drawn prior to odor exposure,
involved the assessment of responses to both strong and after subjects had remained in a sitting position for
weak fruit and floral odors. We excluded individuals who 20–30 min.
were taking cardiovascular medications (statins, beta Immediately prior to odor exposure, participants opened
blockers, etc.) or medications or health problems with a sealed letter with blind or primed instructions (manipula-
obvious immunological or endocrinological consequences. tion described below), and then a cotton ball with the
Additional exclusion criteria included perfume allergies, essential oil or distilled water (depending on the day’s
smoking, problems with smell or taste, respiratory symp- randomized odor assignment) was taped under their nose (as
toms or problems, asthma, and excessive alcohol or caffeine described in more detail below). Next participants rated the
use. extent to which they expected that the odor would affect
The average age of the final sample of 21 men and 35 their mood and physiological responses, as well as the odor’s
women was 24.41 (SD ¼ 6.05, range ¼ 18–43); 36 were
pleasantness, familiarity, and intensity using a 1–10 scale
white, 10 were African American, and 10 were Asian, and 54 (Doty, 1986). Following about 1.25 h of odor exposure with
had at least some college education. The Ohio State periodic endocrine blood/saliva samples, they completed a
University Biomedical Research Review Committee approved second immune blood draw and the first of two tape-
the project; all subjects gave written informed consent stripping sessions (Denda et al., 2000).
prior to participation. After undergoing the cold pressor test, participants had a
third (and final) blood draw for immunological assays, and
2.2. Screening session the second of two tape-stripping sessions. Following these
tasks, subjects provided another salivary cortisol sample,
and completed a thought-listing task. Subjects remained in
We screened for general anosmia using phenylethyl alcohol
the GCRC until 1:00 PM, with additional measurements of
and amyl acetate with a forced-choice staircase procedure
the tape-stripped sites, as well as salivary and blood
(Cowart, 1989). To assure that participants did not have a
sampling for hormone measurements.
specific anosmia for lavender or lemon oils, they were given
three sets of three bottles—two with distilled water, and
the third which contained the essential oil—and asked to 2.4. Essential oils
choose the one that differed from the other two. To be
eligible for the study, subjects had to choose the correct A yellow-tinted cotton ball containing 100 ml of the essential
response in all three trials for both odors. oil or distilled water was taped between the nose and upper
 ARTICLE IN PRESS
Olfactory influences on mood and physiology 331

Figure 1 Timeline for experimental participation during each of the three GCRC sessions.

lip on top of a piece of surgical tape; use of the barrier tape heart rate and production of stress-related hormones, and
avoided percutaneous absorption (Denda et al., 2000). This that it might provoke positive or warm memories. The lemon
method provided continuous and uniform exposure across condition letter said that lemon is a stimulating and
subjects that would not have been possible with ambient activating odor and its scent might make participants feel
room inhalation, and helped maintain experimenter blind- more alert and energized; this letter also suggested that the
ness. Figure 1 shows the intervals when the cotton was scent might lift mood, increase heart rate, help them think
replaced throughout the session to maintain odor strength. more clearly, and perhaps also provoke positive memories
Our Lavandula angustifolia (true lavender) and Citrus and emotions. The water condition letter said that water
limonum were essential oils steam extracted from plants was being used as a no-odor control to see how participants
(Aldrich, Milwaukee, WI). Both oils were initially character- responded in the absence of an odor.
ized using mass spectrometry and gas chromatography by
the vendor and confirmed by the Ohio State Chemical
Instrumentation Center every 8–10 months to test variability 2.6. Experimenter blinding
in thawed aliquots (all from the same original lot, frozen at
80 1C), focusing on the key active components of each oil. Only the researcher who added the odor to the cotton ball
knew the order in which participants were being exposed to
the different odors, and she did not have any subject
2.5. Expectancies: blind vs. primed group contact. She prepared the stimulus just prior to the
information scheduled time for use, and she gave the GCRC nurse the
cotton balls in sealed glass vials. The nurses were told never
Before being exposed to the odor for the first time in the to sniff the vials or the cotton as they taped the cotton
session, all participants were given a letter in a sealed under the subjects’ noses.
envelope. The blind group’s letter thanked them for their A cream, Odor Perception Inhibitor (O-P-I)s, was painted
participation and reminded them not to share any odor- under the nose of all nurses and research assistants who had
related thoughts with staff members. Primed group letters subject contact during the screening session and the
included the same information given to blind group admission to help keep personnel blind regarding the visit’s
participants; in addition, they named the odor to which odor assignment. The research assistants also wore a
the participant would be exposed and provided specific surgical mask, which enhanced the product’s efficacy by
physiological and psychological priming information. The trapping the O-P-Is odor. Further, sessions were run in a
lavender condition letter said that lavender is a calming special GCRC room with reverse air ventilation, a Smok-
odor which is sometimes used as a sleep aid for its relaxing Eters system; use of the fan on high speed after each odor
qualities; further, it suggested that the scent might lower application minimized any traces of ambient odor.
 ARTICLE IN PRESS
332 J.K. Kiecolt-Glaser et al.

2.7. Self-report measures on both acute pain and physiological recovery after a
stressor (Blandini et al., 1995; Hirsch and Liebert, 1998).
Health-related behaviors assessed at screening and each After sitting quietly for a 15-min adaptation period,
visit included recent medication use, exercise, caffeine and participants immersed their right foot up to their ankle for
alcohol intake, sleep, and recent weight changes (Kiecolt- 1 min in warm (37 1C) water, and then immediately
Glaser and Glaser, 1988). Physical activity was assessed with immersed their foot in a pan of 4 1C water for 1 min (Hirsch
questions from Baecke et al. (1982) which have adequate and Liebert, 1998). Participants rated pain intensity on a
reliability (.74–.88). 1–10 scale at the end of the cold pressor.
State mood was assessed with the two 10-item scales of
the positive and negative affect schedule (PANAS) (Watson 2.10. Immunological data
et al., 1988) which are largely uncorrelated. The scales
show good convergent and discriminant validity with other Delayed hypersensitivity to Candida: DTH memory responses
state mood measures (Watson et al., 1988). to a common infectious agent provided a measure of T-cell
immunity (Sokal, 1975). Nurses inoculated subjects’ arm
2.8. Indirect mood assessments with .1 ml Candida (stock solution diluted 1:20 in saline,
Greer Labs, NC) intradermally. The wheal diameter (two
Picture ratings: At the screening session we used computer- dimensions) was self-assessed at 24, 48, and 72 h by
presented positive and neutral pictures from the International participants given detailed instructions and templates for
Affective Picture System (IAPS) (Lang et al., 1999). Using IAPS measurement.
normative data, we constructed three picture series, each Blastogenesis: In all, 2  106 PBLs isolated from whole
lasting 12 min, with equivalent standardized valence and blood preparations were exposed to 2.5, 5.0, and 10.0 mg/ml
arousal data. Our participants rated the slides using the Con A and PHA. The cells were incubated at 371 for 65 h and
original normative study pictographs; for valence ratings then pulsed with MTS/PMS using the Celltiter 96 Aqueous
participants continuously modulated a figure’s face from an nonradioactive cell proliferation assay (Promega). Cells
extreme frown to a broad smile, while for arousal they incubated in media alone were used as controls.
modulated a figure that showed inactivity to rapid movement. Stimulated cytokine production: In all, 4  106 PBLs were
Stroop: For the emotional Stroop, participants name the isolated from whole blood preparations and incubated at
color in which negative or threatening words are printed 37 1C for 72 h in 2 ml of RPMI-1640 media (Gibco) supple-
(Williams et al., 1996). Following work by Mogg et al. (1993) mented with 10% human male serum (Sigma) and 5 mg/ml of
we used 60 words from their lists from each of the following Con A (Sigma). Control cells were prepared in media without
categories: anxiety-relevant negative words, depression- Con A. Supernatants were harvested and frozen at 86 1C
relevant negative words, positive words, and neutral until assayed for IL-6 and IL-10 by ELISA using OptEIA kits
household words; the word types were matched for length from BD Biosciences.
and frequency based on published norms (Carroll et al.,
1971). The 60 words were divided into three sets of 20, with 2.11. Endocrine data
the three sets matched for length and frequency, and words
were presented in random order. Saliva was collected using a salivette (Sarstedt, Newton,
The depression word interference scores for each subject North Carolina), an untreated sterile cotton roll that was
and assessment time point were calculated by subtracting the placed in the subject’s mouth for 2 min to ensure
mean latency for the neutral household words from those for saturation, and assayed using the Cortisol Coat-A-Count
depression-related words, and the same procedure was used RIA (Siemens Medical Solutions Diagnostics, Los Angeles,
for anxiety and positive words (Mogg et al., 1993). Thus, CA). Plasma catecholamine samples were frozen at 70 1C
larger values indicate slower responses to emotion-relevant and assayed by HPLC with ElectroChemical Detection using
words than neutral household words (Mogg et al., 1993). Standards and Chemistry (alumina extraction) from Thermo-
Thought-listing: Using a variation of a common thought- Alko (Beverly, MA). All cortisol and catecholamine samples
listing procedure (Cacioppo and Petty, 1981), the experi- for a subject were frozen after collection and analyzed
menter left the room as subjects spoke into a tape recorder within the same assay run once the participant had
for 2 min, describing their thoughts during their experi- completed the study.
mental participation. Participants were asked to describe
any of their thoughts, including those about the odor, the 2.12. Skin barrier assessment
setting, or the experimenter, and were told that we wanted
an honest and complete a listing of what they actually
Cellophane tape stripping, a common dermatological para-
thought about. The Linguistic Inquiry and Word Count (LIWC)
digm for studying restoration of the skin barrier, allowed us
program provided frequencies of positive and negative
to examine whether the time necessary for recovery from
emotion word use from the tape transcripts (Pennebaker
minor physical insults varied by odor exposure. Measure-
and Francis, 1999).
ment of the rate of transepidermal water loss (TEWL)
through human skin provides a noninvasive method to
2.9. Cold pressor monitor changes in the stratum corneum barrier function
of the skin (Choi et al., 2005). After obtaining baseline
Widely used in behavioral and psychophysiological research, measurements on the volar forearm, cellophane tape
the cold pressor provided a way to assess the impact of odor (3M Scotch-type; St. Paul, MN) was applied repeatedly
 ARTICLE IN PRESS
Olfactory influences on mood and physiology 333

(6–50 times) to remove the superficial layer of cornified skin 3.3. Mood
cells. Tape stripping stopped when the TEWL was elevated
from the basal level of 5–7 g/h/m2 to at least 20 g/h/m2. PANAS: As shown in Figure 2a, the significant time by odor
The number of strips required to reach TEWL X20 g/m2/h interaction for the PANAS-positive affect scores,
was the measure of ‘‘barrier integrity’’ (Choi et al., 2005). F(6, 240) ¼ 4.43, po.001, reflected a greater increase from
TEWL was measured several times during the session using a baseline to post-odor when subjects smelled lemon than
computerized evaporimetry instrument, the DermaLabs when they received lavender (adjusted p ¼ .001) or water
(CyberDERM, Media, PA). (adjusted po.004). After the cold pressor, positive affect
increased for all odors relative to the post-odor assessment,
but remained highest for lemon.
2.13. Statistical methods PANAS-negative affect scores were not normally distrib-
uted, and thus were categorized into scores of 10, 11, and
Unless otherwise noted, analyses consisted of repeated- greater than 11. A repeated-measures cumulative logistic
measures linear models. These models accounted for the regression model fit to this three-level outcome showed a
correlation in measurements from the same subject across marginal expectancy condition by odor interaction,
time and visits. Independent variables assessed in each w2(2) ¼ 5.16, p ¼ .076, as well as significant change over
model were odor, time, gender, and expectancy group time, w2(3) ¼ 12.70, p ¼ .005. For the time effect, negative
(primed vs. blind), as well as their interactions; the baseline affect was higher at baseline then at the three later time
level of the dependent variable was included as a covariate. points combined, OR ¼ 2.80 (1.62, 4.85), p ¼ .001.
Log-transformations were used for right skewed distribu- Emotional Stroop: Analysis of Stroop interference scores
tions. Post hoc tests were performed to investigate with all negative emotion words (both anxiety and depres-
significant interactions and pairwise differences, using sion related) demonstrated a significant interaction be-
Holm’s or Tukey’s procedure as appropriate to adjust for tween time and odor, F(2, 188) ¼ 5.07, p ¼ .007. The
multiple comparisons. A two-sided significance level of decreased reaction time from post-odor to post-stressor
a ¼ .05 was used. All analyses were performed in SASs for lemon contrasted with the increased reaction times for
version 9.1. Nonsignificant data are summarized briefly both lavender (adjusted p ¼ .01) and water (adjusted
because of space constraints. p ¼ .04), shown in Figure 2b.
Secondary analyses demonstrated a significant time by
odor interaction for depression-related words,
3. Results F(2, 188) ¼ 4.25, p ¼ .015, but not for anxiety-related
words, F(2, 188) ¼ .88, p ¼ .42. The decrease in reaction
3.1. Blind vs. primed time to depressive words from post-odor to post-stressor
was greater for lemon than for lavender (adjusted p ¼ .04)
Subjects’ ratings of expected mood and expected physiolo- or water (adjusted p ¼ .02). The time by odor interac-
gical change were highly correlated (r ¼ .84), and thus we tion was marginal for positive words, F(2, 188) ¼ 2.83,
used the mean of these two ratings to evaluate expectancies p ¼ .062.
following each odor application. The significant interaction Positive and negative emotion word use, via LIWC: The
between expectancy group and odor reflected the success of percentage of positive and negative words used in thought
the priming manipulation, F(2, 53) ¼ 52.34, po.001; primed listings was modeled by replacing time with emotion word
subjects had relatively higher expectations for lavender use (positive or negative) in the general model described
(4.82, SEM ¼ .36) and lemon (4.94, SEM ¼ .37) than blind previously. The significant odor by word-type interaction,
subjects (lavender ¼ 3.99, SEM ¼ .33; lemon ¼ 4.69, F(2, 103) ¼ 4.84, p ¼ .01, reflected the fact that although
SEM ¼ .35); ratings for water showed the expected reversal the percentage of positive words used by participants was
of this pattern, with primed subjects reporting lower higher than that of negative words at each visit, the
expectations (1.46, SEM ¼ .26) than blind subjects (2.35, difference was greater on the days when subjects smelled
SEM ¼ .25). lemon (adjusted p ¼ .02) or lavender (adjusted p ¼ .03)
Participants randomized to the blind or primed expec- than the day they smelled water (Figure 2c).
tancy conditions did not differ on age, proportion of men IAPS picture ratings: Emotional valence ratings from the
and women, education, sense of smell, baseline negative or screening session differed by odor, F(2, 108) ¼ 8.13,
positive affect, or health behaviors. po.001, as shown in Figure 2d. Compared to lemon,
subjects’ slide ratings were significantly less positive when
subjects smelled lavender (adjusted po.001) and water
3.2. Gender and health behaviors (adjusted p ¼ .04). Arousal ratings did not differ by odor,
F(2, 108) ¼ .96, p ¼ .39.
The men and women who participated did not differ on age,
education, sense of smell, health behaviors, or negative or
positive affect. There were not reliable differences in any 3.4. Catecholamines and cortisol
health behaviors related to odor, with the exception of
alcohol intake over the prior 48 h, F(2, 102) ¼ 3.48, p ¼ .03. A significant time by odor interaction for norepinephrine is
Subjects reported fewer drinks before sessions in which they illustrated in Figure 3, F(10, 456) ¼ 2.35, p ¼ .01. The
were exposed to lavender (.36, SEM ¼ .30) compared to decrease in norepinephrine following the cold pressor to
lemon (1.12, SEM ¼ .28) and water (1.01, SEM ¼ .28). the first recovery time point was smaller when subjects
 ARTICLE IN PRESS
334 J.K. Kiecolt-Glaser et al.

28 5
PANAS (positive affect) total

Positive
Negative
26 4

% Word Use, LIWC

24 3

22 Lemon 2
Lavender
Water
20 1
Baseline Post-Odor Post-Stress Session End Lemon Lavender Water

50 8

25
Interference Score (ms)

7.5

0 IAPS Valence Ratings

-25

6.5
-50
Lemon
Lavender
Water
-75 6
Baseline, Pre-Odor Post-Odor Post-Stressor Lemon Lavender Water

Figure 2 (a–d) Differences in one self-report and three unobtrusive mood measures as a function of odor condition. Baseline-
adjusted mean (7SEM) changes in self-reported positive affect on the PANAS as a function of time and odor (a) showed the greatest
difference immediately following the first odor application. Subjects showed greater recovery following the stressor when smelling
lemon compared to lavender or water as reflected in baseline-adjusted mean (7SEM) interference in responses to negative emotion
words on the Stroop (b). The differences in mean (7SEM) positive and negative emotion word use in thought listings were greater on
the days when subjects smelled lemon or lavender than the day they smelled water (c). IAPS picture valence ratings from the
screening session were significantly lower for lavender and water compared to lemon (d).

smelled lemon compared to water (adjusted p ¼ .019). 2.8

Norepinephrine returned to pre-stressor levels when sub-
Lemon
Norepinephrine, pg/ml (log10)

jects smelled water and lavender, but remained elevated

when they smelled lemon. Analysis of epinephrine showed Lavender
nonsignificant effects for odor, F(2, 101) ¼ 1.89, p ¼ .16, Water
2.7
expectancy group, F(1, 54) ¼ .09, p ¼ .76, and the time by
odor interaction, F(10, 450) ¼ .61, p ¼ .80.
For salivary cortisol, odor, F(2, 98) ¼ .19, p ¼ .83, expec-
tancy group, F(1, 53) ¼ .59, p ¼ .45, and the time by odor 2.6
interaction, F(8, 380) ¼ .64, p ¼ .74, were nonsignificant.

2.5
3.5. Heart rate and blood pressure
or

or

in

in
so
in

m
od

od

pp

es

70
20
e-

st

ri

pr
pr

st

,+
Po

+
ld
e-

y,
e,

ry

There was a marginal time by odor interaction for diastolic

co
p

er
in

e
ta

ov
ov
l

st
se

blood pressure, F(12, 577) ¼ 1.67, p ¼ .07, shown in Figure

st

Po

ec
ec
Ba

Po

R
R

4. Diastolic blood pressure decreased from post-tape

stripping to post-cold pressor in subjects receiving lavender Figure 3 Baseline-adjusted mean (7SEM) norepinephrine
and increased in subjects receiving lemon or water. For levels from pre-odor exposure through the end of the admission
systolic blood pressure, the odor effect, Fo1, p ¼ .50, the as a function of odor exposure.
 ARTICLE IN PRESS
Olfactory influences on mood and physiology 335

odor by time interaction, Fo1, p ¼ .99, and expectancy 3.6. Immunological data
group, F(1, 53) ¼ 1.70, p ¼ .20, were nonsignificant.
There was a significant expectancy group by time DTH: Using the maximum DTH measurement obtained for
interaction for heart rate, F(5, 266) ¼ 2.48, p ¼ .03. Tests each odor at 24, 48, or 72 h produced a significant difference
for the contrast for the quadratic trends across the pre- in mean DTH wheal size across odors, F(2, 92) ¼ 4.47,
stressor, post-stressor, and recovery time points (times 3–5 p ¼ .014, reflecting the fact that subjects had larger DTH
in Figure 5) showed the quadratic trends were significantly responses when they smelled water compared to lavender
different (p ¼ .002) between the primed and blind groups. (adjusted p ¼ .02) or lemon (adjusted p ¼ .06), shown in
Primed subjects showed higher heart rate increases follow- Figure 6; blind and primed groups did not differ, p ¼ .35.
ing administration of the cold water stressor than blind Two participants were excluded from DTH analyses because
subjects, but the former also had larger decreases during they never showed a positive DTH reaction to Candida, i.e.,
recovery. wheals o5 mm at all time points for each session (Sokal,
1975). Using only the 72 h measurements produced the
same significant odor effect with water the highest,
F(2, 80) ¼ 4.71, p ¼ .01.
74 Production of IL-6 and IL-10: There was a marginal
Diastolic Blood Pressure, mm Hg

Lemon expectancy group effect for stimulated IL-10 production,

72 Lavender F(1, 192) ¼ 3.20, p ¼ .08, with a trend toward higher
Water production in the primed group compared to blind subjects.
70
None of odor, the odor by time, or the odor by expectancy
68 interactions were significant for IL-6 (ps4.10) or IL-10
(p’s4.50).
66 Blastogenesis: There were significant odor by expectancy
group by time interactions for proliferative responses to
64 PHA, F(2, 81) ¼ 7.67, po.001, and Con A, F(2, 95) ¼ 4.37,
p ¼ .015, shown in Figure 7. Controlling for pre-odor
62 baseline, for both mitogens the effect of expectancy group
or

or

in

in

in

on change from post-odor to post-cold pressor was sig-

so
in

m
od

od

pp

es

60
1

3
e-

nificantly different for lavender than for both water (adj

st

+
tri

pr
pr

>
Po

-s

y,

y,
ld

y,
e,

er

er
pe

p ¼ .001 for PHA, p ¼ .016 for Con A) and lemon (adj

co

er
lin

ov

ov
ta

ov
st
se

ec

ec

p ¼ .009 for PHA, p ¼ .094 for Con A). When subjects

st

Po

ec
Ba

R
Po

smelled lavender, the blastogenic responses for both Con A

Figure 4 Baseline-adjusted mean (7SEM) diastolic blood and PHA increased in the primed group from pre- to post-
pressure from pre-odor exposure through the end of the stressor and decreased in the blind group, while the effect
admission as a function of odor exposure. This marginal time by primed group was in the opposite direction for water and
by odor interaction, p ¼ .07, shows that blood pressure lemon.
decreased from post-tape stripping to post-cold pressor when
subjects received lavender and increased when they receiving
3.7. Skin barrier repair
lemon or water.

Using the method to calculate percent recovery as described

by Denda (Denda et al., 2000), recovery from the first tape
80 stripping series improved with time, as would be expected,
Primed
F(1, 55) ¼ 43.40, po.001, but neither odor nor the odor by
Blind
Heart Rate, beats/min

time interaction was significant, p’s4.60. Comparisons of

75
the speed of recovery at 75 min using both the first and
second tape stripping series showed no difference between
70
1.4
DTH Wheal, mm (log10)

65
1.3

60 1.2
or

in

in

in
do

so
n

m
od

pi

es
to

+1

+3
rip

0
e-

pr

>6
s

1.1
pr

st
Po

y,

y,
ld

er

er
-

y,
e,

pe

co

er
ov

ov
lin

ta

ov
st

ec

ec
se

st

Po

ec
R

R
Ba

Po

1
R

Lemon Lavender Water

Figure 5 Baseline-adjusted mean (7SEM) heart rate from pre-
odor exposure through the end of the admission as a function of Figure 6 Mean (7SEM) of the maximum DTH measurement for
expectancy group assignment. each odor at 24, 48, or 72 h after the GCRC session.
 ARTICLE IN PRESS
336 J.K. Kiecolt-Glaser et al.

0.3

0.16
Optical Density (O.D.)

Optical Density (O.D.)

0.25
0.14

0.2
0.12

0.15 0.1
Baseline, Pre-Odor Post Odor Post Cold Pressor Baseline, Pre-Odor Post Odor Post Cold Pressor

Lemon - blind Lavender - blind Water - blind Lemon - blind Lavender - blind Water - blind
Lemon - primed Lavender - primed Water - primed Lemon - primed Lavender - primed Water - primed

Figure 7 (a, b) Baseline-adjusted mean (7SEM) changes in PHA (6a) and Con A (6b) related to both odor exposure and expectancy
group assignment. The values shown are means of responses to the 2.5, 5.0, and 10.0 mg/ml concentrations.

the two, and no odor or odor by time interactions, ps4.20. with negative affect on the PANAS, or with the physiological
Skin barrier integrity was poorer after the stressor than data.
before, reflected in the fewer tape strips required to meet
the TEWL criterion post-stressor, F(1, 54) ¼ 36.18, po.001, 3.11. Expectancies and odor awareness
but neither the odor nor the odor by time interaction was
significant, p’s4.14.
In addition to differences related to the subject’s blind/
primed assignment, secondary analyses examined the
3.8. Pain extent to which individuals’ ratings of expected mood and
physiological change following the initial application of each
Pain intensity ratings following the cold pressor did not odor were related to their subsequent pattern of responses.
differ among the three odors, Fo1, p ¼ .58. Participants’ ratings of expected change were not signifi-
cantly associated with changes in affect or physiological
data.
3.9. Experimenter blindness We also asked subjects to name the odor they smelled, to
see if odor awareness made a difference for ‘‘blind’’
We initially did not evaluate experimenter blindness (i.e., subjects that might have influenced the expectancy manip-
whether or not the experimenter guessed the day’s odor ulation. Lemon was correctly guessed by 77% of the blind
correctly), but later assessed the variable and evaluated group, compared to 7% for lavender and 46% for water (‘‘no
results with 30 subjects. Analyses were performed using only smell’’ was counted as correct for the latter). Participants’
these 30 subjects, with one model equivalent to the primary accuracy in identifying the odor was not significantly
analysis and one model with an indicator for experimenter associated with changes in affect or physiological data.
blindness added. The results of these two models were
consistent for affect and the physiological data; the results
3.12. Aromatherapy users vs. nonusers
were not influenced by whether or not the experimenter
correctly guessed the day’s odor.
Responses to the scripted interview administered during
screening identified 21 subjects who had previously used
3.10. General affective or reflectorial effect theory aromatherapy for reasons other than enjoyment of the scent
(to relax, to improve mood, etc.). Of the 21 users, 16 had
To address the general affective or reflectorial effect theory used some form of aromatherapy at least monthly and rated
(Hirsch, 2001), we first examined the association between their aromatherapy experience positively; these partici-
odor pleasantness and affect. Adjusting for odor, higher pants were considered enthusiastic users. Analyses of IAPS
PANAS-positive affect scores were associated with higher picture rating data showed a significant odor by user
odor pleasantness ratings, F(1, 555) ¼ 4.43, p ¼ .04. On interaction, F(2, 106) ¼ 5.45, p ¼ .006; in the user group,
average, PANAS-positive affect scores increased .41 points arousal was higher in response to lemon than in response to
for each one unit increase in pleasantness rating. However, water or lavender, while arousal did not differ by odor for
this association did not differ significantly between odors. non-users. For the remaining variables, inclusion of user or
Odor pleasantness ratings were not significantly associated enthusiastic user groupings did not significantly alter
 ARTICLE IN PRESS
Olfactory influences on mood and physiology 337

differences related to odor, primed group membership, In a finding relevant to clinical efficacy, pain ratings
time, or their interactions. following the cold pressor were not modulated by either
lemon or lavender; by limiting the cold pressor time to a
minute we had a very mild stressor that nonetheless reliably
4. Discussion elevated catecholamines, heart rate and blood pressure,
altered lymphocyte proliferation, and delayed skin barrier
Lemon oil’s positive effects on mood were reflected in both repair. The finding that odor exposure was unrelated to pain
self-reports (PANAS-positive mood scale) as well as three ratings is in accord with two well-controlled studies that
unobtrusive mood assessments (IAPS valence ratings, emo- used much more challenging pain stimuli than our brief cold
tional Stroop responses, and emotion word use in thought pressor. For example, men and women who placed their
listings). In contrast, lavender’s effects on mood were no hand and forearm in ice water for up to 15 min reported
better than water (and sometimes more negative) on the more pain when the ambient odor was either pleasant
PANAS, the Stroop, and IAPS valence ratings. (lemon) or unpleasant (machine oil) compared to the no-
The immunological data did not support the purported odor condition (Martin, 2006). In another study researchers
clinical efficacy of lemon or lavender oil. In fact, DTH skin compared pain responses to contact heat, pressure, and
responses were larger following inhalation of water than ischemic pain in a randomized crossover design with more
lemon or lavender, and greater DTH responses are typically varied and taxing pain stimuli, e.g., ischemic pain was
thought to indicate a more robust or better immune terminated either after the subject rated the pain as
response to the antigen (Segerstrom, 2006). Odor exposure maximally unpleasant, after 15 min, or on the subject’s
was unrelated to IL-6 and IL-10 production. The sole request; their odors (lavender, rosemary, and distilled
potentially supportive immunological finding was equivocal, water) were not reliably related to pain ratings immediately
as it also reflected the impact of expectancies: When following the stimuli (Gedney et al., 2004). The absence of
subjects smelled lavender, the blastogenic response to any analgesic benefits across all three studies is notable,
mitogens increased in the primed group from pre-stressor because pain reduction is a primary reason for aromather-
to post-stressor and decreased in the blind group, while the apy’s widespread use in health-related applications ranging
priming effect was in the opposite direction for water and from labor pain to post-surgical discomfort (Price and Price,
lemon. 1999).
If lemon oil inhalation had such reliably positive effects We did not find gender differences in odor responsiveness
on mood, why not physiology? Marshaling evidence from even though women have a more acute sense of smell than
experimental studies, a recent review suggested that men as assessed by standardized tests of odor identification
inducing an activated state of positive affect may trigger and detection (Bartoshuk and Beauchamp, 1994). However,
short-term rises in various physiological systems that have our use of standardized detection tests as part of our
transient effects on immune and autonomic function (Press- selection criteria, an important control not found in most
man and Cohen, 2005). For example, induction of positive other aromatherapy studies, likely minimized subsequent
affect via hypnosis or through a movie enhanced catecho- gender differences.
lamine secretion in two laboratory studies (Levi, 1965; Repeated or prolonged exposure to an odor can produce
Zachariae et al., 2001). In our study the catecholamine time decrements in sensitivity to that odorant, and adaptation
by odor interaction reflected the fact that norepinephrine may also diminish behavioral responses; indeed, 30 min of
remained elevated 20 min after the stressor when subjects exposure to lemon-smelling citral resulted in reductions in
were smelling lemon, but returned to pre-stressor levels subsequent ratings of its pleasantness (Dalton and Wysocki,
when they were smelling lavender or water. Indeed, lemon 1996). Accordingly, we may have had some adaptation over
oil’s effects on both norepinephrine and mood are consistent the interval of exposure in the GCRC. Nonetheless, in
with the systemic effect theory which suggests that specific clinical applications of aromatherapy, the same odor is used
scents can evoke catecholamine changes (Feller, 1997; Price over a period of several hours, and thus our paradigm
and Price, 1999). Moreover, catecholamines can inhibit the provided relevant data on results that might be expected
antigen-presenting capacity of Langerhans cells in the skin from normal aromatherapy practices. However, this study
(Seiffert et al., 2002); thus, lemon’s enhanced norepinephr- did not test the long-term effects of aromatherapy.
ine response could have contributed to the lower DTH Many complementary/alternative therapies have not
responses observed following lemon compared to water. been subjected to well-controlled tests. The data from this
Our study was designed to contrast diverse conceptual randomized controlled trial are important because they
perspectives, clarify mechanisms, and assess possible directly address both potential mechanisms and clinical
clinical efficacy. In terms of conceptual perspectives, the efficacy. We chose lemon and lavender because they are
lemon-enhanced norepinephrine and mood responses sup- widely used purported stimulant and relaxant odors, and
ported the systemic effect theory (Feller, 1997; Price and health benefits have been repeatedly ascribed to them,
Price, 1999). In contrast, the only evidence that expectan- particularly lavender. Our sample included regular aro-
cies impacted responses to odors was the lymphocyte matherapy users as well as skeptics, so we also investigated
proliferation differences for lavender; we found no other the possibility that ‘‘true believers’’ might show greater
significant odor-related expectancy effects in other do- benefits. We found clear and consistent evidence that lemon
mains. In accord with the general affective theory, lemon oil inhalation enhances positive mood and also boosts
was perceived most positively and had the greatest positive norepinephrine release (in line with its activating proper-
effect on mood; however, positive mood did not translate ties), but no other obvious physiological or health-related
into enhanced physical well being. benefits from either lemon or lavender; indeed, the finding
 ARTICLE IN PRESS
338 J.K. Kiecolt-Glaser et al.

that both odors appeared to depress DTH responses relative Doty, R.L., 1986. Gender and endocrine-related influences on
to water suggests that the immunomodulatory effects of human olfactory perception. In: Meiselman, H.L., Rivlin, R.S.
these odors were negative, at least for this aspect of the (Eds.), Clinical Measurement of Taste and Smell. Macmillan
immune response. Publishing Company, New York, pp. 377–413.
Feller, R.M., 1997. Practical Aromatherapy: Understanding and
Using Essential Oils to Heal the Mind and Body. Berkley Books,
Role of funding sources New York.
Gedney, J.J., Glover, T.L., Fillingim, R.B., 2004. Sensory and
This research was supported by Grant AT002122 from the affective pain discrimination after inhalation of essential oils.
National Center for Complementary and Alternative Medi- Psychosom. Med. 66, 599–606.
cine (NCCAM) at the National Institutes of Health (NIH), NIH Goel, N., Kim, H., Lao, R.P., 2005. An olfactory stimulus modifies
nighttime sleep in young men and women. Chronobiol. Int. 22,
Training Grant AI55411, by General Clinical Research Center
889–904.
Grant MO1-RR-0034, and by Ohio State Comprehensive Grace, K., 1999. Aromatherapy Pocketbook. Llewellyn Publications,
Cancer Center Core Grant CA16058; the NIH had no further St. Paul, MN.
role in study design; in the collection, analysis and Hardy, M., Kirk-Smith, M.D., Stretch, D.D., 1995. Replacement of
interpretation of data; in the writing of the report; and in drug treatment for insomnia by ambient odour. Lancet 346, 701.
the decision to submit the paper for publication. Hirsch, A.R., 2001. Aromatherapy: art, science, or myth? In:
Weintraub, M.I. (Ed.), Alternative and Complementary Treat-
ment in Neurologic Illness. Churchill Livingstone, Philadelphia,
Conflict of interest PA, pp. 128–150.
Hirsch, M.S., Liebert, R.M., 1998. The physical and psychological
None declared. experience of pain: the effects of labeling and cold pressor
temperature on three pain measures in college women. Pain 77,
41–48.
Acknowledgments Jellinek, J.S., 1997. Psychodynamic odor effects and their mechan-
isms. Cosmet. Toilet. 112, 61–71.
We appreciate the helpful assistance of Cathie Atkinson, Keville, K., Green, M., 1995. Aromatherapy: A Complete Guide to
Michael DiGregorio, Bryon Laskowski, and Laura Von Hoene. the Healing Art. The Crossing Press, Fredom, CA.
Kiecolt-Glaser, J.K., Glaser, R., 1988. Methodological issues in
behavioral immunology research with humans. Brain. Behav.
References Immun. 2, 67–78.
Knasko, S.C., 1992. Ambient odor’s effect on creativity, mood, and
Baecke, J.A.H., Burema, J., Frijters, J.E.R., 1982. A short perceived health. Chem. Senses 17, 27–35.
questionnaire for the measurement of habitual physical activity Knasko, S.C., Gilbert, A.N., Sabini, J., 1990. Emotional state,
in epidemiological studies. Am. J. Clin. Nutr. 36, 936–942. physical well-being, and performance in the presence of feigned
Bartoshuk, L.M., Beauchamp, G.K., 1994. Chemical senses. Annu. ambient odor. J. Appl. Soc. Psychol. 20, 1345–1357.
Rev. Psychol. 45, 419–449. Komori, T., Fujiwara, H., Tanida, M., Nomura, J., Yokoyama, M.M.,
Blandini, F., Martignoni, E., Sances, E., Bono, G., Nappi, G., 1995. 1995. Effects of citrus fragrance on immune function and
Combined response of plasma and platelet catecholamines to depressive states. Neuroimmunomodulation 2, 174–180.
different types of short-term stress. Life Sci. 56, 1113–1120. Lang, P.J., Bradley, M.M., Cuthbert, B.N., 1999. International
Buchbauer, G., Jirovetz, L., Jager, W., Plank, C., Dietrich, H., 1993. Affective Picture System (IAPS): Technical Manual and Affective
Fragrance compounds and essential oils with sedative effects Ratings. The Center for Research in Psychophysiology, University
upon inhalation. J. Pharm. Sci. 82, 660–664. of Florida, Gainesville, FL.
Cacioppo, J.T., Petty, R.E., 1981. Electromyographic specificity Levi, L., 1965. The urinary output of adrenalin and noradrenalin
during covert information processing. Psychophysiology 18, during pleasant and unpleasant emotional states: a preliminary
518–523. report. Psychosom. Med. 27, 80–85.
Carroll, J.B., Davies, P., Richman, B., 1971. Word Frequency Book. Lis-Balchin, M., Hart, S., 2002. Chemical profiles of lavender oils
Duke University Press, Durham, NC. and pharmacology. In: Lis-Balchin, M. (Ed.), Lavender: The
Choi, E.-H., Brown, B.E., Crumrine, D., Chang, S., Man, M.-Q., Elias, Genus Lavandula. Taylor & Francis, New York, pp. 243–250.
P.M., Feingold, K.R., 2005. Mechanisms by which psychologic Martin, G.N., 2006. The effect of exposure to odor on the
stress alters cutaneous permeability barrier homeostasis and perception of pain. Psychosom. Med. 68, 613–616.
stratum corneum integrity. J. Invest. Dermatol. 124, 587–595. Mogg, K., Bradley, B.P., Williams, R., Mathews, A., 1993. Subliminal
Cowart, B., 1989. Relationships between taste and smell across the processing of emotional information in anxiety and depression.
life-span. In: Murphy, C., Cain, W.S., Hegsted, D.M. (Eds.), J. Abnorm. Psychol. 102, 304–311.
Nutrition and the Chemical Senses in Aging: Recent Advances and Pennebaker, J.W., Francis, M.E., 1999. Linguistic Inquiry and Word
Current Research Needs. Annals of the New York Academy of Count: LIWC. Erlbaum, Mahwah, NJ.
Sciences, New York, pp. 39–55. Pressman, S., Cohen, S., 2005. Does positive affect influence health?
Dalton, P., Wysocki, C.J., 1996. The nature and duration of Psychol. Bull. 131, 925–971.
adaptation following long-term odor exposure. Percept. Psycho- Price, S., Price, L., 1999. Aromatherapy for Health Professionals.
phys. 58, 781–792. Churchill Livingstone, Edinburgh.
Denda, M., Tsuchiya, T., Shoji, K., Tanida, M., 2000. Odorant Segerstrom, S.C., 2006. How does optimism suppress immunity?
inhalation affects skin barrier homeostasis in mice and humans. Evaluation of three affective pathways. Health Psychol. 25,
Br. J. Dermatol. 142, 1007–1010. 653–657.
Diego, M.A., Jones, N.A., Field, T., Hernandez-Reif, M., Schanberg, Seiffert, K., Hosoi, J., Torii, H., Ozawa, H., Ding, W.H., Campton,
S., Kuhn, C., McAdam, V., Galamaga, R., Galamaga, M., 1998. K., Wagner, J.A., Granstein, R.D., 2002. Catecholamines inhibit
Aromatherapy positively affects mood, EEG patterns of alertness the antigen-presenting capability of epidermal Langerhans cells.
and math computations. Int. J. Neurosci. 96, 217–224. J. Immunol. 168, 6128–6135.
 ARTICLE IN PRESS
Olfactory influences on mood and physiology 339

Sokal, J.E., 1975. Measurement of delayed skin-test responses. N. Watson, D., Clark, L.A., Tellegen, A., 1988. Development and
Engl. J. Med. 293, 501–502. validation of brief measures of positive and negative affect: the
Tisserand, R., Balacs, T., 1995. Essential Oil Safety: A Guide for PANAS scales. J. Pers. Soc. Psychol. 54, 1063–1070.
Health Care Professionals. Churchhill Livingstone, Edinburgh. Williams, J.M.G., Mathews, A., MacLeod, C., 1996. The emotional
Torii, S., Fukuda, H., Kanemoto, H., Miyanchi, R., Hamauzu, Y., Stroop task and psychopathology. Psychol. Bull. 120, 3–24.
Kawasaki, M., 1988. Contingent negative variation (CNV) and the Zachariae, R., Jorgensen, M.M., Egekvist, H., Bjerring, P., 2001. Skin
psychological effects of odour. In: van Toller, S., Dodd, G.H. reactions to histamine of healthy subjects after hypnotically
(Eds.), Perfumery: The Psychology and Biology of Fragrance. induced emotions of sadness, anger, and happiness. Allergy 56,
Chapman & Hall, London, pp. 107–120. 734–740.