Food Stress
Food Stress
Food Stress
Department of Food Science and Nutrition, University of Minnesota, 1334 Eckles Ave., St. Paul, MN 55108, USA
Department of Psychology, University of Minnesota, 75 East River Road, Minneapolis, MN 55455, USA
c
Marketing Department, Carlson School of Management, University of Minnesota, 321 19th Ave S., Minneapolis, MN 55455, USA
b
a r t i c l e
i n f o
Article history:
Received 13 January 2014
Received in revised form 1 August 2014
Accepted 3 August 2014
Available online 10 August 2014
Keywords:
Stress
Emotion
Mood
Food choice
Trier Social Stress Task
Cortisol
a b s t r a c t
Many people experience stress as a part of their daily lives. Chronic stress can have an impact on physical
and mental health. Since food and eating are generally associated with positive moods, we explored how
aspects of meal preparation can relieve stress and improve measures related to mood.
Our main objectives were to determine whether choosing meal components and/or preparing a meal
would improve measures related to mood and reduce stress.
Participants came individually to our lab at dinner time. We measured stress (salivary cortisol, heart
rate and blood pressure) and took measures related to mood on arrival. We then induced stress (Trier
Social Stress Task) and took measures related to stress and mood again. Each participant was assigned
to one of four experimental conditions. In the prepare-choice condition participants prepared a meal
(pasta + sauce + inclusions) and had control over selection of meal components. In the prepare-no-choice
condition participants prepared their meal, but had no control over the menu. In the choice-no-prepare
condition participants had control over the menu, but the meal was prepared by someone else. In the
no-prepare-no-choice condition participants were provided with a meal prepared by someone else. Food
preference questionnaires conducted before the stress induction ensured that all participants received
foods they liked.
Having no choice produced greater reductions in the mood-related measures of anxiety and anger compared with the choice condition. Systolic blood pressure was reduced more in the no choice than in the
choice condition after the meal. Preparing versus not preparing had little effect on measures related to
stress and mood.
People may nd choosing to be a depleting task on their limited psychological resources; hence, choosing can add to their general stress. Not faced with choosing, one avoids this unnecessary stress. Consuming a meal without the burden of choosing has potential as a stress-reduction strategy.
2014 Elsevier Ltd. All rights reserved.
Introduction
Importance of the food-mood relationship
Eating behaviors, stress, and negative mood1 all affect physical
and mental health, but their interactions are complex and not well
dened. Similar to unhealthy eating behaviors, negative mood and
Corresponding author. Address: 225 Food Science and Nutrition, 1334 Eckles
Ave, St. Paul, MN 55108, USA. Tel.: +1 507 382 8449.
E-mail addresses: [email protected] (K.E. Osdoba), [email protected] (T. Mann),
[email protected] (J.P. Redden), [email protected] (Z. Vickers).
1
Although there are clear distinctions in terms of psychological constructs, the
words mood and emotion are often used synonymously in the literature. For the
remainder of this paper, the words mood and emotion will be used interchangeably.
http://dx.doi.org/10.1016/j.foodqual.2014.08.001
0950-3293/ 2014 Elsevier Ltd. All rights reserved.
242
in liking could presumably improve mood and stress, limited evidence suggests that this actually happens. When the participants
of Garg and Lerner (2013) were given a choice of reward (chocolates vs. a ballpoint pen, with the idea that this would be an easy
choice and most people would choose the chocolates) after induction of sad mood, sadness was reduced more than if the participants were just presented with chocolates as a gift. The work of
Garg and Lerner (2013) and Iyengar and Lepper (2000) showed that
simple choices, such as those with few options and/or trivial consequences, may result in less negative consequences for mood
and stress. The detrimental effects of too many choices, however,
especially when distressed, may outweigh the benets of having
the freedom to choose.
Food preparation
The alleviation of stress and improvement of mood are likely
outcomes of food preparation, although limited evidence suggests
that food preparation itself can be stressful. Benson, Beary, and
Carol (1974) suggested that activities involving mindless, repetitive
tasks elicit a relaxation response. Food preparation, which entails
such tasks as chopping vegetables and repeated stirring, may t
well into this category. Food preparation may also result in
improved mood when it is done out of a sense of duty (i.e., to feed
the family) or to please others (Daniels, Glorieux, Minnen, & van
Tienoven, 2012). Building on this, Costa (2013) found that people
ascribe strong, positive feelings towards cooking hot meals at
home, whereas they feel guilty (along with other negative emotions) when they do not cook at home. Food preparation allows
for a certain amount of autonomy and control. Control in general
is related to well-being and life satisfaction (Tangney, Baumeister,
& Boone, 2004). Knowing the ingredients and processes that go into
ones meal may be an easy way to exercise control and reap the psychological benets. On the other hand, food preparation can be
stressful, especially when hunger, distractions, and time constraints
come into play (Daniels et al., 2012). In the case of mood improvement and stress relief, the advantages of preparing food may, under
many circumstances, outweigh its detriments.
Objectives and hypotheses
The main objective of this study was to explore whether choice
of meal ingredients (vs. no choice) and/or preparation of a meal (vs.
someone else preparing) inuence the stress-reducing and moodlifting effects of food and eating.
Given the stressful consequences inherent to making choices,
we expected choosing ingredients to have detrimental effects on
mood and stress. We specically hypothesized that if people did
not choose their meal ingredients, they would show a greater
improvement in measures related to mood and larger reduction
in stress after eating than if they did choose their meal ingredients.
Given the positive consequences from preparing food, we
expected preparing food to produce improvements in mood and
stress. We specically hypothesized that if people prepared the
meal themselves, they would show a greater improvement in measures related to mood and larger reduction in stress after eating
than if someone else prepared the meal for them.
Materials and methods
Participants
One hundred eighteen participants (36% male, mean age = 28,
SD age = 11, range = 1863) were recruited via email listserv and
posted yers. They were screened for availability and liking of meal
243
would happen later in the study when it was time for the meal.
This was done to familiarize the participant with the setting, so
as to minimize any additional stress invoked by new surroundings
and uncertain tasks. Then the participant and experimenter sat at a
table with a computer. A blood pressure cuff was attached to the
participants non-dominant arm (so he/she could easily work the
computer mouse with the dominant hand). A practice blood pressure measurement and saliva sample (for cortisol measurement)
were taken. At this point, the experimenter left the room for
20 min (habituation period), during which time the participant
answered computerized questionnaires.
After 20 min, the experimenter returned and obtained baseline
blood pressure/heart rate measurements and a saliva sample. Next,
the participant completed a questionnaire containing mood-related
words (see Questionnaires) to get baseline measurements.
Instructions were then given for the stress task.
The stress task closely followed the protocol of the Trier Social
Stress Task (TSST) outlined by Kirschbaum, Pirke, and
Helhammer (1993). The experimenter told the participants that
they would have ve minutes to prepare a speech. They were told
to pretend they were a job applicant interviewing for a position in
a company. They were asked to explain why they would be the
perfect candidate for the job. They were given paper and pen to
take notes, but were told that they could not use the notes during
the speech. The experimenter indicated that two people would
come in to evaluate the speech, and then the experimenter left
the room. This TSST approach has been widely used in laboratory
settings to reliably induce both physiological and psychological
stress. We should note that we slightly modied the original task
in the following ways: the experimenter was unaware of the treatment group; the speech was not videotaped; the speech was
shorted from 10 min to 5 min, and the arithmetic task began at a
different number each minute.
When the ve minutes had passed, two people (one male, one
female; hereafter referred to as confederates) in white lab coats
and holding clipboards entered. They asked the participant to
stand up and deliver the speech into a microphone. The confederates told the participant that the speech would be recorded and a
voice frequency analysis would be done on the recording. (The
speech was not actually recorded.) One of the confederates was
introduced as being specially trained to monitor nonverbal behavior. This confederate would be taking notes throughout the task.
The participant was then told to begin the speech. The participant
was required to speak for the entire ve minutes. Specic verbal
Table 1
Meal components.
Ingredient
Type
Manufacturer
Address
Portion (g)
Pasta
Olive oil
Salt
Rotini
Extra-virgin
Iodized
Creamette
Pompeian
Roundys
Allentown, PA
Baltimore, MD
Milwaukee, WI
224*
3.4**
4.5***
Sauces
Alfredo sauce
Marinara sauce
Four Cheese
Traditional
Roundys
Prego
Milwaukee, WI
Camden, NJ
150
150
Ortega
Bella Sun Luci
Parsippany, NJ
Chico, CA
15
20
Olives
Mushrooms
Mezzetta
Roundys
American Canyon, CA
Milwaukee, WI
20
25
Seasonings
Parmesan cheese
Basil
Black pepper
Grated
Dried
Dried
Roundys
McCormick
Roundys
Milwaukee, WI
Sparks, MD
Milwaukee, WI
7
<1
<1
Inclusions
Green chilies
Sun-dried tomatoes
*
**
***
Cooked: This amount is approximately two servings (based on the Nutrition Facts panel on the pasta box).
Tossed with cooked pasta (13.5 g/4 servings).
Added to pasta cooking water (18 g/4 servings).
244
Consent form,
Pracce
measurements
Time
(min)
Quesonnaires
(20 min)
Baseline mood,
BP/HR, corsol
25
30
Stress
Task
Post-stress mood,
BP/HR
45
Post-stress corsol,
Post-meal mood, BP/HR
50
70
75
Quesonnaires
(30 min)
Meal
Post-meal corsol,
Final mood, BP/HR
105
110
End
115
Comics, debrief,
payment
instruction sheet and the experimenter left the room. The participant sat back down at the table with the pasta and was given
ten minutes to eat.
At the end of the eating period, the experimenter returned and
removed the plate from in front of the participant. Another saliva
sample was taken (post-stress), as well as blood pressure/heart
rate measurements (post-meal). The participant then began
answering questionnaires again (including the post-meal measures
related to mood). The experimenter then left the room for 30 min.
When the 30 min elapsed, the experimenter returned and took
nal blood pressure/heart rate measurements and the post-meal
saliva sample. After that, the participant lled out the nal measures related to mood. Finally, the participant read through a set
of comic strips and rated how funny they were. This was done to
ensure that any residual stress after the meal was gone by the time
they left. Then the participant was debriefed and introduced to the
two confederates who had administered the stress task. Lastly, the
participant was paid and thanked for coming.
Meal
The meal consisted of a hot pasta dish prepared on site and a
glass of water. The participants in the No Choice groups were given
a subset of the ingredients that they had rated six or higher on the
liking scales during the prescreening process. For these participants, menu sheets were lled out in advance by the experimenter.
Participants in the Choice groups lled out their own menu sheets
as described above in Participant visit protocol. Each participants meal consisted of one of two sauces, two or three of four
possible vegetable inclusions, and one of three seasonings. However, all ingredient choices were presented at the meal preparation
area to reinforce the fact that the No Choice participants were missing out on the other ingredients. See Table 1 for ingredients and
portion sizes. Meal ingredients were pre-portioned and set out
prior to the start of the study. Labels were placed behind each
ingredient so each participant was exposed to the names of all
the ingredients. Pasta was precooked according to the package
directions with the addition of 18 g salt to the cooking water.
Cooked pasta was tossed with 13.5 g olive oil, portioned, and
refrigerated until needed. Cooked pasta was held refrigerated for
no longer than 24 h.
When it was time for the meal (right after the stress task), the
experimenter brought the participant over to the meal preparation
area. The experimenter removed the ingredients that the participant was not going to eat and placed them on a tray. For the No
Prepare groups, the experimenter then added the selected ingredients to the bowl of pasta, stirred it, and microwaved it for two minutes. When it was heated, the experimenter poured it onto a plate.
For the Prepare groups, the experimenter gave the participant an
instruction sheet and asked if there were any questions about
the meal preparation. The experimenter then left the room, taking
the tray of extra ingredients. The participant was instructed to
245
Table 2
Words in mood questionnaire (adapted from the Prole of Mood States; McNair et al.,
1971).
Category*
Emotion words
Anxiety
Anxious
Awkward
Discouraged
Embarrassed
Intimidated
On edge
Pressured
Self-conscious
Uneasy
Anger
Angry
Annoyed
Resentful
Threatened
Fatigue
Exhausted
Fatigued
Worn out
Positivity
Calm
Cheerful
Content
Lively
Satised
Sadness
uncategorized
*
Hopeless
Sad
**
Vigorous
Source
Arnett (1994)
Rotter (1966)
Schwartz et al. (2002)
Cohen, Kamarck, and
Mermelstein (1983)
Zuckerman, Kolin, Price,
and Zoob (1964)
Neuberg and Newsom
(1993)
Glynn and Ruderman
(1986)
Tangney et al. (2004)
Van Trijp, Lhteenmki, and
Tuorila (1992)
Stunkard and Messick
(1985)
Rosenberg (1965)
Scheier, Carver, and Bridges
(1994)
Polivy, Herman, and Warsh
(1978)
Raju (1980)
Farmer and Sundber (1986)
Van Strien, Frijters, Bergers,
and Defares (1986)
246
Table 4
Changes in stress and mood responses from before to after the stress task. Least squares means for each factor level are given (standard errors in parentheses). Yes under the
Choice heading indicates Choice group and No indicates No Choice group. Yes under the Prepare heading indicates Prepare group and No indicates No Prepare group. Positive
means indicate increases in a response and negative means indicate a decrease. F-statistics and p-values are for 2 2 ANOVAs for each response and factor.
Responses
Anxietya
Anger
Fatigue
Positivity
Sadness
Systolic BP (mmHg)
Diastolic BP (mmHg)
Heart rate (bpm)
Cortisol (lg/dL)
a
Choice
Prepare
Yes
No
Yes
No
1.0 (0.2)
0.6 (0.2)
0.2 (0.2)
1.0 (0.1)
0.1 (0.1)
9.1 (1.2)
6.3 (0.9)
0.3 (1.0)
0.1 (0.04)
1.8 (0.2)
1.2 (0.2)
0.3 (0.2)
1.4 (0.1)
0.3 (0.1)
11.3 (1.2)
6.5 (0.9)
2.3 (1.0)
0.2 (0.04)
8.6
5.5
0.2
4.1
1.7
1.7
0.0
1.8
0.0
0.00
0.02
0.66
0.05
0.20
0.20
0.86
0.18
0.91
1.3 (0.2)
0.8 (0.2)
0.2 (0.2)
1.1 (0.1)
0.1 (0.1)
9.6 (1.2)
5.9 (0.9)
0.3 (1.0)
0.1 (0.04)
1.4 (0.2)
0.9 (0.2)
0.2 (0.2)
1.3 (0.1)
0.4 (0.1)
10.7 (1.2)
7.0 (0.9)
2.2 (1.0)
0.2 (0.04)
0.0
0.1
0.0
0.5
3.1
0.4
0.7
1.7
1.5
0.85
0.78
0.88
0.47
0.08
0.53
0.40
0.20
0.23
Mood words were rated on a 7-point scale, from 1 = not at all to 7 = extremely.
247
Table 5
Changes in stress and mood responses from after the stress test to after the meal. Least squares means for each factor level are given (standard errors in parentheses). Yes under
the Choice heading indicates Choice group and No indicates No Choice group. Yes under the Prepare heading indicates Prepare group and No indicates No Prepare group.
Positive means indicate increases in a response and negative means indicate a decrease. F-statistics and p-values (one-sided) are for 2 2 ANOVAs for each response and factor.
Responses
Anxietya
Anger
Fatigue
Positivity
Sadness
Systolic BP (mmHg)
Diastolic BP (mmHg)
Heart rate (bpm)
Cortisol (lg/dL)
a
Choice
Prepare
Yes
No
Yes
No
1.0 (0.1)
0.4 (0.1)
0.4 (0.1)
0.4 (0.1)
0.2 (0.1)
3.3 (1.0)
1.2 (0.8)
2.4 (1.0)
0.1 (0.03)
1.5 (0.1)
0.9 (0.1)
0.4 (0.1)
0.6 (0.1)
0.4 (0.1)
7.1 (1.0)
1.6 (0.7)
0.0 (1.0)
0.1 (0.03)
7.0
6.6
0.0
1.2
2.0
7.0
0.2
2.9
0.5
0.00
0.01
0.42
0.14
0.08
0.00
0.35
0.05
0.24
1.3 (0.1)
0.6 (0.1)
0.3 (0.1)
0.6 (0.1)
0.2 (0.1)
4.0 (1.0)
0.7 (0.7)
2.7 (1.0)
0.1 (0.03)
1.3 (0.1)
0.6 (0.1)
0.5 (0.1)
0.4 (0.1)
0.4 (0.1)
6.4 (1.0)
2.1 (0.8)
0.3 (1.0)
0.1 (0.03)
0.0
0.1
0.9
0.9
1.4
2.8
1.8
4.3
0.9
0.50
0.41
0.18
0.18
0.12
0.05
0.09
0.02
0.17
Mood words were rated on a 7-point scale, from 1 = not at all to 7 = extremely.
increased more after the TSST for those in the No Choice groups versus the Choice groups (Table 4). Positivity ratings decreased more
for those in the No Choice groups versus the Choice groups. No differences in stress increase or deterioration in measures related to
mood were observed between those who prepared the meal and
those who did not prepare the meal. No signicant interaction
effects were seen between the Choice and Prepare factors (data
not shown).
Preparing a meal versus not preparing a meal had little effect on
reducing stress or improving measures related to mood. Heart rate
increased more for those in the Prepare groups than in the No Prepare groups after the meal (Table 5). Systolic blood pressure
decreased more for those in the No Prepare groups versus the Prepare groups. Prepare versus No Prepare had no effect on the other
responses.
In agreement with our hypothesis, those in the No Choice groups
showed a greater decrease in anxiety, anger, and systolic blood
pressure after the meal than those in the Choice groups (Table 5).
After the meal, heart rate increased more for those in the Choice
versus the No Choice groups. Choice versus No Choice had no effect
on the other responses.
Discussion
The No Choice groups reacted more strongly to the TSST
Stress and measures related to negative mood would have been
predicted to have increased similarly for all groups of participants,
but in the context of this study the No Choice groups reacted more
strongly to the TSST than the Choice groups. Those in the No Choice
groups had greater increases in anxiety and anger and greater
decreases in positivity after the TSST (Table 4) than did the Choice
groups. The experimental protocol was the same for all participants through the end of the TSST. The manipulated differences
occurred only during the meal portion of the study. However, at
the very beginning of the experiment, participants were briefed
on what would happen during the meal. This was done to prevent
possible stress increases from worry about preparing a meal in an
unfamiliar setting. Verbal cues were subtle, but there was a difference between what was said to the Choice groups versus what was
said to the No Choice groups. Those in the Choice groups heard:
You will ll out a menu where you get to choose the ingredients
for your pasta, and those in the No Choice groups heard: You will
receive a list of ingredients that you will need to add to your
pasta. It is possible that these differences in wording were enough
to induce the choice manipulation before it was intended and
allow for the No Choice groups to be more susceptible to stress
increases because they felt they would not be in control of their
meal choices. This potential susceptibility was possibly reected
Meanpoststress Meanbaseline
.
SDbaseline
248
stress induction was effective and the HPA axis was activated, as evidenced by high cortisol levels after the TSST.
Cortisol directly affects appetite and food-related brain activity,
which may have blurred differences in the effects of choosing and
preparing a meal on cortisol levels after eating. Consumption of
high-carbohydrate foods may increase HPA-axis activity (indicated
by elevated cortisol levels) (Lemmens, Martens, Born, Martens, &
Westerterp-Plantenga, 2011). The pasta meal in this study may
have had this effect, resulting in post-meal cortisol levels that were
higher than they would have been had the meal been lower in carbohydrates. Percent carbohydrate intake was calculated based on
each participants pasta dish composition (from nutrition labels of
each ingredient), taking into account amounts of each ingredient
added and total amount of food consumed. In fact, cortisol change
after the meal was negatively correlated with percent carbohydrate
intake (Pearson correlation coefcient = 0.21, p = 0.04), indicating
that the higher a participants meal was in carbohydrates, the
greater was the participants cortisol decrease. Increased cortisol
during stress can also cause increased food intake (Martens,
Rutters, Lemmens, Born, & Westerterp-Plantenga, 2010). However,
this was not seen in the present study, as cortisol increase after the
TSST was not signicantly correlated with food intake (Pearson correlation coefcient = 0.14, p = 0.18).
Acknowledgements
This material is based upon work supported by NASA under
award No. NNX12AE56G. This research has been supported in part
by the Minnesota Agricultural Experiment Station. We thank Matt
Niezgoda, Sara Olson, Samantha Bzdawka, Claire Burrington, Sean
Lee, Dana Osdoba, Zach Baggio, Amanda Schlink, and Stephanie Elsbernd for their help conducting the studies; Britt Ahlstrom, Rachel
Burns, and Heather Scherschel for their input on the experimental
design; and Megan Heyman and Zhongnan Jin for their help with
statistical analysis.
All authors have approved the nal version of this paper.
249
250
Walsh, E. M., & Kiviniemi, M. T. (2013). Changing how I feel about the food:
Experimentally manipulated affective associations with fruits change fruit
choice behaviors. Journal of Behavioral Medicine. http://dx.doi.org/10.1007/
s10865-012-9490-5.
Wansink, B., Payne, C. R., & North, J. (2007). Fine as North Dakota Wine: Sensory
expectations and the intake of companion foods. Physiology & Behavior, 90,
712716.
Zuckerman, M., Kolin, E. A., Price, L., & Zoob, I. (1964). Development of a sensationseeking scale. Journal of Consulting Psychology, 28(6), 477482.