Healthy food choices are happy food choices: Evidence from a real life sample using smartphone based assessments

Abstract

Research suggests that “healthy” food choices such as eating fruits and vegetables have not only physical but also mental health benefits and might be a long-term investment in future well-being. This view contrasts with the belief that high-caloric foods taste better, make us happy, and alleviate a negative mood. To provide a more comprehensive assessment of food choice and well-being, we investigated in-the-moment eating happiness by assessing complete, real life dietary behaviour across eight days using smartphone-based ecological momentary assessment. Three main findings emerged: First, of 14 different main food categories, vegetables consumption contributed the largest share to eating happiness measured across eight days. Second, sweets on average provided comparable induced eating happiness to “healthy” food choices such as fruits or vegetables. Third, dinner elicited comparable eating happiness to snacking. These findings are discussed within the “food as health” and “food as well-being” perspectives on eating behaviour.

Full text

1
SCientifiC REPORTS | 7: 17069 | DOI:10.1038/s41598-017-17262-9
www.nature.com/scientificreports
Healthy food choices are happy
food choices: Evidence from a real
life sample using smartphone based
assessments
Deborah R. Wahl, Karoline Villinger, Laura M. König , Katrin Ziesemer, Harald T. Schupp &
Britta Renner
Research suggests that “healthy” food choices such as eating fruits and vegetables have not only
physical but also mental health benets and might be a long-term investment in future well-being.
This view contrasts with the belief that high-caloric foods taste better, make us happy, and alleviate
a negative mood. To provide a more comprehensive assessment of food choice and well-being, we
investigated in-the-moment eating happiness by assessing complete, real life dietary behaviour across
eight days using smartphone-based ecological momentary assessment. Three main ndings emerged:
First, of 14 dierent main food categories, vegetables consumption contributed the largest share
to eating happiness measured across eight days. Second, sweets on average provided comparable
induced eating happiness to “healthy” food choices such as fruits or vegetables. Third, dinner elicited
comparable eating happiness to snacking. These ndings are discussed within the “food as health” and
“food as well-being” perspectives on eating behaviour.
When it comes to eating, researchers, the media, and policy makers mainly focus on negative aspects of eating
behaviour, like restricting certain foods, counting calories, and dieting. Likewise, health intervention eorts,
including primary prevention campaigns, typically encourage consumers to trade o the expected enjoyment of
hedonic and comfort foods against health benets1. However, research has shown that diets and restrained eating
are oen counterproductive and may even enhance the risk of long-term weight gain and eating disorders2,3. A
promising new perspective entails a shi from food as pure nourishment towards a more positive and well-being
centred perspective of human eating behaviour1,4,5. In this context, Block et al.4 have advocated a paradigm shi
from “food as health” to “food as well-being” (p. 848).
Supporting this perspective of “food as well-being”, recent research suggests that “healthy” food choices,
such as eating more fruits and vegetables, have not only physical but also mental health benets6,7 and might
be a long-term investment in future well-being8. For example, in a nationally representative panel survey of
over 12,000 adults from Australia, Mujcic and Oswald8 showed that fruit and vegetable consumption predicted
increases in happiness, life satisfaction, and well-being over two years. Similarly, using lagged analyses, White and
colleagues9 showed that fruit and vegetable consumption predicted improvements in positive aect on the sub-
sequent day but not vice versa. Also, cross-sectional evidence reported by Blanchower et al.10 shows that eating
fruits and vegetables is positively associated with well-being aer adjusting for demographic variables including
age, sex, or race11. Of note, previous research includes a wide range of time lags between actual eating occasion
and well-being assessment, ranging from 24 hours9,12 to 14 days6, to 24 months8. us, the ndings support the
notion that fruit and vegetable consumption has benecial eects on dierent indicators of well-being, such as
happiness or general life satisfaction, across a broad range of time spans.
e contention that healthy food choices such as a higher fruit and vegetable consumption is associated with
greater happiness and well-being clearly contrasts with the common belief that in particular high-fat, high-sugar,
or high-caloric foods taste better and make us happy while we are eating them. When it comes to eating, people
usually have a spontaneous “unhealthy = tasty” association13 and assume that chocolate is a better mood booster
Department of Psychology, University of Konstanz, Konstanz, Germany. Deborah R. Wahl and Karoline Villinger
contributed equally to this work. Correspondence and requests for materials should be addressed to D.R.W. (email:
deborah.wahl@uni-konstanz.de) or B.R. (email: britta.renner@uni-konstanz.de)
Received: 5 June 2017
Accepted: 23 November 2017
Published: xx xx xxxx
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
2
SCientifiC REPORTS | 7: 17069 | DOI:10.1038/s41598-017-17262-9
than an apple. According to this in-the-moment well-being perspective, consumers have to trade o the expected
enjoyment of eating against the health costs of eating unhealthy foods1,4.
A wealth of research shows that the experience of negative emotions and stress leads to increased consump-
tion in a substantial number of individuals (“emotional eating”) of unhealthy food (“comfort food”)14–17. However,
this research stream focuses on emotional eating to “smooth” unpleasant experiences in response to stress or neg-
ative mood states, and the mood-boosting eect of eating is typically not assessed18. One of the few studies testing
the eectiveness of comfort food in improving mood showed that the consumption of “unhealthy” comfort food
had a mood boosting eect aer a negative mood induction but not to a greater extent than non-comfort or neu-
tral food19. Hence, even though people may believe that snacking on “unhealthy” foods like ice cream or chocolate
provides greater pleasure and psychological benets, the consumption of “unhealthy” foods might not actually be
more psychologically benecial than other foods.
However, both streams of research have either focused on a single food category (fruit and vegetable con-
sumption), a single type of meal (snacking), or a single eating occasion (aer negative/neutral mood induction).
Accordingly, it is unknown whether the boosting eect of eating is specic to certain types of food choices and
categories or whether eating has a more general boosting eect that is observable aer the consumption of both
“healthy” and “unhealthy” foods and across eating occasions. Accordingly, in the present study, we investigated
the psychological benets of eating that varied by food categories and meal types by assessing complete dietary
behaviour across eight days in real life.
Furthermore, previous research on the impact of eating on well-being tended to rely on retrospective
assessments such as food frequency questionnaires8,10 and written food diaries9. Such retrospective self-report
methods rely on the challenging task of accurately estimating average intake or remembering individual eating
episodes and may lead to under-reporting food intake, particularly unhealthy food choices such as snacks7,20.
To avoid memory and bias problems in the present study we used ecological momentary assessment (EMA)21
to obtain ecologically valid and comprehensive real life data on eating behaviour and happiness as experienced
in-the-moment.
In the present study, we examined the eating happiness and satisfaction experienced in-the-moment, in real
time and in real life, using a smartphone based EMA approach. Specically, healthy participants were asked to
record each eating occasion, including main meals and snacks, for eight consecutive days and rate how tasty their
meal/snack was, how much they enjoyed it, and how pleased they were with their meal/snack immediately aer
each eating episode. is intense recording of every eating episode allows assessing eating behaviour on the level
of dierent meal types and food categories to compare experienced eating happiness across meals and categories.
Following the two dierent research streams, we expected on a food category level that not only “unhealthy” foods
like sweets would be associated with high experienced eating happiness but also “healthy” food choices such as
fruits and vegetables. On a meal type level, we hypothesised that the happiness of meals diers as a function of
meal type. According to previous contention, snacking in particular should be accompanied by greater happiness.
Results
Eating episodes. Overall, during the study period, a total of 1,044 completed eating episodes were reported
(see also Table1). On average, participants rated their eating happiness with M = 77.59 which suggests that over-
all eating occasions were generally positive. However, experienced eating happiness also varied considerably
between eating occasions as indicated by a range from 7.00 to 100.00 and a standard deviation of SD = 16.41.
Food categories and experienced eating happiness. All eating episodes were categorised according
to their food category based on the German Nutrient Database (German: Bundeslebensmittelschlüssel), which
covers the average nutritional values of approximately 10,000 foods available on the German market and is a
validated standard instrument for the assessment of nutritional surveys in Germany. As shown in Table1, eating
happiness diered signicantly across all 14 food categories, F(13, 2131) = 1.78, p = 0.04. On average, experi-
enced eating happiness varied from 71.82 (SD = 18.65) for sh to 83.62 (SD = 11.61) for meat substitutes. Post
hoc analysis, however, did not yield signicant dierences in experienced eating happiness between food cat-
egories, p ≥ 0.22. Hence, on average, “unhealthy” food choices such as sweets (M = 78.93, SD = 15.27) did not
dier in experienced happiness from “healthy” food choices such as fruits (M = 78.29, SD = 16.13) or vegetables
(M = 77.57, SD = 17.17). In addition, an intraclass correlation (ICC) of ρ = 0.22 for happiness indicated that less
than a quarter of the observed variation in experienced eating happiness was due to dierences between food
categories, while 78% of the variation was due to dierences within food categories.
However, as Figure1 (left side) depicts, consumption frequency differed greatly across food categories.
Frequently consumed food categories encompassed vegetables which were consumed at 38% of all eating occa-
sions (n = 400), followed by dairy products with 35% (n = 366), and sweets with 34% (n = 356). Conversely,
rarely consumed food categories included meat substitutes, which were consumed in 2.2% of all eating occasions
(n = 23), salty extras (1.5%, n = 16), and pastries (1.3%, n = 14).
Amount of experienced eating happiness by food category. To account for the frequency of con-
sumption, we calculated and scaled the absolute experienced eating happiness according to the total sum score.
As shown in Figure1 (right side), vegetables contributed the biggest share to the total happiness followed by
sweets, dairy products, and bread. Clustering food categories shows that fruits and vegetables accounted for
nearly one quarter of total eating happiness score and thus, contributed to a large part of eating related happiness.
Grain products such as bread, pasta, and cereals, which are main sources of carbohydrates including starch and
bre, were the second main source for eating happiness. However, “unhealthy” snacks including sweets, salty
extras, and pastries represented the third biggest source of eating related happiness.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
3
SCientifiC REPORTS | 7: 17069 | DOI:10.1038/s41598-017-17262-9
Experienced eating happiness by meal type. To further elucidate the contribution of snacks to eating
happiness, analysis on the meal type level was conducted. Experienced in-the-moment eating happiness signif-
icantly varied by meal type consumed, F (4, 1039) = 11.75, p < 0.001. Frequencies of meal type consumption
ranged from snacks being the most frequently logged meal type (n = 332; see also Table1) to aernoon tea being
the least logged meal type (n = 27). Figure2 illustrates the wide dispersion within as well as between dier-
ent meal types. Aernoon tea (M = 82.41, SD = 15.26), dinner (M = 81.47, SD = 14.73), and snacks (M = 79.45,
SD = 14.94) showed eating happiness values above the grand mean, whereas breakfast (M = 74.28, SD = 16.35)
Meal type N M (SD)Sum Min. Max. Mcwc (SD)
Meals 1,044 77.59 (16.41) 81,004 7.00 100.00 —
Breakfast 237 74.28 (16.35) 17,604 25.00 100.00 −3.04 (13.41)
Lunch 203 73.09 (18.99) 14,838 7.00 100.00 −4.59 (16.52)
Aern oon tea 27 82.41 (15.26) 2,225 39.00 100.00 5.49 (13.81)
Dinner 245 81.47 (14.73) 19,959 11.00 100.00 4.09 (13.4)
Snack 332 79.45 (14.94) 26,378 13.33 100.00 1.52 (13.93)
Food category (according to the German Nutrient Database)
Vegetables 400 77.57 (17.17) 27,995 11.00 100.00 1.16 (15.14)
Fruits 218 78.29 (16.13) 15,659 15.67 100.00 −0.65 (13.21)
Sweets 356 78.93 (15.27) 26,443 13.33 100.00 1.68 (13.74)
Salty extras 16 80.40 (10.35) 1,126 57.67 95.33 −0.07 (8.01)
Pastries 14 78.67 (19.25) 1,023 22.67 95.33 −2.39 (18.26)
Bread 284 75.52 (16.33) 19,407 19.33 100.00 −1.55 (13.46)
Pasta 226 77.89 (16.43) 16,123 22.33 100.00 0.39 (15.93)
Cereals 133 75.05 (16.63) 9,082 29.67 100.00 −3.01 (14.13)
Potatoes 61 80.47 (19.07) 4,426 7.00 100.00 1.91 (16.82)
Dairy products 366 75.46 (16.53) 25,127 22.33 100.00 −1.37 (14.49)
Meat 194 78.26 (16.01) 13,382 22.33 100.00 0.26 (14.19)
Eggs 38 79.22 (16.21) 2,852 36.00 100.00 0.95 (15.2)
Meat substitutes 23 83.62 (11.61) 1,672 59.67 100.00 5.39 (10.44)
Fish 26 71.82 (18.65) 1,580 34.33 98.67 −4.58 (16.84)
Table 1. Descriptive statistics for eating happiness by meal type and food category. Note: Eating happiness
ranged from 1 (low) to 100 (high). Mcwc = person-mean centred average happinessscore.
Figure 1. Le side: Average experienced eating happiness (colour intensity: darker colours indicate greater
happiness) and consumption frequency (size of the cycle) for the 14 food categories. Right side: Absolute share
of the 14 food categories in total experienced eating happiness.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
4
SCientifiC REPORTS | 7: 17069 | DOI:10.1038/s41598-017-17262-9
and lunch (M = 73.09, SD = 18.99) were below the eating happiness mean. Comparisons between meal types
showed that eating happiness for snacks was significantly higher than for lunch t(533) = −4.44, p = 0.001,
d = −0.38 and breakfast, t(567) = −3.78, p = 0.001, d = −0.33. However, this was also true for dinner, which
induced greater eating happiness than lunch t(446) = −5.48, p < 0.001, d = −0.50 and breakfast, t(480) = −4.90,
p < 0.001, d = −0.46. Finally, eating happiness for aernoon tea was greater than for lunch t(228) = −2.83,
p = 0.047, d = −0.50. All other comparisons did not reach signicance, t ≤ 2.49, p ≥ 0.093.
Control Analyses. In order to test for a potential confounding eect between experienced eating happiness,
food categories, and meal type, additional control analyses within meal types were conducted. Comparing expe-
rienced eating happiness for dinner and lunch suggested that dinner did not trigger a happiness spill-over eect
specic to vegetables since the foods consumed at dinner were generally associated with greater happiness than
those consumed at other eating occasions (Supplementary TableS1). Moreover, the relative frequency of vegeta-
bles consumed at dinner (73%, n = 180 out of 245) and at lunch were comparable (69%, n = 140 out of 203), indi-
cating that the observed happiness-vegetables link does not seem to be mainly a meal type confounding eect.
Since the present study focuses on “food eects” (Level 1) rather than “person eects” (Level 2), we ana-
lysed the data at the food item level. However, participants who were generally overall happier with their eating
could have inated the observed happiness scores for certain food categories. In order to account for person-level
eects, happiness scores were person-mean centred and thereby adjusted for mean level dierences in happi-
ness. e person-mean centred happiness scores (Mcwc) represent the dierence between the individual’s average
happiness score (across all single in-the-moment happiness scores per food category) and the single happiness
scores of the individual within the respective food category. e centred scores indicate whether the single
in-the-moment happiness score was above (indicated by positive values) or below (indicated by negative values)
the individual person-mean. As Table1 depicts, the control analyses with centred values yielded highly similar
results. Vegetables were again associated on average with more happiness than other food categories (although
people might dier in their general eating happiness). An additional conducted ANOVA with person-centred
happiness values as dependent variables and food categories as independent variables provided also a highly sim-
ilar pattern of results. Replicating the previously reported analysis, eating happiness diered signicantly across
all 14 food categories, F(13, 2129) = 1.94, p = 0.023, and post hoc analysis did not yield signicant dierences in
experienced eating happiness between food categories, p ≥ 0.14. Moreover, fruits and vegetables were associated
with high happiness values, and “unhealthy” food choices such as sweets did not dier in experienced happiness
Figure 2. Experienced eating happiness per meal type. Small dots represent single eating events, big circles
indicate average eating happiness, and the horizontal line indicates the grand mean. Boxes indicate the middle
50% (interquartile range) and median (darker/lighter shade). e whiskers above and below represent 1.5 of the
interquartile range.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
5
SCientifiC REPORTS | 7: 17069 | DOI:10.1038/s41598-017-17262-9
from “healthy” food choices such as fruits or vegetables. e only dierence between the previous and control
analysis was that vegetables (Mcwc = 1.16, SD = 15.14) gained slightly in importance for eating-related happiness,
whereas fruits (Mcwc = −0.65, SD = 13.21), salty extras (Mcwc = −0.07, SD = 8.01), and pastries (Mcwc = −2.39,
SD = 18.26) became slightly less important.
Discussion
is study is the rst, to our knowledge, that investigated in-the-moment experienced eating happiness in real
time and real life using EMA based self-report and imagery covering the complete diversity of food intake. e
present results add to and extend previous ndings by suggesting that fruit and vegetable consumption has imme-
diate benecial psychological eects. Overall, of 14 dierent main food categories, vegetables consumption con-
tributed the largest share to eating happiness measured across eight days. us, in addition to the investment in
future well-being indicated by previous research8, “healthy” food choices seem to be an investment in the in-the
moment well-being.
Importantly, although many cultures convey the belief that eating certain foods has a greater hedonic and
mood boosting eect, the present results suggest that this might not reect actual in-the-moment experiences
accurately. Even though people oen have a spontaneous “unhealthy = tasty” intuition13, thus indicating that a
stronger happiness boosting eect of “unhealthy” food is to be expected, the induced eating happiness of sweets
did not dier on average from “healthy” food choices such as fruits or vegetables. is was also true for other
stereotypically “unhealthy” foods such as pastries and salty extras, which did not show the expected greater boost-
ing eect on happiness. Moreover, analyses on the meal type level support this notion, since snacks, despite
their overall positive eect, were not the most psychologically benecial meal type, i.e., dinner had a comparable
“happiness” signature to snacking. Taken together, “healthy choices” seem to be also “happy choices” and at least
comparable to or even higher in their hedonic value as compared to stereotypical “unhealthy” food choices.
In general, eating happiness was high, which concurs with previous research from eld studies with generally
healthy participants. De Castro, Bellisle, and Dalix22 examined weekly food diaries from 54 French subjects and
found that most of the meals were rated as appealing. Also, the observed dierences in average eating happiness
for the 14 dierent food categories, albeit statistically signicant, were comparable small. One could argue that
this simply indicates that participants avoided selecting bad food22. Alternatively, this might suggest that the type
of food or food categories are less decisive for experienced eating happiness than oen assumed. is relates to
recent ndings in the eld of comfort and emotional eating. Many people believe that specic types of food have
greater comforting value. Also in research, the foods eaten as response to negative emotional strain, are typically
characterised as being high-caloric because such foods are assumed to provide immediate psycho-physical bene-
ts18. However, comparing dierent food types did not provide evidence for the notion that they diered in their
provided comfort; rather, eating in general led to signicant improvements in mood19. is is mirrored in the
present ndings. Comparing the eating happiness of “healthy” food choices such as fruits and vegetables to that
of “unhealthy” food choices such as sweets shows remarkably similar patterns as, on average, they were associated
with high eating happiness and their range of experiences ranged from very negative to very positive.
is raises the question of why the idea that we can eat indulgent food to compensate for life’s mishaps is so
prevailing. In an innovative experimental study, Adriaanse, Prinsen, de Witt Huberts, de Ridder, and Evers23 led
participants believe that they overate. ose who characterised themselves as emotional eaters falsely attributed
their over-consumption to negative emotions, demonstrating a “confabulation”-eect. is indicates that peo-
ple might have restricted self-knowledge and that recalled eating episodes suer from systematic recall biases24.
Moreover, Boelsma, Brink, Staeu, and Hendriks25 examined postprandial subjective wellness and objective
parameters (e.g., ghrelin, insulin, glucose) aer standardised breakfast intakes and did not nd direct corre-
lations. is suggests that the impact of dierent food categories on wellness might not be directly related to
biological eects but rather due to conditioning as food is oen paired with other positive experienced situations
(e.g., social interactions) or to placebo eects18. Moreover, experimental and eld studies indicate that not only
negative, but also positive, emotions trigger eating15,26. One may speculate that selective attention might contrib-
ute to the “myth” of comfort food19 in that people attend to the consumption eect of “comfort” food in negative
situation but neglect the eect in positive ones.
e present data also show that eating behaviour in the real world is a complex behaviour with many dierent
aspects. People make more than 200 food decisions a day27 which poses a great challenge for the measurement
of eating behaviour. Studies oen assess specic food categories such as fruit and vegetable consumption using
Food Frequency Questionnaires, which has clear advantages in terms of cost-eectiveness. However, focusing on
selective aspects of eating and food choices might provide only a selective part of the picture15,17,22. It is important
to note that focusing solely on the “unhealthy” food choices such as sweets would have led to the conclusion that
they have a high “indulgent” value. To be able to draw conclusions about which foods make people happy, the
relation of dierent food categories needs to be considered. e more comprehensive view, considering the whole
dietary behaviour across eating occasions, reveals that “healthy” food choices actually contributed the biggest
share to the total experienced eating happiness. us, for a more comprehensive understanding of how eating
behaviours are regulated, more complete and sensitive measures of the behaviour are necessary. Developments
in mobile technologies hold great promise for feasible dietary assessment based on image-assisted methods28.
As fruits and vegetables evoked high in-the-moment happiness experiences, one could speculate that these
cumulate and have spill-over eects on subsequent general well-being, including lifesatisfaction across time.
Combing in-the-moment measures with longitudinal perspectives might be a promising avenue for future studies
for understanding the pathways from eating certain food types to subjective well-being. In the literature dierent
pathways are discussed, including physiological and biochemical aspects of specic food elements or nutrients7.
e present EMA based data also revealed that eating happiness varied greatly within the 14 food categories
and meal types. As within food category variance represented more than two third of the total observed variance,
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
6
SCientifiC REPORTS | 7: 17069 | DOI:10.1038/s41598-017-17262-9
happiness varied according to nutritional characteristics and meal type; however, a myriad of factors present in
the natural environment can aect each and every meal. us, widening the “nourishment” perspective by includ-
ing how much, when, where, how long, and with whom people eat might tell us more about experienced eating
happiness. Again, mobile, in-the-moment assessment opens the possibility of assessing the behavioural signature
of eating in real life. Moreover, individual factors such as eating motives, habitual eating styles, convenience, and
social norms are likely to contribute to eating happiness variance5,29.
A key strength of this study is that it was the rst to examine experienced eating happiness in non-clinical
participants using EMA technology and imagery to assess food intake. Despite this strength, there are some
limitations to this study that aect the interpretation of the results. In the present study, eating happiness was
examined on a food based level. is neglects dierences on the individual level and might be examined in future
multilevel studies. Furthermore, as a main aim of this study was to assess real life eating behaviour, the “natural”
observation level is the meal, the psychological/ecological unit of eating30, rather than food categories or nutri-
ents. erefore, we cannot exclude that specic food categories may have had a comparably higher impact on the
experienced happiness of the whole meal. Sample size and therefore Type I and Type II error rates are of concern.
Although the total number of observations was higher than in previous studies (see for example, Boushey et al.28
for a review), the number of participants was small but comparable to previous studies in this eld20,31–33. Small
sample sizes can increase error rates because the number of persons is more decisive than the number of nested
observations34. Specially, nested data can seriously increase Type I error rates, which is rather unlikely to be the
case in the present study. Concerning Type II error rates, Aarts et al.35 illustrated for lower ICCs that adding
extra observations per participant also increases power, particularly in the lower observation range. Considering
the ICC and the number of observations per participant, one could argue that the power in the present study is
likely to be sucient to render the observed null-dierences meaningful. Finally, the predominately white and
well-educated sample does limit the degree to which the results can be generalised to the wider community; these
results warrant replication with a more representative sample.
Despite these limitations, we think that our study has implications for both theory and practice. e cumu-
lative evidence of psychological benets from healthy food choices might oer new perspectives for health pro-
motion and public-policy programs8. Making people aware of the “healthy = happy” association supported by
empirical evidence provides a distinct and novel perspective to the prevailing “unhealthy = tasty” folk intuition
and could foster eating choices that increase both in-the-moment happiness and future well-being. Furthermore,
the present research lends support to the advocated paradigm shi from “food as health” to “food as well-being”
which entails a supporting and encouraging rather constraining and limiting view on eating behaviour.
Methods
e study conformed with the Declaration of Helsinki. All study protocols were approved by University of
Konstanz’s Institutional Review Board and were conducted in accordance with guidelines and regulations. Upon
arrival, all participants signed a written informed consent.
Participants. irty-eight participants (28 females: average age = 24.47, SD = 5.88, range = 18–48 years)
from the University of Konstanz assessed their eating behaviour in close to real time and in their natural envi-
ronment using an event-based ambulatory assessment method (EMA). No participant dropped out or had to be
excluded. irty-three participants were students, with 52.6% studying psychology. As compensation, partici-
pants could choose between taking part in a lottery (4 × 25€) or receiving course credits (2 hours).
Procedure. Participants were recruited through leaflets distributed at the university and postings on
Facebook groups. Prior to participation, all participants gave written informed consent. Participants were invited
to the laboratory for individual introductory sessions. During this rst session, participants installed the applica-
tion movisensXS (version 0.8.4203) on their own smartphones and downloaded the study survey (movisensXS
Library v4065). In addition, they completed a short baseline questionnaire, including demographic variables like
age, gender, education, and eating principles. Participants were instructed to log every eating occasion immedi-
ately before eating by using the smartphone to indicate the type of meal, take pictures of the food, and describe its
main components using a free input eld. Fluid intake was not assessed. Participants were asked to record their
food intake on eight consecutive days. Aer nishing the study, participants were invited back to the laboratory
for individual nal interviews.
Measures. Immediately before eating participants were asked to indicate the type of meal with the follow-
ing ve options: breakfast, lunch, aernoon tea, dinner, snack. In Germany, “aernoon tea” is called “Kaee &
Kuchen” which directly translates as “coee & cake”. It is similar to the idea of a traditional “aernoon tea” meal
in UK. Specically, in Germany, people have “Kaee & Kuchen” in the aernoon (between 4–5 pm) and typically
coee (or tea) is served with some cake or cookies. Dinner in Germany is a main meal with mainly savoury food.
Aer each meal, participants were asked to rate their meal on three dimensions. ey rated (1) how much they
enjoyed the meal, (2) how pleased they were with their meal, and (3) how tasty their meal was. Ratings were given
on a scale of one to 100. For reliability analysis, Cronbach’s Alpha was calculated to assess the internal consist-
ency of the three items. Overall Cronbach’s alpha was calculated with α = 0.87. In addition, the average of the 38
Cronbach’s alpha scores calculated at the person level also yielded a satisfactory value with α = 0.83 (SD = 0.24).
irty-two of 38 participants showed a Cronbach’s alpha value above 0.70 (range = 0.42–0.97). An overall score
of experienced happiness of eating was computed using the average of the three questions concerning the meals’
enjoyment, pleasure, and tastiness.
Analytical procedure. e food pictures and descriptions of their main components provided by the partic-
ipants were subsequently coded by independent and trained raters. Following a standardised manual, additional
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
7
SCientifiC REPORTS | 7: 17069 | DOI:10.1038/s41598-017-17262-9
components displayed in the picture were added to the description by the raters. All consumed foods were cate-
gorised into 14 dierent food categories (see Table1) derived from the food classication system designed by the
German Nutrition Society (DGE) and based on the existing food categories of the German Nutrient Database
(Max Rubner Institut). Liquid intake and preparation method were not assessed. erefore, fats and additional
recipe ingredients were not included in further analyses, because they do not represent main elements of food
intake. Further, salty extras were added to the categorisation.
No participant dropped out or had to be excluded due to high missing rates. Missing values were below 5%
for all variables. e compliance rate at the meal level cannot be directly assessed since the numbers of meals and
snacks can vary between as well as within persons (between days). As a rough compliance estimate, the numbers
of meals that are expected from a “normative” perspective during the eight observation days can be used as a com-
parison standard (8 x breakfast, 8 × lunch, 8 × dinner = 24 meals). On average, the participants reported M = 6.3
breakfasts (SD = 2.3), M = 5.3 lunches (SD = 1.8), and M = 6.5 dinners (SD = 2.0). In comparison to the “norma-
tive” expected 24 meals, these numbers indicate a good compliance (approx. 75%) with a tendency to miss six
meals during the study period (approx. 25%). However, the “normative” expected 24 meals for the study period
might be too high since participants might also have skipped meals (e.g. breakfast). Also, the present compliance
rates are comparable to other studies. For example, Elliston et al.36 recorded 3.3 meal/snack reports per day in an
Australian adult sample and Casperson et al.37 recorded 2.2 meal reports per day in a sample of adolescents. In the
present study, on average, M = 3.4 (SD = 1.35) meals or snacks were reported per day. ese data indicate overall
a satisfactory compliance rate and did not indicate selective reporting of certain food items.
To graphically visualise data, Tableau (version 10.1) was used and for further statistical analyses, IBM SPSS
Statistics (version 24 for Windows).
Data availability. e dataset generated and analysed during the current study is available from the corre-
sponding authors on reasonable request.
References
1. Cornil, Y. & Chandon, P. Pleasure as an ally of healthy eating? Contrasting visceral and epicurean eating pleasure and their
association with portion size preferences and wellbeing. Appetite 104, 52–59 (2016).
2. Mann, T. et al. Medicare’s search for eective obesity treatments: Diets are not the answer. American Psychologist 62, 220–233 (2007).
3. van Strien, T., Herman, C. P. & Verheijden, M. W. Dietary restraint and body mass change. A 3-year follow up study in a
representative Dutch sample. Appetite 76, 44–49 (2014).
4. Bloc, L. G. et al. From nutrients to nurturance: A conceptual introduction to food well-being. Journal of Public Policy & Marketing
30, 5–13 (2011).
5. enner, B., Sproesser, G., Strohbach, S. & Schupp, H. T. Why we eat what we eat. e eating motivation survey (TEMS). Appetite 59,
117–128 (2012).
6. Conner, T. S., Brooie, . L., Carr, A. C., Mainvil, L. A. & Vissers, M. C. Let them eat fruit! The effect of fruit and vegetable
consumption on psychological well-being in young adults: A randomized controlled trial. PloS one 12, e0171206 (2017).
7. ooney, C., Mcinley, M. C. & Woodside, J. V. e potential role of fruit and vegetables in aspects of psychological well-being: a
review of the literature and future directions. Proceedings of the Nutrition Society 72, 420–432 (2013).
8. Mujcic, . & Oswald, A. J. Evolution of well-being and happiness aer increases in consumption of fruit and vegetables. American
Journal of Public Health 106, 1504–1510 (2016).
9. White, B. A., Horwath, C. C. & Conner, T. S. Many apples a day eep the blues away – Daily experiences of negative and positive
aect and food consumption in young adults. British Journal of Health Psychology 18, 782–798 (2013).
10 . Blanchower, D. G., Oswald, A. J. & Stewart-Brown, S. Is psychological well-being lined to the consumption of fruit and vegetables?
Social Indicators Research 114, 785–801 (2013).
11. Grant, N., Wardle, J. & Steptoe, A. e relationship between life satisfaction and health behavior: A Cross-cultural analysis of young
adults. International Journal of Behavioral Medicine 16, 259–268 (2009).
12 . Conner, T. S., Brooie, . L., ichardson, A. C. & Pola, M. A. On carrots and curiosity: Eating fruit and vegetables is associated with
greater ourishing in daily life. British Journal of Health Psychology 20, 413–427 (2015).
13. aghunathan, ., Naylor, . W. & Hoyer, W. D. e unhealthy = tasty intuition and its eects on taste inferences, enjoyment, and
choice of food products. Journal of Marketing 70, 170–184 (2006).
14. Evers, C., Sto, F. M. & de idder, D. T. Feeding your feelings: Emotion regulation strategies and emotional eating. Personality and
Social Psychology Bulletin 36, 792–804 (2010).
15. Sproesser, G., Schupp, H. T. & enner, B. e bright side of stress-induced eating: eating more when stressed but less when pleased.
Psychological Science 25, 58–65 (2013).
16. Wansin, B., Cheney, M. M. & Chan, N. Exploring comfort food preferences across age and gender. Physiology & Behavior 79,
739–747 (2003).
17. Taut, D., enner, B. & Baban, A. eappraise the situation but express your emotions: impact of emotion regulation strategies on ad
libitum food intae. Frontiers in Psychology 3, 359 (2012).
18. Tomiyama, J. A., Finch, L. E. & Cummings, J. . Did that brownie do its job? Stress, eating, and the biobehavioral eects of comfort
food. Emerging Trends in the Social and Behavioral Sciences: An Interdisciplinary, Searchable, and Linkable Resource (2015).
19. Wagner, H. S., Ahlstrom, B., edden, J. P., Vicers, Z. & Mann, T. e myth of comfort food. Health Psychology 33, 1552–1557
(2014).
20. Schüz, B., Bower, J. & Ferguson, S. G. Stimulus control and aect in dietary behaviours. An intensive longitudinal study. Appetite 87,
310–317 (2015).
21. Shiman, S. Conceptualizing analyses of ecological momentary assessment data. Nicotine & Tobacco Research 16, S76–S87 (2014).
22. de Castro, J. M., Bellisle, F. & Dalix, A.-M. Palatability and intae relationships in free-living humans: measurement and
characterization in the French. Physiology & Behavior 68, 271–277 (2000).
23. Adriaanse, M. A., Prinsen, S., de Witt Huberts, J. C., de idder, D. T. & Evers, C. ‘I ate too much so I must have been sad’: Emotions
as a confabulated reason for overeating. Appetite 103, 318–323 (2016).
24. obinson, E. elationships between expected, online and remembered enjoyment for food products. Appetite 74, 55–60 (2014).
25. Boelsma, E., Brin, E. J., Stafleu, A. & Hendris, H. F. Measures of postprandial wellness after single intae of two
protein–carbohydrate meals. Appetite 54, 456–464 (2010).
26. B oh, B. et al. Indulgent thining? Ecological momentary assessment of overweight and healthy-weight participants’ cognitions and
emotions. Behaviour Research and erapy 87, 196–206 (2016).
27. Wansin, B. & Sobal, J. Mindless eating: e 200 daily food decisions we overloo. Environment and Behavior 39, 106–123 (2007).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

www.nature.com/scientificreports/
8
SCientifiC REPORTS | 7: 17069 | DOI:10.1038/s41598-017-17262-9
28. Boushey, C., Spoden, M., Zhu, F., Delp, E. & err, D. New mobile methods for dietary assessment: review of image-assisted and
image-based dietary assessment methods. Proceedings of the Nutrition Society, 1–12 (2016).
29. Sto, F. M. et al. e DONE framewor: Creation, evaluation, and updating of an interdisciplinary, dynamic framewor 2.0 of
determinants of nutrition and eating. PLoS ONE 12, e0171077 (2017).
30. Pliner, P. & ozin, P. In Dimensions of the meal: e science, culture, business, and art of eating (ed H Meiselman) 19–46 (Aspen
Publishers, 2000).
31. Inauen, J., Shrout, P. E., Bolger, N., Stadler, G. & Scholz, U. Mind the gap? Anintensive longitudinal study of between-person and
within-person intention-behaviorrelations. Annals of Behavioral Medicine 50, 516–522 (2016).
32. Zepeda, L. & Deal, D. in before you eat: photographic food diaries asintervention tools to change dietary decision maing and
attitudes. InternationalJournal of Consumer Studies 32, 692–698 (2008).
33. Stein, . F. & Corte, C. M. Ecologic momentary assessment of eatingdisordered behaviors. International Journal of Eating Disorders
34, 349–360 (2003).
34. Bolger, N., Stadler, G. & Laurenceau, J. P. Power analysis for intensive longitudinal studies in Handbook of research methods for
studying daily life (ed. Mehl, M. . & Conner, T. S.) 285–301 (New Yor: e Guilford Press, 2012).
35. Aarts, E., Verhage, M., Veenvliet, J. V., Dolan, C. V. & Van Der Sluis, S. A solutionto dependency: using multilevel analysis to
accommodate nested data. Natureneuroscience 17, 491–496 (2014).
36. Elliston, . G., Ferguson, S. G., Schüz, N. & Schüz, B. Situational cues andmomentary food environment predict everyday eating
behavior in adults withoverweight and obesity. Health Psychology 36, 337–345 (2017).
37 . Casperson, S. L. et al. A mobile phone food record app to digitally capture dietary intae for adolescents in afree-living environment:
usability study. JMIR mHealth and uHealth 3, e30 (2015).
Acknowledgements
is research was supported by the Federal Ministry of Education and Research within the project SmartAct
(Grant 01EL1420A, granted to B.R. & H.S.). e funding source had no involvement in the study’s design; the
collection, analysis, and interpretation of data; the writing of the report; or the decision to submit this article for
publication. We thank Gudrun Sproesser, Helge Giese, and Angela Whale for their valuable support.
Author Contributions
B.R. & H.S. developed the study concept. All authors participated in the generation of the study design. D.W., K.V.,
L.K. & K.Z. conducted the study, including participant recruitment and data collection, under the supervision
of B.R. & H.S.; D.W. & K.V. conducted data analyses. D.W. & K.V. prepared the rst manuscript dra, and B.R. &
H.S. provided critical revisions. All authors approved the nal version of the manuscript for submission.
Additional Information
Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-017-17262-9.
Competing Interests: e authors declare that they have no competing interests.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2017
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com