- Email: [email protected]
See food diet? Cultural differences in estimating fullness and intake as a function of plate size
Accepted Manuscript See food diet? Cultural differences in estimating fullness and intake as a function of plate size Mei Peng, Sarah Adam, Michael J. Hautus, Myoungju Shin, Lisa M. Duizer, Huiquan Yan PII:
S0195-6663(17)30074-0
DOI:
10.1016/j.appet.2017.06.032
Reference:
APPET 3538
To appear in:
Appetite
Received Date: 17 January 2017 Revised Date:
30 June 2017
Accepted Date: 30 June 2017
Please cite this article as: Peng M., Adam S., Hautus M.J., Shin M., Duizer L.M. & Yan H., See food diet? Cultural differences in estimating fullness and intake as a function of plate size, Appetite (2017), doi: 10.1016/j.appet.2017.06.032. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1 2
See food diet? Cultural differences in estimating fullness and intake as a function of plate size
3 4 5
Mei Peng a*; Sarah Adam a; Michael J. Hautus b; Myoungju Shin c; Lisa M. Duizer d; Huiquan Yan e
RI PT
6 7 8 9
SC
10
a. Department of Food Science, University of Otago, Dunedin, New Zealand
12
b. School of Psychology, The University of Auckland, Auckland, New Zealand
13
c. School of Psychology, Charles Sturt University, Bathurst, Australia
14
d. Department of Food Science, University of Guelph, Guelph, Ontario, Canada
15 16
e. The School of Trade and Tourism Management, Liuzhou Vocational & Technical College, Liuzhou, China
M AN U
11
17
TE D
18 19 20 21
EP
22
24 25 26
AC C
23
27
Correspondence to:
28
Mei Peng (PhD)
29
Department of Food Science, University of Otago
30
Postal Address: PO Box 56, Dunedin, 9054, New Zealand
31
Email: [email protected]
32
Telephone: +64 3 4794052
1
ACCEPTED MANUSCRIPT Abstract
34
Previous research has suggested that manipulations of plate size can have a direct impact on
35
perception of food intake, measured by estimated fullness and intake. The present study,
36
involving 570 individuals across Canada, China, Korea, and New Zealand, is the first
37
empirical study to investigate cultural influences on perception of food portion as a function
38
of plate size. The respondents viewed photographs of ten culturally diverse dishes presented
39
on large (27cm) and small (23cm) plates, and then rated their estimated usual intake and
40
expected fullness after consuming the dish, using 100-point visual analog scales. The data
41
were analysed with a mixed-model ANCOVA controlling for individual BMI, liking and
42
familiarity of the presented food. The results showed clear cultural differences: (1)
43
manipulations of the plate size had no effect on the expected fullness or the estimated intake
44
of the Chinese and Korean respondents, as opposed to significant effects in Canadians and
45
New Zealanders (p<.05); (2) Canadian (88.91±0.42) and New Zealanders (90.37±0.41)
46
reported significantly higher estimated intake ratings than Chinese (80.80±0.38) or Korean
47
(81.69±0.44; p<.05), notwithstanding the estimated fullness ratings from the Western
48
respondents were comparable or even higher than those from the Asian respondents. Overall,
49
these findings, from a cultural perspective, support the notion that estimation of fullness and
50
intake are learned through dining experiences, and highlight the importance of considering
51
eating environments and contexts when assessing individual behaviours relating to food
52
intake.
53
Keywords: cross-country comparison; plate size effect; expected fullness; estimated intake;
54
portion-size perception
AC C
EP
TE D
M AN U
SC
RI PT
33
2
ACCEPTED MANUSCRIPT 1
Introduction
56
Eating is a multi-sensory experience that is often pleasurable. Today, as foods become more
57
diverse and accessible in many countries, identifying factors that control food intake is an
58
important goal. While knowledge is expanding about physiological mechanisms involved in
59
eating (1), research has also recognised some deep-seated perceptual mechanisms for
60
controlling food intake (2-5).
61
Decisions on portion size largely depend on perception of food portion (6, 7), which is often
62
measured by the estimated satiation (or fullness) or estimated intake of the food. By
63
definition, the expected fullness is the relative feeling of fullness to be expected after
64
consuming different foods when compared on a calorie-for-calorie basis, which is acquired
65
by a learned association between sensory properties of food and their abilities to promote
66
satiation. Estimated intake is referred to as estimated consumption volume of food (8). While
67
it has been suggested that people are extremely adept at estimating these entities (5), research
68
has also found that these estimations are highly susceptible to environmental factors (9, 10).
69
Manipulations of plate size have been shown to have a direct effect on perception of food
70
intake (8, 11-13). Often, people underestimate the quantity of food if it is presented on a large
71
plate (i.e., a low Food/Plate ratio), or conversely overestimate the food quantity on a small
72
plate (i.e., a high Food/Plate ratio). This effect is referred to as the plate-size-effect (14).
73
Although it remains unclear whether the plate-size-effect can be related to reduction of food
74
intake (15-18), this effect is widely demonstrated with estimated fullness or intake (19-22).
75
The underlying mechanism that underpins the plate-size-effect is yet to be distinguished (14).
76
Previous research has considered the Delboeuf illusion – when the surrounding context
77
influences the perception of object size – to be a potential explanation for the plate-size-effect
78
(see 11). It is postulated that people use plate size as a yardstick when estimating their food
79
intake, and in turn changes in plate size impact on perception of food portion. This theory
80
highlights a learned association between eating environments and estimations of food intake.
81
Notably, the plate-size-effect has only been investigated in Western countries. A typical
82
Western dining setting (see Figure 1) involves planning food portion size based on the
83
expected fullness elicited by the presentation of food, and serving food onto individual
84
dishware prior to the start of a meal, (5). With this dining approach, visual cues are critical
85
for portion selection, and for indicating the amount of food consumed (23). By contrast,
86
Asian countries employ a communal dining approach (24), in which individuals share a
AC C
EP
TE D
M AN U
SC
RI PT
55
3
ACCEPTED MANUSCRIPT 87
variety of dishes, with food served in bite-sized portions over the course of a meal. In this
88
case, planning for food intake is not at the beginning of a meal, instead it is a continuous task
89
over the course of a meal.
M AN U
SC
RI PT
90
Figure 1: Illustration of a typical Western (A) and Asian (B) dining setting for four people.
93
Cultural differences between the Western and Asian dining styles provide an interesting
94
outlet for testing the association between eating environments and people’s perception of
95
food portion. Specifically, the present study tests for cultural effects on using plate size as a
96
perceptual cue for estimating expected fullness and estimated intake. With the Asian dining
97
approach, plate size is thought to be a less salient perceptual cue for estimating food portion,
98
as an individual’s food intake and the plate size are not as strongly correlated as they are in
99
the Western eating environment. We hypothesise that individuals who are used to the
100
communal dining approach are less susceptible to manipulations of plate sizes, in comparison
101
to individuals who are accustomed to the Western dining approach. Findings from this study
102
are expected to provide additional insights into cultural influences on people’s ability to
103
estimate satiation and food intake, and more broadly implicate general effects of
104
environmental factors on eating behaviours.
AC C
EP
TE D
91 92
4
ACCEPTED MANUSCRIPT 105
2
Method
106
2.1
107
Respondents were recruited in medium-sized cities in two Western countries (Canada and
108
New Zealand; NZ) and two Asian countries (China and Korea), through the networks of the
109
associated researchers in each of the countries. Recruiting processes included using university
110
websites, previous participant databases, student associations, and local social media.
111
Participants contributed to this study without receiving a reimbursement.
112
A total of 570 individuals, residing in New Zealand (N=149), China (N=161), Korea (N=131)
113
and Canada (N=129), participated in this study. The respondents were informed about the
114
study and provided their consent by completing an online questionnaire. Thus, data from
115
incomplete questionnaires were eliminated. In addition, data given by respondents who lived
116
in their current country of residence for less than 20 years were excluded from the analysis.
117
This served to minimise cases where the respondent’s dining approach was not representative
118
of the country assessed. Originally, a total of 1120 participants activated the questionnaire
119
link, giving an overall completion rate of 56.3%.
120
This study complied with ethical considerations in all of the assessed countries (Otago
121
Human Ethics Committee: 16/01B; University of Guelph Research Ethics Board: 16AU005).
122
2.2
123
The online questionnaire was developed and distributed using Qualtrics© (USA, 2016). The
124
questionnaire consisted of 20 image-based questions and a series of demographic questions.
125
The online format of this type of study has been supported by previous studies, which found
126
high correlations between expected satiety and ideal portion size made from two-dimensional
127
pictures and actual food intake (25).
128 129
Table 1: Information of the ten meals presented as food images, along with the ratio of food area to plate area.
TE D
M AN U
SC
RI PT
Respondents
AC C
EP
Questionnaire design
Name
Chicken Caesar Salad Dumplings Beef Stir fry Fish Linguine Chicken Fried Rice Korean Beef Plate Lamb and Potato Meatballs and Potato Teriyaki Chicken Mushroom Risotto
Energy (kcal)
398.4 378.8 406.2 407.6 374.6 379.0 377.8 436.3 374.4 421.8
Weight (g)
346.0 168.4 321.2 257.1 262.3 337.2 301.2 337.6 265.9 148.2
5
Food/Plate Ratio 27cm Plate
23cm Plate
0.53 0.31 0.35 0.32 0.46 0.48 0.41 0.32 0.32 0.32
0.74 0.43 0.49 0.45 0.64 0.66 0.56 0.45 0.45 0.44
ACCEPTED MANUSCRIPT Note: The Food/Plate ratio was calculated as the food area divided by the entire plate area (in cm2).
131
Ten dishes were made, with either Asian or Western origin (Table 1). All dishes contained
132
energy levels within a range of 375 – 440 kcal, with balanced sources of protein, fibre and
133
carbohydrates. Each meal was placed and photographed on a 23cm-diameter plate and then
134
transferred to a 27cm-diameter plate. In order to minimise variation in the food area between
135
plates, care was taken to ensure each dish was mounted to a similar height. All photos
136
included cutlery (a knife and fork set or a pair of chopsticks, depending on the ethnic origin
137
of the dish) and a standard 355ml drink can, to serve as visual references to the size of the
138
dish. Photographs of each dish were taken under constant lighting conditions (Canon SX50
139
HS, 180 dpi). Figure 2 displays an example of a pair of photographs. For each image, the area
140
of the food and the plate was measured three times with image measurement software
141
(Klonk, USA). The averaged measures of the food area and the plate area were used to
142
compute the Food/Plate ratio (Table 1). The small plates were associated with a higher
143
Food/Plate ratio than the large plates.
AC C
144
EP
TE D
M AN U
SC
RI PT
130
145 146 147
Figure 2: An example of a photograph of the same dish on the large plate (A; 27cm diameter) and the small plate (B; 23cm diameter). Aside from the plates, sizes of other components are constant. Respondents were specifically instructed not to consider the can of drink in their responses.
148
These 20 images and their associated questions were presented to the respondents in
149
randomised orders. Consecutive presentations of the same dish were avoided by adopting a
150
block design. Accompanied with each image, the respondents were asked to respond, on a
151
100-point visual analog scale (VAS), to the question ‘How full will you be after consuming
152
the entire dish?’ with the end points anchored with ‘not full at all’ and ‘extremely full’; and
153
the question ‘As your main meal of the day, what portion of the plate would you usually
154
consume?’ with the end points anchored with ‘none’ and ‘all of the food’. Response data to 6
ACCEPTED MANUSCRIPT these two questions are referred to as “expected fullness ratings” and “estimated intake
156
ratings” in the following text. In addition, the respondents were asked to rate their liking and
157
familiarity for each of the ten meals on a VAS anchored with ‘extremely dislike / extremely
158
like’ and ‘not familiar at all / extremely familiar’, respectively.
159
The demographic questions included information about sex, age, ethnicity, current country
160
and length of residence in the county. Self-reported height and weight was also acquired for
161
computing body-mass index (BMI).
162
The questionnaire was originally developed in English and subsequently translated by the
163
associated researchers. The translated questionnaires were validated by preliminary trials
164
with bilingual respondents who completed this questionnaire twice, one-week apart, in
165
different languages (Chinese questionnaire: N=28; Korean questionnaire: N=6). Their
166
responses on the separate versions were compared. No systematic differences due to the use
167
of language were found.
168
2.3
169
Prior to the main analyses, a series of univariate analyses was performed on the demographic
170
datasets, as well as on the liking and familiarity scores, to test heterogeneity across countries.
171
Subsequently, effects of plate size on the expected fullness and estimated intake were
172
analysed separately using a mixed-model analysis of covariance (ANCOVA). Plate Size and
173
Dish type were within-subject factors, and Country was the between-subject factor. The
174
participant’s BMI, liking and familiarity scores were included in this model as continuous
175
covariates. Multiple comparison tests, with Bonferroni correction, were employed post-hoc
176
where appropriate to disentangle higher-order effects. All data analyses were performed using
177
SPSS (23.0; IBM®).
178
3
179
The results suggested that age and gender did not differ significantly across the four countries
180
(see Table 2). On average, the four countries gave similar liking and familiarity scores across
181
the ten dishes. Country differences were found for BMI, according to results from a one-way
182
ANOVA (F(3, 565)=7.90, p<.001). The averaged BMI from NZ was significantly higher than
183
the indices from China and Korea. The index of Canada was not significantly different from
184
any of the other three countries.
185
Table 2: Respondents information of the four countries included in this study.
M AN U
SC
RI PT
155
AC C
EP
TE D
Data Analysis
Results
Country
N
Age
7
Female %
BMI
ACCEPTED MANUSCRIPT 129 161 131 149
Canada China Korea New Zealand
26.7±4.8 24.8±5.4 27.6±4.9 23.8±3.8
68.2 63.4 58.8 65.8
24.00±0.47 23.18±0.25 23.45±0.30 25.22±0.42
186
The mixed-model ANCOVA on expected fullness ratings found a significant 3-way
188
interaction was obtained across Plate Size, Country and Dish (F(27,5657)=10.0, p<.001, see
189
supplementary Table 1). The 3-way interaction was in part attributable to the significant
190
interaction between Plate Size and Country (F(3,5657)=12.5, p<.001). Post-hoc tests showed
191
that the effect of Plate Size on expected fullness was only present with data from the Western
192
countries. Specifically, the small plates were associated with significantly higher expected
193
fullness ratings than the large plates, reported by the Canadian (p<.001; Cohen’s d=0.42) and
194
NZ (p=.002; Cohen’s d=0.45) respondents. For the Asian respondents, no significant
195
difference was observed between the large and small plates. On average, all four countries
196
rated similarly for the large plates, but not for the small plates (p<.05). Mean ratings for each
197
country and each plate size are presented in Table 3.
198
Notably, the covariates of liking and familiarity were shown to be significantly interacted
199
with Plate Size (liking: F(1,5657)=36.9, p<.001; familiarity: F(1,5657)=22.8, p<.001). These
200
results consolidated the significant interaction effect due to Dish × Plate Size, by confirming
201
the characteristics of dish having mediation effects. BMI, however, was not shown to play a
202
role in the Plate Size effect.
203 204 205
Table 3: Summary of the mean and standard error (SE) of expected fullness ratings and estimated intake ratings, for the four countries, for food presented on large and small plates. Different letters within the “post-hoc” column indicate that the associated means are significantly different.
EP
TE D
M AN U
SC
RI PT
187
Country
Expected Fullness Ratings Small plate postSE Mean SE hoc 0.566 A 88.07 0.539 0.554 A 86.58 0.528 0.522 A 80.78 0.498 0.585 A 82.17 0.558
AC C
Large plate Mean
Canada New Zealand China Korea
206
80.34 80.69 78.48 79.00
Estimated Intake Ratings Small plate postSE Mean SE hoc 0.577 A 89.86 0.559 0.414 A 91.29 0.421 0.563 B 79.64 0.562 0.676 B 79.79 0.642
Large plate posthoc A A B B
Mean 91.64 92.73 79.30 80.08
207
The between-subject comparison revealed significant differences in the averaged fullness
208
ratings across the assessed countries (F(3,5657)=8.6, p<.001). Post-hoc tests, based on tests of
209
simple effects, showed that Canada (M=84.2, SE=0.40) and NZ (M=83.6, SE=0.39) gave
210
significantly higher ratings for the expected fullness than either China (M=79.6, SE=0.36) or
8
posthoc A A B B
ACCEPTED MANUSCRIPT Korea (M=80.6, SE=0.41; p<.05), with no significant difference detected within the pairs of
212
Western or of Asian countries.
213
In addition, Dish was also found to have a significant main effect on the expected fullness
214
ratings (F(9, 5657)=80.2, p<.001). Averaged fullness ratings for each dish are presented in Table
215
3. Post-hoc tests were applied to assess the differences in the expected fullness across Dishes
216
(see Table 4).
217 218 219
Table 4: Descriptive statistics, including the means and standard deviations, of expected fullness and estimated intake ratings for each of the tested dishes. Means with different letters in the “post-hoc” column are statistically different. Expected Fullness Rating
Chicken Caesar Salad Dumplings Beef Stir fry Fish Linguine Chicken Fried Rice Korean Beef Plate Lamb and Potato Meatballs and Potato Teriyaki Chicken Mushroom Risotto
Mean 73.17 73.68 81.37 80.45 88.88 88.47 87.49 79.48 83.30 82.73
Post-hoc A A B B D D CD B C BC
Mean 87.10 87.78 84.66 86.47 82.82 85.67 84.01 86.72 84.23 84.98
M AN U
SE 0.61 0.61 0.62 0.61 0.61 0.62 0.62 0.61 0.61 0.61
Estimated Intake Rating
SC
Dish name
RI PT
211
SE 0.66 0.66 0.67 0.65 0.66 0.65 0.66 0.65 0.65 0.66
Post-hoc C C B C A BC AB C AB B
Results from the ANCOVA of estimated intake ratings also suggested a significant 3-way
221
interaction was obtained (F(27,5657)=4.4, p<.001; see supplementary Table 2) , which was
222
largely accounted for by the significant interaction between Plate Size and Country
223
(F(3,5657)=12.4, p<.001). Post-hoc tests of the interaction effect suggested that Plate Size had
224
significant effect on the estimated intake rating for Canada (p<.001; Cohen’s d=0.11) and NZ
225
(p<.001; Cohen’s d=0.16), notwithstanding the small effect sizes. For both countries, the
226
estimated intake was significantly higher for the large plate compared to the small plate. By
227
contrast, Plate Size showed no significant impact on the estimated intake given by either
228
Chinese or Korean respondents (p>0.5). Cross-country comparisons with respect to Plate Size
229
showed significant differences between the pairs of Asian and Western countries, as shown in
230
Table 3.
231
On average, significant differences were observed across countries (F(3,5657)=141.4, p<.001).
232
Post-hoc tests, based on tests of simple effects, revealed that NZ (M=90.37; SE=0.41) and
233
Canada (M=88.91; SE=0.42) gave significantly higher estimated intake ratings than China
234
(M=80.80; SE=0.38) and Korea (M=81.69; SE=0.44). Further post-hoc tests suggested that
AC C
EP
TE D
220
9
ACCEPTED MANUSCRIPT the estimated intake ratings also varied significantly depending on Dishes (F(9,5657)=6.2,
236
p<.001; see Table 4).
237
All three covariates of this ANCOVA model were shown to be significantly interacted with
238
Plate Size (liking: F(1,5657)=105.2, p<.001; familiarity: F(1,5657)=53.6, p<.001; BMI:
239
F(1,5657)=5.7, p=.017).
240
4
241
The present study investigated cultural differences in perception of food portion in relation to
242
plate size, using an image-based questionnaire. The findings suggested that the Western
243
respondents estimated the same amount of food to be more satiating when placed on a small
244
plate, and correspondingly estimated lower intake (11, 22). On the contrary, this perceptual
245
bias related to plate size was not evident for the Chinese or Korean respondents, supporting
246
the notion that the Asian respondents are less susceptible to the plate-size-effect.
247
While there is no reason to believe there are cultural differences in susceptibilities to the
248
Delboeuf illusion, an implication from the current results is that individuals in the Western
249
eating culture are more responsive to the Delboeuf illusion in the context of eating, relative to
250
the Asian culture. As discussed previously, the Western dining approach, involving meal
251
planning on individual plates, potentially reinforces the association between plate size and
252
food intake. On the contrary, plate sizes in the context of Asian dining provides little
253
relevance to estimations of either fullness or intake, which in turn have little effect on those
254
estimates.
255
These findings also open a broad discussion about available perceptual cues that can mediate
256
people’s food intake decisions. In the present study, estimates from the Asian respondents
257
were comparable for the large and small plates. A possible explanation to these findings is
258
that the Asian respondents estimate fullness and intake using information from a broader
259
visual context, like the whole dinning setting. More broadly, the Western and Asian dining
260
environment also differ in terms of levels of social interaction. Presumably, the communal
261
dining approach involves more interactions with the other diners. Previous research has
262
highlighted that social interactions can be an important factor in food intake regulation (17,
263
26). Further studies should assess the role of social interaction as a mechanism for intake
264
control in various cultures, in order to explain the cultural differences observed in the present
265
study.
RI PT
235
AC C
EP
TE D
M AN U
SC
Discussion
10
ACCEPTED MANUSCRIPT Another possible explanation to these cultural differences is related to BMI. According to
267
previous research (14, 27), as well as the present data, individual BMI could be another a
268
mediator for the plate-size-effect on estimated intake. It is plausible that the observed
269
differences in the plate-size-effect on estimated intake are attributable to the heterogeneous
270
BMI measures across the assessed countries. However, it should be noted that the significant
271
plate-size-effect on the estimated fullness was intendent of BMI measures, given the non-
272
significant relationship them.
273
Furthermore, the study found that the Western respondents estimated higher estimated intake
274
than did the Asian respondents, despite the fact that the former group reported higher
275
expected fullness from consuming the entire meal. These results suggested that the Western
276
respondents showed a relatively high tendency of “cleaning plate”, consistent with previous
277
studies (8). Conceivably, these contrasting responses are attributable to different cultural
278
expectations for finishing the plate. With the Western dining approach, there is a high
279
expectation for individuals to consume the self-selected food in their entirety (28). By
280
contrast, the expectation for “cleaning plate” is lessened with the communal dining approach,
281
as it would require collective rather than individual effort. These results also indicate that
282
“estimated fullness” and “estimated intake” measure different entities with regards to the
283
perception of food portion.
284
Some of the experimental factors should be considered in interpretation of the current
285
findings. Although all respondents were primarily recruited in medium-sized cities in all four
286
countries, these cities differ considerably in terms of population, demographic, and socio-
287
economic environment, relative to their respective countries. Given the current trend of food
288
globalisation, the respondents may be, to a different degree, exposed to food and eating styles
289
of another culture. Prior exposure to food and dining styles of another culture was not
290
considered specifically in this study. In addition, the caveat of the present study is that the
291
assessments were only based on the Western dining style. Future studies should investigate
292
cultural differentiations in responding to the communal dining style, in order to further
293
understand cultural impacts on portion judgements in various environments.
294
The prevalence of obesity varies considerably across regions of the world (29). While the
295
national differences in obesity rate are attributable to a variety of factors, such as genetics,
296
socio-economics, and health policies, culturally-influenced eating behaviour represents an
297
important yet relatively understudied area in the literature. The present study provides some
AC C
EP
TE D
M AN U
SC
RI PT
266
11
ACCEPTED MANUSCRIPT clear evidence for cultural differences in estimating fullness and intake based on food images.
299
These findings, from a cultural perspective, support the notion that estimation of fullness and
300
intake are learned through dining experiences, (6) and highlight the importance of
301
considering eating environments and contexts when assessing individual behaviours relating
302
to food intake.
303
5
304
The authors would like to thank William Pickup for his help with making the meals, and Patti
305
Aguilar for her assistance in data collection in Canada.
306
6
307
The authors declare no conflict of interest.
Acknowledgements
SC
Conflict of Interests
RI PT
298
AC C
EP
TE D
M AN U
308
12
ACCEPTED MANUSCRIPT 309
7
Supplementary material
310 311 312 313 314
Table 1: Summary of results from post-hoc tests (tests of simple effects with Bonferroni correction) for the 3-way interaction effect (i.e., Country, Plate Size, and Dish) on expected fullness ratings. The mean difference refers to the difference in the fullness rating given to the small and large plate. The pvalue, adjusted by Bonferroni correction, indicates a significant difference if it is less than .05. Significant p-values are in bold. Canada
Chicken Caesar Salad Dumplings Beef Stir fry Fish Linguine Chicken Fried Rice Korean Beef Plate Lamb and Potato Meatballs and Potato Teriyaki Chicken Mushroom Risotto
Mean difference 1.24 1.91 -3.33 -1.31 -0.60 -6.24 -2.84 0.36 -2.78 -5.68
NZ
SE
p-value
1.31 1.38 1.29 1.30 1.30 1.33 1.29 1.30 1.29 1.29
0.342 0.167 0.010 0.312 0.645 <0.001 0.028 0.784 0.031 <0.001
Mean difference -1.87 -1.06 -5.40 2.07 2.76 -3.22 -2.44 1.24 -1.20 -3.42
China
SE
p-value
1.02 0.99 1.02 0.98 0.99 0.99 0.99 0.98 1.00 0.98
0.066 0.237 <0.001 0.035 0.005 0.001 0.014 0.230 0.231 0.001
Mean difference 0.81 0.20 -0.85 -0.01 -1.94 0.74 0.45 0.31 -0.39 1.49
Korea
SE
p-value
1.26 1.29 1.27 1.25 1.24 1.24 1.24 1.24 1.24 1.25
0.520 0.877 0.504 0.991 0.118 0.550 0.718 0.805 0.750 0.412
Mean difference -4.12 -2.43 0.23 -1.63 0.42 0.88 1.81 0.49 0.06 -2.75
RI PT
Dish name
SE
p-value
1.88 1.81 1.81 1.88 1.79 1.82 1.93 1.78 1.81 1.83
<0.001 0.180 0.897 0.384 0.814 0.629 0.230 0.785 0.974 0.135
Note: N(Canada) = 129; N(New Zealand) = 149; N(China) = 161; N(Korea) = 131. Degrees of freedom are N-2.
316 317 318 319 320
Table 2: Summary of results from post-hoc tests (tests of simple effects with Bonferroni correction) for the 3-way interaction effect (i.e., Country, Plate Size, and Dish) on estimated intake ratings. The mean difference refers to the difference in the fullness rating given to the small and large plate. The pvalue, adjusted by Bonferroni correction, indicates a significant difference if it is less than .05. Significant p-values are in bold. Canada
Chicken Caesar Salad Dumplings Beef Stir fry Fish Linguine Chicken Fried Rice Korean Beef Plate Lamb and Potato Meatballs and Potato Teriyaki Chicken Mushroom Risotto
SE 1.45 1.53 1.43 1.44 1.44 1.48 1.43 1.44 1.43 1.43
NZ
p-value 0.436 0.024 <0.001 0.391 0.030 <0.001 0.001 <0.001 0.093 <0.001
Mean difference 8.06 6.30 0.53 10.08 18.35 5.36 0.67 3.78 4.67 0.32
SE 1.22 1.20 1.23 1.20 1.20 1.20 1.19 1.19 1.21 1.19
p-value <0.001 <0.001 0.662 <0.001 <0.001 <0.001 0.576 0.002 <0.001 0.790
Mean difference 3.36 4.42 1.06 -1.28 0.23 1.72 1.87 -1.39 -0.65 1.92
China SE 1.22 1.25 1.23 1.21 1.21 1.22 1.20 1.20 1.20 1.21
p-value 0.006 0.005 0.390 0.288 0.849 0.160 0.120 0.251 0.587 0.112
Korea Mean difference 4.56 2.54 1.95 1.19 -2.46 2.02 2.48 -2.74 4.59 1.36
EP
Note: N(Canada) = 129; N(New Zealand) = 149; N(China) = 161; N(Korea) = 131. Degrees of freedom are N-2.
AC C
321
Mean difference 1.12 3.33 9.16 1.23 3.13 12.35 4.73 8.71 -2.40 7.71
TE D
Dish name
M AN U
SC
315
13
SE 1.64 1.64 1.68 1.65 1.66 1.74 1.82 1.66 1.67 1.67
p-value 0.010 0.113 0.248 0.244 0.142 0.249 0.140 0.100 0.012 0.419
ACCEPTED MANUSCRIPT 8
References
323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370
1. Chambers Adam P, Sandoval Darleen A, Seeley Randy J. Integration of Satiety Signals by the Central Nervous System. Current Biology. 2013;23(9):R379-R88. 2. Sorensen LB, Moller P, Flint A, Martens M, Raben A. Effect of sensory perception of foods on appetite and food intake: a review of studies on humans. Int J Obes Relat Metab Disord. 2003;27(10):1152-66. 3. Sobal J, Wansink B. Kitchenscapes, tablescapes, platescapes, and foodscapes influences of microscale built environments on food intake. Environment and Behavior. 2007;39(1):124-42. 4. Brunstrom JM. The control of meal size in human subjects: a role for expected satiety, expected satiation and premeal planning. The Proceedings of the Nutrition Society. 2011;70(2):155-61. 5. Brunstrom J. Mind over platter: pre-meal planning and the control of meal size in humans. Int J Obes. 2014;38(S1):S9-S12. 6. Brunstrom J, Rogers PJ. How Many Calories Are on Our Plate? Expected Fullness, Not Liking, Determines Meal-size Selection. Obesity. 2009;17(10):1884-90. 7. Brunstrom JM, Shakeshaft NG. Measuring affective (liking) and non-affective (expected satiety) determinants of portion size and food reward. Appetite. 2009;52(1):108-14. 8. Wansink B, Painter JE, North J. Bottomless Bowls: Why Visual Cues of Portion Size May Influence Intake. Obesity Research. 2005;13(1):93-100. 9. Rolls BJ, Morris EL, Roe LS. Portion size of food affects energy intake in normalweight and overweight men and women. The American Journal of Clinical Nutrition. 2002;76(6):1207-13. 10. Wansink B. Environmental factors that increase the food intake and consumption volume of unknowing consumers. Annual Review of Nutrition. 2004;24(1):455-79. 11. Van Ittersum K, Wansink B. Plate Size and Color Suggestibility: The Delboeuf Illusion’s Bias on Serving and Eating Behavior. Journal of Consumer Research. 2012;39(2):215-28. 12. van Ittersum K, Wansink B. Do Children Really Prefer Large Portions? Visual Illusions Bias Their Estimates and Intake. Journal of the American Dietetic Association. 2007;107(7):1107-10. 13. Wansink B, Kim J. Bad Popcorn in Big Buckets: Portion Size Can Influence Intake as Much as Taste. Journal of Nutrition Education and Behavior. 2005;37(5):242-5. 14. Holden SS, Zlatevska N, Dubelaar C. Whether Smaller Plates Reduce Consumption Depends on Who’s Serving and Who’s Looking: A Meta-Analysis. Journal of the Association for Consumer Research. 2016;1(1):134-46. 15. Libotte E, Siegrist M, Bucher T. The influence of plate size on meal composition. Literature review and experiment. Appetite. 2014;82:91-6. 16. Robinson E, Nolan S, Tudur-Smith C, Boyland EJ, Harrold JA, Hardman CA, et al. Will smaller plates lead to smaller waists? A systematic review and meta-analysis of the effect that experimental manipulation of dishware size has on energy consumption. Obesity Reviews. 2014;15(10):812-21. 17. Koh J, Pliner P. The effects of degree of acquaintance, plate size, and sharing on food intake. Appetite. 2009;52(3):595-602. 18. Rolls BJ, Roe LS, Halverson KH, Meengs JS. Using a smaller plate did not reduce energy intake at meals. Appetite. 2007;49(3):652-60. 19. McClain AD, van den Bos W, Matheson D, Desai M, McClure SM, Robinson TN. Visual illusions and plate design: the effects of plate rim widths and rim coloring on perceived food portion size. Int J Obes. 2014;38(5):657-62.
AC C
EP
TE D
M AN U
SC
RI PT
322
14
ACCEPTED MANUSCRIPT
SC
RI PT
20. van Kleef E, Shimizu M, Wansink B. Serving Bowl Selection Biases the Amount of Food Served. Journal of Nutrition Education and Behavior. 2012;44(1):66-70. 21. Wansink B, van Ittersum K. Portion size me: Plate-size induced consumption norms and win-win solutions for reducing food intake and waste. Journal of Experimental Psychology: Applied. 2013;19(4):320-32. 22. Wansink B, van Ittersum K, Painter JE. Ice Cream Illusions: Bowls, Spoons, and SelfServed Portion Sizes. American Journal of Preventive Medicine. 2006;31(3):240-3. 23. Wadhera D, Capaldi-Phillips ED. A review of visual cues associated with food on food acceptance and consumption. Eating Behaviors. 2014;15(1):132-43. 24. Belk R. Sharing. Journal of Consumer Research. 2010;36(5):715-34. 25. Wilkinson LL, Hinton EC, Fay SH, Ferriday D, Rogers PJ, Brunstrom J. Computerbased assessments of expected satiety predict behavioural measures of portion-size selection and food intake. Appetite. 2012;59(3):933-8. 26. Herman CP, Roth DA, Polivy J. Effects of the Presence of Others on Food Intake: A Normative Interpretation. Psychological Bulletin. 2003;129(6):873-86. 27. Peng M. How does plate size affect estimated satiation and intake for individuals in normal-weight and overweight groups? Obesity Science & Practice. 2017:n/a-n/a. 28. Brunstrom J, Collingwood J, Rogers PJ. Perceived volume, expected satiation, and the energy content of self-selected meals. Appetite. 2010;55(1):25-9. 29. Font JC, Fabbri D, Gil J. Decomposing cross-country differences in levels of obesity and overweight: Does the social environment matter? Social Science & Medicine. 2010;70(8):1185-93.
M AN U
371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
AC C
EP
TE D
393
15
S0195-6663(17)30074-0
DOI:
10.1016/j.appet.2017.06.032
Reference:
APPET 3538
To appear in:
Appetite
Received Date: 17 January 2017 Revised Date:
30 June 2017
Accepted Date: 30 June 2017
Please cite this article as: Peng M., Adam S., Hautus M.J., Shin M., Duizer L.M. & Yan H., See food diet? Cultural differences in estimating fullness and intake as a function of plate size, Appetite (2017), doi: 10.1016/j.appet.2017.06.032. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1 2
See food diet? Cultural differences in estimating fullness and intake as a function of plate size
3 4 5
Mei Peng a*; Sarah Adam a; Michael J. Hautus b; Myoungju Shin c; Lisa M. Duizer d; Huiquan Yan e
RI PT
6 7 8 9
SC
10
a. Department of Food Science, University of Otago, Dunedin, New Zealand
12
b. School of Psychology, The University of Auckland, Auckland, New Zealand
13
c. School of Psychology, Charles Sturt University, Bathurst, Australia
14
d. Department of Food Science, University of Guelph, Guelph, Ontario, Canada
15 16
e. The School of Trade and Tourism Management, Liuzhou Vocational & Technical College, Liuzhou, China
M AN U
11
17
TE D
18 19 20 21
EP
22
24 25 26
AC C
23
27
Correspondence to:
28
Mei Peng (PhD)
29
Department of Food Science, University of Otago
30
Postal Address: PO Box 56, Dunedin, 9054, New Zealand
31
Email: [email protected]
32
Telephone: +64 3 4794052
1
ACCEPTED MANUSCRIPT Abstract
34
Previous research has suggested that manipulations of plate size can have a direct impact on
35
perception of food intake, measured by estimated fullness and intake. The present study,
36
involving 570 individuals across Canada, China, Korea, and New Zealand, is the first
37
empirical study to investigate cultural influences on perception of food portion as a function
38
of plate size. The respondents viewed photographs of ten culturally diverse dishes presented
39
on large (27cm) and small (23cm) plates, and then rated their estimated usual intake and
40
expected fullness after consuming the dish, using 100-point visual analog scales. The data
41
were analysed with a mixed-model ANCOVA controlling for individual BMI, liking and
42
familiarity of the presented food. The results showed clear cultural differences: (1)
43
manipulations of the plate size had no effect on the expected fullness or the estimated intake
44
of the Chinese and Korean respondents, as opposed to significant effects in Canadians and
45
New Zealanders (p<.05); (2) Canadian (88.91±0.42) and New Zealanders (90.37±0.41)
46
reported significantly higher estimated intake ratings than Chinese (80.80±0.38) or Korean
47
(81.69±0.44; p<.05), notwithstanding the estimated fullness ratings from the Western
48
respondents were comparable or even higher than those from the Asian respondents. Overall,
49
these findings, from a cultural perspective, support the notion that estimation of fullness and
50
intake are learned through dining experiences, and highlight the importance of considering
51
eating environments and contexts when assessing individual behaviours relating to food
52
intake.
53
Keywords: cross-country comparison; plate size effect; expected fullness; estimated intake;
54
portion-size perception
AC C
EP
TE D
M AN U
SC
RI PT
33
2
ACCEPTED MANUSCRIPT 1
Introduction
56
Eating is a multi-sensory experience that is often pleasurable. Today, as foods become more
57
diverse and accessible in many countries, identifying factors that control food intake is an
58
important goal. While knowledge is expanding about physiological mechanisms involved in
59
eating (1), research has also recognised some deep-seated perceptual mechanisms for
60
controlling food intake (2-5).
61
Decisions on portion size largely depend on perception of food portion (6, 7), which is often
62
measured by the estimated satiation (or fullness) or estimated intake of the food. By
63
definition, the expected fullness is the relative feeling of fullness to be expected after
64
consuming different foods when compared on a calorie-for-calorie basis, which is acquired
65
by a learned association between sensory properties of food and their abilities to promote
66
satiation. Estimated intake is referred to as estimated consumption volume of food (8). While
67
it has been suggested that people are extremely adept at estimating these entities (5), research
68
has also found that these estimations are highly susceptible to environmental factors (9, 10).
69
Manipulations of plate size have been shown to have a direct effect on perception of food
70
intake (8, 11-13). Often, people underestimate the quantity of food if it is presented on a large
71
plate (i.e., a low Food/Plate ratio), or conversely overestimate the food quantity on a small
72
plate (i.e., a high Food/Plate ratio). This effect is referred to as the plate-size-effect (14).
73
Although it remains unclear whether the plate-size-effect can be related to reduction of food
74
intake (15-18), this effect is widely demonstrated with estimated fullness or intake (19-22).
75
The underlying mechanism that underpins the plate-size-effect is yet to be distinguished (14).
76
Previous research has considered the Delboeuf illusion – when the surrounding context
77
influences the perception of object size – to be a potential explanation for the plate-size-effect
78
(see 11). It is postulated that people use plate size as a yardstick when estimating their food
79
intake, and in turn changes in plate size impact on perception of food portion. This theory
80
highlights a learned association between eating environments and estimations of food intake.
81
Notably, the plate-size-effect has only been investigated in Western countries. A typical
82
Western dining setting (see Figure 1) involves planning food portion size based on the
83
expected fullness elicited by the presentation of food, and serving food onto individual
84
dishware prior to the start of a meal, (5). With this dining approach, visual cues are critical
85
for portion selection, and for indicating the amount of food consumed (23). By contrast,
86
Asian countries employ a communal dining approach (24), in which individuals share a
AC C
EP
TE D
M AN U
SC
RI PT
55
3
ACCEPTED MANUSCRIPT 87
variety of dishes, with food served in bite-sized portions over the course of a meal. In this
88
case, planning for food intake is not at the beginning of a meal, instead it is a continuous task
89
over the course of a meal.
M AN U
SC
RI PT
90
Figure 1: Illustration of a typical Western (A) and Asian (B) dining setting for four people.
93
Cultural differences between the Western and Asian dining styles provide an interesting
94
outlet for testing the association between eating environments and people’s perception of
95
food portion. Specifically, the present study tests for cultural effects on using plate size as a
96
perceptual cue for estimating expected fullness and estimated intake. With the Asian dining
97
approach, plate size is thought to be a less salient perceptual cue for estimating food portion,
98
as an individual’s food intake and the plate size are not as strongly correlated as they are in
99
the Western eating environment. We hypothesise that individuals who are used to the
100
communal dining approach are less susceptible to manipulations of plate sizes, in comparison
101
to individuals who are accustomed to the Western dining approach. Findings from this study
102
are expected to provide additional insights into cultural influences on people’s ability to
103
estimate satiation and food intake, and more broadly implicate general effects of
104
environmental factors on eating behaviours.
AC C
EP
TE D
91 92
4
ACCEPTED MANUSCRIPT 105
2
Method
106
2.1
107
Respondents were recruited in medium-sized cities in two Western countries (Canada and
108
New Zealand; NZ) and two Asian countries (China and Korea), through the networks of the
109
associated researchers in each of the countries. Recruiting processes included using university
110
websites, previous participant databases, student associations, and local social media.
111
Participants contributed to this study without receiving a reimbursement.
112
A total of 570 individuals, residing in New Zealand (N=149), China (N=161), Korea (N=131)
113
and Canada (N=129), participated in this study. The respondents were informed about the
114
study and provided their consent by completing an online questionnaire. Thus, data from
115
incomplete questionnaires were eliminated. In addition, data given by respondents who lived
116
in their current country of residence for less than 20 years were excluded from the analysis.
117
This served to minimise cases where the respondent’s dining approach was not representative
118
of the country assessed. Originally, a total of 1120 participants activated the questionnaire
119
link, giving an overall completion rate of 56.3%.
120
This study complied with ethical considerations in all of the assessed countries (Otago
121
Human Ethics Committee: 16/01B; University of Guelph Research Ethics Board: 16AU005).
122
2.2
123
The online questionnaire was developed and distributed using Qualtrics© (USA, 2016). The
124
questionnaire consisted of 20 image-based questions and a series of demographic questions.
125
The online format of this type of study has been supported by previous studies, which found
126
high correlations between expected satiety and ideal portion size made from two-dimensional
127
pictures and actual food intake (25).
128 129
Table 1: Information of the ten meals presented as food images, along with the ratio of food area to plate area.
TE D
M AN U
SC
RI PT
Respondents
AC C
EP
Questionnaire design
Name
Chicken Caesar Salad Dumplings Beef Stir fry Fish Linguine Chicken Fried Rice Korean Beef Plate Lamb and Potato Meatballs and Potato Teriyaki Chicken Mushroom Risotto
Energy (kcal)
398.4 378.8 406.2 407.6 374.6 379.0 377.8 436.3 374.4 421.8
Weight (g)
346.0 168.4 321.2 257.1 262.3 337.2 301.2 337.6 265.9 148.2
5
Food/Plate Ratio 27cm Plate
23cm Plate
0.53 0.31 0.35 0.32 0.46 0.48 0.41 0.32 0.32 0.32
0.74 0.43 0.49 0.45 0.64 0.66 0.56 0.45 0.45 0.44
ACCEPTED MANUSCRIPT Note: The Food/Plate ratio was calculated as the food area divided by the entire plate area (in cm2).
131
Ten dishes were made, with either Asian or Western origin (Table 1). All dishes contained
132
energy levels within a range of 375 – 440 kcal, with balanced sources of protein, fibre and
133
carbohydrates. Each meal was placed and photographed on a 23cm-diameter plate and then
134
transferred to a 27cm-diameter plate. In order to minimise variation in the food area between
135
plates, care was taken to ensure each dish was mounted to a similar height. All photos
136
included cutlery (a knife and fork set or a pair of chopsticks, depending on the ethnic origin
137
of the dish) and a standard 355ml drink can, to serve as visual references to the size of the
138
dish. Photographs of each dish were taken under constant lighting conditions (Canon SX50
139
HS, 180 dpi). Figure 2 displays an example of a pair of photographs. For each image, the area
140
of the food and the plate was measured three times with image measurement software
141
(Klonk, USA). The averaged measures of the food area and the plate area were used to
142
compute the Food/Plate ratio (Table 1). The small plates were associated with a higher
143
Food/Plate ratio than the large plates.
AC C
144
EP
TE D
M AN U
SC
RI PT
130
145 146 147
Figure 2: An example of a photograph of the same dish on the large plate (A; 27cm diameter) and the small plate (B; 23cm diameter). Aside from the plates, sizes of other components are constant. Respondents were specifically instructed not to consider the can of drink in their responses.
148
These 20 images and their associated questions were presented to the respondents in
149
randomised orders. Consecutive presentations of the same dish were avoided by adopting a
150
block design. Accompanied with each image, the respondents were asked to respond, on a
151
100-point visual analog scale (VAS), to the question ‘How full will you be after consuming
152
the entire dish?’ with the end points anchored with ‘not full at all’ and ‘extremely full’; and
153
the question ‘As your main meal of the day, what portion of the plate would you usually
154
consume?’ with the end points anchored with ‘none’ and ‘all of the food’. Response data to 6
ACCEPTED MANUSCRIPT these two questions are referred to as “expected fullness ratings” and “estimated intake
156
ratings” in the following text. In addition, the respondents were asked to rate their liking and
157
familiarity for each of the ten meals on a VAS anchored with ‘extremely dislike / extremely
158
like’ and ‘not familiar at all / extremely familiar’, respectively.
159
The demographic questions included information about sex, age, ethnicity, current country
160
and length of residence in the county. Self-reported height and weight was also acquired for
161
computing body-mass index (BMI).
162
The questionnaire was originally developed in English and subsequently translated by the
163
associated researchers. The translated questionnaires were validated by preliminary trials
164
with bilingual respondents who completed this questionnaire twice, one-week apart, in
165
different languages (Chinese questionnaire: N=28; Korean questionnaire: N=6). Their
166
responses on the separate versions were compared. No systematic differences due to the use
167
of language were found.
168
2.3
169
Prior to the main analyses, a series of univariate analyses was performed on the demographic
170
datasets, as well as on the liking and familiarity scores, to test heterogeneity across countries.
171
Subsequently, effects of plate size on the expected fullness and estimated intake were
172
analysed separately using a mixed-model analysis of covariance (ANCOVA). Plate Size and
173
Dish type were within-subject factors, and Country was the between-subject factor. The
174
participant’s BMI, liking and familiarity scores were included in this model as continuous
175
covariates. Multiple comparison tests, with Bonferroni correction, were employed post-hoc
176
where appropriate to disentangle higher-order effects. All data analyses were performed using
177
SPSS (23.0; IBM®).
178
3
179
The results suggested that age and gender did not differ significantly across the four countries
180
(see Table 2). On average, the four countries gave similar liking and familiarity scores across
181
the ten dishes. Country differences were found for BMI, according to results from a one-way
182
ANOVA (F(3, 565)=7.90, p<.001). The averaged BMI from NZ was significantly higher than
183
the indices from China and Korea. The index of Canada was not significantly different from
184
any of the other three countries.
185
Table 2: Respondents information of the four countries included in this study.
M AN U
SC
RI PT
155
AC C
EP
TE D
Data Analysis
Results
Country
N
Age
7
Female %
BMI
ACCEPTED MANUSCRIPT 129 161 131 149
Canada China Korea New Zealand
26.7±4.8 24.8±5.4 27.6±4.9 23.8±3.8
68.2 63.4 58.8 65.8
24.00±0.47 23.18±0.25 23.45±0.30 25.22±0.42
186
The mixed-model ANCOVA on expected fullness ratings found a significant 3-way
188
interaction was obtained across Plate Size, Country and Dish (F(27,5657)=10.0, p<.001, see
189
supplementary Table 1). The 3-way interaction was in part attributable to the significant
190
interaction between Plate Size and Country (F(3,5657)=12.5, p<.001). Post-hoc tests showed
191
that the effect of Plate Size on expected fullness was only present with data from the Western
192
countries. Specifically, the small plates were associated with significantly higher expected
193
fullness ratings than the large plates, reported by the Canadian (p<.001; Cohen’s d=0.42) and
194
NZ (p=.002; Cohen’s d=0.45) respondents. For the Asian respondents, no significant
195
difference was observed between the large and small plates. On average, all four countries
196
rated similarly for the large plates, but not for the small plates (p<.05). Mean ratings for each
197
country and each plate size are presented in Table 3.
198
Notably, the covariates of liking and familiarity were shown to be significantly interacted
199
with Plate Size (liking: F(1,5657)=36.9, p<.001; familiarity: F(1,5657)=22.8, p<.001). These
200
results consolidated the significant interaction effect due to Dish × Plate Size, by confirming
201
the characteristics of dish having mediation effects. BMI, however, was not shown to play a
202
role in the Plate Size effect.
203 204 205
Table 3: Summary of the mean and standard error (SE) of expected fullness ratings and estimated intake ratings, for the four countries, for food presented on large and small plates. Different letters within the “post-hoc” column indicate that the associated means are significantly different.
EP
TE D
M AN U
SC
RI PT
187
Country
Expected Fullness Ratings Small plate postSE Mean SE hoc 0.566 A 88.07 0.539 0.554 A 86.58 0.528 0.522 A 80.78 0.498 0.585 A 82.17 0.558
AC C
Large plate Mean
Canada New Zealand China Korea
206
80.34 80.69 78.48 79.00
Estimated Intake Ratings Small plate postSE Mean SE hoc 0.577 A 89.86 0.559 0.414 A 91.29 0.421 0.563 B 79.64 0.562 0.676 B 79.79 0.642
Large plate posthoc A A B B
Mean 91.64 92.73 79.30 80.08
207
The between-subject comparison revealed significant differences in the averaged fullness
208
ratings across the assessed countries (F(3,5657)=8.6, p<.001). Post-hoc tests, based on tests of
209
simple effects, showed that Canada (M=84.2, SE=0.40) and NZ (M=83.6, SE=0.39) gave
210
significantly higher ratings for the expected fullness than either China (M=79.6, SE=0.36) or
8
posthoc A A B B
ACCEPTED MANUSCRIPT Korea (M=80.6, SE=0.41; p<.05), with no significant difference detected within the pairs of
212
Western or of Asian countries.
213
In addition, Dish was also found to have a significant main effect on the expected fullness
214
ratings (F(9, 5657)=80.2, p<.001). Averaged fullness ratings for each dish are presented in Table
215
3. Post-hoc tests were applied to assess the differences in the expected fullness across Dishes
216
(see Table 4).
217 218 219
Table 4: Descriptive statistics, including the means and standard deviations, of expected fullness and estimated intake ratings for each of the tested dishes. Means with different letters in the “post-hoc” column are statistically different. Expected Fullness Rating
Chicken Caesar Salad Dumplings Beef Stir fry Fish Linguine Chicken Fried Rice Korean Beef Plate Lamb and Potato Meatballs and Potato Teriyaki Chicken Mushroom Risotto
Mean 73.17 73.68 81.37 80.45 88.88 88.47 87.49 79.48 83.30 82.73
Post-hoc A A B B D D CD B C BC
Mean 87.10 87.78 84.66 86.47 82.82 85.67 84.01 86.72 84.23 84.98
M AN U
SE 0.61 0.61 0.62 0.61 0.61 0.62 0.62 0.61 0.61 0.61
Estimated Intake Rating
SC
Dish name
RI PT
211
SE 0.66 0.66 0.67 0.65 0.66 0.65 0.66 0.65 0.65 0.66
Post-hoc C C B C A BC AB C AB B
Results from the ANCOVA of estimated intake ratings also suggested a significant 3-way
221
interaction was obtained (F(27,5657)=4.4, p<.001; see supplementary Table 2) , which was
222
largely accounted for by the significant interaction between Plate Size and Country
223
(F(3,5657)=12.4, p<.001). Post-hoc tests of the interaction effect suggested that Plate Size had
224
significant effect on the estimated intake rating for Canada (p<.001; Cohen’s d=0.11) and NZ
225
(p<.001; Cohen’s d=0.16), notwithstanding the small effect sizes. For both countries, the
226
estimated intake was significantly higher for the large plate compared to the small plate. By
227
contrast, Plate Size showed no significant impact on the estimated intake given by either
228
Chinese or Korean respondents (p>0.5). Cross-country comparisons with respect to Plate Size
229
showed significant differences between the pairs of Asian and Western countries, as shown in
230
Table 3.
231
On average, significant differences were observed across countries (F(3,5657)=141.4, p<.001).
232
Post-hoc tests, based on tests of simple effects, revealed that NZ (M=90.37; SE=0.41) and
233
Canada (M=88.91; SE=0.42) gave significantly higher estimated intake ratings than China
234
(M=80.80; SE=0.38) and Korea (M=81.69; SE=0.44). Further post-hoc tests suggested that
AC C
EP
TE D
220
9
ACCEPTED MANUSCRIPT the estimated intake ratings also varied significantly depending on Dishes (F(9,5657)=6.2,
236
p<.001; see Table 4).
237
All three covariates of this ANCOVA model were shown to be significantly interacted with
238
Plate Size (liking: F(1,5657)=105.2, p<.001; familiarity: F(1,5657)=53.6, p<.001; BMI:
239
F(1,5657)=5.7, p=.017).
240
4
241
The present study investigated cultural differences in perception of food portion in relation to
242
plate size, using an image-based questionnaire. The findings suggested that the Western
243
respondents estimated the same amount of food to be more satiating when placed on a small
244
plate, and correspondingly estimated lower intake (11, 22). On the contrary, this perceptual
245
bias related to plate size was not evident for the Chinese or Korean respondents, supporting
246
the notion that the Asian respondents are less susceptible to the plate-size-effect.
247
While there is no reason to believe there are cultural differences in susceptibilities to the
248
Delboeuf illusion, an implication from the current results is that individuals in the Western
249
eating culture are more responsive to the Delboeuf illusion in the context of eating, relative to
250
the Asian culture. As discussed previously, the Western dining approach, involving meal
251
planning on individual plates, potentially reinforces the association between plate size and
252
food intake. On the contrary, plate sizes in the context of Asian dining provides little
253
relevance to estimations of either fullness or intake, which in turn have little effect on those
254
estimates.
255
These findings also open a broad discussion about available perceptual cues that can mediate
256
people’s food intake decisions. In the present study, estimates from the Asian respondents
257
were comparable for the large and small plates. A possible explanation to these findings is
258
that the Asian respondents estimate fullness and intake using information from a broader
259
visual context, like the whole dinning setting. More broadly, the Western and Asian dining
260
environment also differ in terms of levels of social interaction. Presumably, the communal
261
dining approach involves more interactions with the other diners. Previous research has
262
highlighted that social interactions can be an important factor in food intake regulation (17,
263
26). Further studies should assess the role of social interaction as a mechanism for intake
264
control in various cultures, in order to explain the cultural differences observed in the present
265
study.
RI PT
235
AC C
EP
TE D
M AN U
SC
Discussion
10
ACCEPTED MANUSCRIPT Another possible explanation to these cultural differences is related to BMI. According to
267
previous research (14, 27), as well as the present data, individual BMI could be another a
268
mediator for the plate-size-effect on estimated intake. It is plausible that the observed
269
differences in the plate-size-effect on estimated intake are attributable to the heterogeneous
270
BMI measures across the assessed countries. However, it should be noted that the significant
271
plate-size-effect on the estimated fullness was intendent of BMI measures, given the non-
272
significant relationship them.
273
Furthermore, the study found that the Western respondents estimated higher estimated intake
274
than did the Asian respondents, despite the fact that the former group reported higher
275
expected fullness from consuming the entire meal. These results suggested that the Western
276
respondents showed a relatively high tendency of “cleaning plate”, consistent with previous
277
studies (8). Conceivably, these contrasting responses are attributable to different cultural
278
expectations for finishing the plate. With the Western dining approach, there is a high
279
expectation for individuals to consume the self-selected food in their entirety (28). By
280
contrast, the expectation for “cleaning plate” is lessened with the communal dining approach,
281
as it would require collective rather than individual effort. These results also indicate that
282
“estimated fullness” and “estimated intake” measure different entities with regards to the
283
perception of food portion.
284
Some of the experimental factors should be considered in interpretation of the current
285
findings. Although all respondents were primarily recruited in medium-sized cities in all four
286
countries, these cities differ considerably in terms of population, demographic, and socio-
287
economic environment, relative to their respective countries. Given the current trend of food
288
globalisation, the respondents may be, to a different degree, exposed to food and eating styles
289
of another culture. Prior exposure to food and dining styles of another culture was not
290
considered specifically in this study. In addition, the caveat of the present study is that the
291
assessments were only based on the Western dining style. Future studies should investigate
292
cultural differentiations in responding to the communal dining style, in order to further
293
understand cultural impacts on portion judgements in various environments.
294
The prevalence of obesity varies considerably across regions of the world (29). While the
295
national differences in obesity rate are attributable to a variety of factors, such as genetics,
296
socio-economics, and health policies, culturally-influenced eating behaviour represents an
297
important yet relatively understudied area in the literature. The present study provides some
AC C
EP
TE D
M AN U
SC
RI PT
266
11
ACCEPTED MANUSCRIPT clear evidence for cultural differences in estimating fullness and intake based on food images.
299
These findings, from a cultural perspective, support the notion that estimation of fullness and
300
intake are learned through dining experiences, (6) and highlight the importance of
301
considering eating environments and contexts when assessing individual behaviours relating
302
to food intake.
303
5
304
The authors would like to thank William Pickup for his help with making the meals, and Patti
305
Aguilar for her assistance in data collection in Canada.
306
6
307
The authors declare no conflict of interest.
Acknowledgements
SC
Conflict of Interests
RI PT
298
AC C
EP
TE D
M AN U
308
12
ACCEPTED MANUSCRIPT 309
7
Supplementary material
310 311 312 313 314
Table 1: Summary of results from post-hoc tests (tests of simple effects with Bonferroni correction) for the 3-way interaction effect (i.e., Country, Plate Size, and Dish) on expected fullness ratings. The mean difference refers to the difference in the fullness rating given to the small and large plate. The pvalue, adjusted by Bonferroni correction, indicates a significant difference if it is less than .05. Significant p-values are in bold. Canada
Chicken Caesar Salad Dumplings Beef Stir fry Fish Linguine Chicken Fried Rice Korean Beef Plate Lamb and Potato Meatballs and Potato Teriyaki Chicken Mushroom Risotto
Mean difference 1.24 1.91 -3.33 -1.31 -0.60 -6.24 -2.84 0.36 -2.78 -5.68
NZ
SE
p-value
1.31 1.38 1.29 1.30 1.30 1.33 1.29 1.30 1.29 1.29
0.342 0.167 0.010 0.312 0.645 <0.001 0.028 0.784 0.031 <0.001
Mean difference -1.87 -1.06 -5.40 2.07 2.76 -3.22 -2.44 1.24 -1.20 -3.42
China
SE
p-value
1.02 0.99 1.02 0.98 0.99 0.99 0.99 0.98 1.00 0.98
0.066 0.237 <0.001 0.035 0.005 0.001 0.014 0.230 0.231 0.001
Mean difference 0.81 0.20 -0.85 -0.01 -1.94 0.74 0.45 0.31 -0.39 1.49
Korea
SE
p-value
1.26 1.29 1.27 1.25 1.24 1.24 1.24 1.24 1.24 1.25
0.520 0.877 0.504 0.991 0.118 0.550 0.718 0.805 0.750 0.412
Mean difference -4.12 -2.43 0.23 -1.63 0.42 0.88 1.81 0.49 0.06 -2.75
RI PT
Dish name
SE
p-value
1.88 1.81 1.81 1.88 1.79 1.82 1.93 1.78 1.81 1.83
<0.001 0.180 0.897 0.384 0.814 0.629 0.230 0.785 0.974 0.135
Note: N(Canada) = 129; N(New Zealand) = 149; N(China) = 161; N(Korea) = 131. Degrees of freedom are N-2.
316 317 318 319 320
Table 2: Summary of results from post-hoc tests (tests of simple effects with Bonferroni correction) for the 3-way interaction effect (i.e., Country, Plate Size, and Dish) on estimated intake ratings. The mean difference refers to the difference in the fullness rating given to the small and large plate. The pvalue, adjusted by Bonferroni correction, indicates a significant difference if it is less than .05. Significant p-values are in bold. Canada
Chicken Caesar Salad Dumplings Beef Stir fry Fish Linguine Chicken Fried Rice Korean Beef Plate Lamb and Potato Meatballs and Potato Teriyaki Chicken Mushroom Risotto
SE 1.45 1.53 1.43 1.44 1.44 1.48 1.43 1.44 1.43 1.43
NZ
p-value 0.436 0.024 <0.001 0.391 0.030 <0.001 0.001 <0.001 0.093 <0.001
Mean difference 8.06 6.30 0.53 10.08 18.35 5.36 0.67 3.78 4.67 0.32
SE 1.22 1.20 1.23 1.20 1.20 1.20 1.19 1.19 1.21 1.19
p-value <0.001 <0.001 0.662 <0.001 <0.001 <0.001 0.576 0.002 <0.001 0.790
Mean difference 3.36 4.42 1.06 -1.28 0.23 1.72 1.87 -1.39 -0.65 1.92
China SE 1.22 1.25 1.23 1.21 1.21 1.22 1.20 1.20 1.20 1.21
p-value 0.006 0.005 0.390 0.288 0.849 0.160 0.120 0.251 0.587 0.112
Korea Mean difference 4.56 2.54 1.95 1.19 -2.46 2.02 2.48 -2.74 4.59 1.36
EP
Note: N(Canada) = 129; N(New Zealand) = 149; N(China) = 161; N(Korea) = 131. Degrees of freedom are N-2.
AC C
321
Mean difference 1.12 3.33 9.16 1.23 3.13 12.35 4.73 8.71 -2.40 7.71
TE D
Dish name
M AN U
SC
315
13
SE 1.64 1.64 1.68 1.65 1.66 1.74 1.82 1.66 1.67 1.67
p-value 0.010 0.113 0.248 0.244 0.142 0.249 0.140 0.100 0.012 0.419
ACCEPTED MANUSCRIPT 8
References
323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370
1. Chambers Adam P, Sandoval Darleen A, Seeley Randy J. Integration of Satiety Signals by the Central Nervous System. Current Biology. 2013;23(9):R379-R88. 2. Sorensen LB, Moller P, Flint A, Martens M, Raben A. Effect of sensory perception of foods on appetite and food intake: a review of studies on humans. Int J Obes Relat Metab Disord. 2003;27(10):1152-66. 3. Sobal J, Wansink B. Kitchenscapes, tablescapes, platescapes, and foodscapes influences of microscale built environments on food intake. Environment and Behavior. 2007;39(1):124-42. 4. Brunstrom JM. The control of meal size in human subjects: a role for expected satiety, expected satiation and premeal planning. The Proceedings of the Nutrition Society. 2011;70(2):155-61. 5. Brunstrom J. Mind over platter: pre-meal planning and the control of meal size in humans. Int J Obes. 2014;38(S1):S9-S12. 6. Brunstrom J, Rogers PJ. How Many Calories Are on Our Plate? Expected Fullness, Not Liking, Determines Meal-size Selection. Obesity. 2009;17(10):1884-90. 7. Brunstrom JM, Shakeshaft NG. Measuring affective (liking) and non-affective (expected satiety) determinants of portion size and food reward. Appetite. 2009;52(1):108-14. 8. Wansink B, Painter JE, North J. Bottomless Bowls: Why Visual Cues of Portion Size May Influence Intake. Obesity Research. 2005;13(1):93-100. 9. Rolls BJ, Morris EL, Roe LS. Portion size of food affects energy intake in normalweight and overweight men and women. The American Journal of Clinical Nutrition. 2002;76(6):1207-13. 10. Wansink B. Environmental factors that increase the food intake and consumption volume of unknowing consumers. Annual Review of Nutrition. 2004;24(1):455-79. 11. Van Ittersum K, Wansink B. Plate Size and Color Suggestibility: The Delboeuf Illusion’s Bias on Serving and Eating Behavior. Journal of Consumer Research. 2012;39(2):215-28. 12. van Ittersum K, Wansink B. Do Children Really Prefer Large Portions? Visual Illusions Bias Their Estimates and Intake. Journal of the American Dietetic Association. 2007;107(7):1107-10. 13. Wansink B, Kim J. Bad Popcorn in Big Buckets: Portion Size Can Influence Intake as Much as Taste. Journal of Nutrition Education and Behavior. 2005;37(5):242-5. 14. Holden SS, Zlatevska N, Dubelaar C. Whether Smaller Plates Reduce Consumption Depends on Who’s Serving and Who’s Looking: A Meta-Analysis. Journal of the Association for Consumer Research. 2016;1(1):134-46. 15. Libotte E, Siegrist M, Bucher T. The influence of plate size on meal composition. Literature review and experiment. Appetite. 2014;82:91-6. 16. Robinson E, Nolan S, Tudur-Smith C, Boyland EJ, Harrold JA, Hardman CA, et al. Will smaller plates lead to smaller waists? A systematic review and meta-analysis of the effect that experimental manipulation of dishware size has on energy consumption. Obesity Reviews. 2014;15(10):812-21. 17. Koh J, Pliner P. The effects of degree of acquaintance, plate size, and sharing on food intake. Appetite. 2009;52(3):595-602. 18. Rolls BJ, Roe LS, Halverson KH, Meengs JS. Using a smaller plate did not reduce energy intake at meals. Appetite. 2007;49(3):652-60. 19. McClain AD, van den Bos W, Matheson D, Desai M, McClure SM, Robinson TN. Visual illusions and plate design: the effects of plate rim widths and rim coloring on perceived food portion size. Int J Obes. 2014;38(5):657-62.
AC C
EP
TE D
M AN U
SC
RI PT
322
14
ACCEPTED MANUSCRIPT
SC
RI PT
20. van Kleef E, Shimizu M, Wansink B. Serving Bowl Selection Biases the Amount of Food Served. Journal of Nutrition Education and Behavior. 2012;44(1):66-70. 21. Wansink B, van Ittersum K. Portion size me: Plate-size induced consumption norms and win-win solutions for reducing food intake and waste. Journal of Experimental Psychology: Applied. 2013;19(4):320-32. 22. Wansink B, van Ittersum K, Painter JE. Ice Cream Illusions: Bowls, Spoons, and SelfServed Portion Sizes. American Journal of Preventive Medicine. 2006;31(3):240-3. 23. Wadhera D, Capaldi-Phillips ED. A review of visual cues associated with food on food acceptance and consumption. Eating Behaviors. 2014;15(1):132-43. 24. Belk R. Sharing. Journal of Consumer Research. 2010;36(5):715-34. 25. Wilkinson LL, Hinton EC, Fay SH, Ferriday D, Rogers PJ, Brunstrom J. Computerbased assessments of expected satiety predict behavioural measures of portion-size selection and food intake. Appetite. 2012;59(3):933-8. 26. Herman CP, Roth DA, Polivy J. Effects of the Presence of Others on Food Intake: A Normative Interpretation. Psychological Bulletin. 2003;129(6):873-86. 27. Peng M. How does plate size affect estimated satiation and intake for individuals in normal-weight and overweight groups? Obesity Science & Practice. 2017:n/a-n/a. 28. Brunstrom J, Collingwood J, Rogers PJ. Perceived volume, expected satiation, and the energy content of self-selected meals. Appetite. 2010;55(1):25-9. 29. Font JC, Fabbri D, Gil J. Decomposing cross-country differences in levels of obesity and overweight: Does the social environment matter? Social Science & Medicine. 2010;70(8):1185-93.
M AN U
371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392
AC C
EP
TE D
393
15