1 How Packaging Makes us Fat: Volume Estimation Heuristics and Size Preferences (with Nailya Ordabayeva, Erasmus ) 16/5/2013 Pierre CHANDON L’Oréal Chaired Professor of Marketing, Innovation and Creativity. Director, INSEAD Social Science Research Center Overview of presentation How Packaging Makes us Fat Background Trends in package and portion sizes. The psychophysics of volume estimation. Perception of 1D vs. 3D Supersizing (Chandon & Ordabayeva 2009) Stronger underestimation of 3D vs. 1D supersizing Influences consumption and size choices. Predicting & Managing Downsizing Impressions (Ordabayeva & Chandon 2013) Additive change model predicts perceived downsizing. Additive change model can be used for “smart downsizing”. Asymmetries in judging size increases vs. decreases (ongoing work) Estimations, production of size increases, decreases. Effects of valence, reference level, expertise. Conclusions
26
Embed
How Packaging Makes us Fat: Volume Estimation Heuristics ...faculty.insead.edu/pierre-chandon/documents/Chandon volume... · How Packaging Makes us Fat: Volume Estimation Heuristics
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
How Packaging Makes us Fat: Volume Estimation Heuristics and Size Preferences
(with Nailya Ordabayeva, Erasmus )16/5/2013
Pierre CHANDON
L’Oréal Chaired Professor of Marketing, Innovation and Creativity.
Director, INSEAD Social Science Research Center
Overview of presentation
How Packaging Makes us Fat
Background Trends in package and portion sizes.
The psychophysics of volume estimation.
Perception of 1D vs. 3D Supersizing (Chandon & Ordabayeva 2009)
No systematic product differences (confounds of healthiness and size)
20
Study 3
Decoupling Size Direction and Valence
Valence manipulation (tasting)
Sugary or salty iced tea
Pleasant or disgusting JB
N = 145 participants
Magnitude estimation vs. production
Add or subtract 33%, then 67% (iced tea)
Add or subtract 45%, then 90% (JB)
Study 3
Decoupling Size Direction and Valence
?(-33%)
120ml
Iced tea Jelly Beans
Downsizing
Supersizing
?(-67%)
?(+33%)
120ml?
(+67%)
?(-45%)
105g?
(-90%)
?(+45%)
105g?
(+90%)
21
0.0
0.5
1.0
1.5
2.0
0.0 0.5 1.0 1.5 2.0
Est
imat
ed p
ort
ion
(m
ult
iple
of
refe
ren
ce p
ort
ion
)
Actual portion (multiple of reference portion)
Supersizing: EST = 1.0×(ACT^.65)
Downsizing: EST = 1.01×(ACT^1.16)
Study 3
Size Estimates
Asymmetry (bdown/bup) = 1.78
0.0
0.5
1.0
1.5
2.0
0.0 0.5 1.0 1.5 2.0
Est
imat
ed p
ort
ion
(m
ult
iple
of
refe
ren
ce p
ort
ion
)
Actual portion (multiple of reference portion)
estimation: unpleasant observed
estimation: pleasant observed
Pleasant Supersizing: EST = .98×(ACT^.70)
Pleasant Downsizing: EST = .97×(ACT^.98)
Unpleasant Supersizing: EST = 1.01×(ACT^.61)
Unpleasant Downsizing: EST = 1.01×(ACT^1.24)
Study 3
Pleasantness does not Influence Size Estimates
22
0.0
0.5
1.0
1.5
2.0
0.0 0.5 1.0 1.5 2.0
Pro
du
ced
po
rtio
n (
mu
ltip
le o
f re
fere
nce
po
rtio
n)
Actual portion (multiple of reference portion)
Supersizing: EST = 1.0×(ACT^1.04)
Downsizing: EST = .92×(ACT^.89)
Study 3
Lower Asymmetry for Size Production Task
Asymmetry (bup/bdown) = 1.17
0.0
0.5
1.0
1.5
2.0
0.0 0.5 1.0 1.5 2.0
Pro
du
ced
po
rtio
n (
mu
ltip
le o
f re
fere
nce
po
rtio
n)
Actual portion (multiple of reference portion)
dosage: unpleasant observed
dosage: pleasant observed
Unpleasant Supersizing: EST = .92×(ACT^1.12)
Pleasant Supersizing: EST = .99×(ACT^1.02)
Unpleasant Downsizing: EST = .92×(ACT^.88)
Pleasant Downsizing: EST = .87×(ACT^.77)
Study 3
Pleasantness does not Influence Size Production
23
Study 4
Expertise
136 cooking school students
Pasta:
165g (x 3): 55g or 495g
Sugar:
140g + 75% (35g) or – 75% (245g)
Study 4
Lower Asymmetry for Size Estimates of Experts
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Pre
dict
ed p
erce
ived
vol
ume
Actual volume
estimation: observed
estimation: predicted
b = .84
b = .90
Asymmetry (bdown/bup) = 1.07
No effects of location, cooking experience, math aptitude
24
Study 4
Lower Asymmetry for Size Production of Experts
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Pre
dict
ed p
erce
ived
vol
ume
Actual volume
production: observed
production: predicted
b = .88
b = .80
Asymmetry (bup/bdown) = 1.10
No effects of location, cooking experience, math aptitude
Explanations?
Attention? Drawing attention to height, width, length did not help.
Lower asymmetry for cooks, especially for magnitude production.
Future studies with training and feedback.
Estimation strategy? Multiplication is easier than fractionation.
Future studies with eye-movement analyses.
Motivation? Manipulating product valence did not help.
Approach/avoidance Approaching objects seem closer than receding ones (Lewin 1935).
Approaching noises have a higher pitch (Doppler 1942)
Approaching events seem closer than past ones (Caruso et al. 2013).
25
Summary
Key results
People underestimate size changes because they fail to compound changes across dimensions.
Not because of incorrect attention to changing dimensions.
Not because of failure to estimate each dimension.
This happens because people add instead of multiplying the changes in each dimension.
Size change perceptions are steeper and more linear for downsizing than for supersizing. Why?
Summary
This is how we can help: Provide salient size information in units that people understand and
complement with other sensory information (haptics).
To improve accuracy: linearize the volume estimation task from 3D to 2D to 1D.
To improve acceptance of downsizing: Elongate products. This works because people fail to compound the effects of two diminishing dimensions and compensate them with the steep increase in one dimension.