20% reduction in the typical portion size is tolerated and generally goes unnoticed.
When portions are cut by 30% or more, people feel deprived. Polivy and Herman
( 2006 ) suggest that psychological stress is engendered by restrained eating (de-
priving oneself of food or dieting) in the presence of food. The difficulty of
maintaining food restrictions leading to disinhibition and overconsumption may be
rooted in evolutionary ecology and psychology. Self-indulgence may have been an
adaptive behavior when food was available (Polivy et al. 2005 ; Polivy and Herman
2006 ). Therefore, one strategy is to reduce energy density of foods while increasing
volume size so that individuals do not think or feel deprived (Rolls and Barnett
2000 ; Wansink 2010 ). In these studies, psychologists and nutritionists use
self-report measures and food consumption patterns to assess stress. Rarely have
biomarkers of stress such as salivary cortisol (Putterman and Linden 2006 ) or 24-h
free-cortisol (Bedford et al. 2010 ) been employed.
Portion sizes and perception of portion size are influenced by the size and shape
of packages, serving vessels and utensils. From 1970 to 2000, the number of larger
sized packages increased tenfold (Young and Nestle 2002 ). A recent counter-trend
has been to reduce package size, for example, in 100-calories packages to control
serving size and caloric intake. However, a disincentive is that the cost per ml or g
is much higher than for the equivalent product in larger sizes (Wansink 2010 ).
When items are packaged or presented as if they are a single serving, they are
generally consumed as a single serving even though 20 oz (591 ml) of cola is 2 1/2
servings, a supersized 6 oz (170 g) blueberry muffin is 3 servings, and a large bagel
is about 4 servings of bread (Nestle 2002 ; USDA 2000 ). Supersizing restaurant
portions leads to the consumption of many servings during a single meal. For
example, Italian casual dining restaurants serve pasta portions of 2–4 cups which is
four to six USDA servings (Nestle 2002 ; USDA 2000 ; Young and Nestle 2002 ). To
accommodate these larger portions, plates and bowls have also become larger, and
this leads to an underestimation of the amount of food on the plate and greater
consumption (Wansink 2010 ).
An experiment measured the influence of different-size bowls and scoops on
portion size of self-served ice cream by nutrition faculty and graduate students in
nutritional sciences (Wansink et al.2006b). People were given a small bowl
(502 ml) or a large bowl (1005 ml) and a small (59 ml) or large scoop (89 ml).
Both influenced serving sizes. The nutrition experts with the large bowls served
themselves 31% more (127 calories) than their colleagues with the small bowls and
people with both large bowls and scoops served 57% more than those using small
bowls and scoops. The large scoop alone increased serving sizes by 14.5%. What is
most surprising is these nutritional scientists were unaware of the differences in
serving sizes. Observations at a Chinese buffet confirmed the plate-size bias. Diners
with large plates served 52% more, ate 45% more, and wasted 135% more food
than those with smaller plates (Wansink and van Ittersum 2013 ). Furthermore, even
education immediately preceding a dining event was not effective in reducing the
bias of plate size on serving size.
In contrast to Wansink and colleagues’findings of consistent effects of the
influence of serving dish size on consumption volume, a meta-analysis of 9
10 Objective and Subjective Aspects of the Drive to Eat in... 203