573440
research-article2015
HPQ0010.1177/1359105315573440Journal of Health PsychologyPalascha et al.
Article
How does thinking in Black and
White terms relate to eating
behavior and weight regain?
Journal of Health Psychology
2015, Vol. 20(5) 638–648
© The Author(s) 2015
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1359105315573440
hpq.sagepub.com
Aikaterini Palascha, Ellen van Kleef and
Hans CM van Trijp
Abstract
This study explores the role of dichotomous thinking on eating behavior and its association with restraint
eating and weight regain in a wide range of people. In a web-based survey with 241 adults, dichotomous
thinking and behavioral outcomes related to eating (restraint eating, weight regain, body mass index, dieting)
were assessed. Results showed that eating-specific dichotomous thinking (dichotomous beliefs about food
and eating) mediates the association between restraint eating and weight regain. We conclude that holding
dichotomous beliefs about food and eating may be linked to a rigid dietary restraint, which in turn impedes
people’s ability to maintain a healthy weight.
Keywords
body mass index, dichotomous thinking, overweight, restraint eating, weight regain
Introduction
Maintaining a healthy body weight is challenging, particularly considering the obesogenic
environment in which we live (Johnson et al.,
2012). Today, a large proportion of the population exercises dietary restraint in order to lose
weight. However, most of the times, weight loss
is achieved only in the short-term, and evidence
shows that in the long-term, 50 percent of this
weight is regained within 1 year (Wadden et al.,
1989). Several factors are implicated in weight
maintenance including behavior, physiology,
psychology, and the environment (Ohsiek and
Williams, 2011). In this study, we focus on a cognitive factor (dichotomous thinking) that may
associate with restraint eating and weight regain
and, in turn, with the ability to control weight.
Dichotomous thinking is defined as the tendency
to think in terms of binary oppositions such as
“good or bad,” “black or white,” “healthy or
unhealthy” (Oshio, 2009) and could be regarded
as a form of cognitive rigidity, reflecting thus a
less complex thinking style (Tiggemann, 2000).
Dichotomous thinking can manifest itself in
many different ways, from extreme appraisals
of food (“good” or “bad”) to extreme appraisals
of the diet (“on” or “off”) and the weight status
Wageningen University, The Netherlands
Corresponding author:
Aikaterini Palascha, Wageningen University, Marketing
and Consumer Behaviour Group, Hollandseweg 1,
Wageningen, The Netherlands.
Email: [email protected]
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
639
Palascha et al.
(“acceptable” or “unacceptable”) (Dove et al.,
2009). It has been shown that consumers tend to
employ simplistic thinking styles, as they categorize foods as good/bad or healthy/unhealthy
(Dean et al., 2011; Rozin et al., 1996) and make
unfounded inferences about the food attributes
based on these crude categorizations (Rozin
et al., 1996). Dichotomous thinking was found
to associate positively with restraint eating in a
non-clinical sample of males and females
(Tiggemann, 2000). It is believed that an “all or
nothing” response to minor dietary transgressions may lead to temporary abandonment of
the diet and overeating (Lethbridge et al., 2011;
Ramacciotti et al., 2008). Furthermore, a prospective study with weight regainers, maintainers, and people with healthy weight showed that
dichotomous thinking was one of the best predictors of weight regain in obesity (Byrne et al.,
2004). Together, evidence suggests that there is
a link between dichotomous thinking, restraint
eating, and weight regain, which is further
investigated in this study.
Restraint eating is often initiated as a
response to weight gain (Johnson et al., 2012).
Evidence from epidemiological studies shows
that exercising dietary restraint is a helpful
strategy to successful long-term weight maintenance (McGuire et al., 1999; Vogels et al., 2005;
Vogels and Westerterp-Plantenga, 2007; Wing
et al., 2008), but that the decrease in dietary
restraint that typically follows the achievement
of weight loss is directly associated with weight
regain (McGuire et al., 1999). The restraint theory proposes that dietary restraint may induce a
counter-regulatory response leading to disinhibited eating (Herman and Mack, 1975).
Two forms of restraint eating can be distinguished: the “rigid” and the “flexible” restraint
(Westenhoefer, 1991). The rigid form is characterized by a dichotomous, rule-based, “all-or-nothing”
approach to eating, dieting, and weight. The flexible restraint represents a more graduated approach,
in which fattening foods are part of the diet in
smaller quantities without triggering emotions of
guilt and anxiety. This difference may be a key
factor in distinguishing between restraint eaters
who are able to efficiently control their intake
(flexible restraint) from those who are prone to
overeating and weight regain (rigid restraint)
(Johnson et al., 2012). The study of Sairanen et al.
(2014) revealed that an increase in flexible
restraint and a reduction in rigid restraint during a
weight loss intervention were related to better
weight loss maintenance and well-being (Sairanen
et al., 2014). Dichotomous thinking, which seems
to play an important role in distinguishing between
the two forms of restraint eating, may impact on
the ability to achieve and maintain a desirable
weight.
Despite the potential relevance of dichotomous thinking in the food and eating domain,
the relationship between dichotomous thinking
and behavioral parameters such as dietary
restraint and weight regain has been explored in
very few studies so far and mainly within the
areas of obesity and/or eating disorders research
(Alberts et al., 2012; Byrne et al., 2003, 2004;
Dove et al., 2009; Lethbridge et al., 2011;
Lingswiler et al., 1989; Ramacciotti et al., 2008;
Seamoore et al., 2006). The relationship
between cognitive and behavioral outcomes is
expected to be more pronounced in people with
eating disorders, who are more likely to hold
dysfunctional cognitions related to food and
eating (Epstein and Meier, 1989; Teasdale et al.,
2001). This partly explains why most research
on dichotomous thinking is targeted to obese
individuals and people with eating disorders.
However, there is evidence that a wider range
of people think in dichotomous terms (Oakes,
2005; Oakes and Slotterback, 2005; Rozin
et al., 1996; Tiggemann, 2000).
In order to examine how dichotomous thinking
relates to restraint eating and weight regain in a
broader sample of people, we performed a crosssectional study in which cognitive (dichotomous
thinking) and behavioral parameters (restraint eating, weight regain, dieting) were assessed. Apart
from the focal objective of the study which is to
investigate whether restraint eating and weight
regain associate with dichotomous thinking, we
also examine whether the eating-specific dichotomous thinking (dichotomous cognitions regarding
food and dieting) mediates the association of
dichotomous thinking, as a general personality
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
640
Journal of Health Psychology 20(5)
Restraint eating
General
Eating-specific
Dichotomous thinking
Dichotomous thinking
DTEDS-General, DTI
DTEDS-Eating
Weight regain
Figure 1. Direct and mediation effects of dichotomous thinking on restraint eating and weight regain
(research framework).
trait, with restraint eating and weight regain, as
well as the association between restraint eating
and weight regain. The research framework is
reflected in Figure 1. We expect that dichotomous
thinking, both as a general personality trait but
also as the dichotomous cognitions regarding food
and dieting, positively relates to restraint eating
and weight regain. Moreover, we hypothesize that
the relationship of general dichotomous thinking
with restraint eating and weight regain is mediated
by the eating-specific dichotomous thinking, in
that people who think in dichotomous terms in
general tend to exercise restraint eating to a greater
extent and regain more weight after weight loss
particularly because they tend to hold more
dichotomous beliefs about food and dieting.
Eating-specific dichotomous thinking is further
expected to mediate the relationship between
restrained eating and weight regain in a recursive
way. People who exercise restraint eating to a
greater extent are expected to regain more weight
after weight loss because they tend to hold more
dichotomous beliefs about food and dieting.
Weight regain may then lead to higher levels of
restraint eating in a compensatory way. This leads
to the following hypotheses:
H1: Dichotomous thinking (general and
eating-specific) is positively associated with restraint eating.
H2:
Dichotomous thinking (general and
eating-specific) is positively associated with weight regain.
H3: Eating-specific dichotomous thinking
mediates the association between general dichotomous thinking and restraint
eating.
H4:
Eating-specific dichotomous thinking
mediates the association between general dichotomous thinking with
weight regain.
H5:
Eating-specific dichotomous thinking
mediates the association between
restraint eating and weight regain.
Methods
Participants
A total of 328 Dutch-speaking adults agreed to
take part in this study. Of those, 241 completed
the entire survey (73.4%). The sample was
recruited through a mailing list of Wageningen
University, which included primarily people
who have given their consent to be invited in
future studies. In order to reach the wider population, the survey was also posted in social
media (Facebook and Twitter accounts of the
authors). No specific inclusion and exclusion
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
641
Palascha et al.
criteria were used in this study to ensure higher
representativeness of the Dutch population,
with the exception of the language criterion
(only Dutch-speaking individuals were
included) that was used to minimize cultural
variations with respect to understanding and
responding to questions. Participants were
naive to the actual purpose of the study. They
were only told that the survey is part of an MSc
thesis regarding eating behavior. No measures
were taken in order to prevent people filling in
the survey more than once. However, given the
length of the survey (approximately 20 minutes), such likelihood seems rather small.
Study design
Data were collected by means of an online survey. Only those respondents who gave their
informed consent participated in the study.
Participants were asked to respond to a set of
independent self-reports. The order of dichotomous thinking and restraint eating measures
was randomized and preceded the weight regain
and control measures.
Measures
Dichotomous Thinking in Eating Disorders
Scale. The Dichotomous Thinking in Eating
Disorders Scale (DTEDS) is a validated 11-item
self-report, comprising two subscales: the Eating subscale (DTEDS-Eating) and the General
subscale (DTEDS-General) (Byrne et al.,
2004). Questions such as “I view my attempts
to diet as either successes or failures” and “I
think of food as either good or bad” are included
in the eating subscale, while the general subscale contains questions such as “I think of
myself as either good or bad” and “I think of
things in ‘black and white’ terms.” Questions
are administered as 4-point Likert scales
(1 = “not at all true for me” and 4 = “very true
for me”). Internal consistency, as measured in
this study, was very good for the total measure
(α = .88), as well as for both of its subscales
(Eating subscale: α = .80, General subscale:
α = .84). Responses on individual questions
were averaged, and final scores ranged from 1
to 4, with higher scores indicating higher levels
of dichotomous thinking.
Dichotomous Thinking Inventory. The dichotomous
Thinking Inventory (DTI) is a 15-item self-report
scale which assesses dichotomous thinking in a
more general context (Oshio, 2009). Questions
such as “It works out best when even ambiguous
things are made clear-cut” and “All questions
have either a right answer or a wrong answer” are
assessed and administered with a 6-point Likert
scale (1 = “disagree strongly” and 6 = “agree
strongly”). The questionnaire consists of three
constructs, namely, “Preference for dichotomy”
(α = .76), “Dichotomous beliefs” (α = .82), and
“Profit-and-loss thinking” (α = .75). The last construct (Profit-and-loss thinking) was not included
in this study because it was considered irrelevant
in the specific context. Higher average scores,
ranging from 1 to 6, indicate higher dichotomous
thinking. The inclusion of two measures of
dichotomous thinking in this study allows for
direct comparison of those measures.
Measures of eating behavior and weight status. Respondents filled out the Restraint Eating
subscale of the Dutch Eating Behavior Questionnaire (DEBQ) (α = .93). Based on the DEBQ
cut-off points, each respondent was categorized
in one of the following categories: “Very low,”
“Low,” “Below average,” “Above average,”
“High,” “Very high” (Van Strien et al., 1986).
Moreover, respondents reported whether they
had lost weight during the last 5 years (yes/no)
and whether they had regained some or all of
this weight afterward. Those who lost weight in
the past 5 years and regained 4 kg or more were
classified as “weight regainers,” while those
who regained less than 4 kg were classified as
“weight maintainers” (Manson et al., 1995).
People who did not lose weight in the past
5 years were not included in this categorization.
Voluntary and involuntary weight loss was not
distinguished in this study.
Control variables. Finally, a set of control variables was measured. Those included age, sex,
education level, self-reported weight (kg), and
height (cm) in order to calculate body mass
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
642
Journal of Health Psychology 20(5)
index (BMI) and dieting (“Are you currently on
a diet?”) (yes/no).
Statistical analysis
Data were analyzed with SPSS 20. Correlation
coefficients between all measures were calculated using Pearson’s r. For dichotomous variables, Phi coefficients were estimated instead of
Pearson’s r. Linear or logistic regression analyses were further performed depending on the
type of dependent variable (continuous vs
dichotomous). Simple and adjusted models were
produced to take into account the influence of
control variables (sex, age, educational level,
BMI, and dieting). To test mediation, the method
of Baron and Kenny was used (Baron and
Kenny, 1986). Significance of mediation was
tested with Sobel’s test (Preacher and Hayes,
2004). Independent samples t-test was used to
test the differences in mean dichotomous thinking scores between dieters and non-dieters.
but no significant association with the measures
of general dichotomous thinking was observed.
A positive correlation with DTEDS-Eating
(r = .315, p < .001) was evident though. DTEDSEating positively correlated with all other variables measured in the study. In contrast, general
dichotomous thinking correlated positively
only with restraint eating and DTEDS-Eating.
Dichotomous thinking and restraint
eating
All measures of dichotomous thinking significantly predicted restraint eating. These positive
associations persisted in the presence of control variables (sex, age, BMI, educational
level, and dieting). Specifically, for each 1-unit
increase in dichotomous thinking, as measured
by DTEDS-General, DTI, and DTEDS-Eating,
restraint eating increased by .386, .209, and
.533, respectively.
Dichotomous thinking and weight
regain
Results1
A mixed population of 49 males and 192 females
was finally assessed. Participants aged 15–
74 years (32.3 ± 15.3), with mean BMI (23 ± 4.5)
falling into the normal range and mean restraint
eating scores (2.5 ± 0.8) lying below the average
based on norms for non-obese people (2.06–
2.70) (Van Strien et al., 1986). In addition,
64.7 percent of the sample was classified as
highly educated, 28.2 percent as middle educated, and 7.1 percent as low educated. As
regards dieting, 34 percent of the participants
were currently on a diet and 22.8 percent reported
to have regained weight after weight loss in the
past 5 years.
DTEDS-Eating significantly predicted weight
regain (β = .885, p = .003). For each 1-unit
increase in DTEDS-Eating, there was a 142.4 percent increase in the odds of being weight regainer
compared to being weight maintainer. DTEDSEating was only marginally associated with
weight regain (β = .586, p = .067) in the presence
of control variables (sex, age, BMI, educational
level, and dieting). Associations between general
dichotomous thinking measures and weight
regain were not significant; therefore no mediation analysis was performed.
Mediation effect of eatingspecific dichotomous thinking on
Correlations between cognitive and
the association between general
behavioral measures
dichotomous thinking and restraint
Restraint eating positively correlated with
eating
weight regain (r = .256, p < .001), dieting
(r = .495, p < .001), BMI (r = .262, p < .001), and
all measures of dichotomous thinking (Table 1).
Weight regain similarly correlated with dieting
(r = .364, p < .001) and BMI (r = .385, p < .001),
Eating-specific dichotomous thinking partially
mediated the relationship between DTEDS-G
and restraint eating, and fully mediated the relationship between DTI and restraint eating.
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
643
Palascha et al.
Table 1. Matrix of bivariate correlation coefficients.
Restraint
eating
Restraint eating
BMI
Dieting
Weight regain
DTEDS-General
DTI
DTEDS-Eating
1
.262**
.495**
.256*
.263**
.201**
.600**
BMI
1
.416**
.385**
.057
.066
.260**
Dieting
Weight
regain
DTEDSGeneral
DTI
DTEDSEating
1
.363**,1
<.001
−.012
.393**
1
.082
−.027
.315**
1
.464**
.613**
1
.312**
1
BMI: body mass index; DTEDS-General: Dichotomous Thinking in Eating Disorders Scale–General subscale; DTEDSEating: Dichotomous Thinking in Eating Disorders Scale–Eating subscale; DTI: Dichotomous Thinking Inventory.
Weight regain was a dichotomous variable with weight regainers defined as those who lost weight in the past 5 years
and regained 4 kg or more and weight maintainers defined as those who lost weight in the past 5 years and regained less
than 4 kg.
1Phi coefficient is reported instead of Pearson’s r for correlation between dichotomous variables.
*p < .05; **p < .001.
β=.738*
DTEDS-Eating
β=.294*
β=.75*
β=.635*
DTEDS-General
β=.337* (Adjusted model: β=-.216*)
DTI
β=.202* (Adjusted model: β=.15)
Restraint eating
Figure 2. Regression coefficients for mediation analysis (eating-specific dichotomous thinking mediating
the association between general dichotomous thinking and restraint eating).
Specifically, both measures of general dichotomous thinking were significantly associated with
restraint eating (DTEDS-General: β = .337,
p < .001, DTI: β = .202, p = .002) (first assumption) (Figure 2). DTEDS-Eating was significantly associated with both DTEDS-General
(β = .738, p < .001) and DTI (β = .294, p < .001)
(second assumption), as well as with restraint
eating in both adjusted models (DTEDS-General:
β = .75, p < .001, DTI: β = .635, p < .001) (third
assumption). Finally, the coefficients of both
DTEDS-General and DTI were lower in the
adjusted models compared to the simple models
(DTEDS-General: β = −.216, p = .01 vs β = .337,
p < .001, DTI: β = .15, p = .781 vs β = .202,
p = .002) (fourth assumption). A full mediation
effect was observed in DTI, since its association
with restraint eating was not evident anymore in
the adjusted model. The mediation effects, as
tested with Sobel’s tests, were significant in both
cases (DTEDS-General: confidence interval (CI)
[0.41, 0.72]; DTI: CI [0.11, 0.28]).
Mediation effect of eating-specific
dichotomous thinking on the
association between restraint eating
and weight regain
Eating-specific dichotomous thinking fully
mediated the relationship between restraint eating and weight regain. Specifically, restraint eating was significantly associated with weight
regain (β = .671, p = .013) (first assumption)
(Figure 3). DTEDS-Eating was significantly
associated with restraint eating (β = .631, p < .001)
(second assumption), as well as with weight
regain in the adjusted model (β = .724, p = .041)
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
644
Journal of Health Psychology 20(5)
DTEDS-Eating
β=.724*
β=.631*
β=.671* (Adjusted model: β=-.266)
Restraint eating
Weight regain
Figure 3. Regression coefficients for mediation analysis (eating-specific dichotomous thinking mediating
the association between restraint eating and weight regain).
(third assumption). Finally, the association of
restrain eating with weight regain was not evident anymore in the adjusted model compared to
the simple model (β = −.266, p = .423 vs β = .671,
p = .013) (fourth assumption). The mediation
effect, as tested with Sobel’s test, was significant
(CI [.03, .95]).
Additionally, an independent samples t-test
showed that weight regainers (n = 55) compared
to weight maintainers (n = 45) had higher scores
on restraint eating (t(98) = −2.619, p = .01, mean
difference = −.42, standard error of the mean
(SEM) = .16) and DTEDS-Eating (t(98) = −3.283,
p = .001, mean difference = −.51, SEM = .16).
Eating-specific dichotomous thinking,
BMI and dieting
Regression analysis showed that BMI significantly predicted DTEDS-Eating (β = .043,
p < .001). Specifically, for each 1-unit increase
in BMI, there was an increase in DTEDS-Eating
by 0.043. This association only marginally persisted in the presence of confounders (sex, age,
educational level, restraint eating, and dieting)
(β = .019, p = .054). Furthermore, an independent samples t-test showed that DTEDS-Eating
was significantly higher among those who were
currently dieting (2.3 ± .09), compared to nondieters (1.7 ± .05) (t(239) = −6.119, p < .001).
Discussion
This study investigated how a cognitive factor
(dichotomous thinking) relates to eating behavior
factors such as dietary restraint and weight regain.
A special focus was the role of eating-specific
dichotomous thinking in mediating those relationships and explaining whether dichotomous
thinking could have an impact on the ability to
maintain a healthy weight.
Increasing levels of dichotomous thinking as
a general personality trait was related to higher
levels of dietary restraint in our sample, confirming our initial hypothesis (H1). This finding
was quite robust, since both measures of general dichotomous thinking (DTEDS-General
and DTI) were positively associated with
restraint eating. The study of Tiggemann (2000)
similarly showed that dichotomous thinking,
measured by the Dysfunctional Attitudes Scale
(DAS), was positively related to restraint eating. Furthermore, higher levels of eatingspecific dichotomous thinking were related to
higher levels of restraint eating in our study
(H1). When we explored these relationships in
more detail, we found that the eating-specific
rather than the general dichotomous thinking
was mainly responsible for the high levels of
dietary restraint (H3 was confirmed).
In regard to weight regain, the positive association with eating-specific dichotomous thinking was only marginally significant in the
presence of control variables (age, sex, BMI,
education level, dieting). Besides, no association with the measures of general dichotomous
thinking was evident. These findings were
rather unexpected (H2 and H4 not confirmed).
The study of Byrne et al. (2004) revealed that
dichotomous thinking (and particularly the general rather than the eating-specific trait) was
one of the best prospective predictors of weight
regain in obese people. However, in this cohort
study, weight regain was measured at 1-year
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
645
Palascha et al.
follow-up after the weight loss intervention. It
is likely that the effect of dichotomous thinking
on weight regain diminishes for longer time
intervals (e.g. 4 years follow-up). Dove et al.
(2009) found that dichotomous thinking was
not related to short-term weight loss in a sample
of 76 obese and overweight women who participated in a weight management program for
12 weeks. The authors suggest that dichotomous thinking may have a different effect
among people. In some individuals, it may lead
to overeating (and thus weight regain), but in
others, it may translate to a focused and determined approach to dieting that is sustainable for
at least a short period of time.
In our study, a strong positive association
between restraint eating and weight regain was
observed. This implies that an increase in
restraint eating is accompanied by an increase
in weight regain and vice versa. It has been previously proposed that restraint eating may have
an inverse effect by leading to disinhibited eating, when restraint eaters have the feeling that
they have violated their diets (Herman and
Mack, 1975). According to our findings, this
counter-regulation may have long-term implications and associate with weight regain.
Moreover, the study of Nguyen and Polivy
(2014), in which data from 17 studies were reanalyzed, revealed that individuals who consider themselves as successful dieters were
more likely to be unrestrained eaters than
restraint eaters (Nguyen and Polivy, 2014).
However, other studies have shown that an
increase in restraint eating is associated with a
reduction in weight regain (McGuire et al.,
1999; Vogels et al., 2005; Vogels and WesterterpPlantenga, 2007; Wing et al., 2008), suggesting,
thus, that dietary restraint is a helpful strategy to
weight loss maintenance.
Further analysis of the data showed that in
line with our hypothesis (H5), eating-specific
dichotomous thinking mediates the association
between restraint eating and weight regain. This
implies that an increase in restraint eating may
relate to more weight regain due to the tendency
of restraint eaters to think about food and dieting in dichotomous terms. This “all or nothing”
thinking style may be the reason why people
fail to adhere to their diets and regain weight in
the long-term.
Eating-specific dichotomous thinking was
finally found to relate to BMI, with people who
have higher BMI reporting higher dichotomous
thinking scores, providing thus support to findings from Dove et al. (2009). Cultural discourses
about moral aspects of eating should be taken
into account when interpreting this finding.
People make moral judgments about the eating
behavior and weight status of other people but
also for themselves (Delaney and McCarthy,
2014). Thus, for overweight and obese people
who demonstrate “immoral” consumption practices more frequently, these moral judgments of
“good” and “bad” may be more prevalent.
Finally, we found that current dieters had higher
scores on eating-specific dichotomous thinking
compared to non-dieters. Tiggemann (2000)
also found that current and not past dieters have
more dysfunctional cognitive attitudes (including dichotomous thinking).
The study findings imply that people who
are inclined to think in black and white terms in
general aspects of their lives, also tend to
restrain their food intake, and the dichotomous
beliefs about food and dieting play a key role in
this effect. Therefore, we could argue that rigid
beliefs about food and dieting (e.g. “I think of
food as either good or bad” or “I view my
attempts to diet as either successes or failures”)
is what underpins restraint eating behavior.
Taking into account the strong positive association between restraint eating and weight
regain, we could also argue that dichotomous
thinking and its relevance to restraint eating
may be risky not only for people with eating
disorders but also for everyone who is trying to
restrain in order to maintain their weight. We
showed that it was the dichotomous beliefs
about food and dieting that predicted weight
regain, rather than the restraint eating behavior
per se. This shows that a rigid form of dietary
restraint that encompasses a cognitive dysfunctionality (in this case, dichotomous thinking)
may impede people’s ability to control their
intake and lead to weight regain. In line with
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
646
Journal of Health Psychology 20(5)
this observation, Sairanen et al. (2014) showed
that adopting a more flexible restraint eating
behavior (as opposed to a rigid restraint eating)
could lead to a better weight loss maintenance
and well-being.
The main limitation of the study was the use of
a convenience sampling method, which resulted
in a non-balanced sample with females being
over-represented in the study population.
Moreover, the cross-sectional design of the study
does not allow us to prove causality of the
observed associations. Intervention studies are
needed to establish causality. This cross-sectional
design further brings an uncertainty regarding the
direction of the observed associations. For example, the longitudinal association between restraint
eating and weight regain observed in this study
could be explained by either restraint eating leading to weight regain or weight regain leading to
restrained eating. Another limitation was the reliance on self-reports and the risk of biased
responding (midpoint, extreme, or socially desirable responding), which could have an influence
on the study findings. Furthermore, in this study,
we did not make a distinction between voluntary
and involuntary weight loss (for reasons other
than dieting). These two segments of people
could have different mindsets regarding dieting
and eating in general, which should be taken into
account in future studies. Also, the profile of people who gave up filling in the survey could not be
assessed, since demographic data were asked at
the end of the questionnaire. Therefore, we cannot ascertain whether the non-response was associated with other variables and what influence
this could have on our results. Finally, the likelihood of duplicate responses in the data can not be
ruled out.
The present study adds to the existing scientific literature related to the role of dysfunctional cognitive styles on eating behavior, and
particularly to the cognitive factors that may
impede or enhance people’s ability to control
their eating. While outcomes of previous studies on dichotomous thinking can be generalized
only to obese people or people with eating disorders, our findings have wider implications
concerning a more general population.
Conclusion
In summary, our results suggest that dichotomous thinking, and particularly the dichotomous cognitions related to food and dieting,
may have an adverse effect on restraint eating
behavior with possible implications on weight
loss maintenance in the long-term. Such simplified and dysfunctional thinking styles should be
avoided, since they have the potential to induce
a rigid response to dietary transitions, and therefore impede people’s ability to maintain a
healthy body weight. Implications of these findings concern the general population and not
only obese people, or people with eating
disorders.
Acknowledgements
The authors wish to thank Dr Gerry Jager who gave
feedback on the thesis report that framed this article. All
authors provided feedback on drafts of this article, and
read and approved the final version of this article.
Funding
No external funding was received by the authors to
conduct the study. This paper was based on results
from the thesis that was made in the framework of
implementation of a postgraduate program (MSc),
which has been co-financed through the Action “State
Scholarships Foundation’s Grants Program following
a procedure of individualized evaluation for the academic year 2012–2013”, from resources of the operational program “Education and Lifelong Learning” of
the European Social Fund and the National Strategic
Reference Framework 2007–2013.
Note
1.
Tables including demographic data of the study
population and statistics on the regression analyses performed in the study are available online
as supplementary material.
References
Alberts HJ, Thewissen R and Raes L (2012) Dealing
with problematic eating behaviour. The effects
of a mindfulness-based intervention on eating
behaviour, food cravings, dichotomous thinking and body image concern. Appetite 58(3):
847–851. DOI:10.1016/j.appet. 2012.01.009.
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
647
Palascha et al.
Atsushi O (2009) Development and validation of the
dichotomous thinking inventory. Social Behavior
and Personality: An International Journal 37(6):
729–741. DOI:10.2224/sbp. 2009.37.6.729.
Baron RM and Kenny DA (1986) The moderator
mediator variable distinction in social psychological-research—Conceptual, strategic, and
statistical considerations. Journal of Personality
and Social Psychology 51(6): 1173–1182.
DOI:10.1037/0022-3514.51.6.1173.
Byrne S, Cooper Z and Fairburn C (2003) Weight
maintenance and relapse in obesity: A qualitative study. International Journal of Obesity
27(8): 955–962.
Byrne SM, Cooper Z and Fairburn CG (2004)
Psychological predictors of weight regain in
obesity. Behaviour Research and Therapy
42(11): 1341–1356.
Dean M, Lähteenmäki L and Shepherd R (2011)
Nutrition communication: Consumer perceptions and predicting intentions. Proceedings of
the Nutrition Society 70(1): 19–25.
Delaney M and McCarthy MB (2014) Saints, sinners
and non-believers: The moral space of food. A
qualitative exploration of beliefs and perspectives on healthy eating of Irish adults aged
50–70. Appetite 73: 105–113. DOI:10.1016/j.
appet.2013.10.017.
Dove ER, Byrne SM and Bruce NW (2009) Effect
of dichotomous thinking on the association
of depression with BMI and weight change
among obese females. Behaviour Research
and Therapy 47(6): 529–534. DOI:10.1016/j.
brat.2009.02.013.
Epstein S and Meier P (1989) Constructive thinking: A broad coping variable with specific
components. Journal of Personality and Social
Psychology 57(2): 332–350.
Herman CP and Mack D (1975) Restrained and unrestrained eating. Journal of Personality 43(4):
647–660.
Johnson F, Pratt M and Wardle J (2012) Dietary
restraint and self-regulation in eating behavior. International Journal of Obesity 36(5):
665–674. DOI:10.1038/ijo.2011.156.
Lethbridge J, Watson HJ, Egan SJ, et al. (2011) The
role of perfectionism, dichotomous thinking,
shape and weight overvaluation, and conditional goal setting in eating disorders. Eating
Behaviors 12(3): 200–206.
Lingswiler VM, Crowther JH and Stephens MAP
(1989) Affective and cognitive antecedents
to eating episodes in bulimia and binge eating. International Journal of Eating Disorders
8(5): 533–539. DOI:10.1002/1098-108X(198909)
8:5<533::AID-EAT2260080505>3.0.CO;2-O.
McGuire MT, Wing RR, Klem ML, et al. (1999)
What predicts weight regain in a group of successful weight losers? Journal of Consulting
and Clinical Psychology 67(2): 177–185.
Manson JE, Willett WC, Stampfer MJ, et al. (1995)
Body weight and mortality among women.
New England Journal of Medicine 333(11):
677–685.
Nguyen C and Polivy J (2014) Eating behavior,
restraint status, and BMI of individuals high and
low in perceived self-regulatory success. Appetite
75: 49–53. DOI:10.1016/j.appet.2013.12.016.
Oakes ME (2005) Stereotypical thinking about foods
and perceived capacity to promote weight gain.
Appetite 44(3): 317–324.
Oakes ME and Slotterback CS (2005) Too good to be
true: Dose insensitivity and stereotypical thinking of foods’ capacity to promote weight gain.
Food Quality and Preference 16(8): 675–681.
Ohsiek S and Williams M (2011) Psychological factors
influencing weight loss maintenance: An integrative literature review. Journal of the American
Academy of Nurse Practitioners 23(11): 592–
601. DOI:10.1111/j.1745-7599.2011.00647.x.
Preacher KJ and Hayes AF (2004) SPSS and SAS
procedures for estimating indirect effects in
simple mediation models. Behavior Research
Methods, Instruments, & Computers 36(4):
717–731. DOI:10.3758/BF03206553.
Ramacciotti CE, Coli E, Bondi E, et al. (2008)
Shared psychopathology in obese subjects with
and without binge-eating disorder. International
Journal of Eating Disorders 41(7): 643–649.
DOI:10.1002/eat.20544.
Rozin P, Ashmore M and Markwith M (1996) Lay
American conceptions of nutrition: Dose insensitivity, categorical thinking, contagion, and
the monotonic mind. Health Psychology 15(6):
438–447.
Sairanen E, Lappalainen R, Lapveteläinen A, et al.
(2014) Flexibility in weight management.
Eating Behaviors 15(2): 218–224. Available at:
http://dx.doi.org/10.1016/j.eatbeh.2014.01.008
Seamoore D, Buckroyd J and Stott D (2006) Changes
in eating behaviour following group therapy for
women who binge eat: A pilot study. Journal of
Psychiatric and Mental Health Nursing 13(3): 337–
346. DOI:10.1111/j.1365-2850.2006.00960.x.
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015
648
Journal of Health Psychology 20(5)
Teasdale JD, Scott J, Moore RG, et al. (2001) How
does cognitive therapy prevent relapse in residual depression? Evidence from a controlled trial.
Journal of Consulting and Clinical Psychology
69(3): 347–357.
Tiggemann M (2000) Dieting and cognitive style.
The role of current and past dieting behaviour
and cognitions. Journal of Health Psychology
5(1): 17–24.
Van Strien T, Frijters JER, Bergers GPA, et al. (1986)
The Dutch Eating Behavior Questionnaire
(DEBQ) for assessment of restrained, emotional,
and external eating behavior. International
Journal of Eating Disorders 5(2): 295–315.
DOI:10.1002/1098-108X(198602)5:2<295::AIDEAT2260050209>3.0.CO;2-T.
Vogels N and Westerterp-Plantenga MS (2007)
Successful long-term weight maintenance: A
2-year follow-up. Obesity 15(5): 1258–1266.
DOI:10.1038/oby.2007.147.
Vogels N, Diepvens K and Westerterp-Plantenga MS
(2005) Predictors of long-term weight maintenance. Obesity Research 13(12): 2162–2168.
DOI:10.1038/oby.2005.268.
Wadden TA, Sternberg JA, Letizia KA, et al. (1989)
Treatment of obesity by very low calorie diet,
behaviour therapy, and their combination: a
five year perspective. International Journal of
Obesity 13(Suppl. 2): 39–46.
Westenhoefer J (1991) Dietary restraint and disinhibition: Is restraint a homogeneous construct?
Appetite 16(1): 45–55.
Wing RR, Papandonatos G, Fava JL, et al. (2008)
Maintaining large weight losses: The role of
behavioral and psychological factors. Journal
of Consulting and Clinical Psychology 76(6):
1015–1021. DOI:10.1037/a0014159.
Downloaded from hpq.sagepub.com at University of Birmingham on June 4, 2015