Understanding successful behaviour change: the role
of intentions, attitudes to the target and motivations
and the example of diet
Jane Ogden*, Lubna Karim, Abida Choudry and Kerry Brown
Abstract
Although many attempts to change health
behaviour fail, some individuals do show suc-
cessful behaviour change. This study assessed
the role of behavioural intentions, motivations
and attitudes to the target in explaining success-
ful changes in diet with a particular focus on
positive and negative intentions and positive
and negative attitudes. Participants (n 5 282)
completed a questionnaire describing a recent
change in eating behaviour (becoming a vege-
tarian, cutting out a food group, eating fewer
calories), their intentions, their attitudes to the
food being avoided, a range of motivations and
their degree of success. The results showed that
the three behaviour change groups differed in
terms of their cognitions with those trying to
eat fewer calories reporting less success in
changing their behaviour. Successful vegetari-
anism was associated with a lower positive
attitude; successfully cutting out a food group
was related to ethical motivations, a lower
positive attitude and greater positive and neg-
ative intentions, and reducing calorie intake
was associated with greater positive intentions
and a lower positive attitude. Therefore, success
was associated with different cognitions de-
pending upon the type of change being made,
although cognitions such as ‘I will eat more
vegetables’ and ‘I no longer find high fat foods
palatable’ were consistently most predictive of
success. Suggestions for the development of
more effective interventions to change health
behaviours are made.
Introduction
Most people attempt to change an aspect of their
health behaviour at some time whether it be to stop
smoking, drink less alcohol, exercise more often or
practice safer sex. For some, these intentions are
translated in successful behaviour change. For
many, however, such intentions never result in
actual behaviour change or may do so only in the
short term. Eating behaviour is especially difficult to
change particularly when this change involves
weight loss-related dieting. Although some studies
indicate that dieters may successfully reduce their
energy intake [1, 2], other studies have found that
dieters rarely reduce energy intake enough to be
successful at weight loss [3–10]. Research exploring
the links between dieting and eating has been de-
veloped within the framework of restraint theory and
was based upon an early study by Herman and
Mack [3]. Restraint theory explored the impact of
imposing cognitive restraint on eating and showed
that paradoxically trying to eat less can result in
overeating. Subsequent studies have explored the
conditions under which overeating eating occurs and
has highlighted a role for factors such as smoking
abstinence [4], food cues [5], cognitive shifts [6]
and lowered mood [7]. In addition, reviews of the
success of weight-loss interventions illustrate that,
although the percentage of people who initially lose
Department of Psychology, University of Surrey,
Guildford, Surrey GU2 7XH, UK
*Correspondence to: J. Ogden.
E-mail: [email protected]
HEALTH EDUCATION RESEARCH Vol.22 no.3 2007
Pages 397–405
Advance Access publication 13 September 2006
� The Author 2006. Published by Oxford University Press. All rights reserved.
For permissions, please email: [email protected]
doi:10.1093/her/cyl090
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
weight has recently increased, a large majority of
them regain this weight on 5-year follow-up evalua-
tions [8–10]. Therefore, the majority of dieters find
it difficult to change their eating behaviour. Never-
theless, a small minority do succeed in sustaining
weight loss after 5 years. Similarly, successful
changes in eating behaviour can also be found in
populations other than dieters including vegetarians,
those who chose to fast at specific times of the year
and those who avoid specific foods for religious
reasons. What factors therefore predict successful
changes in eating behaviour? Some research has
addressed the correlates of successful dieting and
has pointed to a role for previous dieting history,
the intensity and length of any intervention and
the individual’s beliefs about the causes and con-
sequences of obesity [e.g. 10, 11]. However, other
research findings point to a potential role for cog-
nitions which may explain successful changes in
eating behaviour beyond that shown by the minor-
ity of dieters. In particular, research in health and
social psychology has drawn upon social cognition
models such as the theory of planned behavi-
our (TPB) [12] and protection motivation theory
(PMT) [13] and highlights a role for behavioural
intentions, attitudes and motivations.
First, in terms of behavioural intentions, research
has shown consistently that the intention to perform
a behaviour can be translated into actual behaviour.
For example, research indicates that the intention
to use condoms predicts condom use, that the in-
tention to exercise correlates with this behaviour
and the intention to attend for cervical or breast
screening practices predicts actual attendance [e.g.
14–16]. In terms of eating behaviour, research has
also shown that the intention to eat healthily is
a successful predictor of subsequent behaviour [e.g.
17]. Therefore, the cognition ‘I intend to …’ seems
to translate into ‘I did’. Sutton [18] carried out an
analysis of the association between behavioural
intentions and behaviour across a series of studies
and concluded that intentions generally predict
between 19% and 38% of the actual variance in
behaviour. This suggests that behavioural inten-
tions may be useful predictors of successful dietary
change. In direct contrast to the social cognition
literature, however, research drawing upon restraint
theory as a means of understanding eating behav-
iour implicitly indicates that the intention to under
eat does not predict actual under eating but has the
paradoxical effect of being associated with over-
eating. For example, studies show that dieters who
make an intention to eat less or to eat more healthily
may end up eating less healthily and overeating
[3, 5]. This approach finds reflection in the work
of Wegner [19] who describe the ‘theory of ironic
processes of mental control’ and suggest that
attempting not to think about something can have
the opposite effect to that which is desired. Further,
it is similar to the work describing different types
of goals and the differential effect of goals and anti
goals [20]. It may not, therefore, be behavioural
intentions per se which predict successful dietary
change but the direction of this intention. Most
research exploring intentions and health behaviours
focuses on positive intentions such as ‘I intend to
use a condom’ and ‘I intend to exercise’, whereas
eating research implicitly focuses on negative
intentions such as ‘I intend not to eat food high in
calorie’. In line with this, the present study aimed
to explore the impact of the type and direction of
behavioural intention of successful changes in
eating behaviour and to examine whether positive
and negative intentions differentially impact upon
the success of behaviour change.
Second, research shows a role for motivational
factors. For example, a study examining the motiv-
ations behind the choice to adopt a vegetarian diet
found a role for moral, health, gustatory and eco-
logical factors [21]. Similarly, Mooney and
Wahlbourn [22] investigated reasons for food
rejection by college students and reported that stu-
dents who avoided meat placed more importance
on ethical factors. Furst et al. [23] also explored
what motivations influenced food choice and found
a role for issues of availability and cost and Steptoe
and Pollard [24] used the ‘food choice question-
naire’ to explore the factors that motivate food
selection and reported a role for familiarity with a
food and convenience. Changes in diet are, there-
fore, the result of a range of motivations. Further,
it would seem that successful food avoidance
J. Ogden et al.
398
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
shown in line with religious rules or by vegetarians
is motivated by different factors by the attempted
food avoidance shown by dieters. However, re-
search to date has been descriptive in its focus and
has not explored whether some forms of motivation
are more predictive of success than others. So as to
lend clarity to this issue, the present study examined
a range of motivational factors and their association
with successful dietary change.
Finally, social cognition models such as the TPB
and PMT [12, 13] also include a role for attitudes
which are defined as ‘a function of a person’s salient
beliefs, which represent perceived consequences of
the behaviour’ [25]. These are usually operational-
ized by asking an individual to rate a particular
behaviour using differentials such as ‘unpleasant/
pleasant’. However, such attitudes are generally
directed at the behaviour per se such as ‘dieting is
effective’ and ‘dieting is pleasant’ [e.g. 26] rather
than at the target of the behaviour. Attitudes to the
target have been described by Eagly and Chaiken
[27] and would seem to be particularly pertinent in
the area of eating behaviour as different foods are
embedded with different meanings and can gener-
ate both positive and negative responses. Examples
of this would include an individual’s attitudes to
cigarettes (for the person intending to stop smok-
ing) or attitudes to meat by the person intending to
become a vegetarian. However, to date, although
research has incorporated attitudes in social cog-
nitive model-based research of behaviour, such
research has focused on attitudes towards the
behaviour rather than to the object. Furthermore,
no research neither has explored the direction of
attitude nor has examined the impact of the direction
of attitude on successful changes in eating behav-
iour. This would seem to be particularly relevant
given that some forms of dietary change such as
vegetarianism may be accompanied by negative
attitudes to the target (‘I will avoid meat and I find
it disgusting’), whereas others may involve posi-
tive attitudes to the target (‘I will eat low fat foods
but I still prefer high fat alternatives’).
In summary, research suggests that while some
attempts to change behaviour are successful, many
are not. Further, some types of changes in eating
behaviour seem to be more consistently associated
with success than others. The literature exploring
behaviour change has drawn upon restraint theory
and social cognition models and has highlighted
a role for a range of cognitions. In particular, re-
search points to a role for behavioural intentions
in explaining this variability but has treated this con-
struct as unidimensional neglecting the potential for
a differential effect of the positive and negative
versions of this variable. Further, the literature has
also pointed to a role for attitudes but has focused
on attitudes to the behaviour rather than attitudes
to the target which would seem to be particularly
pertinent to a behaviour such as eating which is
concerned with a target so embedded with meaning.
Finally, while previous research has explored the
type of motivations linked with food choice, the
relationship between different type of motivation
and success remains unexplored. The present study
therefore aimed to assess differences in cognitions
between different types of changes in eating be-
haviour. In addition, the study aimed to assess
the relative impact of the positive and negative
versions of both intentions and attitudes to the
target and a range of motivations in predicting the
degree of success in actual behaviour change. In
particular, given that dieting seems to generally in-
volve negative intentions, positive attitudes to the
target and weight and health-related motivations,
it was predicted that the use of such cognitions
would help to explain why dieting is a univers-
ally unsuccessful form of behaviour change.
Methods
Design
The study used an anonymous, cross-sectional
questionnaire survey examining a recent episode
of attempted change in eating behaviour.
Participants
Questionnaires (n = 350) were given out to un-
dergraduate students at two universities through
lectures. Completed questionnaires were received
from 282 (response rate = 80.5%).
Successful changes in diet
399
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
Measures
Participants were asked to describe a recent change
in eating behaviour and then to complete questions
describing behavioural intentions, attitudes to the
target, motivations and success which were rated
on a five-point Likert scale ranging from ‘not at
all’ to ‘totally’. Reliability was assessed using
Cronbach’s alphas for each scale within each
behaviour change group.
Change in eating behaviour
Participants were asked to ‘think about a time
recently that you decided to change what you eat’
and to identify if it was one of the following: fol-
lowing a vegetarian diet, cutting out on a particular
food group or food or eating fewer calories. These
were selected as they represented changes in eating
behaviour which were expected to be present in the
population being studied. In addition, they reflected
types of behaviour change which were hypothe-
sized to range from successful (becoming a vege-
tarian) to less successful (eating fewer calories).
They were then asked to go to the section relating
to this behaviour and to rate a series of statements
describing behaviour intentions.
Behaviour intentions
Participants were asked to rate six statements
describing three positive and three negative inten-
tions which were specific to their type of behaviour
change. Positive intention statements included ‘I
would eat more vegetables’ for the vegetarian’s
section, ‘I would eat more low calorie foods’ for
the fewer calorie section and ‘I will eat more foods
from other food groups’ for the cutting out a food
section. Negative intention statements included
‘I will not eat meat’ for the vegetarian’s section,
‘I will not eat sweet foods’ for the fewer calories
section and ‘I will not purchase food from this food
group’ for the cutting out a food group section.
These items were summated to compute a total
negative intention score and a total positive in-
tention score. All alphas for negative intention
were >0.7. The alphas for positive intention for
those becoming a vegetarian and those trying to eat
fewer calories were all >0.5. However, the alpha for
positive intention for those cutting out a food group
was 0.04. On removing one item, the correlation
between the remaining two items was 0.4. There-
fore, for positive intentions for those cutting out
a food group, a two-item scale was used.
Attitudes to the target
Participants were asked to rate the ‘foods you were
trying to avoid’ using a series of four positive adjec-
tives (pleasant, desirable, appetizing, tasty) and four
negative adjectives (disgusting, sickening, horrible,
revolting). These were summated to create a total
positive attitude score and a total negative attitude
score. The alphas for positive and negative attitudes
for all three groups were >0.9.
Motivations
Participants were asked ‘to what extent was your
decision to make this change motivated by the
following reasons’ and then rated the following
motivations: taste, religion, weight, finance, ethics,
health and availability.
Success
Finally, participants were asked to rate their success
at food avoidance using three items ‘I was success-
ful at making this change in my behaviour’, ‘I was
able to stick to my decision to change’ and ‘I kept
breaking my rules’ using a five-point Likert scale
ranging from ‘totally disagree’ to ‘totally agree’.
These were summated to create a total success score.
The alphas for the three groups were all >0.9.
Higher scores reflected higher intentions (posi-
tive and negative), stronger attitudes (positive and
negative), a greater endorsement of the motivation
and greater success.
Participants also described their age, sex and
ethnicity (White/Black/Asian/Other).
Results
The data were analysed to describe the participants’
profile characteristics and type of change in eating
behaviour and to explore differences in intentions,
J. Ogden et al.
400
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
attitude and motivations between the different types
of dietary change. The results were then analysed to
assess the role of positive and negative behavio-
ural intentions, positive and negative attitudes and
a range of motivations in predicting successful
dietary change using multiple regression analysis
within each type of dietary change. A blocked
method of entry was used to enable the impact of
motivations (block 1), intentions (block 2) and
attitudes (block 3) to be assessed separately.
Profile characteristics
Participants’ profile characteristics are shown in
Table I.
The majority of participants was women and
Caucasian, although a large minority described
themselves as Asian. The most common form of
dietary change was an attempt to eat fewer calories.
Difference between different types of
dietary change
The means (and SDs) for the different types of
dietary change are shown in Table II.
The three groups were comparable in terms of
positive and negative intentions and motivations
relating to finance, religion and availability. Trying
to consume fewer calories was associated with
greater motivation relating to health and weight and
trying to become a vegetarian was associated with
greater negative attitudes, lower positive attitudes
and higher motivations relating to taste and ethical
issues. Trying to eat fewer calories resulted in a
lower level of success than the other two types of
behaviour change.
The best predictors of successful dietary
change
The results were analysed to explore the best
predictors of successful behaviour change within
each behaviour change group using multiple re-
gression analysis.
Predicting successfully becoming
a vegetarian
Due to the smaller sample size within this group
(n = 39), initial univariate correlations were carried
out to assess which variables could be entered into
the multiple regression analysis. The results showed
that motivations relating to health (r = �0.44) and
ethics (r = 0.4), positive attitudes (r = �0.59) and
negative attitudes (r = 0.36) were significantly cor-
related with success and were therefore entered into
the multiple regression analysis (P-values < 0.05).
The remaining motivations and positive and nega-
tive intentions were not entered into the analysis.
The results from this analysis showed that a positive
attitude significantly predicted success (b = �0.51,
P = 0.005) accounting for 39% of the variance.
Health motivation (b = �0.24, P = 0.143), ethical
motivation (b = 0.16, P = 0.32) and negative
attitude (b = �0.05, P = 0.76) were unrelated to
success. Successful behaviour change for becom-
ing a vegetarian was associated with lower ratings
of the foods to be avoided as ‘pleasant’, ‘tasty’,
‘desirable’ and ‘appetizing’.
Predicting successfully cutting out
a food group
The best predictors for cutting out a food group
were a higher ethical motivation, greater positive
and negative intentions and a lower positive attitude
towards the food being avoided. Success at cutting
out a food group was therefore associated with an
ethical motivation, cognitions such as ‘I will eat
more foods from other food groups’ and ‘I will not
Table I. Profile characteristics
Variables n % Mean SD Range
Age 182 25.26 9.92 16–66
years
Sex
Male 77 27.3
Female 204 72.3
Ethnicity
White 150 53.2
Black 20 7.1
Asian 83 29.4
Other 28 9.9
Type of change
Vegetarian 39 13.8
Cut out a food group 90 31.9
Eat fewer calories 153 54.3
Successful changes in diet
401
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
purchase foods from this food group’ and lower
beliefs that the foods being avoided were ‘tasty’ or
‘pleasant’ (see Table III).
Predicting successful calorie reduction
Successfully eating fewer calories was associated
with higher positive intentions and a lower positive
attitude to the foods being avoided. Therefore,
greater endorsement of beliefs such as ‘I will only
eat foods that are low in calories’ and less beliefs
that high-calorie foods are ‘tasty’ or ‘pleasant’ were
predictive of success (see Table IV).
Discussion
The present study aimed to assess differences
between different types of changes in eating
behaviour and to evaluate the best predictors of
successful behaviour change with a focus on
behavioural intentions, attitude to the target and
motivations. However, there are some problems
with the study that need to be addressed. First, the
study was cross sectional in design. It is therefore
possible that rather than predicting changes in
behaviour, the cognitions identified are post hoc
justifications of behaviour change. But, why differ-
ent types of change should generate different types
of justification is an interesting question in itself.
Future research should employ a prospective design
to clarify issues of causality. Second, participants
were asked to select the behaviour themselves and
it is possible that they chose the one which resulted
in most success which could shift the results in
a socially desirable direction. The study, however,
was anonymous and participants did not have to
complete the questionnaire if they did not wish
to or if they felt they had not changed their be-
haviour. Pressure to answer in a particular way was
therefore kept to a minimum. The results how-
ever do provide some insights into which cogni-
tions are related to which type of change in eating
behaviour.
The study focused on three types of changes in
eating behaviour, namely, becoming a vegetarian,
cutting out a food group and eating fewer calories.
The results showed that eating fewer calories was
associated with motivations relating to weight and
health and that becoming a vegetarian was associ-
ated with ethical and taste-related motivations,
a lower positive attitude and a greater negative
attitude for the foods being avoided. In addition,
eating fewer calories was associated with lower
success than the other two groups. Previous re-
search suggests that dieting is often unsuccessful
resulting in weight maintenance or weight gain,
Table II. Differences according to type of dietary change (means and SDs)
Variables Vegetarian (n = 39) Cut out a food group (n = 90) Eat fewer calories (n = 153) F
Intentions
Positive intentions 3.41 6 0.83 3.45 6 1.11 3.28 6 0.89 0.38
Negative intentions 3.65 6 1.09 3.54 6 1.19 3.41 6 0.95 1.04
Attitude
Positive attitude 2.79 6 1.4 3.60 6 1.07 3.62 6 1.08 8.70**
Negative attitude 2.47 6 1.41 1.65 6 1.02 1.86 6 1.14 6.96**
Motivations
Financial motivation 1.44 6 0.99 1.32 6 0.7 1.44 6 0.85 1.02
Taste motivation 2.31 6 1.55 1.63 6 1.05 1.58 6 0.93 7.14**
Religious motivation 1.67 6 1.35 1.41 6 1.09 1.25 6 0.7 2.64
Weight motivation 2.17 6 1.67 3.13 6 1.56 4.01 6 1.26 30.78**
Health motivation 2.97 6 1.67 3.84 6 1.41 4.22 6 0.98 14.09**
Availability motivation 1.53 6 1.0 1.41 6 0.85 1.69 6 1.04 1.3
Ethical motivation 3.17 6 1.65 1.66 6 1.35 1.4 6 0.84 35.25**
Success 3.71 6 1.44 3.35 6 1.19 3.16 6 1.27 3.51*
*P < 0.05, **P < 0.001.
J. Ogden et al.
402
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
failed under eating and actual overeating [3, 7, 10].
The present study suggests that this lack of success
may be due to the kinds of cognitions associated
with this behaviour.
To explore this further, the results were then
analysed to assess the best predictors of successful
changes in eating behaviour within each behaviour
change group. For becoming a vegetarian, success-
ful behaviour change was associated with lower
ratings of the positive attributes of the food being
avoided. For cutting out a food group, the best
predictors of success were an ethical motivation,
both positive and negative intentions and a lower
positive attitude, while successfully eating fewer
calories was predicted by higher positive intentions
and a lower positive attitude. Previous research has
highlighted that attempts to change eating behav-
iour are related to a range of motivations includ-
ing ethics, availability, weight and health [21–23].
These findings support this variability in motiva-
tions and suggest that different motivations are
associated with different forms of successful be-
haviour outcomes. Previous research also suggests
a role for behavioural intentions but has tended to
treat this construct as unidimensional in nature and
has not assessed the possibility of a differential
impact from either positive or negative intentions
[e.g. 15]. The results from this study support the
importance of assessing both positive and negative
behavioural intentions for better predictions of
successful outcomes. Further, they indicate that
positive rather than negative intentions may be
more predictive of success. This is line with
Table III. Predicting successful cutting out a food group
(n = 90)
Block Variables Standard
beta
Adjusted
R2
F change
1 Financial motivation �0.20
Taste motivation 0.19
Religious motivation 0.17
Weight motivation �0.24*
Health motivation �0.14
Availability motivation �0.02
Ethical motivation 0.27* 0.27 5.20***
2 Financial motivation �0.20
Taste motivation 0.19
Religious motivation 0.19
Weight motivation �0.27*
Health motivation �0.16
Availability motivation 0.02
Ethical motivation 0.26*
Positive intentions 0.28**
Negative intentions 0.22* 0.36 5.9**
3 Financial motivation �0.19
Taste motivation 0.13
Religious motivation 0.20
Weight motivation �0.19
Health motivation �0.12
Availability motivation 0.03
Ethical motivation 0.33**
Positive intentions 0.24**
Negative intentions 0.19*
Positive attitude �0.23*
Negative attitude �0.01 0.38 5.5*
*P < 0.05, **P < 0.01, ***P < 0.001.
Table IV. Predicting successful calorie reduction (n = 153)
Block Variables Standard
Beta
Adjusted
R2
F change
1 Financial motivation 0.05
Taste motivation 0.12
Religious motivation �0.04
Weight motivation 0.00
Health motivation 0.00
Availability motivation �0.09
Ethical motivation 0.02 �0.04 0.31
2 Financial motivation 0.07
Taste motivation 0.10
Religious motivation �0.03
Weight motivation �0.03
Health motivation �0.00
Availability motivation �0.09
Ethical motivation �0.00
Positive intentions 0.35**
Negative intentions 0.08 0.11 11.98***
3 Financial motivation 0.05
Taste motivation 0.00
Religious motivation �0.04
Weight motivation 0.04
Health motivation 0.03
Availability motivation �0.07
Ethical motivation 0.05
Positive intentions 0.31***
Negative intentions 0.03
Positive attitude �0.32**
Negative attitude 0.12 0.26 14.25***
**P < 0.01, ***P < 0.001.
Successful changes in diet
403
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
Wegner’s work on ironic processes of control [19]
and reflects the concepts of goals and anti goals
described within the literature on self-regulation
[20]. While cutting out a food group was related to
stronger intentions regardless of their direction,
successfully eating fewer calories required stronger
positive intentions. Finally, most previous research
has also highlighted a role for attitudes but has
emphasized attitudes to the behaviour rather than
the target of that behaviour [e.g. 12]. The results
from the present study suggest that attitudes to the
target are related to the degree of success and that
the absence of a positive attitude is consistently
most related to outcome.
In summary, successful changes in three types of
eating behaviour are related to different types of
cognitions. However, an ethical motivation, posi-
tive rather than negative intentions and the absence
of a positive attitude to the target all emerge as
factors most likely to predict success. Accordingly,
cognitions such as ‘eating meat is unethical’, ‘I will
eat more vegetables’ and ‘I no longer find high fat
foods palatable’ are more predictive of success than
motivations relating to other factors such as health
and weight, negative intentions which involve
avoiding foods (‘I will not eat sweet foods’) and
liking the food which is to be avoided (‘I still like
sweet foods’).
These results have implications for both research
and practice. In terms of research, the results
suggest that neither intentions nor attitudes to the
target are unidimensional constructs making it
worthwhile to differentiate between the positive
and negative versions of these variables and that
attitudes to the target rather than just attitudes to the
behaviour are predictive of outcome. It is possible
that the moderate associations between intentions
and behaviour described in the literature [18] may
be due to either the absence of a measure of the
different types of intention or the ability of the two
opposing types of intention to cancel each other
out. The results may also explain the apparent con-
tradiction between the social cognition literature
which suggests that intentions are predictive of
behaviour and the literature based upon restraint
theory which suggests that intentions to under eat
may paradoxically result in overeating. The social
cognition literature may report a positive associa-
tion between intentions and behaviour because it
primarily focuses on positive intentions. In contrast,
restraint literature may report a negative associ-
ation between intentions and behaviour because its
focus is on dieting which is implicitly associated
with negative intentions.
In terms of practice, the results from this study
have implications for the development of effective
behaviour change interventions. At present, many
interventions designed to improve health-related
behaviours such as diet, smoking and drinking
encourage the avoidance of these behaviours. Such
approaches emphasize health as a motivator for
behaviour change and encourage individuals to
develop negative intentions (e.g. ‘I will not eat
fatty foods’ and ‘I will not smoke’), while leaving
them with a positive attitude for these objects of
their behaviour (‘I like fatty foods’, ‘I like ciga-
rettes’). The results from the present study suggest
that such an approach is unlikely to result in actual
behaviour change as it promotes the cognitions
which are linked with failure. However, if an inter-
vention could encourage individuals to be motiv-
ated by factors other than health (such as ethics), to
focus on what they intend to do rather than what
they are going to avoid and to develop a dislike
for the object to be avoided, then perhaps such an
approach would be more predictive of positive
outcomes. Further, such interventions could be
facilitated by social and structural changes de-
signed to promote healthier lifestyles. For example,
changes in social norms towards seeing unhealthy
behaviours such as smoking and an unhealthy diet
as unattractive and socially unacceptable may
reduce an individual’s positive attitudes to the
unhealthy behaviours. Furthermore, the provision
of more easily accessible healthy foods and banning
smoking in public areas make many positive in-
tentions easier both to make and adhere to. Intend-
ing ‘to do’ a healthy behaviour and ‘not liking’ the
unhealthy behaviour seem to be the key to success.
Such an approach should be incorporated into
behaviour change interventions which could
be complimented by social and policy changes
J. Ogden et al.
404
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
designed to provide a better environment in which
a decision to be healthier can be more easily
translated into actual healthier behaviour.
Conflict of interest statement
None declared.
References
1. Thompson JP, Palmer RL, Petersen SA. Is there a metab-
olic component to counterregulation? Int J Eat Disord
1988; 7: 307–19.
2. Van Strien T, Cleven A, Schippers G. Restraint, tendency to
overeating and ice cream consumption. Int J Eat Disord
2000; 28: 333–8.
3. Herman P, Mack D. Restrained and unrestrained eating.
J Pers 1975; 43: 646–60.
4. Ogden J. The effects of smoking cessation, restrained eating,
and motivational states on food intake in the laboratory.
Health Psychol 1994; 13: 114–21.
5. Fedoroff IC, Polivy J, Herman CP. The effect of pre-
exposure to food cues on the eating behavior of restrained
and unrestrained eaters. Appetite 1997; 28: 33–47.
6. Ogden J, Greville L. Cognitive changes to preloading in
restrained and unrestrained eaters as measured by the
Stroop task. Int J Eat Disord 1993; 14: 185–95.
7. Polivy J, Herman CP, McFarlane T. Effects of anxiety on
eating: does palatability moderate distress-induced over-
eating in dieters? J Abnorm Psychol 1994; 103: 505–10.
8. Wilson GT. Behavioral treatment of obesity: thirty years and
counting. Adv Behav Res Ther 1995; 16: 31–75.
9. Wadden TA. Treatment of obesity by moderate and severe
calorie restriction: results of clinical research trials. Ann
Intern Med 1993; 119: 688–93.
10. NHS Centre for Reviews and Dissemination, University
of York. Systematic Review of Interventions in the Treat-
ment and Prevention of Obesity, 1997.
11. Ogden J. The correlates of long-term weight loss: a group
comparison study of obesity. Int J Obesity 2000; 24:
1018–25.
12. Ajzen I. From intention to actions: a theory of planned
behavior. In: Kuhl J, Beckman J (eds). Action-Control:
From Cognition to Behavior. Heidelberg, Germany:
Springer, 1985, 11–39.
13. Rogers RW. A protection motivation theory of fear appeals
and attitude change. J Psychol 1975; 91: 93–114.
14. Yzer MC, Siero FW, Buunk BP. Bringing up condom use
and using condoms with new sexual partners: intentional
or habitual? Psychol Health 2001; 16: 409–21.
15. Plotnikoff RC, Higginbotham N. Protection motivation
theory and the prediction of exercise and low-fat diet
behaviours among Australian cardiac patients. Psychol
Health 1998; 13: 411–29.
16. Sheeran P, Orbell S. Using implementation intentions to
increase attendance for cervical cancer screening. Health
Psychol 2000; 19: 283–9.
17. Povey R, Conner M, Sparks P et al. The theory of planned
behaviour and healthy eating: examining additive and
moderating effects of social influence variables. Psychol
Health 2000; 14: 991–1006.
18. Sutton S. Predicting and explaining intentions and behav-
iour: how well are we doing? J Appl Soc Psychol 1998; 28:
1317–38.
19. Wegner DM. Ironic processes of mental control. Psychol
Rev 1994; 101: 34–52.
20. Wrosch C, Scheier MF, Miller GE et al. Adaptive self-
regulation of unattainable goals: goal disengagement, goal
reengagement, and subjective well-being. Pers Soc Psychol
Bull 2003; 29: 1494–508.
21. Santos MLS, Booth DA. Influences on meat avoidance
among British students. Appetite 1996; 27: 197–205.
22. Mooney KM, Wahlbourn L. When college students
reject food: not just a matter of taste. Appetite 2001; 36:
41–50.
23. Furst T, Connors M, Bisogni CA et al. Food choice:
a conceptual model of the process. Appetite 1996; 26:
247–66.
24. Steptoe A, Pollard TM, Wardle J. Development of a measure
of the motives underlying the selection of food: the food
choice questionnaire. Appetite 1995; 25: 267–84.
25. Conner M, Norman P. Predicting Health Behaviours,
2nd edition. Buckingham, England: Open University
Press, 2005.
26. Fishbein M, Ajzen I. Belief, Attitude, Intention, and
Behaviour: An Introduction to Theory and Research.
Reading, MA: Addison-Wesley, 1975.
27. Eagly AH, Chaiken S. The Psychology of Attitudes. Fort
Worth, TX: Harcourt Brace Jovanovich, 1993.
Received on November 13, 2005; accepted on July 7, 2006
Successful changes in diet
405
D
ow
nloaded from
https://academ
ic.oup.com
/her/article/22/3/397/596170 by guest on 01 M
ay 2022
