Temptation and restraint: The role of reward and punishment sensitivity in disorder eating

The current study will focus on why patients with Anorexia Nervosa are able to
restrict their food intake while being in a state of starvation, whereas
individuals who are obese systematically fail to restrict their food intake and
overeat.
The reinforcing effects of drugs and food are proposed to be partly regulated
by common biological mechanisms sensitive to reward (Davis & Carter, 2009).
Consistent with this view, and with the findings of studies on drug users,
obese individuals have been found to show an attentional bias to food (e.g.,
Yokum et al., 2011), suggesting that a tendency to prioritize reward-related
information might be a risk factor that lowers the threshold for becoming obese
(Davis et al., 2007). On the other hand, heightened sensitivity for cues of
punishment (e.g. undesirable weight gain) might help people to resist
temptation and restrain from food intake. In line with this, patients with
Anorexia Nervosa have been found to report to be more sensitive to punishment
than healthy controls (e.g., Glashouwer et al., 2014).
The general hypotheses for the current study is as follows. (1) Risk for
obesity is proposed to increase with high general appetitive and low general
aversive processing biases. (2) Risk for Anorexia is thought to increase with
high general aversive and low general appetitive processing biases.

The primary objective of the current study is to (further) investigate the role
of reward and punishment sensitivity in Anorexia Nervosa and obesity by
examining general and disorder-specific appetitive and aversive biases. The
secondary objective of the current study is to, exploratory, investigate the
relationship between general as well as disorder-specific appetitive/aversive
biases at intake and subsequent treatment outcome at 1 year follow up, and
whether specifically the disorder-specific biases change in the course of
treatment, for patients with Anorexia Nervosa specifically. The third objective
is to test whether executive control plays a moderating role in the
relationships between appetitive/aversive biases and eating disorder severity
and/or BMI.

This is a longitudinal observational study for Anorexia patients, and a
cross-sectional observational study for obesity patients. Both patient groups
will perform the first measure, which consists of computer tasks and
questionnaires. The computer tasks measure general and disorder-relevant reward
and punishment related attentional biases and executive control. The
questionnaires will measure self-reported, eating disorder symptomatology,
hunger and time since eaten, positive affect, general reward and punishment
sensitivity, and, symptoms of depression and anxiety. The anorexia patients
will also be approached for a follow-up after a year during which they will do
the same computer tasks and will complete the same questionnaires. There will
also be a control group matched on age and educational level, that will perform
the first measure.

The only burden for patients who participate in this study is the time
investment. There are no risks related to participation in this study. Patients
with Anorexia who participate in this study will be tested three times. Twice
at the centre of eating disorders, and once either at home or also at the
centre of eating disorders, depending on their preferences. The first
measurement will take 115 minutes, the second 130, and the last 60 minutes.
Patients with obesity will only participate in the first measurement and their
time investment will therefore be limited to 115 minutes in total.
This study is expected to lead to a better understanding of developmental and
maintaining factors of eating disorders. Since Anorexia Nervosa mainly develops
during adolescence this is an important age group to investigate the role of
reward and punishment sensitivity in the disorder in. The same holds for
obesity, during adolescence more and more food choices are made independently,
and the role of these characteristics in eating behavior are therefore best
studied in this age group.