The investigators will examine whether compulsivity in those who score above-average (but below clinical cut-off) on an eating disorder questionnaire can be altered by 9 days of N-acetyl cysteine (NAC). N-acetyl cysteine has been shown to be of some benefit in individuals with other compulsive disorders, such as trichotillomania and addiction, so this research investigates whether a short period of time taking N-acetyl cysteine changes compulsivity, measured using cognitive tasks and questionnaires.

Anorexia nervosa is a psychiatric disorder with a high mortality rate and for which there is very little evidence for pharmacological interventions of value. The picture is similar in bulimia nervosa, with a high mortality rate and with mixed outcomes from studies testing Selective Serotonin Reuptake Inhibitors (SSRIs). It is possible to conceptualise eating disorders as akin to compulsive disorders, especially noting that repetitive weighing, calorie counting and compulsive exercise and compensatory mechanisms might fit into a hypothetic obsessive-compulsive spectrum (OC spectrum). Some disorders on this OC spectrum are associated with oxidative stress. This may also be true in eating disorders: it has recently been found in a meta-analysis that there is evidence of increased oxidative stress markers in AN.

Glutathione is the main cellular anti-oxidant in mammals and it is possible to increase glutathione levels (and thus potentially combat oxidative stress) with a dietary supplement called N-Acetyl Cysteine (NAC). NAC is an acetylated form of the naturally occurring amino acid cysteine, and is converted to cysteine in the body. Cysteine is a key precursor of glutathione. Oral administration of cysteine itself does not increase brain glutathione levels because cysteine is poorly absorbed from the gastro-intestinal tract. However NAC is bioavailable and thus administration of it does increase glutathione in the brain.

Using NAC to increase glutathione levels has been shown to be beneficial in some compulsive disorders such as addictions, Obsessive-Compulsive Disorder (OCD) and related disorders such as trichotillomania in preliminary studies. There have been positive indications for use and evidence in pathological gambling, trichotillomania, OCD and cocaine addiction. However, NAC has not to our knowledge ever been considered for the treatment of eating disorders. This research therefore aims to investigate whether increasing glutathione can reduce neuropsychological markers of compulsivity in a group at risk for eating disorders.

The researchers have chosen to use a female non-patient group which is thought to be highly compulsive, but which will also be without the possible confounds of medication (as in OCD or addiction disorders) and without malnutrition (as in clinical eating disorders). Those recruited will be females, because there is evidence that there may be different risk factors and thus mechanisms underlying the development of eating disorders in males and females, such that recruiting both might produce a confound. The investigators have also chosen to use a short duration of NAC administration. This not only allows the use of a cross-over design without fears of high attrition rates and non-compliance, but it also will allow the detection of subtle and early changes in compulsivity. The hypothesis is thus that improvements will be seen in tasks measuring compulsivity, and also potentially in tasks which measure related constructs such as impulsivity, habit-based learning, and delay discounting. The researchers do not expect improvement in self-report symptom questionnaires. However, a possible future study would be a randomized clinical trial looking at increasing glutathione in a clinical group over time and how it affects clinical symptomatology.

The questionnaires chosen will identify eating disorder symptom levels (the Eating Attitudes Test - 26 (EAT-26) is a validated way of doing this) and look at compulsive starvation using the self-starvation scale (Godier & Park 2015). The Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders-IV (SCID) will be used to rule out those with a current Axis I diagnosis including an eating disorder, to prevent the complications of starvation and medication in this initial pilot study. The Eysenck Personality Questionnaire (EPQ) will be used to ascertain personality traits, and the National Adult Reading Test (NART) will be used to estimate Intelligence Quotient (IQ).

The rationale for choosing the tasks was based on the hypothesis that compulsivity may be improved by increasing glutathione levels. Thus, the tasks chosen are designed to pick up differences in different facets of compulsivity. The Wisconsin Card Sorting Task (WCST) is able to quantitatively measure set shifting, which is known to be impaired in eating disorders (Roberts et al. 2007) and is a facet of compulsivity. The Affective Go/No go picks up on attention bias within the umbrella of compulsivity; and the Attention Switching task picks up the ability to switch attention between different task demands, within the same compulsivity umbrella. The Cambridge Gambling Task is a good measure of disadvantageous decision making within the umbrella of impulsivity (which is thought to be orthogonally related to, rather than opposite to, compulsivity) (Fineberg et al. 2014). The sequential learning task has been used in OCD, where it is able to disentangle goal-directed learning from habitual learning based on two computational algorithms. The hypothesis is that compulsivity is related to a disturbance in goal-directed learning, with habit learning preferred leading to rigid behaviours (Gillan & Robbins 2014) (Voon et al. 2015). Thus it will be useful to see if more goal-directed and less habit learning occurs after increases in glutathione levels. The delay discount task measures the ability to ignore delays in time when considering reward value, which is relevant in eating disorders as often delay discounting is reduced so performance is better (Steinglass et al. 2012). This has an interesting interaction with impulsivity as it seems almost to be opposite to it, yet anorexic patients have shown a better performance. Thus this task might be useful as an exploratory measure. The Facial Expression Recognition Task (FERT) is very sensitive to early changes in emotional biases after only small pharmacological changes, and as the profile of individuals high eating disorder symptoms is different in this task compared to healthy controls (Jones et al. 2008) - it can operate as a positive control to identify if increased glutathione is having any effect on general emotional biases, even if it is not on compulsivity.

Because the aim of the study is to see whether NAC may lower compulsivity by increasing glutathione levels, a researcher will take a small venous blood sample (5mls) and a saliva sample prior to testing in both arms of the study to check to what to what extent NAC treatment elevates glutathione levels over placebo. Participants who do not wish to give blood will not be excluded but asked to provide a saliva sample only. Blood and saliva will be rendered acellular by centrifugation and stored until assay in the University Departments of Chemistry and Pharmacology. Once assayed any remaining sample will be discarded according to Standard Operating Procedures for sample disposal.

Inclusion Criteria:

• Participant is willing and able to give informed consent for participation in the study.
• Females aged 18 to 40 years.
• Score between 9 and 19 on the EAT-26.
• Participant is a fluent English speaker

Exclusion Criteria:

• Any current psychotropic medications.
• Any other significant disease or disorder which, in the opinion of the Investigator, may either put the participants at risk because of participation in the trial, or may influence the result of the trial, or the participant's ability to participate in the study.
• BMI below 18.5, indicating that they are underweight.
• Any Axis I psychiatric disorder.
• History of allergies to drugs or vaccines or any component of the NAC or placebo capsule (gelatine, magnesium stearate, NAC).
• Pregnant, could be pregnant, breast feeding, or high risk of pregnancy (no reliable contraception use but sexual activity).
• Has taken part in a psychological or medical experiment involving taking any kinds of drugs within the last 6 weeks.
• Has had recent surgery.
• Taking selenium, vitamin E, or NAC.
• Current regular cigarette smoking of over 5 cigarettes per day.