Citalopram and Self Emotional Processing

The Effect of Acute Citalopram on Self-referential Emotional Processing and Social Cognition in Healthy Volunteers

This study is investigating the effect of an acute dose of citalopram on emotional processing about the self. Using a parallel-group double-blind design, participants will be randomised to receive either an acute dose of citalopram or placebo. Participants will then complete a number of widely used computer-based cognitive tasks measuring emotional processing biases towards the self. This study has also been registered on OSF: https://osf.io/nhjvs/?view_only=b39c49bddfd543b99b627dc992e49b45

Antidepressants are thought to operate by changing the way patients process emotional information. After a single dose of citalopram or fluoxetine healthy volunteers have been found to display an increased recognition of happy facial expressions and a reduced recognition of sad faces, in the absence of changes in mood. Studies using depressed participants have produced similar results. However, there has been comparatively little research on changes in emotional processing biases about the self following antidepressant administration. Sense of self has been proposed as fundamental for mental health, with self-schemas acting as a focus through which valence and reward influenced perception, memory and decision-making. Antidepressants may increase learning of positive information about the self, potentially remediating negative self-schema and subsequently reducing depression symptoms. In this study, the investigators aim to examine whether acute administration of citalopram is associated with an increase in positive emotional learning biases about the self. Using a parallel-group double-blind design, participants will be randomised to receive either an acute dose of citalopram or placebo. Participants will then complete a number of widely used computer-based cognitive tasks measuring emotional processing biases. Identifying early changes in cognition and behaviour following antidepressant treatment will increase our knowledge of how antidepressants operate, and provide putative targets to identify early response to antidepressants. This study has also been registered on OSF: https://osf.io/nhjvs/?view_only=b39c49bddfd543b99b627dc992e49b45 Starting from the 8th November 2019 an additional task (the Oxford Cognition Stress Task (OCST)) was included in the test battery. This task has been developed by the Psychopharmacology and Emotion Research Laboratory (PERL), Department of Psychiatry, University of Oxford. This is an acute psychosocial stress induction paradigm, comprised of computerised cognitive tasks with an induced failure component. An algorithm varies task timing/difficulty to be just beyond participants' ability, accompanied by aversive feedback. The OCST induces mild, transient increases in stress and arousal, as indexed by heart rate, skin conductance, salivary cortisol and self-reported subjective state measures. Data for this task will be collected, analysed and published by PERL and will not be included in any publications relating to the previous registration for this study. The OCST task has been included at the end of the test battery and is therefore not expected to influence data relating to any self-report measures or tasks outlined in the previous registration This section of the study has been registered separately on ClinicalTrials.gov (titled 'Citalopram and Stress Reactivity') to reflect the separate research questions and study team involvement.

Molecular Mechanisms of Pharmacological Action, Depression, Depressive Disorder, Mental Disorder, Antidepressive Agents, Cognition

Following the screening, eligible participants will be randomised to receive either a single 20mg oral dose of citalopram or a matched lactose placebo tablet using an online randomisation tool. Note that the study is not assessing the safety or efficacy of citalopram, but rather using it as a probe to understand the role of serotonin in self-referential emotional processing biases.

An overall index of positive or negative bias will be calculated for each referential condition (self, friend, stranger) using errors to criterion (the number of errors made before 8 rule-congruent responses). Bias is calculated by subtracting errors to criterion made when learning the dislike rule from errors to criterion made when learning the like rule. A positive value indicates a negative bias, as fewer errors are made learning the dislike rule compared to the like rule. Conversely, a negative value indicates a positive bias, as fewer errors are made learning the like rule compared to the dislike rule. The minimum possible value is - 24 (complete bias towards being liked), and the maximum value is + 24 (complete bias towards being disliked).

Mean reaction times will be calculated for each referential condition (self, friend, stranger), reward condition (high, medium, low) and valence condition (positive, neutral, negative) for each respective task.

Mean accuracy will be calculated for each referential condition (self, friend, stranger), reward condition (high, medium, low) and valence condition (positive, neutral, negative) for each respective task.

Discriminative accuracy (d') will be calculated through applying Z-score transformations, and subtracting hit z-scores from false alarm z-scores. Z-scores are adjusted by adding or subtracting .005 if hit or false-alarm rates are 0 or 1. d' -values can then be compared for each possible categorical combination to examine implicit self-biases.

The main outcome for this task is the proportion of rounds on which participants choose to cooperate. The conditional probability of cooperating will be calculated according to the proportion of rounds on which participants cooperated following each of the four possible outcomes.

The mean number of positive and negative words categorised as describing or not describing the participant/the other will be recorded.

Inclusion Criteria: Male or Female Aged 18 -45 years Fluent in written and spoken English at a sufficient level to understand and complete the tasks Body Mass Index (BMI) 18-30 Participant is willing and able to give informed consent for participation in the study Not currently taking any regular medications (expect the contraceptive pill) Exclusion Criteria: Any past or current Axis 1 DSM-V psychiatric disorder Current use of psychoactive medication (except the contraceptive pill, the Depo-Provera injection or the progesterone implant) Current or past history of drug or alcohol dependency History of current significant neurological condition (e.g. epilepsy) Known hypersensitivity to the study drug Currently pregnant or breast feeding Previous participation in a study that uses the same or similar computer tasks as those used in the present study Previous participation in a study that involves the use of a medication within the last three months Significant medical condition Smokers consuming > 5 cigarettes per day Individuals consuming > 6 caffeinated drinks per day Lactose Intolerance (due to the study involving administration of a lactose placebo tablet)

An anonymised dataset will be published as open access data on a secure online repository, such as Open Science Framework (https://osf.io/). Participant IDs will be randomly reassigned in this dataset to ensure complete removal of any linkage between anonymised and personal data. Self-report questionnaire data and task outcomes will be included.

An anonymised dataset will be shared after full anonymisation of study data and publication of findings. Data will be available indefinitely.