Evaluating a Novel Mobile App for Social Cognition in Psychosis

Evaluating the Feasibility and Acceptability of a Novel Mobile Intervention Targeting Social Cognition in Individuals With Psychosis

To develop, and then evaluate a mobile phone app to deliver therapy homework activities between group sessions (social cognition intervention) in individuals with psychosis. The investigators are interested in whether offering homework via an app is a) feasible, and b) acceptable. The investigators will also assess whether there is an initial indication that offering homework via the app improves outcomes following the group therapy.

This study is a pilot, feasibility trial of a clinical intervention. The investigators will initially pilot the app with three clinical participants, who will be asked to use the app for a period of three days and provide feedback as to whether there are any difficulties with usability, glitches, display, etc. Two therapy groups will then be run consecutively. The therapy delivered will be a modified version of Group Training for Social Cognition in Psychosis (GRASP). Both groups will receive their homework tasks delivered via an app. Participants will undergo an assessment before and after the therapy, and a follow up interview as detailed in the measures section below. Primary objective: To evaluate a mobile app to deliver therapy homework between sessions of a social cognition therapy group for individuals with psychosis. The investigators are interested in whether offering homework via an app is a) feasible, and b) acceptable. Secondary objective: To assess whether there is an initial indication that our intervention improves social cognition skills.

Each session lasts 90 minutes and includes three 20-minute work packages tapping different social cognition skills; specifically affect recognition and theory of mind/mental state attribution in this study. Examples of the work packages include guessing the emotions of others based on their facial expressions, activities such as indicating how confident participants feel about guessing emotions when looking at more ambiguous pictures, and watching/discussing video clips with content relevant to the group.

The novel homework app was designed to deliver tasks to the participants between group therapy sessions. Three types of task were delivered by the app; 'Stories', 'Emotions', and 'Facts and Guesses'. The 'Stories' task was designed to encourage participants to think about how thoughts, emotions and actions interact. Participants were presented with a short vignette and asked to identify the emotion or behaviour of a character. 'Emotions' aimed to target affect perception by presenting faces of different emotion presentations and asking clients to select the correct emotion from a list of three, whilst 'Facts and Guesses' aimed to address difficulties with jumping to conclusions in social situations by showing clients a photograph of a social situation and asking them to determine whether a given statement about the photo was a "fact" or a "guess". Confidence in answers was also assessed in order to encourage flexible thinking.

Measures of affect recognition. . Participants are invited to watch 21 videos lasting ten seconds each, of a male actor displaying seven different emotions: happy, sad, afraid, surprised, angry, disgusted or no emotion, and then identify which emotion they think it being portrayed. This is the emotion perception measure recommended by the Social Cognition Psychometric Evaluation (SCOPE) group for studies evaluating social cognition interventions

Measure of attributional bias. Participants are read five, hypothetical negative situations with ambiguous causes, asked to imagine that this is happening to them, and then come up with an explanation for how the situation occurred. Participants are asked whether the character in the story was acting purposefully, how angry they feel, and how much they blame the character, using Likert type responses. This measure takes approximately 6 minutes to administer.

Measure of Theory of Mind. The participants are read ten short passages portraying an interaction between two people, at the end of which one character drops a hint. The participant is asked what the character really meant, as a way to assess how well they can infer meaning from indirect speech. This is the measure for theory of mind/mental state attribution recommended by SCOPE. This measure takes approximately 6 minutes to administer.

Measure of empathising abilities/emotional intelligence. This can be viewed as a more 'global' measure of social abilities as opposed to domain specific (i.e. emotion perception, theory of mind). This is a 22-item self-report measure, and takes approximately 6 minutes to administer.

Measure of Community Functioning. This is a five item self-report measure which asks participants how much their mental health problems effect their ability to work, manage their home, engage in leisure activities and have social relationships, using a nine-point scale (from 0 - not at all, to 8- very much so). Additionally, participants are asked how likely they feel this is to change in the future using the same scale.

Inclusion Criteria: A diagnosis on the psychosis spectrum (using either DSM 5, DSM-IV, or ICD-10 [F20-F29]) Age 18-65 years Good command of the English language Premorbid IQ of over 70 Owns a smartphone or willing to use a study smartphone for the required period Likely to benefit from social cognition intervention as assessed by clinical team [for intervention part]. Exclusion Criteria: Lacks capacity to give informed consent to participate in research Poses significant risk to self or others Inability to understand verbal or written English (i.e. inability to understand information sheet or requires an interpreter). High levels of psychotic symptoms which precludes meaningful participation Unsuitable to attend groups (group-therapy part only)