Hostile interpretation bias may be a feature of severe, chronic irritability in children, one of the most common psychiatric symptoms of childhood. Interpretation bias training (IBT) is a computer-based training program that may reduce irritability in youths. This trial lays the groundwork for a test IBT on irritability.
This trial lays the foundation for a preliminary test of efficacy of IBT on irritability by establishing IBT's neurocognitive treatment targets: hostile interpretation bias and response in the neural threat-learning system. The design is a single-blinded, randomized controlled trial of IBT on its targets. The study will have four arms, with 25 participants in each arm for all four conditions of training (active versus sham) and scanning (in MRI or out of MRI scanner). During IBT, participants judge as happy or angry facial expressions which are on a continuum between happy and angry. The point at which judgments shift from predominantly happy to angry on this continuum is the indifference point. During training feedback encourages no change in the indifference point or a change in the indifference point towards more happy judgments of ambiguous faces. A shift in indifference point towards more benign judgments is interpreted as a reduction in hostile interpretation. The design will test whether active relative to sham IBT shifts the indifference point towards more benign judgments. Neural response to active versus sham IBT will be measured in half the sample.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
BASIC_SCIENCE
Masking
SINGLE
Enrollment
97
Interpretation bias training
University of Colorado, School of Medicine
Aurora, Colorado, United States
Change in Interpretation Bias
Interpretation bias is measured by the indifference point of happy-angry judgments of a continuum of 15 images of face emotion expression that are morphed between completely happy=1 to completely angry=15 facial expressions. Indifference point is represented on this scale of 1 to 15 faces. It represents the point on the face continuum where judgements change from predominantly happy to predominantly angry. Indifference points have a minimum value of 1 and a maximum value of 15, though are typically in the middle of the face-emotion continuum, e.g. 7. Lower indifference points represent a higher bias towards hostile (or angry) judgments. Change in interpretation bias is the difference between the post- and pre-training indifference points. This can range from -14 to 14, with positive numbers reflecting the degree to which the happy/angry decision bias is shifted towards more happy judgements. High positive numbers reflect a greater increase in happy judgments.
Time frame: Up to one day before and after the single 1 day training session of Interpretation bias training session.
Neural Response in Threat Learning System.
Amygdala functional connectivity to ventromedial prefrontal cortex is measured by covariance in fluctuations in blood oxygen-level dependent signal between these regions during functional magnetic resonance imaging (fMRI) during the course of training. The outcome measure is a count of these randomized participants for whom the fMRI measure is valid for the planned connectivity analysis. These participants completed the task in scanner with acceptable behavioral data and fMRI data. Quality control for fMRI included visual inspection of brain images for quality, coverage, and proper alignment across functional, anatomic and template images. The fMRI images passed analysis benchmarks of \<20% censored volumes and \<0.25 mm Euclidean distance average censored motion across the task. Censor thresholds are 0.3 mm Euclidean distance and 5% voxels with outlying intensity values.
Time frame: During the course of the 1 day single session of interpretation bias training session.
Learning Rate
Learning rate as measured by a reinforcement learning computational model during IBT, please see the reference Stoddard et al., 2023, for model details. As in general reinforcement learning models, it is the proportion of prediction error that is applied to future expectations. In this trial, when a participant received feedback about a judgement they made of a face, the participant implicitly compares that feedback to their expectations about that face. The difference between the feedback and their expectations is called prediction error. The learning rate the proportion of this prediction error that is applied to future expectations across the learning session. In that way it represents learning. Learning rate is a continuous, unitless value. It ranges from 0, representing no learning, to 1, representing perfect learning. It is generally small. The learning rate reported here is its form estimated across face emotions (see Stoddard et al., 2023).
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Time frame: During the course of the 1 day single interpretation bias training session.
Irritability
Irritability is the tendency towards anger and temper outburst as measured by the Affective Reactivity Index, a seven item scale. Each item has 3 response options which are coded as follows: "not true"=0, "somewhat true"=1, and "certainly true"=2. The first six items are summed to generate a total score that reflects the degree over irritability over the past week. The Affective Reactivity Index ranges from 0 to 12. Higher values reflect more irritability. What is reported here is the average post- minus pre-treatment score, by parent report. In these differences, negative values are in the unit of this scale and the absolute value represents the degree of irritability reduction.
Time frame: Within one week prior to Interpretation bias training and one week +/- 3 days after Interpretation bias training.