Trial Outcomes & Findings for Mobile Attention Retraining in Overweight Female Adolescents (NCT NCT02977403)

NCT ID: NCT02977403

Last Updated: 2025-05-31

Results Overview

AB was obtained for each stimulus pairing (High-Palatability Food \[HPF\] minus Non-Food \[NF\] image, Low-Palatability Food \[LPF\] minus NF image, HPF minus LPF image). Trials where the probe appeared behind the more food-salient cue (e.g., a HPF image, or LPF vs NF image) were considered congruent trials. Trials where the probe appeared behind the less salient cue (e.g., NF image, or LPF image when the other image was a HPF image) were considered incongruent trials. The average reaction time during incongruent trials was subtracted from reaction time during during congruent trials. Positive scores represent a quicker reaction time for (and bias towards) the more palatable stimulus, and negative scores represent a slower reaction time for (and bias away from) the more palatable stimulus. A difference score of 0 represents no bias towards or away from the more palatable stimulus. Only trials with correct responses for the direction of the probe were included in computations.

• Recruitment status: COMPLETED
• Study phase: NA
• Target enrollment: 82 participants
• Primary outcome timeframe: 2-weeks
• Results posted on: 2025-05-31

Participant Flow

Participants were recruited from 2/10/2017 to 4/27/2023 from the local community by advertisements sent by mail and posted flyers.

Participant milestones

Measure	AB Control Control condition - the probe is equally likely to replace the food picture and the neutral picture. There is no correlation between picture type and probe location, and no training of attention should occur. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	AB Retraining Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.
Overall Study STARTED	36	32
Overall Study COMPLETED	29	29
Overall Study NOT COMPLETED	7	3

Reasons for withdrawal

Withdrawal data not reported

Baseline Characteristics

Mobile Attention Retraining in Overweight Female Adolescents

Baseline characteristics by cohort

Measure	AB Control n=36 Participants Control condition - the probe is equally likely to replace the food picture and the neutral picture. There is no correlation between picture type and probe location, and no training of attention should occur. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	AB Retraining n=32 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Total n=68 Participants Total of all reporting groups
Age, Continuous	14.9 years STANDARD_DEVIATION 1.7 • n=5 Participants	15.0 years STANDARD_DEVIATION 1.6 • n=7 Participants	14.9 years STANDARD_DEVIATION 1.6 • n=5 Participants
Sex: Female, Male Female	36 Participants n=5 Participants	32 Participants n=7 Participants	68 Participants n=5 Participants
Sex: Female, Male Male	0 Participants n=5 Participants	0 Participants n=7 Participants	0 Participants n=5 Participants
Ethnicity (NIH/OMB) Hispanic or Latino	6 Participants n=5 Participants	3 Participants n=7 Participants	9 Participants n=5 Participants
Ethnicity (NIH/OMB) Not Hispanic or Latino	27 Participants n=5 Participants	27 Participants n=7 Participants	54 Participants n=5 Participants
Ethnicity (NIH/OMB) Unknown or Not Reported	3 Participants n=5 Participants	2 Participants n=7 Participants	5 Participants n=5 Participants
Race (NIH/OMB) American Indian or Alaska Native	0 Participants n=5 Participants	0 Participants n=7 Participants	0 Participants n=5 Participants
Race (NIH/OMB) Asian	1 Participants n=5 Participants	0 Participants n=7 Participants	1 Participants n=5 Participants
Race (NIH/OMB) Native Hawaiian or Other Pacific Islander	0 Participants n=5 Participants	0 Participants n=7 Participants	0 Participants n=5 Participants
Race (NIH/OMB) Black or African American	19 Participants n=5 Participants	18 Participants n=7 Participants	37 Participants n=5 Participants
Race (NIH/OMB) White	14 Participants n=5 Participants	8 Participants n=7 Participants	22 Participants n=5 Participants
Race (NIH/OMB) More than one race	2 Participants n=5 Participants	5 Participants n=7 Participants	7 Participants n=5 Participants
Race (NIH/OMB) Unknown or Not Reported	0 Participants n=5 Participants	1 Participants n=7 Participants	1 Participants n=5 Participants
Region of Enrollment United States	36 participants n=5 Participants	32 participants n=7 Participants	68 participants n=5 Participants
Fat Mass	33.6 Kg STANDARD_DEVIATION 11.3 • n=5 Participants	33.8 Kg STANDARD_DEVIATION 10.4 • n=7 Participants	33.7 Kg STANDARD_DEVIATION 10.8 • n=5 Participants
Body Mass Index Z-score	1.88 Z-score STANDARD_DEVIATION 0.64 • n=5 Participants	1.80 Z-score STANDARD_DEVIATION 0.55 • n=7 Participants	1.84 Z-score STANDARD_DEVIATION 0.60 • n=5 Participants
Recent Loss Of Control Eating	11 Participants n=5 Participants	9 Participants n=7 Participants	20 Participants n=5 Participants

PRIMARY outcome

Timeframe: 2-weeks

Population: Computer failures led to missing data for many participants. There were interruptions in data recording during magnetoencephalography sessions that were not predictable due to equipment failures.

AB was obtained for each stimulus pairing (High-Palatability Food [HPF] minus Non-Food [NF] image, Low-Palatability Food [LPF] minus NF image, HPF minus LPF image). Trials where the probe appeared behind the more food-salient cue (e.g., a HPF image, or LPF vs NF image) were considered congruent trials. Trials where the probe appeared behind the less salient cue (e.g., NF image, or LPF image when the other image was a HPF image) were considered incongruent trials. The average reaction time during incongruent trials was subtracted from reaction time during during congruent trials. Positive scores represent a quicker reaction time for (and bias towards) the more palatable stimulus, and negative scores represent a slower reaction time for (and bias away from) the more palatable stimulus. A difference score of 0 represents no bias towards or away from the more palatable stimulus. Only trials with correct responses for the direction of the probe were included in computations.

Outcome measures

Measure	AB Retraining n=22 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=21 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Changes in Food-cue Visual Probe Task Attention Bias (AB) Reaction Time	1.85 Change in milliseconds Interval -8.44 to 12.13	4.27 Change in milliseconds Interval -6.25 to 14.8

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power at the caudate left hemisphere during attention capture (0-250ms following stimulus). Oscillatory power was normalized as per NeuroImage 39 (2008) pp 1788-1802, by estimating noise power as ρθ = WθTΣWθ (where Wθ is a (M × 1) column vector of weighting parameters that are tuned specifically to the location and orientation represented by θ, Σ represents the noise covariance matrix and ρθ is the beamformer-projected sensor noise power at the location and orientation θ). Within each stimuli-pairing and attention phase, oscillatory power during the incongruent trials was divided by oscillatory power during the congruent trials, then log transformed. Given a ratio was used, the oscillator power outcomes are unitless. Change in power (post-intervention minus pre-intervention) was calculated. Positive changes represent an increase in oscillatory power from pre- to post intervention.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudate Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.010 unitless Interval -0.027 to 0.048	.021 unitless Interval -0.014 to 0.056

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudate right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudate Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.032 unitless Interval -0.011 to 0.075	0.023 unitless Interval -0.017 to 0.063

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pallidum left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pallidum Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.007 unitless Interval -0.055 to 0.041	0.038 unitless Interval -0.006 to 0.083

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pallidum right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pallidum Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.028 unitless Interval -0.022 to 0.078	0.009 unitless Interval -0.037 to 0.056

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the putamen left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Putamen Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.013 unitless Interval -0.056 to 0.03	0.046 unitless Interval 0.006 to 0.086

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the putamen right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Putamen Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.017 unitless Interval -0.029 to 0.063	0.020 unitless Interval -0.022 to 0.063

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudal anterior cingulate cortex left hemisphere - during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudal Anterior Cingulate Cortex Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.035 unitless Interval -0.009 to 0.078	-0.041 unitless Interval -0.081 to -0.001

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudal anterior cingulate cortex right hemisphere - during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudal Anterior Cingulate Cortex Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.053 unitless Interval 0.011 to 0.095	0.002 unitless Interval -0.037 to 0.041

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the rostral anterior cingulate cortex left hemisphere - during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Rostral Anterior Cingulate Cortex Left Hemisphere - During Attention Capture (0-250ms Following Stimulus)	0.012 unitless Interval -0.03 to 0.055	0.012 unitless Interval -0.027 to 0.051

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the rostral anterior cingulate cortex right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Rostral Anterior Cingulate Cortex Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.002 unitless Interval -0.034 to 0.039	-0.024 unitless Interval -0.058 to 0.01

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Neural activity during a food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the lateral orbitofrontal cortex left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Lateral Orbitofrontal Cortex Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.010 unitless Interval -0.05 to 0.03	0.026 unitless Interval -0.011 to 0.063

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the lateral orbitofrontal cortex right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Lateral Orbitofrontal Cortex Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.026 unitless Interval -0.065 to 0.014	-0.001 unitless Interval -0.037 to 0.036

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the medial orbitofrontal cortex left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Medial Orbitofrontal Cortex Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.010 unitless Interval -0.048 to 0.029	0.018 unitless Interval -0.017 to 0.054

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the medial orbitofrontal cortex right hemisphere during attention capture (0-250ms following stimulus).The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Medial Orbitofrontal Cortex Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.018 unitless Interval -0.053 to 0.018	-0.011 unitless Interval -0.044 to 0.022

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudal dorsolateral prefrontal cortex left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudal Dorsolateral Prefrontal Cortex Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.026 unitless Interval -0.006 to 0.058	0.020 unitless Interval -0.01 to 0.049

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudal dorsolateral prefrontal cortex right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudal Dorsolateral Prefrontal Cortex Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.027 unitless Interval -0.01 to 0.065	0.031 unitless Interval -0.003 to 0.066

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the rostral dorsolateral prefrontal cortex left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Rostral Dorsolateral Prefrontal Cortex Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.006 unitless Interval -0.04 to 0.029	0.033 unitless Interval 0.001 to 0.066

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the rostral dorsolateral prefrontal cortex right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Rostral Dorsolateral Prefrontal Cortex Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.014 unitless Interval -0.018 to 0.046	0.008 unitless Interval -0.021 to 0.038

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the superior dorsolateral prefrontal cortex left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Superior Dorsolateral Prefrontal Cortex Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.021 unitless Interval -0.004 to 0.046	-0.013 unitless Interval -0.036 to 0.01

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the superior dorsolateral prefrontal cortex right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Superior Dorsolateral Prefrontal Cortex Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.041 unitless Interval 0.017 to 0.064	0.019 unitless Interval -0.002 to 0.041

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars opercularis left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Opercularis Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.010 unitless Interval -0.048 to 0.029	0.069 unitless Interval 0.034 to 0.105

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars opercularis right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Opercularis Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.037 unitless Interval -0.007 to 0.08	0.031 unitless Interval -0.009 to 0.071

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars orbitalis left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Orbitalis Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.019 unitless Interval -0.06 to 0.023	0.019 unitless Interval -0.019 to 0.058

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars orbitalis right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Orbitalis Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.011 unitless Interval -0.053 to 0.03	0.016 unitless Interval -0.023 to 0.055

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars triangularis left hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Triangularis Left Hemisphere During Attention Capture (0-250ms Following Stimulus)	-0.015 unitless Interval -0.059 to 0.029	0.064 unitless Interval 0.024 to 0.105

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars triangularis right hemisphere during attention capture (0-250ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Triangularis Right Hemisphere During Attention Capture (0-250ms Following Stimulus)	0.022 unitless Interval -0.02 to 0.063	0.016 unitless Interval -0.023 to 0.054

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the the caudate left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudate Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.012 unitless Interval -0.025 to 0.049	-0.052 unitless Interval -0.086 to -0.017

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudate right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudate Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.030 unitless Interval -0.013 to 0.073	-0.031 unitless Interval -0.07 to 0.009

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pallidum left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pallidum Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	-0.011 unitless Interval -0.058 to 0.036	-0.051 unitless Interval -0.095 to -0.008

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pallidum right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pallidum Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.037 unitless Interval -0.009 to 0.084	-0.026 unitless Interval -0.069 to 0.017

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the putamen left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Putamen Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.0005 unitless Interval -0.04 to 0.041	-0.062 unitless Interval -0.1 to -0.025

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the putamen right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Putamen Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.018 unitless Interval -0.024 to 0.061	-0.031 unitless Interval -0.07 to 0.009

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudal anterior cingulate cortex left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudal Anterior Cingulate Cortex Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.026 unitless Interval -0.012 to 0.064	-0.069 unitless Interval -0.104 to -0.033

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudal anterior cingulate cortex right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudal Anterior Cingulate Cortex Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.058 unitless Interval 0.011 to 0.105	-0.028 unitless Interval -0.072 to 0.016

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the rostral anterior cingulate cortex left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Rostral Anterior Cingulate Cortex Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	-0.005 unitless Interval -0.048 to 0.038	-0.030 unitless Interval -0.07 to 0.01

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the rostral anterior cingulate cortex right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Rostral Anterior Cingulate Cortex Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.032 unitless Interval -0.01 to 0.074	-0.012 unitless Interval -0.051 to 0.027

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the lateral orbitofrontal cortex left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Lateral Orbitofrontal Cortex Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.021 unitless Interval -0.022 to 0.063	-0.062 unitless Interval -0.102 to -0.023

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the lateral orbitofrontal cortex right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Lateral Orbitofrontal Cortex Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.012 unitless Interval -0.029 to 0.052	0.002 unitless Interval -0.035 to 0.04

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the medial orbitofrontal cortex left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Medial Orbitofrontal Cortex Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.018 unitless Interval -0.023 to 0.058	-0.038 unitless Interval -0.076 to -0.001

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the medial orbitofrontal cortex right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Medial Orbitofrontal Cortex Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.041 unitless Interval 0.001 to 0.08	-0.009 unitless Interval -0.046 to 0.028

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudal dorsolateral prefrontal cortex left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudal Dorsolateral Prefrontal Cortex Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.016 unitless Interval -0.024 to 0.056	-0.012 unitless Interval -0.049 to 0.025

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the caudal dorsolateral prefrontal cortex right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Caudal Dorsolateral Prefrontal Cortex Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.005 unitless Interval -0.041 to 0.05	0.006 unitless Interval -0.036 to 0.048

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the rostral dorsolateral prefrontal cortex left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Rostral Dorsolateral Prefrontal Cortex Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.015 unitless Interval -0.017 to 0.047	-0.031 unitless Interval -0.061 to -0.002

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the rostral dorsolateral prefrontal cortex right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Rostral Dorsolateral Prefrontal Cortex Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.010 unitless Interval -0.024 to 0.044	-0.006 unitless Interval -0.037 to 0.026

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the superior dorsolateral prefrontal cortex left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Superior Dorsolateral Prefrontal Cortex Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.004 unitless Interval -0.025 to 0.033	-0.022 unitless Interval -0.049 to 0.005

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the superior dorsolateral prefrontal cortex right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Superior Dorsolateral Prefrontal Cortex Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.023 unitless Interval -0.009 to 0.055	0.010 unitless Interval -0.02 to 0.039

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars opercularis left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Opercularis Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.043 unitless Interval 0.003 to 0.082	-0.054 unitless Interval -0.09 to -0.017

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars opercularis right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Opercularis Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.001 unitless Interval -0.037 to 0.039	-0.018 unitless Interval -0.053 to 0.017

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars orbitalis left hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Orbitalis Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.013 unitless Interval -0.026 to 0.053	-0.052 unitless Interval -0.089 to -0.016

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars orbitalis right hemisphere during attention deployment (250-500ms following stimulus).The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Orbitalis Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	-0.027 unitless Interval -0.067 to 0.012	0.023 unitless Interval -0.014 to 0.06

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars triangularis left hemisphere during attention deployment (250-500ms following stimulus).The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Triangularis Left Hemisphere During Attention Deployment (250-500ms Following Stimulus)	0.023 unitless Interval -0.015 to 0.062	-0.071 unitless Interval -0.107 to -0.036

PRIMARY outcome

Timeframe: 2-weeks

Population: Equipment failure meant many participants did not have data for this analysis

Change in beta band (13-35 Hz) oscillatory power during food-cue visual probe attention bias task completed at the baseline laboratory visit vs. post-EMA intervention visit (conducted 2 weeks after the baseline visit) at the pars triangularis right hemisphere during attention deployment (250-500ms following stimulus). The same analysis procedure was followed as described in detail for the first primary outcome.

Outcome measures

Measure	AB Retraining n=12 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=14 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Change in Beta Band (13-35 Hz) Oscillatory Power During Food-cue Visual Probe Attention Bias Task in the Pars Triangularis Right Hemisphere During Attention Deployment (250-500ms Following Stimulus)	-0.009 unitless Interval -0.055 to 0.037	0.013 unitless Interval -0.03 to 0.056

SECONDARY outcome

Timeframe: 2-weeks

Population: Some subjects withdrew before completion of the post-smartphone visit.

Frequency of self-reported loss-of-control eating episodes measured via the Eating Disorder Examination Interview at the baseline visit and post-EMA intervention visit (conducted 2 weeks after the baseline visit).

Outcome measures

Measure	AB Retraining n=26 Participants Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=26 Participants Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Frequency of Loss-of-control Eating Episodes	.731 eating episode (count) Standard Deviation 2.3189	.827 eating episode (count) Standard Deviation 1.968

Adverse Events

AB Retraining

Serious events: 0 serious events

Other events: 4 other events

Deaths: 0 deaths

Control Sham

Serious events: 0 serious events

Other events: 2 other events

Deaths: 0 deaths

Serious adverse events

Adverse event data not reported

Other adverse events

Measure	AB Retraining n=32 participants at risk Active treatment - the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location. Attention Bias Retraining: Attention retraining program on smartphone where the probe always replaces the neutral picture. There is a perfect correlation between picture type and probe location.	Control Sham n=36 participants at risk Sham Comparator program - the probe randomly replaces the neutral or food picture. There is no correlation between picture type and probe location Sham Comparator: AB Control: Sham Comparator "training" where the probe randomly replaces the neutral or food pictures. There is no correlation between picture type and probe location
Musculoskeletal and connective tissue disorders Headache	3.1% 1/32 • Number of events 1 • 2 weeks	2.8% 1/36 • Number of events 1 • 2 weeks
Gastrointestinal disorders Nausea	9.4% 3/32 • Number of events 3 • 2 weeks	0.00% 0/36 • 2 weeks
Musculoskeletal and connective tissue disorders Neck Pain	0.00% 0/32 • 2 weeks	2.8% 1/36 • Number of events 1 • 2 weeks

Additional Information

Dr. Jack Yanovski

Eunice Kennedy Shriver National Institute of Health, NIH

Phone: 301-496-0858

Email: jy15i@nih.gov

Results disclosure agreements

• Principal investigator is a sponsor employee
• Publication restrictions are in place