Modif-i Tobacco Trial: A clinical trial of the efficacy of a “brain-training” smartphone app to assist tobacco cessation

The intervention will be delivered using the app "Modif-i", which has been designed specifically for the trial. When participants download Modif-i, the app will prompt them to answer questions about past-week and past-month tobacco use and then complete a short (~3-minute) test of their approach/avoidance bias to smoking-related and non-smoking related images. Following these tasks, instructions for the training task will be displayed and participants will be prompted to complete their first training task. Modif-i delivers a cognitive bias modification (CBM) training task in which participants respond to tobacco and non-tobacco images that appear on their smartphone screen. Each image presentation is surrounded by a white "frame" that is either in portrait or landscape orientation. Participants are instructed to swipe images "away" (by swiping their finger upwards, causing the image to shrink and disappear) if images are displayed surrounded by a portrait frame, or towards themselves (causing the image to expand) if surrounded by a landscape-oriented frame. Each training session lasts approximately 3-5 minutes. Modif-i will remind participants (using app notifications) to complete training sessions regularly for 4 weeks. At the end of each week, it will also prompt them to answer questions about past-week tobacco use. After the end of the 4th week, it will also prompt them to complete the test of approach/avoidance bias for a second time. After the 4-week intervention period, participants will have free access to the training task, but will not receive further notifications from the app reminding them to use it. Participants' adherence to the intervention will be assessed using app data regarding the number of training sessions completed and the times they were completed at. To maintain blinding, additional details of the task are not currently made available to the public. However a full protocol including all details will be uploaded when data collection is complete.

The intervention will be delivered using the app "Modif-i", which has been designed specifically for the trial. When participants open Modif-i for the first time, they will be shown a short animated video which explains that “cognitive bias modification” is designed to train people to avoid tobacco and approach positive, healthy goals and activities. Prior to commencing the intervention, participants will be prompted to select 6 tobacco-related pictures that most strongly remind them of their tobacco use. Participants can either select from their own photographs, if they have photographs of tobacco products or paraphernalia stored on their phone; take photographs of tobacco products/paraphernalia using the app (by allowing the app access to their phone’s camera); or select from a library of tobacco-related images provided in the app. Participants will then be prompted to select 6 pictures that “represent your goals and motivations”. Again, participants can either select their own photographs, take photographs, or select from a library of images representing a range of healthy activities or positive goals and sources of pleasure (e.g., family or friends enjoying time together; financial success; employment; exercise, sports, and recreational activities; healthy foods; pets; travel and holidays) which do not contain any depictions of tobacco. If participants take/use their own photographs, they will be reminded to ensure that positive images do not contain depictions of tobacco. Once the participant selects their 12 images, they will complete a pre-training craving rating before they are presented with instructions for the ApBM task. Images (200 x 200 pixels) will be displayed with a white “frame” around them which will be in either landscape or portrait orientation. When the frame is in landscape orientation, the participant is required to swipe downwards (i.e., towards themself), which (if the movement of their finger on the screen traverses at least 40 pixels downward) causes the image to expand, increasing to 600 x 600 pixels over a period of 500 milliseconds (ms), as if the participant has “pulled” the image “towards” themselves. When the frame is in portrait orientation, the participant is instructed to swipe upwards (i.e., away from themself), which (if the finger’s movement is at least 40 pixels upwards) causes the image to shrink until it disappears over a period of 500 ms, as if they have “pushed” it “away”. If the participant swipes (at least 40 pixels) in the wrong direction, a red “X” is displayed for 500 ms to inform them that they made an error. After the image expands/shrinks to its maximum/minimum extent (or a red X is displayed in the case of erroneous responses), there is a 250 ms inter-stimulus interval during which the screen is blank before presentation of the next image. Following the display of the instructions, participants complete 10 practise trials (including 5 images in portrait frames and 5 in landscape frames, in random order) to familiarise themselves with the task requirements before commencing the main series of training trials. The training session then consists of 156 trials, comprising 13 presentations of each image. For each tobacco image, 12 of the 13 presentations are framed in portrait orientation, and one is framed in landscape orientation. This is reversed for positive images, such that 12 of the 13 presentations of each positive image are framed in landscape orientation, while one is framed in portrait orientation. Thus, participants are supposed to push away 92.3% of tobacco images and pull 92.3% of positive images towards themselves. If participants make the incorrect response, a red “X” is displayed, but the trial is not repeated. To increase engagement and encourage participants to respond both quickly and accurately, the task is gamified with a scoring system. Each time the participant swipes an image in the correct direction, they are awarded points. If they swipe correctly and within 500 milliseconds (ms) of image onset, they receive 40 points for that trial. If they swipe correctly within 501-1000 ms of image onset they receive 30 points. If they respond correctly within 1001-1500ms they receive 20 points. Correct responses that are slower than 1500 ms earn only 10 points. If they swipe an image incorrectly (i.e., swipe down for portrait or swipe up for landscape), they lose 100 points, regardless of their reaction time. Upon completion of the task, the final point score is displayed. On the second and subsequent sessions, their final score for that session will be shown alongside their previous personal best score, so they can compare their performance. On the second and subsequent sessions, participants are offered the option to either review the task instructions and complete the 10 practise trials, or to skip these steps and commence the training immediately. Each session typically takes approximately 3-5 minutes. Participants are prompted by app notifications to complete at least 2 sessions of the training task each week for 4 weeks. After the 4-week intervention period, participants continue to have free access to the training task, but will no longer receive notifications reminding them to complete it. Participants' adherence to the intervention will be assessed using app data regarding the number of training sessions completed and the times they were completed at.