Person:

Enock, Philip M.

Testing feasibility and efficacy of psychological treatment via mobile devices is important, given its potential benefits for high-dosage treatment delivery, widespread and inexpensive dissemination, and efficient research methods. We conducted the first randomized controlled trial of attention bias modification training delivered via smartphones, comparing this training to control training in a double-blind design, also including a waitlist condition. All participants performed a variant of dot-probe training involving faces with neutral and disgust (representative of social threat) expressions in brief sessions three times daily over 4 weeks on their own smartphones, at home or anywhere they chose. Attention bias modification, also known as cognitive bias modification of attention, training included a contingency to induce attentional deployment away from disgust faces, whereas the control training included no contingency. Participants completed weekly Internet-based self-report symptom assessments as well as smartphone-delivered dot-probe attention bias assessments, whose reliability findings supported the viability of using smartphones for reaction-time based assessments. The between-groups training effect on attention bias scores was small, showing statistical significance in some analyses and not in others. On measures ofsocial anxiety, intention-to-treat analyses (n = 326) revealed significant pre–post treatment declines with medium to large effect sizes in both training groups, whereas small declines in a waitlist group were nonsignificant. Both training groups showed greater reductions in social anxiety than did waitlist; however, the benefits under these two training conditions were statistically indistinguishable. Improvements in the two training conditions beyond those of waitlist could be attributable to any factors common to them, but not to the contingency training specific to active attention bias modification training.

Development of novel candidate interventions to treat anxiety disorders is an important research priority, given the burden of these disorders, barriers to treatment access, and the promising but limited success of current approaches, including attentional bias modification treatment. I created a novel training game paradigm, Intrinsically-Motivating Playable Attentional Control Training (IMPACT), with several potential ways that its design could increase the strength of attentional change and commensurate clinical benefits beyond existing training methods.

In a large online experiment, I randomized participants among three alternative IMPACT training conditions. All involved the same smiling and disgust faces falling down on the screen, and players tapped faces to score points and prevent them from reaching the bottom. In IMPACT-Positive, players tapped smiling faces only, ignoring disgust faces. In IMPACT-Threat, players tapped disgust faces, ignoring smiling faces. In IMPACT-Undirected, players tapped all faces without regard to expression. After training, participants completed flanker tasks, reaction-time measures of general and emotional attentional control and attentional bias toward threat versus neutral stimuli. Participants also confronted an anxiety-provoking stressor and rated their state anxiety before IMPACT, after IMPACT, and after the stressor.

I tested hypotheses regarding differential effects of the training variants on attentional measures and anxiety reactivity, finding that training did not cause group differences in measures of general or emotional attentional control, but they did lead to differences in attentional bias. The anxiety-provoking stressor induced a rapid rise in anxiety, but no differences emerged among the training conditions.

Overall, results show the potential for researchers to abandon the tradition of repeated reaction-time trials in favor of engaging, fluid games that continuously motivate trainees and prompt attentional shifts. Additional testing of the IMPACT paradigm is needed to establish whether this particular game training approach is clinically useful for reducing anxiety.