Consumer Virtual Reality (VR) technology offers a powerful, immersive medium for scalable dissemination of mental health interventions. Decades of research has shown VR exposure therapy to be efficacious in the treatment of anxiety disorders and that the fear reduction generalizes to real-world stimuli. Many studies also report continued improvement over time, after discontinuing VR use. The lowered threshold hypothesis states that this continued improvement is moderated by lowering the threshold to conduct subsequent in-vivo exposure. The current study is the first to formally test this hypothesis, using data from a recent trial on automated VR exposure therapy for spider phobia, in which participants (n = 49) were followed for 1 year, completing assessments 1 week, 3 and 12 months post-treatment. The assessment included validated self-report of phobia symptoms, a standardized behavioral approach test featuring a real spider, and a questionnaire for self-reporting frequency of in-vivo exposures since last assessment. Number of in-vivo exposures was found to be independently associated with greater symptom decrease in longitudinal outcome models. In sequential structural equation models, immediate post-treatment symptom reduction was associated with subsequent in-vivo exposures, which in turn was associated with continued symptom reduction. However, this applied only to self-reported phobia symptoms (not behavioral avoidance) and no associations were found past 3 months. Our findings offer preliminary, partial support for the lowered threshold hypothesis, suggesting that VR exposure interventions may benefit from including explicit in-virtuo to in-vivo transitioning components.