Panic-relevant predictability preferences: a laboratory test.

In the present investigation the authors evaluated the role of verbally instructed safety periods as they relate to predictability preference within a single-session challenge paradigm involving recurrent administrations of 20% carbon dioxide (CO-sub-2) enriched air. Participants included 160 nonclinical young adults (91 women, 69 men) between the ages of 18 and 59 (M=21.6 years; SD=7.23). Results indicated that equivalent levels of anxiety were experienced during predictable and unpredictable administrations of 20% CO-sub-2, yet participants preferred predictable compared with unpredictable trials, with women showing a stronger predictability preference than men. Lower anxiety was reported during known safety periods (predictable room air trials) but not during predictable compared with unpredictable administrations of 20% CO-sub-2. Findings are discussed in relation to theoretical and applied implications for the role of predictability in the nature of anxiety and its disorders.