The Effects of Emotional Exposure on State Anxiety

The design was a repeated-measures randomized crossover trial. Each study participant completed 4 different sessions on 4 different days: (day 1) participant screening; (day 2 or 3) study condition A (seated, quiet rest); (alternate day 2 or 3) study condition B (moderate-intensity YogaFit); and (day 4) a session to rate IAPS pictures (14) and undergo a single DEXA scan [Prodigy/301776R. G.E. Medical Systems-Lunar DEXA Scanner, Fairfield, CT, USA]. Utilizing a simple randomization process, participants were randomized in a 1:1 ratio to participate in either the YogaFit condition or the quiet rest condition. The study conditions were counterbalanced and alternated using odd and even numbers to begin the sequence. Participants were not informed if the first condition was A or B. Self-reported anxiety levels and 10 min of continuous physiological data were collected: (a) baseline; (b) post-condition; and (c) post-exposure for both condition A and B. All participants completed both the psychological and physiological assessments. Both experimental conditions A and B were completed in 90 min. To ensure compliance, participants were observed through a glass window every 10 min during study conditions. Testing time and procedures were identical with the exception of the experimental condition on days two and three.

All participants were familiarized with the videotaped YogaFit session before participating in the study. Participants were instructed to arrive for all visits dressed in loose fitting athletic wear and were asked to refrain from all food, alcohol, caffeine, gum chewing or the use of tobacco products 3 h before their scheduled meeting in the laboratory.

During pre-screening, participants completed: (a) a standard health history questionnaire; (b) the STAI-Y2; (c) the BDI and (d) a custom DEXA safety questionnaire. Qualified participants were then asked to track their menstrual cycle using a menstrual log and e-mail the primary investigator on day 1 of the follicular phase of their menstrual cycle, thus using the first day of the women's menstrual cycle to generate an unpredictable random sequence and to eliminate any sequence bias. The participant was then scheduled for condition A first or condition B first. The conditions were counterbalanced to control for order effects, with odd numbers beginning condition A first and even numbers beginning condition B first. To control for the effects of the menstrual cycle on anxiety, participants completed condition A and condition B within a 7-day period during the follicular phase of their menstrual cycle (specifically days 1-10). Participants were also asked to maintain pre-study physical activity levels during the 7-day period. All testing was completed by day 10 of their menstrual cycle. State anxiety was psychologically quantified using a reliable and validated assessment of anxiety, Spielberger's State-Trait Anxiety Inventory (STAI-YI).

Instruments for testing all physiological variables were calibrated and used by the same investigator in order to control possible inter-tester variation. During both condition A and B, a ProComp Infiniti System with Biograph software version 6.0 was used to measure HR and indices of HRV at a sampling rate of 2048 cycles per sec at the 3 different time points. Three electrodes were placed on the participants to monitor HR as specified by the ProComp Infiniti System. The negative electrode was placed in the right shoulder fossa below the clavicle, the positive electrode was placed medially on the thorax below the sternum, and the ground electrode was placed in the left shoulder fossa below the clavicle. Indices of HRV included the time-domain variable, root mean square of successive differences (RMSSD) and two frequency-domains: the low frequency power bands and the high frequency power bands in normalized units (LFNU and HFNU, respectively).

Measures of HR and HRV were analysed using CardioPro Infinity-HRV Analysis Software Module-SA7J90 [ProComp Infiniti System, Montreal, Quebec, Canada: Cimetra LLC]. An accelerometer using the ActiLife 6 Software and a Polar Heart Rate Sensor were strapped to the participant and used to monitor the intensity level of both study conditions A and B. The accelerometer was used only to record HR data and no other accelerometer data were analysed. Participants were then informed of the experimental condition. For study condition A (quiet, seated rest), participants sat on a yoga mat in a cross-legged position for 30 min in a quiet laboratory setting. For study condition B [YogaFit Vinyasa Flow (referred to as YogaFit in this manuscript)], participants followed, via digital versatile disc, a standardized YogaFit format choreographed by a certified American Council of Exercise Instructor and Registered Yoga Alliance Teacher. Yoga- Fit was performed in the same laboratory setting and lasted 30 min. There are 7 principles of alignment that are applied in every YogaFit session: (1) establish a base and dynamic tension; (2) create core stability; (3) align the spine; (4) soften and align the knees; (5) relax shoulders back and down; (6) hinge at the hips; and (7) shorten the lever. The YogaFit essence is based on: Breathing, Feeling, Listening to the Body, Letting Go of Competition, Judgement, Expectations, and Staying in the Present Moment. The movements included in each specific phase were chosen from Beth Shaw's YogaFit Vinyasa Flow Series. To measure intensity of theYogaFit routine, ratings of perceived exertion (RPE) [24] were recorded following both condition A and B. The accelerometer and heart rate sensor were removed after the study condition. After both study conditions, participants again completed STAI-Y1 and 10 min of post-condition data were collected. Participants then viewed 90 emotionally arousing pictures from the IAPS after both study conditions A and B. Pictures were viewed on a 70" × 70"portable projection screen. The 90 pictures were arranged in 3 blocks of 30; each block contained 10 pictures from each valence category. Among the 90 pictures used, 30 were pleasant (15 erotica and 15 babies, families, and cute animals); 30 were neutral (15 neutral people and 15 neutral objects and scenes); and 30 were unpleasant (15 threat and 15 mutilation) based on normative ratings of valence [5]. No more than two pictures from the same category appeared consecutively. Two different picture orders were constructed and counterbalanced across testing day and experimental condition. Each picture was shown for 4-s, followed by a 12-, 14-, or 16-s inter-picture interval, which consisted of a centrally located fixation cross. The total picture-viewing time, including brief breaks between each picture block, was approximately 30 min. Participants were instructed to look at each picture the entire time it was on the monitor and to subjectively categorize each picture as pleasant, neutral, or unpleasant using a response pad [E-Prime 2.0, Psychology Software Tools, Inc., Pittsburgh, PA, USA] resting on their lap. These ratings were not used to compare to the normative ratings, but rather, the purpose of categorizing the pictures was to ensure that participants concentrated on the pictures. Immediately after viewing the pictures, participants again completed STAI-Y1 and 10 min post-exposure data were collected. In accordance with the Center for the Study of Emotion and Attention, participants were also asked to rate the IAPS images in order to further image standardization. Participants rated each of the 90 pictures (hard copy, one picture per page in a standard, self-paced order) using the Self-Assessment Manikin (SAM) during visit 4 to the laboratory. Additionally, participants underwent a single DEXA scan for body composition assessment.