Effects of Head-Mounted Display on kinematics of the Timed Up and GO (TUG) test: does the addition of a visual stimulus matter?

• ¹ Temple University, Physical Therapy, United States
• ² Stony Brook University, Physical Therapy, United States

The use of virtual reality in clinic settings is becoming more common given the availability of portable, easy to use head-mounted displays (HMDs). HMDs support virtual reality environments that are variable, motivating, yet safe to include in clinical assessment and interventions in different populations (Adamovich, Fluet, Tunik, & Merians, 2009; Saposnik et al., 2010). However, some concerns have been reported about the effect of HMDs on vision and posture that include, but are not limited to, subjective discomfort of the users (Peli, 1998) and modified neck and trunk postures (Knight & Baber, 2007), which may occur due to the added weight on the head, restriction of the field of view, or unfamiliarity with the device (Patterson, Winterbottom, & Pierce, 2006). We aim to determine the influence of wearing an HMD with and without presentation of an augmented visual stimulus during the sub-components of the Timed Up and Go (TUG) test in younger and older adults. The TUG sub-components include sit-to-stand, walk, turn, and stand-to-sit activities. The three questions we wished to address are 1) does the HMD have a biomechanical effect, 2) does augmented visual stimulus affect TUG performance, 3) is there an age interaction. Using the HMD (Oculus Rift Development Kit 2™) a visual scene displaying bright dots (i.e. “snowflakes”) that rotate in pitch-up or pitch-down directions was overlaid on top of the veridical scene, i.e. augmented reality (Fig1). Twelve younger adults (6 males, age 25.9 ± 3.9) and sixteen older adults (8 males, age 69.0 ± 4.4) completed the following four conditions: (1) TUG, (2) TUG with HMD with no augmented visual scene (TUGHMD), (3) TUG with HMD with pitch up rotated visual scene (TUGPU), and (4) TUG with HMD with pitch down rotated visual scene (TUGPD). Motor performance was evaluated using six Trigno™ wireless motion sensors (Delsys Inc.) that were placed on the sternum, lumbar, both wrists and shanks. The dependent variables included turning cadence (step/min), gait speed (m/sec), and peak trunk velocity (PTV) (°/sec) in the TUG sub-components around the anteroposterior (AP), vertical, and mediolateral (ML) axes. There were decrements in performance when wearing the HMD, regardless of the addition of a visual stimulus, compared to viewing the normal view of the room in the standard TUG test. While viewing a rotated visual scene in a pitch up or pitch down direction, adults significantly decreased their PTV around the ML axis in walking and their AP acceleration range in sit-to-stand compared to standard TUG conditions (all ps<0.01). Adults showed a lower vertical acceleration range in sit-to-stand as well while viewing a pitch up rotating visual scene only (p=0.002) compared to TUG. Compared to younger adults, older adults showed a decline in PTV around the AP axis in turning (17.3% in TUG, 12.8% in TUGHMD, 11.9% in TUGPU, 17.6% in TUGPD) (p=0.03). Our results support the conclusion that: 1) wearing an HMD can present biomechanical, perceptual, and unfamiliarity challenges that should be accounted for in clinical research and 2) assessment of spatiotemporal measures in turning could be of a great interest to assess the risk of fall in the elderly. The results from this work can help clinicians to discriminate between the effects of wearing an HMD without a virtual scene versus HMD plus virtual scene. This information may be critical as the use of HMDs in clinics becomes more common. Fig1. The Oculus Rift development kit 2 (™) (on the right side) and the presented visual scene through it with the virtual snowflakes (on the left side).