What is the most effective type of audio-biofeedback for postural motor learning? Marco Dozza a, *, Lorenzo Chiari b,1 , Robert J. Peterka c,2 , Conrad Wall d,3 , Fay B. Horak c,4 a Department of Applied Mechanics at Chalmers University of Technology, SAFER - Box 8077 - S-402 78 - Go ¨teborg, Sweden b Department of Electronics, Computer Science, and Systems at University of Bologna, Viale Risorgimento, 2 - I-40136 - Bologna, Italy c Neurological Science Institute, Oregon Health and Science University, Portland, OR 97006, USA d Department of Otology and Laryngology, Harvard Medical School, Boston, MA 02114, USA 1. Introduction The concept of biofeedback has been well known since the 1950s [1]. In the 1970s and 1980s, biofeedback was extensively applied to improve postural stability [2–4]. However, in the last few years, the interest on biofeedback systems to enhance postural stability has been renewed, partially due to recent advances in technology [5]. This renewed interest is evident by several recent studies showing the efficacy of biofeedback in improving postural stability [6–8]. The first challenge in the development of a biofeedback device is its design [9]. The biofeedback design should optimize three main components: (1) the sensor, which acquires the biological inputs to feed back; (2) the processor, which converts these biological inputs into new information understandable for a user; and (3) the interface, which conveys this information to the user. In addition, to improve the design of the biofeedback device, it is important to determine the amount of information that is actually needed and can be used by the human user. To date, there are no studies on this issue. Another challenge in the development of biofeedback devices is the protocol design for device validation [3]. It is important to distinguish improvements due to biofeedback from improvements due to other mechanisms, such as placebo effects and effects of spontaneous motor learning with practice. This distinction is fundamental to evaluate retention and transfer of motor perfor- mance after exposure to biofeedback. To date, very few studies have considered this distinction, which is well known to be crucial for the evaluation of biofeedback devices [2,3]. Audio-biofeedback (ABF) of body sway has been shown to improve stability by strengthening the closed-loop control of posture without influencing the open-loop (anticipatory) postural control [10]. Further, ABF showed the largest improvement in subjects with bilateral vestibular loss, suggesting that biofeedback information can substitute for lack of vestibular information [11]. Gait & Posture 34 (2011) 313–319 A R T I C L E I N F O Article history: Received 10 December 2010 Received in revised form 12 May 2011 Accepted 20 May 2011 Keywords: Biological feedback control systems Posture control Perturbed stance Motor learning A B S T R A C T Biofeedback is known to improve postural control and reduce postural sway. However, the effects that different biofeedback modes (coding for more or less complex movement information) may have on postural control improvement are still poorly investigated. In addition, most studies do not take into account the effects of spontaneous motor learning from repetition of a task when investigating biofeedback-induced improvement in postural control. In this study, we compared the effects of four different modes of audio-biofeedback (ABF), including direction and/or magnitude of sway information or just a non-specific-direction alarm, on the postural sway of 13 young healthy adults standing on a continuously rotating surface. Compared to the non-specific-direction alarm, ABF of continuous postural sway direction and/or amplitude resulted in larger postural sway reduction in the beginning of the experiment. However, over time, spontaneous postural motor learning flattened the effects of the different modes of ABF so that the alarm was as effective as more complex information about body sway. Nevertheless, motor learning did not make ABF useless, since all modes of ABF further reduced postural sway, even after subjects learned the task. All modes of ABF resulted in improved multi-segmental control of posture and stabilized the trunk-in-space. Spontaneous motor learning also improved multi- segmental control of posture but not trunk-in-space stabilization as much as ABF. In conclusion, although practice standing on a perturbing surface improved postural stability, the more body sway information provided to subjects using ABF, the greater the additional improvement in postural stability. ß 2011 Elsevier B.V. All rights reserved. * Corresponding author. Tel.: +46 31 772 3621. E-mail addresses: marco.dozza@chalmers.se (M. Dozza), lorenzo.chiari@unibo.it (L. Chiari), peterkar@ohsu.edu (R.J. Peterka), cwall@mit.edu (C. Wall), horakf@ohsu.edu (F.B. Horak). 1 Tel: +39 051 209 3014. 2 Tel.: +1 503 418 2616. 3 Tel: +1 617 573 4153. 4 Tel: +1 503 418 2500. Contents lists available at ScienceDirect Gait & Posture jo u rn al h om ep age: ww w.els evier.c o m/lo c ate/g aitp os t 0966-6362/$ – see front matter ß 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2011.05.016