The 21 sr International Conference on Auditory Display (ICAD-2015) July 8-10, 2015, Graz, Austria THE EFFECTS OF VARIOUS PARAMETER COMBINATIONS IN PARAMETER- MAPPING SONIFICATIONS S. Camille Peres 1 , Daniel Verona 2 , Paul Ritchey 1 Texas A&M University 1 Department of Environmental and Occupational Health 2 Department of Biomedical Engineering ABSTRACT This study will be investigating the design of parameter- mapping sonifications and investigating how different combinations of sound parameter mappings affect the user’s ability to understand and interpret sEMG data. The parameter mappings being used are all redundantly mapped and the specific parameter combinations are 1) pitch and loudness, 2) pitch, loudness, and attack time, and 3) loudness and attack time. There will be both spatialized (right and left) and non- spatialized versions of each of these mappings. These mappings will be used to present sonifications of two channels of sEMG data to participants to explore if they can identify muscle activation order (which muscle activates first) and relative muscle exertion levels (which muscle has a higher exertion). It is expected that participants will perform better with the spatialized mappings. It is also expected that the participants will perform better with the mappings that include attack time because this results in greater timbral variety. 1. INTRODUCTION The simplest accepted definition of sonification is “the use of non-speech audio to convey information” [1]. However, as new techniques for sonification have been implemented, a further definition of sonification was offered by Hermann [2]. This definition states that a technique which takes data as input and generates sounds can only be called a sonification if the sound reflects objective properties of the data, the conversion of data to audio is systematic (explicitly defined), the generated sounds are reproducible, and the system can be used with different data sets. Also in [2], Hermann discusses the necessity of defining what type of sonification is being employed: audification, parameter-mapping sonification, or model-based sonification. This study is using parameter-mapping sonification (PMSon) to sonify surface electromyography (sEMG) data. sEMG measures muscle activation and exertion and is used as an index of fatigue processes occurring within a muscle [3], and as a biofeedback tool [4]. PMSon is a common form of sonification [5,6,7] defined as the mapping of data features to acoustic features of sound events or streams [2]. Many parameters of sound have been examined for use in PMSon, including pitch, loudness, harmonics, speed, tremolo, attack time, and spatial location [8,9,10]. However, there is still a lack of objective evaluation of sonification parameters [10]. For some auditory displays, it has been shown that mapping more than one parameter redundantly (such as pitch and loudness) results in better performance than mapping only one parameter at a time [11]. However, the benefit in performance was only found when certain dimensions of sound were used, specifically pitch and loudness. When scatterplots of temperature data were sonified and spatialization (panning) was used to redundantly represent time (x-axis), performance improved compared to a temporal mapping only [12]. The number of octave ranges used was also varied and it was found that participants performed better with wider octave ranges as compared to a one octave range. These findings indicate that for different types of auditory displays, the best parameters used for mapping will likely be different, and thus empirical research needs to be conducted to identify the most appropriate mappings. This study is currently in progress and data collection is expected to begin soon. The study seeks to evaluate four parameters of sound as they relate specifically to sonifying sEMG data: pitch, loudness, attack time, and spatial location. Sonification of EMG data has been shown to have potential clinical application in regards to diagnosing musculoskeletal problems [13] and in rehabilitation for stroke patients [14]; however the pleasantness of EMG sonification needs improvement [13, 15]. The purpose of this study is to identify parameters of sound that are useful for interpreting sEMG data, and to determine which mappings users find to be the most intuitive and the most pleasant. 2. PARAMETER MAPPING Mapping pitch and loudness redundantly has been shown to improve user performance [11]. However, it may be the case that certain redundant mappings do not result in redundancy gains and certain redundant mappings do [11]. With this in mind, we have taken the four parameters of sound mentioned above and combined them to create six different redundant sonification designs: 1. Pitch, Loudness, Non-spatialized 2. Pitch, Loudness, Attack time, Non-spatialized 3. Loudness, Attack time, Non-spatialized 4. Pitch, Loudness, Spatialized 5. Pitch, Loudness, Attack time, Spatialized 6. Loudness, Attack time, Spatialized SuperCollider is being used to create the sonifications for this study. The Pbind function in SuperCollider is used to play 10 tones per second, and the parameters of each tone (pitch, loudness, and attack time) are controlled by the sEMG data. Triangle waves are being used for these sonifications, since they are slightly brighter in tone than sine waves. Pitch and loudness increase as the sEMG amplitude increases and attack time decreases as the sEMG amplitude increases. Each sonification presents two channels of sEMG data simultaneously, and the channels are referred to as Muscle A and Muscle B. To spatialize the sonifications, data from The 21 st International Conference on Auditory Display (ICAD 2015) July 8–10, 2015, Graz, Austria ICAD 2015 - 331