Injury Outcomes in Native and Non-Native American Sign Language Users: An EMG Assessment

Open Access
- Author:
- Miller, Kara
- Area of Honors:
- Biomedical Engineering
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Meghan Vidt, Thesis Supervisor
Jian Yang, Thesis Honors Advisor - Keywords:
- Electromyography
American Sign Language
Fatigue
Muscle activation
Biomechanical assessments - Abstract:
- High percentages of native and non-native American Sign Language users report upper extremity pain. Previous studies observed that non-native signers, those with no deaf or signing parents, reported higher levels of pain than native signers, those with at least one deaf or signing parent. Higher pain prevalence in non-natives suggest that native signers may use their arm muscles differently than non-native signers when signing. Our first objective was to determine if differences exist in activation of muscles for natives and non-natives while signing. The second objective was to determine if differences exist in the time that the muscles are fatigued between the two groups of signers. Nine natives and fifteen non-natives were assessed in a prior study. Briefly, sixteen surface electromyography (EMG) sensors placed on the muscle bellies of right and left upper trapezius, middle trapezius, anterior deltoid, and middle deltoid muscle compartments, along with the wrist flexor, wrist extensor, radial deviation, and ulnar deviation muscle groups, recorded muscle activation while participants signed along with a 7.5-minute video. Before the trial, maximal voluntary contractions (MVC) were measured for each muscle for 5 seconds, using postures known to elicit maximal activity. EMG data was filtered with a 4th order Butterworth filter, rectified, and enveloped with a 2nd order Butterworth filter, then normalized by each corresponding muscle’s MVC using a custom MATLAB program. The average activation across the trial was calculated for each muscle, then averaged across groups. Consistent with prior literature, a 15% MVC threshold was used to identify instances where fatigue could occur. The time spent above the fatigue threshold (fatigued) was calculated for each muscle for each participant. A weighted total, based on the physiological cross-sectional area of the muscles, was used to calculate the percent of time above the fatigue threshold. The weighted total was separately calculated for each joint, including right and left shoulder, and right and left wrist. The EMG data was split into quarters temporally to examine if there were differences in fatigue levels at different time periods during the trial. A Kruskal-Wallis ANOVA was used to assess differences in average activation across native and non-native groups. Another Kruskal-Wallis ANOVA was used to assess differences in the percent of time fatigued across native and non-native groups. Lastly, a Friedman’s Test was completed to compare the time spent fatigued in each quarter between native and non-native signers. Results revealed no differences (p>0.05) in the average activation between groups for any muscles, but the non-native signers have more fatigued left upper trapezius (p=0.0026), and native signers have more fatigued left wrist (p=0.0032). The Friedman’s Test results revealed no significance (p>0.05) in fatigue levels between groups of signers and each quarter of signing. Non-native signers may rely more heavily on shoulder shrugging, leading to overuse injuries, whereas native signers more efficiently use their wrists. Future work will examine the role of EMG along with other biomechanical factors, including work envelope and ballistic signing, to determine their contributions to the increased injury prevalence for non-native compared to native signers.