@article{anderson_mirka_joines_kaber_2009, title={Analysis of Alternative Keyboards Using Learning Curves}, volume={51}, ISSN={["1547-8181"]}, DOI={10.1177/0018720808329844}, abstractNote={Objective : To quantify learning percentages for alternative keyboards (chord, contoured split, Dvorak, and split fixed angle) and understand how physical, cognitive, and perceptual demand affect learning. Background: Alternative keyboards have been shown to offer ergonomic benefits over the conventional, single-plane QWERTY keyboard design, but productivity-related challenges may hinder their widespread acceptance. Method: Sixteen participants repeatedly typed a standard text passage using each alternative keyboard. Completion times were collected and subsequent learning percentages were calculated. Participants were asked to subjectively rate the physical, cognitive, and perceptual demands of each keyboard, and these values were then related to the calculated learning percentages. Results: Learning percentage calculations revealed the percentage for the split fixed-angle keyboard (90.4%) to be significantly different ( p < .05) from the learning percentages for the other three keyboards (chord, 77.3%; contour split, 76.9%; Dvorak, 79.1%). The average task completion time for the conventional QWERTY keyboard was 40 s, and the average times for the fifth trial on the chord, contoured split, Dvorak, and split fixed-angle keyboards were 346, 69, 181, and 42 s, respectively. Conclusions: Productivity decrements can be quickly regained for the split fixed-angle and contour split keyboard but will take considerably longer for Dvorak and chord keyboards. The split fixed-angle keyboard involved physical learning, whereas the others involved some combination of physical and cognitive learning, a result supported by the subjective responses. Application: Understanding the changes in task performance time that come with learning can provide additional information for a cost-benefit analysis when considering the implementation of ergonomic interventions.}, number={1}, journal={HUMAN FACTORS}, author={Anderson, Allison M. and Mirka, Gary A. and Joines, Sharon M. B. and Kaber, David B.}, year={2009}, month={Feb}, pages={35–45} }
 @article{anderson_meador_mcclure_makrozahopoulos_brooks_mirka_2007, title={A biomechanical analysis of anterior load carriage}, volume={50}, ISSN={["0014-0139"]}, DOI={10.1080/00140130701450195}, abstractNote={Front load carriage is a common occupational task in some industries (e.g. agriculture, construction), but, as compared to lifting tasks, relatively little research has been conducted on the biomechanical loading during these activities. The focus of this study was to explore the low back biomechanics during these activities and, specifically, to examine the effects of load height and walking speed on trunk muscle activity and trunk posture. Eleven male participants participated in two separate front load-carriage experiments. The first experiment called for carrying a barbell (with weight corresponding to 20% of elbow flexion strength) at three heights (knuckle height, elbow height and shoulder height) at a constant horizontal distance from the spine. The second experiment called for participants to carry a bucket of potatoes weighing 14 kg at the same three heights, but with no further restrictions in technique. In both experiments, the participants performed this task while either standing still or walking at a self-selected speed. As they performed these tasks, the activity levels of the right-side muscle of the rectus abdominis, external oblique, biceps brachii, anterior deltoid and three levels (T9, T12 and L3) of the erector spinae were sampled. Mid-sagittal plane trunk posture was also quantified using three magnetic field-based motion sensors at T9, T12 and L3. The results showed a significant effect of both walking speed and load height on trunk posture and trunk muscle activity levels in both the barbell and bucket experiments. In the barbell experiment, the walking trials generated 43% more trunk muscle activity than the standing trials. Trials at shoulder height produced 11% more muscle activity than trials at elbow height in the T9 erector spinae muscles and 71% more muscle activity in the anterior deltoid. In the bucket experiment, trunk muscle activity responded in a similar fashion, but the key result here was the quantification of the natural hyperextension posture of the spine used to balance the bucket of potatoes. These results provide insight into muscle activation patterns in dynamic settings, especially (load) carrying biomechanics, and have implications in industrial settings that require workers to carry loads in front of their bodies.}, number={12}, journal={ERGONOMICS}, author={Anderson, A. M. and Meador, K. A. and McClure, L. R. and Makrozahopoulos, D. and Brooks, D. J. and Mirka, G. A.}, year={2007}, pages={2104–2117} }