THESIS
2010
xxv, 254 p. : ill. (some col.) ; 30 cm
Abstract
Humans are essentially probabilistic in nature, so that repeated movements made by different perons and even the same person will vary from trial to trial. This variability is shown in past studies of movements made, using the Fitts paradigm. These studies, which investigate the microstructure of movements, show that a movement is made up of a sequence of submovements and that the number of submovements used under a given experimental condition may vary.
Most derivations of Fitts' law are deterministic, that is, no consideration was made of the probabilistic nature of human beings. In these studies, the variation of movement time with Index of Difficulty (ID) was considered under different conditions and for different tasks. There are several studies that have investigated the numbe...[
Read more ]
Humans are essentially probabilistic in nature, so that repeated movements made by different perons and even the same person will vary from trial to trial. This variability is shown in past studies of movements made, using the Fitts paradigm. These studies, which investigate the microstructure of movements, show that a movement is made up of a sequence of submovements and that the number of submovements used under a given experimental condition may vary.
Most derivations of Fitts' law are deterministic, that is, no consideration was made of the probabilistic nature of human beings. In these studies, the variation of movement time with Index of Difficulty (ID) was considered under different conditions and for different tasks. There are several studies that have investigated the number of submovements, and their duration and variability in a probabilistic way.
This thesis investigates the distribution of movement time (MT) and sub-movements as a function of ID, and interprets Fitts' law by a simple probabilistic derivation. A computer-based mouse movement was adopted. The trajectory of the movement was recorded and analyzed to identify the distribution of sub-movements and MTs. Results showed the probabilistic characteristics of movements and interpreted Fitts' Law by a model of sub-movements based on the skewness of the MT distributions. Furthermore, previous studies' results were also validated and discussed. In addition, EMG data of hand movement was coupled with on-screen cursor movement, and used to validate the cursor's moving profiles. Inside kinetic and biomechanical characteristics with muscle pulse were correlated with cursor movement, and the relationship was build and modeled.
The contribution of the study in this thesis is to fill the gap between purely deterministic and purely probabilistic models, and introduce a more realistic explanation of the microstructure of the aiming task and Fitts' Law. Moreover, it also contributes to link the virtual output on display with the physical arm muscle control with an explanatory relationship.
Post a Comment