Dynamics of pointing with pointer acceleration

Publikation: Bidrag til bog/antologi/rapport/proceedingKonferencebidrag i proceedingsForskningpeer review

  • Jörg Müller

In this paper we investigate the dynamics (including velocities and accelerations) of mouse pointing when Pointer Acceleration (PA) functions are used. We also propose a simple model for these dynamics from a control theoretic perspective. The model allows us to simulate the effect of PA functions on pointing dynamics. In particular, it reproduces and explains many important phenomena we observe in pointing dynamics with PA functions. These include: (1) Pointer position, velocity, and acceleration over time, (2) Different accelerations when moving in different directions, and the resulting mouse drift when using PA functions, (3) Discontinuous jumps in phase space and associated acceleration peaks in Hooke plots when using Step PA functions. We identify parameters of the model using a reciprocal pointing task with the mouse controlling a pointer on a computer screen using sigmoid and step PA functions and constant gain. Our model explains the human-computer system including the PA function as a closed-loop dynamical system. In particular, we use a second-order model resembling a spring-mass-damper system (second order lag). Our model explains and allows to simulate the role of PA functions in pointing, including the phenomena described above.

TitelHuman-Computer Interaction - INTERACT 2017 - 16th IFIP TC 13 International Conference, Proceedings
RedaktørerRegina Bernhaupt, Girish Dalvi, Anirudha Joshi, Devanuj K. Balkrishan, Jacki O'Neill, Marco Winckler
Antal sider21
Vol/bind10515 LNCS
ForlagSpringer VS
ISBN (trykt)9783319676869
ISBN (Elektronisk)978-3-319-67687-6
StatusUdgivet - 2017
Begivenhed16th IFIP TC13 International Conference on Human–Computer Interaction, INTERACT 2017 - Mumbai, Indien
Varighed: 25 sep. 201729 sep. 2017


Konference16th IFIP TC13 International Conference on Human–Computer Interaction, INTERACT 2017
SerietitelLecture Notes in Computer Science

Se relationer på Aarhus Universitet Citationsformater

ID: 118504879