How should Fitts' Law be applied to human-computer interaction?

Area of Science:

• Human-Computer Interaction
• Cognitive Psychology
• Usability Engineering

Background:

• Fitts' Law is widely applied to predict human movement time in HCI.
• General application of Fitts' Law may overlook task-specific movement characteristics.
• Understanding user movement sequences and target properties is crucial for accurate modeling.

Purpose of the Study:

• To challenge the universal applicability of Fitts' Law in HCI.
• To investigate how movement sequences (point-click vs. point-drag) affect Fitts' Law model fit.
• To determine optimal Fitts' Law parameters for text selection tasks.

Main Methods:

• Conducted two experiments on text selection tasks with controlled movement sequences.
• Applied various Fitts' Law models using different target size definitions (e.g., diagonal, horizontal, vertical extent).
• Analyzed pointing and dragging times for point-click and point-drag sequences.

Main Results:

• For point-click, the diagonal target size provided the best fit for pointing time.
• For point-drag, the vertical target size best fit pointing time, while dragging time fit models using terminal character dimensions.
• Pointing was faster in point-click than point-drag; both pointing actions were faster than dragging.

Conclusions:

• Fitts' Law requires task-specific parameters, including movement sequence and direction, for accurate HCI analysis.
• Interface design and analysis necessitate defining task characteristics before applying Fitts' Law.
• Distinct models are needed for analyzing pointing and dragging movements in text selection.