Accuracy and Frequency Analysis of MultiTouch Interfaces for Individuals with Parkinsonian and Essential Hand Tremor
Eric J. Frett, Kenneth E. Barner · 2005 · Proceedings of the 7th International ACM SIGACCESS Conference on Computers and Accessibility (Assets '05) · doi:10.1145/1090785.1090799
Summary
This 2005 study from the University of Delaware compares four computer pointing devices — an optical mouse, optical trackball, isotonic joystick, and a FingerWorks MultiTouch Surface (MTS) — for users with Parkinsonian tremor and essential tremor. The researchers recruited five participants with Parkinson's disease, five with essential tremor, and eleven controls without any tremor condition. Using WinFitts, a Fitts' law data acquisition program developed at the University of Oregon's HCI Lab, the study collected both temporal and spatial performance data across multiple target distances and sizes at various screen resolutions. The temporal measures included Fitts' law analysis (movement time, throughput, error rate) and Proximity Movement Time, which captures how users slow down as they approach a target when tremor becomes uncontrollable. The spatial measures included Deviation Accuracy (how closely the cursor follows the optimal path), Click Histogram (spatial distribution of click locations relative to the target), and Successive Click Errors (repeated clicks in the starting location due to tremor-induced finger quivering). Additionally, the researchers used a BIOPAC triaxial accelerometer attached to participants' wrists to record resting tremor signals, then applied MUSIC spectral estimation to analyze how each device affected the frequency and amplitude of the tremor during use.
Key findings
The mouse provided the fastest movement times and highest throughput for both tremor groups according to Fitts' law analysis, while the MTS and trackball offered the next best performance levels. However, the error analysis revealed a more nuanced picture: the mouse produced the fewest errors for both groups, but the MTS generated the most errors, attributed to the device's touch sensitivity causing tremor-induced inadvertent clicks. Spatially, the MTS showed the highest deviation accuracy for essential tremor subjects (59.4%) but the lowest for Parkinson's subjects (68.9% for mouse vs. 58.2% for MTS), again linked to the zero-force touch interface. The MTS had the least movement error and movement variability of all devices for both tremor groups, with statistical significance (p<0.05). Spectral analysis showed the mouse was the only device that suppressed tremor for Parkinson's subjects (0.171 Hz frequency decrease), while for essential tremor subjects, the joystick provided the greatest frequency decrease. The trackball produced the most energy loss (14.3 dB) for essential tremor users. These findings demonstrate that no single device is optimal across all measures — the best choice depends on the type of tremor and whether temporal speed or spatial accuracy is prioritized.
Relevance
This study is significant as one of the earliest empirical comparisons of multitouch technology for users with motor tremor, predating the widespread adoption of touchscreen devices. Its finding that the FingerWorks MTS excelled in movement accuracy metrics but suffered from high error rates due to touch sensitivity is particularly prescient given the subsequent proliferation of touchscreen interfaces. The research highlights a fundamental tension in accessible input design: devices that reduce the physical effort of interaction (zero-force touch surfaces) may simultaneously introduce new error modes for users with tremor. For practitioners, the study underscores that device selection for users with motor impairments should be informed by the specific type of tremor (Parkinsonian vs. essential) and the nature of the task. The spectral analysis methodology offers a rigorous framework for evaluating tremor suppression that goes beyond simple task completion metrics. While the small sample sizes (5 per tremor group) limit generalizability, the study provides foundational evidence for the accessible design of touch-based interfaces.
Tags: motor accessibility · input devices · Parkinson's disease · essential tremor · multitouch · pointing devices · tremor suppression · Fitts' law · human-computer interaction
Standards referenced: ISO 9241