Skewed Dual Normal Distribution Model: Predicting Touch Pointing Success Rates for Targets Near Screen Edges and Corners
HCI Today summarized the key points
- •This article presents a new model for predicting touch success rates for targets near the edges and corners of a screen.
- •The existing Dual Gaussian Distribution Model performs well for targets near the center of the screen, but it cannot handle situations near the edges.
- •The researchers report that edges make the tap distribution asymmetric, and they designed a Skewed Dual Normal Distribution Model to capture this effect.
- •Across three experiments—one-dimensional horizontal and vertical tasks, and a two-dimensional corner task—they confirmed that as targets get closer to the edge, the distribution becomes more skewed and the success rate increases as well.
- •This model supports designs that make use of edge regions in scrollable interfaces, and it also explains a user strategy of tapping together with the edge.
This summary was generated by an AI editor based on HCI expert perspectives.
Why Read This from an HCI Perspective
This article is meaningful for both HCI/UX practice and research because it proposes a way to quantitatively predict touch success rates for targets near the edges and corners of a screen. Since it addresses regions that prior models treated as exceptions, it can explain layouts that commonly occur in real scroll environments or dense interfaces. In particular, it helps designers make data-driven judgments about how much they can leverage the screen edge.
CIT's Commentary
From a CIT perspective, an interesting point is that the edge is not merely a constraint, but an interaction resource in which the user’s corrective strategy comes into play. In other words, beyond the asymmetry in the distribution caused by off-screen constraints, the interpretation suggests that users may adopt a strategy of pressing together with the edge. This prompts a re-examination of traditional guidelines that conservatively leave space in the layout. That said, the model fits well when devices have flat bezels and straight boundaries, and its generalization may weaken for cases or protruding bezels—so it should be applied alongside hardware conditions in real product design. From the standpoint of UX tooling, the advantage is that it is an interpretable model that can be directly connected to a layout optimization engine.
Questions to Consider While Reading
- Q.In this model, can we further verify whether the ‘edge-together pressing strategy’ is actually a learned behavior, or instead an immediate adaptation to physical constraints?
- Q.When targets move in a scrollable interface, what computational cost and design constraints should be considered to apply this model to real-time layout reconfiguration or dynamic adjustment of touch target areas?
- Q.On devices with different boundary conditions—such as protruding bezels, protective cases, or curved screens—how would the asymmetric distribution change, and what form would an appropriate extension model take?
This commentary was generated by an AI editor based on HCI expert perspectives.
Please refer to the original for accurate details.
Subscribe to Newsletter
Get the weekly HCI highlights delivered to your inbox every Friday.