Original title: Grounding GUI Design in Computational Psychology
Xianni Wang, Javier Romero Davila, Saku Sourulahti, Torsten Schaub +1 · September 2026
Designing good screen layouts takes a lot of slow, manual trial and error. To fix this, researchers created a system that designs user interfaces automatically. It uses answer set programming, a logic-based computing method that solves complex rule puzzles. Designers give the system their preferences, and the system turns design goals into strict rules. These rules balance looks and function by checking element properties, relative positioning, grid alignment, visual grouping, whitespace, and color harmony. In two user tests, people preferred layouts from this full system over random layouts and designs made with simple rule-of-thumb shortcuts. In a third study, professional designers reported that the tool helped them sketch and explore ideas during the early stages of design.
You are arranging furniture in a new living room, wanting the couch to face the TV, the coffee table within reach, and the colors to match. Moving things around manually to get it just right takes many attempts and often feels frustrating. Here, the researchers create a computer system that follows rules for furniture placement, generating arrangements that look good and work well without constant manual tweaking.
This matters because designing user interfaces by hand is time-consuming and iterative. Automating layout generation can save designers significant effort and help them explore more creative options early on.
The abstract doesn't specify limitations, but the method's effectiveness is demonstrated only for the objectives and conditions tested in the user studies.
Ask anything about this paper. Answers come from its abstract, and say so when the abstract doesn't cover your question.
Grounding GUI Design in Computational Psychology — Xianni Wang, Javier Romero Davila, Saku Sourulahti et al.
Abstract licence: CC0
Explanation generated by Synthesis from the abstract — it is not the original paper. This is a preprint, so it hasn't necessarily been peer reviewed.