Impact of Design Decisions in Scanpath Modeling

TL;DR

This paper systematically examines how various design parameters influence the evaluation of scanpath models in GUI visual saliency, highlighting the importance of these choices for accurate user attention prediction.

Contribution

It provides a comprehensive analysis of how input size, inhibition-of-return decay, and masking radius affect scanpath evaluation metrics across multiple models and datasets.

Findings

Small parameter variations significantly impact evaluation metrics.

Effects are consistent across different models and datasets.

Design decisions critically influence scanpath prediction accuracy.

Abstract

Modeling visual saliency in graphical user interfaces (GUIs) allows to understand how people perceive GUI designs and what elements attract their attention. One aspect that is often overlooked is the fact that computational models depend on a series of design parameters that are not straightforward to decide. We systematically analyze how different design parameters affect scanpath evaluation metrics using a state-of-the-art computational model (DeepGaze++). We particularly focus on three design parameters: input image size, inhibition-of-return decay, and masking radius. We show that even small variations of these design parameters have a noticeable impact on standard evaluation metrics such as DTW or Eyenalysis. These effects also occur in other scanpath models, such as UMSS and ScanGAN, and in other datasets such as MASSVIS. Taken together, our results put forward the impact of design decisions for predicting users' viewing behavior on GUIs.