Attention allocation on mobile app interfaces when human interacts with them

Li Zhu, Gaochao Cui, Yan Li, Jianhai Zhang, Wanzeng Kong, Andrzej Cichocki, Junhua LiCognitive 2021 · DOI

TL;DR

This study uses eye-tracking to investigate how users allocate attention across mobile app interfaces. They found that users tend to focus more on the bottom half of the screen, and that colors and backgrounds significantly influence where attention is initially directed, with users paying more attention to the bottom-left when these elements are present.

What Problem Does This Paper Address?

As smartphones and mobile apps become increasingly ubiquitous, understanding how users interact with these interfaces is crucial for effective design. Specifically, this paper addresses the question of how users allocate their attention across different regions of a mobile app interface. This is particularly relevant for app developers and advertisers aiming to optimize information delivery and user experience, especially considering phenomena like "banner blindness" where users tend to ignore certain types of visual information.

Methods

The researchers designed an experiment with two sessions using an eye-tracking system (Tobii Pro X2-30) to record participants' gaze patterns.

  • Session 1: Participants browsed original app interfaces that included colors and backgrounds.
  • Session 2: Participants browsed the same app interfaces, but with colors and backgrounds removed, leaving only grayscale layouts.

The smartphone screen was divided into four equal areas of interest (AOIs): upper-left, upper-right, bottom-left, and bottom-right. They measured two key metrics:

  • First fixation duration: The time spent staring at a region for the first time.
  • Total fixation duration: The cumulative time spent staring at a region throughout a trial.

Additionally, participants' smartphone operation level (familiarity and accuracy) was assessed through tasks like changing wallpapers and booking rooms, and its correlation with fixation durations was analyzed.

Datasets

The study used app interfaces extracted from apps available on the Apple Store. These interfaces were selected to have approximately even distributions of color, layout, and icons across the four quarters of the screen. No new benchmark datasets were introduced.

Key Insights and Ideas

  • Bottom-heavy attention: Users consistently spent more time looking at the bottom half of the screen, regardless of whether colors and backgrounds were present. This suggests a natural bias towards lower screen regions.
  • Color and background matter for initial focus: The presence of colors and backgrounds significantly influenced where users first looked. In Session 1 (with colors), the bottom-left region attracted the most initial attention. However, in Session 2 (grayscale), this specific preference for the bottom-left diminished, suggesting that visual elements like color and background guide the initial gaze.
  • Skill influences initial attention: For interfaces with colors and backgrounds, users with higher smartphone operation levels showed shorter first fixation durations on the dominant bottom-left AOI. This implies that experienced users can process information more quickly.
  • First impression is key: The ratio of first fixation duration to total fixation duration was similar across AOIs, suggesting that where users look first often dictates where they continue to focus. This highlights the importance of initial visual cues.

Key Results

  • Total Fixation Durations: In Session 1, total fixation durations were significantly different across all four AOIs, with a notable preference for the bottom-left. In Session 2, significant differences were mainly observed between the upper and bottom halves, with the bottom half receiving more attention. The left half of the screen also received more attention than the right half in Session 1, but this difference disappeared in Session 2.
  • First Fixation Durations: Similar to total fixation, first fixation durations were significantly different across AOIs in both sessions, with AOI3 (bottom-left) and AOI4 (bottom-right) attracting longer first fixations than AOI1 (upper-left) and AOI2 (upper-right).
  • Correlation with Operation Level: A significant negative correlation (r = -0.65, p < 0.005) was found between the first fixation duration of the bottom-left AOI (AOI3) and smartphone operation level in Session 1. No significant correlation was found in Session 2 or for total fixation duration in either session.

Limitations and Future Work

  • Limited AOI granularity: The study divided the screen into four equal areas. More granular divisions (e.g., 9 areas) could provide finer insights but might be limited by eye-tracker precision on smaller smartphone screens.
  • Task selection for operation level: The tasks used to assess smartphone operation level were not standardized, potentially introducing bias. Future work could focus on developing standardized assessment methods.
  • Sample size: The study involved 18 participants, which the authors acknowledge is a relatively small cohort. They suggest that a larger sample size would be beneficial for replicating findings.
  • Interface variation: While efforts were made to select interfaces with even distributions of properties, variations in layout and color across different apps still existed. Future research could aim to control these factors more rigorously.

The authors suggest that their findings can inform the design of app interfaces and advertisements by guiding the placement of important information in areas that naturally attract user attention. They also propose that the amount of content displayed should be tailored to the target user's operation level.

Summary updated

Abstract

With the popularity of smartphones and the pervasion of mobile apps, people spend more and more time to interact with a diversity of apps on their smartphones, especially for young population. This raises a question: how people allocate attention to interfaces of apps during using them. To address this question, we, in this study, designed an experiment with two sessions (i.e., Session1: browsing original interfaces; Session 2: browsing interfaces after removal of colors and background) integrating with an eyetracking system. Attention fixation durations were recorded by an eye-tracker while participants browsed app interfaces. The whole screen of smartphone was divided into four even regions to explore fixation durations. The results revealed that participants gave significantly longer total fixation duration on the bottom left region compared to other regions in the session (1) Longer total fixation duration on the bottom was preserved, but there is no significant difference between left side and right side in the session2. Similar to the finding of total fixation duration, first fixation duration is also predominantly paid on the bottom area of the interface. Moreover, the skill in the use of mobile phone was quantified by assessing familiarity and accuracy of phone operation and was investigated in the association with the fixation durations. We found that first fixation duration of the bottom left region is significantly negatively correlated with the smartphone operation level in the session 1, but there is no significant correlation between them in the session (2) According to the results of ratio exploration, the ratio of the first fixation duration to the total fixation duration is not significantly different between areas of interest for both sessions. The findings of this study provide insights into the attention allocation during browsing app interfaces and are of implications on the design of app interfaces and advertisements as layout can be optimized according to the attention allocation to maximally deliver information.
BibTeX
@inproceedings{zhu2021attention,
  author = {Li Zhu and Gaochao Cui and Yan Li and Jianhai Zhang and Wanzeng Kong and Andrzej Cichocki and Junhua Li},
  title = {{Attention allocation on mobile app interfaces when human interacts with them}},
  booktitle = {Cognitive},
  year = {2021},
  doi = {10.1007/s11571-021-09760-4},
}
Figure