Effects of Presentation Mode and Pace Control on Performance in Image Classification.

Area of Science:

• Information Visualization
• Human-Computer Interaction
• Cognitive Psychology

Background:

• Effective visualization is crucial for identifying items in large datasets, especially when metadata is insufficient for automated queries.
• Users often resort to visual inspection, necessitating efficient display modes for rapid item discovery.
• Two distinct display metaphors, static (page-based) and moving (conveyor belt), are commonly considered for such tasks.

Purpose of the Study:

• To compare the effectiveness, efficiency, and user satisfaction of static versus moving display modes for visual search in large datasets.
• To investigate how different display modes influence user search strategies and performance.
• To analyze the impact of predefined versus user-controlled pace on search outcomes.

Main Methods:

• Two experiments were conducted with 18 participants performing visual search tasks on large image collections.
• Participants used either a static (page-based grid) or a moving (continuous flow grid) display mode.
• The pace of image presentation was controlled, either predefined or user-set, across the experiments.

Main Results:

• At a fixed presentation pace, neither display mode significantly impacted the number of target items recalled (recall).
• Users perceived the moving mode as slower, leading them to select a faster real pace compared to the static mode.
• This adjustment in pace for the moving mode resulted in an equal average number of target images found per second across both modes, but at a potential cost to overall recall.

Conclusions:

• Display mode (static vs. moving) has no significant effect on recall when the presentation pace is fixed.
• User perception of pace influences search behavior; moving displays may encourage faster real-time searching at the expense of thoroughness.
• Designers should consider the interplay between display metaphor, perceived pace, and user search strategies to optimize information visualization tools.