On the Factors Causing Processing Difficulty of Multiple-Scene Displays

Abstract

Multiplex viewing of static or dynamic scenes is an increasing feature of screen media. Most existing multiplex experiments have examined detection across increasing scene numbers, but currently no systematic evaluation of the factors that might produce difficulty in processing multiplexes exists. Across five experiments we provide such an evaluation. Experiment 1 characterises difficulty in change detection when the number of scenes is increased. Experiment 2 reveals that the increased difficulty across multiple-scene displays is caused by the total amount of visual information accounts for differences in change detection times, regardless of whether this information is presented across multiple scenes, or contained in one scene. Experiment 3 shows that whether quadrants of a display were drawn from the same, or different scenes did not affect change detection performance. Experiment 4 demonstrates that knowing which scene the change will occur in means participants can perform at monoplex level. Finally, Experiment 5 finds that changes of central interest in multiplexed scenes are detected far easier than marginal interest changes to such an extent that a centrally interesting object removal in nine screens is detected more rapidly than a marginally interesting object removal in four screens. Processing multiple-screen displays therefore seems dependent on the amount of information, and the importance of that information to the task, rather than simply the number of scenes in the display. We discuss the theoretical and applied implications of these findings.

Expanding the Bounds of Seated Virtual Workspaces

“…However, some degree of movement can be beneficial; the movements required for typical dual-monitor usage may provide some protection compared to single display environments, due to the greater variation in muscle activity [27,103,123]. User controls for managing the workspace have the potential to minimize physical consequences if they are used, but as noted by Ball and North in a study of a pan and zoom interface for an array of displays, participants preferred not to use the provided controls, instead choosing to physically navigate [5,100]. In addition, as the desk surface tends to support the peripherals used in daily productivity tasks, there may be a natural inclination to turn the neck, rather than the rest of the body, to keep the hands on peripherals.…”

See no evil: Cognitive challenges of security surveillance and monitoring.

“…A recent study by Stainer et al (2017) revealed that detection performance on a multiple-scene display is not exclusively driven by the number of scenes in the display. The authors showed that more visual information was associated with poorer detection performance regardless of whether this information was presented in a single scene or across multiple scenes.…”

Searching in CCTV: effects of organisation in the multiplex

“… 2008 ) and detecting changes in multiplexes is faster if participants are able to ignore irrelevant scenes within the multiplex (Stainer et al. 2017 ). However, in Experiment 1 it was not simply the case that portions of the display could be ignored because different parts of the multiplex were relevant depending on the identity of the target.…”

“… 2011 ), nine (Stainer et al. 2017 ) or 16 (Tickner and Poulton 1973 ) scenes. The present study employed a task that simulates a common reactive surveillance task in the control room—finding a target scene (Keval and Sasse 2010 )—and used a multiplex of 27 scenes taken from police CCTV surveillance cameras in one Scottish city.…”