Optimal characteristics of inserted graphic objects in stimulating CCTV operator vigilance and performance

DSpace/Manakin Repository

Show simple item record

dc.contributor.author Donald, Fiona Margaret
dc.date.accessioned 2012-01-17T10:49:00Z
dc.date.available 2012-01-17T10:49:00Z
dc.date.issued 2012-01-17
dc.identifier.uri http://hdl.handle.net/10539/11049
dc.description Ph.D. Faculty of Humanities, University of the Witwatersrand, 2011 en_US
dc.description.abstract Vigilance is a key process fundamental for sustained performance in many jobs and in particular those requiring continual detection in visually intensive tasks. This research examined operators’ overall vigilance performance levels and decrements over time in the context of closed circuit television (CCTV) surveillance. The aims of the research were to develop an intervention to enhance the detection of significant events, and to establish the levels of overall vigilance performance and decrements in a CCTV surveillance task. The intervention consisted of electronically inserting graphic objects (IGOs) or images into the video stream with the intention of assisting operators in detecting actual significant events. IGOs could potentially represent an infinite range of visual stimuli, but it was argued that only particular visual characteristics are likely to enhance the detection of real significant events, rather than merely facilitating the detection of the IGOs themselves. In addition, the characteristics of IGOs are likely to influence the extent to which their relationship to significant events is understood. The research identified a range of characteristics that could be incorporated into IGO design, and focused on salience and semantic distance for the empirical part of the research. A matched three-group quasi-experimental design involved a sample consisting of 29 specialised CCTV surveillance operators, 13 control room operators doing surveillance, and 31 novices. The task consisted of observing a ninety-minute CCTV video showing general and target behaviour in a video stream of actual work settings. The control group received no IGOs, one treatment group received generic IGOs, and the second treatment group received IGOs with close semantic distances to target behaviours. There were indications that the IGOs had positive effects on alertness and attention sets, but this did not translate into statistically significant improvements in detection rates. Reasons for this included IGO characteristics, the complex and dynamic nature of CCTV displays and significant events, and the dynamic and spatiotemporal properties of the IGOs. Semantic distance was confirmed as an important IGO characteristic. The research demonstrated a number of critical insights into vigilance dynamics and visual processing and highlighted that there are gaps in the understanding of the attention processes that occur in jobs requiring sustained attention. Only half the target behaviours were detected despite all target behaviours being visible, indicating a concerning underperformance in intensive visual detection tasks involving complex work situations. Responses to vigilance demands were highly individualised, with decrements and surges beginning at different times across individuals. Qualitative analyses of participants’ behaviour also found fluctuations in task engagement, suggesting that sustained attention is unstable. Results did not support a steady, linear vigilance decrement for all sub-samples. An increment in detection rates was found for specialised participants after 60 minutes, while novices to surveillance tasks showed a more linear decrement. Work exposure was an important variable that contributed to detection levels and performance fluctuations over time. The research highlights differences between tasks with simple visual stimuli frequently used in vigilance research versus complex real-world tasks in vigilance intensive jobs. Important insights regarding vigilance processes in complex real-world jobs emerged, including the need for active searching processes, visual analysis, high levels of situation awareness and the importance of operator’s frame of reference and approach to the detection task. The research has likely implications for other visual imaging technologies such as x-rays, infrared and thermal imaging, and technology using newer millimetre wave and terahertz based imaging common in security, policing, and defence. en_US
dc.language.iso en en_US
dc.title Optimal characteristics of inserted graphic objects in stimulating CCTV operator vigilance and performance en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search WIReDSpace


Browse

My Account

Statistics