See No Evil: Cognitive Challenges of Security Surveillance and Monitoring

While the development of intelligent technologies in security surveillance can augment human capabilities, they do not replace the role of the operator entirely; as such, when developing surveillance support it is critical that limitations to the cognitive system are taken into account. The current article reviews the cognitive challenges associated with the task of a CCTV operator: visual search and cognitive/perceptual overload, attentional failures, vulnerability to distraction, and decision-making in a dynamically evolving environment. While not directly applied to surveillance issues, we suggest that the NSEEV (noticing - salience, effort, expectancy, value) model of attention could provide a useful theoretical basis for understanding the challenges faced in detection and monitoring tasks. Having identified cognitive limitations of the human operator, this review sets out a research agenda for further understanding the cognitive functioning related to surveillance, and highlights the need to consider the human element at the design stage when developing technological solutions to security surveillance. General Audience Summary High expectations are placed upon surveillance technologies to protect infrastructure and public places. While technological systems can automate some aspects of the surveillance process, the human operator is still ultimately responsible for detection of suspicious activities and decision-making. Thus, the optimal design and development of new technology should not focus solely on the capabilities of the system itself, but on supporting the operator's cognitive vulnerabilities. Observations and interviews with closed-circuit television personnel have previously highlighted several challenges with the task and control room environment-including information overload, distraction, and long work shifts-which we review in terms of the cognitive science literature. We set out a research agenda to investigate these issues in relation to a recent psychological theory of detection, and suggest ways in which progress can be made regarding training of operators, visualization and workspace design, and the use of intelligent and automated systems.