Lights, Camera, Action! Exploring Effects of Visual Distractions on Completion of Security Tasks

Human errors in performing security-critical tasks are typically blamed on the complexity of those tasks. However, such errors can also occur because of (possibly unexpected) sensory distractions. A sensory distraction that produces negative effects can be abused by the adversary that controls the e...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Berg, Bruce, Kaczmarek, Tyler, Kobsa, Alfred, Tsudik, Gene
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 31.05.2017
Subjects
Automation
Bluetooth
Human error
Human performance
Security
Task complexity
Visual effects
Visual stimuli
Online AccessGet full text

Cover

More Information
Summary:Human errors in performing security-critical tasks are typically blamed on the complexity of those tasks. However, such errors can also occur because of (possibly unexpected) sensory distractions. A sensory distraction that produces negative effects can be abused by the adversary that controls the environment. Meanwhile, a distraction with positive effects can be artificially introduced to improve user performance. The goal of this work is to explore the effects of visual stimuli on the performance of security-critical tasks. To this end, we experimented with a large number of subjects who were exposed to a range of unexpected visual stimuli while attempting to perform Bluetooth Pairing. Our results clearly demonstrate substantially increased task completion times and markedly lower task success rates. These negative effects are noteworthy, especially, when contrasted with prior results on audio distractions which had positive effects on performance of similar tasks. Experiments were conducted in a novel (fully automated and completely unattended) experimental environment. This yielded more uniform experiments, better scalability and significantly lower financial and logistical burdens. We discuss this experience, including benefits and limitations of the unattended automated experiment paradigm.
ISSN:2331-8422