Related Experiment Video
Updated: Jun 8, 2026
13:00
Measuring Attention and Visual Processing Speed by Model-based Analysis of Temporal-order Judgments
Published on: January 23, 2017
Real-time performance modelling of a Sustained Attention to Response Task.
Grégoire S Larue1, Andry Rakotonirainy, Anthony N Pettitt
1Centre for Accident Research and Road Safety - Queensland, Queensland University of Technology, Queensland, Australia. g.larue@qut.edu.au
Ergonomics
|September 25, 2010
Summary
Monotonous tasks significantly impair vigilance and performance. This study models vigilance decline in real time using reaction times, achieving 72% accuracy in detecting lapses.
Area of Science:
- Cognitive Psychology
- Human Factors Engineering
- Neuroscience
Background:
- Vigilance decreases during monotonous tasks, leading to errors.
- Existing research often links vigilance decline to factors like sleep deprivation, not task monotony.
- Monotony offers minimal cognitive and motor stimulation, impacting sustained attention.
Purpose of the Study:
- To model and detect vigilance decline in real time during monotonous tasks.
- To quantify the impact of monotony on performance using reaction times.
- To compare the accuracy of different mathematical models in detecting hypovigilance.
Main Methods:
- A laboratory experiment using a Sustained Attention to Response Task (SART) adaptation.
- Collection and analysis of participant reaction times to model vigilance.
- Comparison of Bayesian models, neural networks, and generalized linear mixed models for detection accuracy.
Main Results:
- Monotonous tasks caused an average performance decline of 45%.
- Vigilance modeling detected decline via reaction times with 72% accuracy and a 29% false alarm rate.
- Bayesian models outperformed neural networks and generalized linear mixed models in detecting vigilance lapses.
Conclusions:
- Monotony negatively affects sustained attention, which can be modeled and predicted in real time.
- Reaction times serve as effective surrogate measures for real-time vigilance monitoring.
- The developed Bayesian modeling framework can detect vigilance decline in humans performing monotonous tasks.
