Cognitive Ergonomics Research Facility San Diego State University
Home About CERF Research Collaborations Commercial

Click on Tabs to read about individual projects
     Index of Cognitive Activity: Measuring the Impact of Augmented Cognition

Principal Investigator: Sandra P. Marshall
Funding Agency: DARPA
Grant Number: (ONR N00014-95-1-0237)

The goal of the project is to validate and implement a new psychophysiological metric, the Index of Cognitive Activity (ICA). ICA assesses cognitive workload by measuring changes in pupil dilation that occur as individuals exert mental effort. ICA is being used in a number of collaborative studies to evaluate the cognitive impact of innovative technologies that are currently being developed in DARPA’s Augmented Cognition Program.

The Index of Cognitive Activity is a psychophysiological measure based on pupil reflexes that occur when an individual experiences effortful cognitive processing. The pupils dilate when an individual engages in different types of cognitive acts such as reading text, listening to instructions, searching a display, or interpreting a graph. Because the underlying phenomenon to be measured is a reflex, it is not under an individual’s conscious control and is not subject to the same confounding factors as subjective measures of cognitive difficulty or external estimates based on performance. The Index has a wide range of application because the required measurement of pupil dilation can be made easily in most environments without interfering with a user’s normal performance in the environment.

The Index of Cognitive Activity (ICA) is implemented with state-of-the-art eye tracking technology and derives from recent advances in the mathematical field of wavelet theory. Already employed to evaluate two projects supported by the Office of Naval Research, the patented procedure in its current form has been validated on tasks with varying cognitive demands and has successfully identified interfaces and software applications that increase or decrease the cognitive workload for the user. Under DARPA’s Augmented Cognition Program, the ICA is being extended to provide real-time assessment of cognitive workload in a number of complex environments.

The ICA is currently being validated through multiple collaborative efforts with other research programs within the Augmented Cognition Program. Through these efforts, the real-time estimate of cognitive workload from ICA is to be integrated with other sensors derived from physiological measures such as EEG and fNIR.

Technology Transition
The patented technology underlying the Index of Cognitive Activity is owned by San Diego State University and licensed for commercial use exclusively to EyeTracking, Inc. It is already being used in some military applications, including a project supported by the Office of Naval Research in the study of better interface designs for weatherforecasters and a project supported by DARPA for the hardware integration of EEG sensors and eye-tracking equipment.
ICA is being used in a number of collaborative studies to evaluate the cognitive impact of innovative technologies that are currently being developed in DARPA’s Augmented Cognition Program.




Application of Neuro-science Technology to
Educational and Social Research, Hong Kong.
S. Marshall, Cognitive and Instructional Applications for New Eye-Tracking Technologies May, 2002.

IEEE 7th Conference on Human Factors and Power Plants, Scottsdale, AZ.
S. Marshall, The Index of Cognitive Activity: Measuring Cognitive Workload. Presented in D. Schmorrow (Chair), Tomorrow’s Human Computer Interaction from Vision to Reality: Building cognitively aware computational systems. Symposium. September 18, 2002.

LHICSS-36 IEEE Hawaii International Conference on System Sciences, Big Island, Hawaii.
S. Marshall, B. Dickson, & C. Pleydell-Pearce, Combining EEG and Pupil Dilation to Measure Cognitive Workload: A Case Study. January 6-9, 2003

Recent Accomplishments

  • Achieved real-time estimate of cognitive workload. The Index of Cognitive Activity can be calculated in real time with only a 2-second lag.
  • Demonstrated feasibility of concurrent data collection with ICA and EEG. Multiple sensors can be used simultaneously to produce high fidelity estimates of cognitive state change.
  • Linked ICA to important changes in performance in two studies. One study shows that ICA correlates highly with shifts in cognitive strategies used to perform task, with more effortful strategies resulting in higher estimates of workload. A second study shows reliable changes in ICA resulting from both task difficulty and task performance, with high ICA resulting from high levels of difficulty and low levels of performance.



    ©2002 Cognitive Ergonomic Research Facility Contact UsCERF HomeEyeTracking IncSDSU