Publication Date


Document Type


Committee Members

Dragana Claflin (Committee Member), Jennie Gallimore (Committee Chair), Yan Liu (Committee Member), Chandler Phillips (Committee Member), Waleed Smari (Committee Member)

Degree Name

Doctor of Philosophy (PhD)


Extreme acceleration maneuvers encountered in modern agile fighter aircraft can wreak havoc on human physiology thereby significantly influencing cognitive task performance. Increased acceleration causes a shift in local arterial blood pressure and profusion causing declines in regional cerebral oxygen saturation. As oxygen content continues to decline, activity of high order cortical tissue reduces to ensure sufficient metabolic resources are available for critical life-sustaining autonomic functions. Consequently, cognitive abilities reliant on these affected areas suffer significant performance degradations. This goal of this effort was to develop and validate a model capable of predicting human cognitive performance under acceleration stress. An Air Force program entitled, "Human Information Processing in Dynamic Environments (HIPDE)" evaluated cognitive performance across twelve tasks under various levels of acceleration stress. Data sets from this program were leveraged for model development and validation. Development began with creation of a proportional control cardiovascular model that produced predictions of several hemodynamic parameters including eye-level blood pressure. The relationship between eye-level blood pressure and regional cerebral oxygen saturation (rSO2) was defined and validated with objective data from two different HIPDE experiments. An algorithm was derived to relate changes in rSO2 within specific brain structures to performance on cognitive tasks that require engagement of different brain areas. Data from two acceleration profiles (3 and 7 Gz) in the Motion Inference experiment were used in algorithm development while the data from the remaining two profiles (5 and 7 Gz SACM) verified model predictions. Data from the "precision timing" experiment were then used to validate the model predicting cognitive performance on the precision timing task as a function of Gz profile. Agreement between the measured and predicted values were defined as a correlation coefficient close to 1, linear best-fit slope on a plot of measured vs. predicted values close to 1, and low mean percent error. Results showed good overall agreement between the measured and predicted values for the rSO2 (Correlation Coefficient: 0.7483-0.8687; Linear Best-Fit Slope: 0.5760-0.9484; Mean Percent Error: 0.75-3.33) and cognitive performance models (Motion Inference Task - Correlation Coefficient: 0.7103-0.9451; Linear Best-Fit Slope: 0.7416-0.9144; Mean Percent Error: 6.35-38.21; Precision Timing Task - Correlation Coefficient: 0.6856 - 0.9726; Linear Best-Fit Slope: 0.5795 - 1.027; Mean Percent Error: 6.30 - 17.28). The evidence suggests that the model is an accurate predictor of cognitive performance under high acceleration stress across tasks, the first such model to be developed. Applications of the model include Air Force mission planning, pilot training, improved adversary simulation, analysis of astronaut launch and reentry profiles, and safety analysis of extreme amusement rides.

Page Count


Department or Program

Ph.D. in Engineering

Year Degree Awarded


Included in

Engineering Commons