Measuring Driving Workload using NASA TLX
During Apex Performance Engineering’s inaugural year of competing in Ontario Time Attack, we emphasized driver development as an important area to improve. We know that high-speed vehicle maneuvering is a complex task that puts the driver under cognitive, physical and psychological stress. What we do not know is how we are affected by these demands in our unique case.
We measure driver workload using the NASA Task Load Index administered during Ontario Time Attack events six and seven. The purpose is to quantify the driver’s subjective experience and correlate it to driver development strategies.
Shannonville Motorsports Park - A New Challenge
Ontario Time Attack events six and seven posed a unique challenge. Because of our limited experience with the race track, the driver faced an additional challenge of learning the course with limited prior knowledge. We participated in a lapping event in ‘long’ configuration one week prior to the Ontario Time Attack weekend which took place on both ‘pro’ and ‘long’ configurations.
As a driver coach, how can I support the driver achieve peak performance as quickly as possible? This question was particularly important because we would only have so much time to deliver on performance objectives.
NASA Task Load Index - Measuring Workload
Because of the challenge of learning a new track, workload as a metric is interesting because the task changes over time with more experience. Session types varied from lapping, qualifying and competition, providing a diverse set of conditions the driver must adapt to.
We used the NASA Task Load Index (TLX) to capture the driver’s subjective evaluation of workload. When permissible, the driver would complete the NASA TLX rating procedure after a session. The procedure asks the subject to rate the relevancy of six factors to the task.
The NASA TLX rating system is unique in that it combines the weighted importance of each factor with the magnitude of their subjective rating. This increases the sensitivity of the rating scale. Overall workload is derived as a combination of the adjusted scores of each factor.
Workload Results
As we developed our driver development program, we questioned whether the competitive aspect added to the perceived workload by the driver. I hypothesized that the driver would experience higher workload due to competitive pressures. To test this hypothesis, lets look at the overall workload and identify the measurements by session type.
Contrary to the hypothesis, lapping sessions were considered the most demanding. Because it was our first time on this track, the additional challenge of learning it increased the overall workload.
This leads us to another question; how did workload change over time as we progressed? Let’s look at the components of workload in chronological order. This give us an estimate of how workload changes as we gained experience.
Majority of the workload is caused by three factors: mental demand, physical demand and temporal demand. Early on, the largest contributor to overall workload was mental demand. However, as we gained more experience, mental demand contributed less to overall workload and is overtaken by physical demand to become the largest contributor. Temporal demand varies between sessions and is less correlated with time.
Since the NASA TLX rating procedure consists of two parts, we wanted to understand how adjusted scores differed from the raw rating by the driver. Let’s update our charts to look at raw scores instead of adjusted scores.
Raw workload scores are relatively consistent between measurements. Some subtleties are noteworthy; performance and mental demand decrease as time progresses. However, the overall consistency of the raw ratings suggests that the variable nature of the overall workload is caused by changes in the importance of each factor, not by the magnitude which the driver scored each demand.
Driver Feedback
Qualitative understanding of the driver’s experience brought much needed insight into the results. Driving is highly experiential, so articulating subjective experiences is not always simple. These results gives us a foundation to discuss our progress.
The challenge of learning the track dominated our discussion. Despite the substantial preparation and research put into the track prior to lapping, the driver found it cognitively demanding to associate the new information with the preparation. It was only until the driver had a high fidelity mental model of the track did he feel ready to maneuver the vehicle at the limit.
As the driver began to take full advantage of the vehicle’s potential, physical demand became the primary contributor of workload. The cause of this increase is two-fold: operating the vehicle at the limit requires more physical movement of the controls which in turn increases the g-forces acting on the driver.
Driver Development Plan - A Personalized Approach
The goal of training to improve fitness levels is to minimize the stress the body experiences during competition. Because each driver is unique in their experience and needs, a personalized approach is necessary to focus on areas of greatest importance.
One aspect of high speed vehicle maneuvering is driving technique. This is not in the scope of our discussion. Rather, we focus on preparing the driver physically and mentally to successfully execute their chosen driving strategy. It is noteworthy that the three largest contributors to overall workload are factors imposed on the driver due to the nature of the task, not due to the interaction between the driver and the task.
Here are a few strategies we brainstormed that may help condition the driver face the imposed demands:
| Demand | Strategy |
|---|---|
| Mental | Perceptual-cognitive training, visualization |
| Physical | Resistance training, heat acclimation, reaction time, diet and nutrition |
| Temporal | Rhythm training, multitasking |
Not all strategies require active training. Combating physical fatigue could involve proper nutrition and hydration. Meeting nutritional requirements can be difficult at the race track; packing your own lunch and snacks could help reduce fatigue and therefore improve driving performance. Reducing temporal demand may be as simple as scheduling lapping sessions well in advance to give yourself enough time to learn the track without feeling pressured.
Active training strategies would replicate driving workloads to prepare your body for high performance driving. Resistance training would focus on strengthening the neck and upper body to allow the driver to maintain vehicle control under high g-loading. Aspects such as balance and coordination would aim to increase the driver’s perceptual sensitivity to vehicle balance.
Closing Remarks
Tweaking the performance of the driver is equally - if not more important as improving the performance of the vehicle. The NASA TLX workload assessment provided insight into demand factors affecting the driver. Being able to communicate these concerns in a concise and quantifiable way allows us to address the concerns in meaningful way and remove barriers to achieving high performance.
Were we successful in achieving our performance goals? At our first lapping day, we set a baseline lap time of 2:06.16. By the end of event seven, we had set a lap time of 2:01.53. We feel there are merits to accelerated driver development, though there is still much opportunity to improve total system performance.
Acknowledgements
Ping Cheng Zhang at Apex Performance Engineering graciously allowed his workload data and subjective feedback comments to be shared. A big thank you for your trust and patience as we discussed at length the specifics of the Shannonville weekend.
References
- Hart, Sandra G., and Lowell E. Staveland. “Development of NASA-TLX (Task Load Index): Results of empirical and theoretical research.” In Advances in psychology, vol. 52, pp. 139-183. North-Holland, 1988.
- Hart, Sandra G. “NASA-task load index (NASA-TLX); 20 years later.” In Proceedings of the human factors and ergonomics society annual meeting, vol. 50, no. 9, pp. 904-908. Sage CA: Los Angeles, CA: Sage publications, 2006.
- Potkanowicz, Edward S., and Ronald W. Mendel. “The Case for Driver Science in Motorsport: A Review and Recommendations.” Sports Medicine 43, no. 7 (July 2013): 565–74. https://doi.org/10.1007/s40279-013-0040-2.
- Faubert, Jocelyn, and Lee Sidebottom. “Perceptual-cognitive training of athletes.” Journal of Clinical Sport Psychology 6, no. 1 (2012): 85-102.
- Hoyes, Kevin, and Dave Collins. “Fit to race: Identifying the balance, type and sources of knowledge in fitness for Motorsport.” International Journal of Sports Science & Coaching 13, no. 5 (2018): 751-760.
- Ferguson, David P. The Science of Motorsport. Routledge, 2018.