HFES Day 1: Individual Difference in Performance

I just attended a very interesting panel discussion of new research approaches to correlating traditional personality trait assessments with cognitive experimental approaches in order to find individual predictions of performance.

Personality assessments tend to not predict individual performance because of the conflation of factors of trait (individual capacities like intellectual capacity or extraversion), state (temporal factors like stress or environment), and task (differences in the type of task being performed). All of the presenters were discussing research approaches to correlating factors such as genetic traits, neurological blood flow, or cognitive abilities and strategies to provide predictive models to assess an individuals performance.

A few interesting tidbits from the presentations:

• There do seem to be genetic markers that correlate with performance levels in certain types of attentiveness and working memory tasks.
• One study participant was a “super attender” and had no degredation of task accuracy over time or higher memory load.
• Increasing automation in our systems leads to operators increasing their roles as monitors of the system, which is a vigilence task (rather than a motor task in more manual assembly for example).
• Extraversion is a weak predictor of vigilence, but a combination of engagement, coping, and increased cerebral blood flow is a better predictor.
• In navigation tasks, people break into two general types: survey (who visually think of an overall map) and route (who use a series of directions or landmarks).
• Survey people can also use route techniques when necessary. Route people can’t generally use survey techniques effectively.
• Survey people are distracted by added auditory routes and perform more poorly when combined than when using either type individually.
• People with good sense of direction (survey) use all techniques more often and more effectively than people with poor sense of direction (route).
• Young people don’t seem to be learning survey techniques like direction cardinality (North, South, East, West), while older people tend to ask for such information to help in the task.
• In past research, individual differences were often looked at as error margins in the data. Only fairly recently have those differences been an area of research in themselves, and it is a rich but complex world of continued research.
• We should design our systems (and training and automation) to support individuals with different capabilities and approaches rather than “designing for the mean.”