Presentation
Examining Personality and Decision-Making Style as Predictors of Automated Aid Efficiency
SessionWednesday Poster Session
DescriptionAutomated diagnostic aids can improve performance in difficult detection tasks, yet people often underuse them and fall short of optimal performance. We tested whether stable individual differences help explain this gap. Participants completed a signal detection task with and without a highly reliable (93%) automated aid, then reported their personality traits (conscientiousness, neuroticism) and decision-making tendencies (satisficing, decision difficulty). We computed each person’s aid-use efficiency by comparing their observed aided performance to the mathematically optimal level predicted by signal detection theory.
Although the aid substantially improved performance overall, individual differences did not predict how efficiently participants used the aid. Bayesian analyses favored the null for all preregistered correlations between aid-use efficiency and conscientiousness, neuroticism, satisficing, and decision difficulty. Exploratory results showed expected relationships among the traits themselves, but these did not translate into differences in aid use.
These findings suggest that the common tendency to underuse decision aids may reflect general cognitive limitations rather than personality-based reliance strategies. For designers and practitioners, this implies that improving human–automation teaming should focus on system design and training that support effective integration for all users, rather than attempting to identify “better” aid users based on personality.
Although the aid substantially improved performance overall, individual differences did not predict how efficiently participants used the aid. Bayesian analyses favored the null for all preregistered correlations between aid-use efficiency and conscientiousness, neuroticism, satisficing, and decision difficulty. Exploratory results showed expected relationships among the traits themselves, but these did not translate into differences in aid use.
These findings suggest that the common tendency to underuse decision aids may reflect general cognitive limitations rather than personality-based reliance strategies. For designers and practitioners, this implies that improving human–automation teaming should focus on system design and training that support effective integration for all users, rather than attempting to identify “better” aid users based on personality.
Contributor
Event Type
Poster
TimeWednesday, October 21st5:30pm - 6:30pm PDT
Location
Similar Presentations