Performance limitations in cognitive workload: does task difficulty impact response time variability?

Cognitive workload theories typically assume linear additive impacts on performance as cognitive workload increases (O'Donnell & Eggemeier, 1986; Tsang & Wilson, 1997). Meanwhile, current trends suggest inclusion of variability analysis may be of benefit (Balota & Spieler, 1999; Brown & Heathcote, 2003; Van Zandt, 2002). The first study replicates Cooper and Shepard's mental rotation study (1973), confirming that mean response times shift in the analysis of variance (ANOVA), but that significant variability changes are also present. The second study adds a grouped-response dual tone discrimination task, revealing similar trends in means shifts and further significant increases in variability as difficulty rises. Implications for the development and analysis of cognitive theories of cognitive workload are discussed.