Reynolds et al. (2006) employed a task that required the participants to memorize four target numbers that indicated a Go response, while a further four numbers were classified as No-go stimuli. Interestingly Reynolds found a larger effect size than that found in the current study, r = .39 vs. .15, despite the fixed nature of the targets 4. Both the study presented here, and the study of Reynolds and colleagues (2006), observed a relationship between the BIS- 11 subscale of Cognitive Complexity and performance on inhibitory performance, which is in contrast to the results obtained using the standard Go/No-go task employed by Keilp et al. (2005) that related impulsivity to the higher-order factors of Attentional and Motor Impulsiveness of the BIS-11. While information on the individual subscales was not reported (Keilp et al., 2005), Cognitive Complexity forms part of the BIS higher-order factor of Non-planning, which did not show a relationship with performance on their Go/No-go inhibition task (Keilp et al., 2005). Taken together these findings suggest that it is the complexity of the Go/No-go tasks used here (alternating rule) and by Reynolds et al. (multiple targets) which is the key manipulation that leads to reduced performance related to high levels of the BIS first order factor of Cognitive Complexity. What remains to be determined is an operational definition of ‘complexity’. Despite both the study here and Reynolds et al. utilising a more complex task, the nature of that complexity is not identical. In Reynolds et al. the complexity is in the form of a higher WM requirement, with 4 items to be responded to, and 4 to inhibit. Here the WM load effect (i.e. going from PGNG level 2 to level 3) did not show any interaction with Cognitive Complexity, but was maximally a WM load of only two items; rather the complexity was derived from the shifting nature of the No-go targets.