Inhibition
The question whether during selection of one thought, competing thoughts are actively inhibited is a hotly debated issue in cognitive science, with far-reaching theoretical and practical consequences. Over the last years, our goal has been to come up with an unambiguous empirical indicator of inhibition. |
Mayr, U., & Keele, S. (2000). Changing internal constraints on action: The role of backward inhibition. Journal of Experimental Psychology: General, 129, 4-26. PDF |
|
In the past, such clear evidence has been hard to come by, as "inhibitory" patterns of results can often be explained in terms of non-inhibitory processes. In our lab, we have come up with a way of assessing inhibition between task sets (Mayr & Keele, 2000). We compare situations in which people need to switch back to a task that they had disengaged from very recently (the color task in the figure) to situations where the switch goes to a less recently abandoned task (the orientation task in the figure). Switching back to a recently abandoned task costs extra time, supposedly because it is still inhibited. This so-called backward-inhibition cost seems to be relatively immune against alternative, non-inhibitory explanations (Mayr, 2002). It also provides a tool for examining cognitive and neural characteristics of inhibition. 
For example, Mayr, Diedrichsen, Ivry, & Keele (2006) showed that left- and right prefrontal patients had very different types of task-switching problems. Specifically, right prefrontal patients (see Figure) showed no task-set inhibition. We also recently reported results suggesting that inhibition of the immediate past may serve the function of clearing the slate for making random, unpredictable choices, a skill that can come very handy in real-life competitive situations (Mayr & Bell, 2006). In ongoing work, we are interested in showing a more direct link between inhibition and "free choice" and we are trying to establish an empirical distinction between two theoretically distinct types of inhibition: executive and associative.