It has been shown that the sooner individuals return to a task that they had previously stopped doing, the longer it will take them to complete that task again. This has been shown in experiments that have looked at the time it takes to do a task, while taking into consideration what task was undertaken two tasks prior to the present task. Such experiments have been designed around continual switching between three tasks, ensuring that a task never immediately repeats itself, giving a number of ABA and CBA sequences. In an ABA
sequence, participants repeat what they were doing two tasks prior to the present task, a lag-2 repeat, where as in a CBA sequence they do not repeat what they were doing two tasks prior to the present task, a lag-2 non repeat sequence. Comparisons of these sequences have shown that reaction time to task A, in an ABA sequence, on average is longer than when task A is in
a CBA sequence.
This difference in response times has been linked to a process of inhibition, where a new task is believed to make the components of a previous task that interfere with the correct response to that new task, less active. This means that the sooner individuals return to a previously inhibited task, the longer it will take them to complete it again. This difference in response times between trials in an ABA or CBA sequence has been labelled as a backward inhibition cost.
There is still a great deal of debate about what components of that task were previously made less active, and this may be because of the variety of methodologies that have been used to study this question. What has become apparent is that the cue and its relationship to the target, may play some part in the backward inhibition that occurs.
The experiments in this study have sought to simplify previous experiments, in order to determine how much the relationship between the cue and target may be affecting backward inhibition. This simplification involved repeating the same task, which was to identify the position of the target on the screen, while only switching between three cues and targets that have a fixed one-to-one relationship with each other. The only difference between conditions and experiments was the transparency of the relationship between the cue and target. The transparency is linked to how obvious the cue's meaning is in relation to what task it specifies should be carried out on the target.
The overall results suggest that when we use this methodology, the increases we see in backward inhibition are linked to a combination of top down inhibition of a language label, and bottom up episodic negative or positive priming of the visual image. These results could give us further insights into how methodological changes, which have previously not been thought as important, may actually be altering backward inhibition.