Suppose you are asked to name a few pictures. You are shown the drawing of a tiger and you say “tiger”. Then a mouse appears and you say “mouse”, and so on. The experimenter avoids pictures of a gerenuk or babirusa, so you are cruising along nicely.

And then this pops up:

The holiday is over and you struggle to say “bird”—whoops, no, actually you are supposed to say “rabbit!”

This “Stroop” effect, named after its discoverer, is arguably one of the largest and most reliable effects in human cognition. The effect comes in many variants, and most often involves either a printed word superimposed on a picture, as in the present case, or it involves a color-naming task in which the color ink in which words are printed must be named. Inconveniently, the words themselves sometimes spell out a conflicting color. For example, the word “green” might be printed in red ink and people have to say “red”. (I would have liked to print the word in red ink, but this editor will not allow that.)

The conventional interpretation of the effect is that it reveals our inability to suppress reading, because reading is an “automatic” process that runs to completion, whether we like it or not. In this instance, when the printed word is in conflict with the primary picture-naming (or color-naming) task, we cannot help but be slowed or confused by what we read.

There are, however, some interesting boundary conditions on this effect. For example, when a person has to identify a word to which they spatially attend—that is, the person focuses exclusively on that location on the screen—then typically another word appearing in a different location at the same time does not influence the processing of the focal word.

This gives rise to an interesting paradox: Why does a printed word interfere with naming of a picture or of the color ink, whereas a printed word does not affect the processing of another printed word when the target word is in the focus of attention?

A recent article in the Psychonomic Bulletin & Review has shed some light on this paradox. Researchers Serje Robidoux and Derek Besner argued that when one spatially focuses on a word, this is highly attention demanding, and that it is this attention demand that prevents another word in a different location from interfering. By extension, Robidoux and Besner argued that if they could design a target stimulus for color naming that required as much attention as word reading, then the Stroop effect should disappear because no attention is “left over” for the color word to interfere with color naming.

The stimulus designed by Robidoux and Besner was a color patch that contained a mixture of colors, one of which was dominant—that is, it occupied most of the patch. For example, in the usual color-naming task the patch below would require the response “red” even though it also contained blue bits:

To make the task even more attention-demanding, the bits with the wrong color would vary from trial to trial. Hence people did not know where to focus their attention ahead of time: they had to work that out anew on each trial.

In a first experiment, Robidoux and Besner showed that the attention demands of this color-naming task were identical to that of word reading. This calibration was a necessary step for their second, crucial experiment in which the color-naming task was combined with distracting stimuli in a different spatial location.

The logic of this second study was as follows: If a distractor effect were found—that is, if the word “green” presented above the patch in the previous picture were to slow down the response “red”—then this would support the view that reading does not require spatial attention and is “unstoppable” in some sense. Conversely, if no such interference were found, then Robidoux and Besner argued it “would support the view that reading requires spatial attention and that the previous demonstrations of unattended processing were the result of color stimuli that did not appropriately control spatial attention.”

The data very clearly showed the latter outcome: The time to name the target color patch was identical, irrespective of whether it was accompanied by an incongruent distractor (“green”) or a congruent distractor (“red”).

The word did not interfere with color naming of the patch that was made up of more than one color.

To complete the story, Robidoux and Besner conducted a final experiment in which the target color patches were simple frames of a single color. As expected, the conventional Stroop effect re-appeared: The word “green” above a uniformly red square slowed color naming compared to a condition in which the word was “red”.

A good rule of thumb in experimental sciences is that if you understand a phenomenon, then you should be able to make it go away. The Stroop effect has frequently resisted such attempts to “make it go away,” but the work by Robidoux and Besner has shown us how this can be accomplished. If all attention is focused on the color-naming stimulus, then an incongruent distractor word no longer interferes. Or, in the authors’ words, “there is no word processing without spatial attention.”