<<Back to Cognition Class Home Page
---------------------------------------------------------------------------------------
THE STROOP EFFECTThe Stroop Effect is one of the most studied phenomena in cognitive psychology. When Colin McLeod wrote his important review of the Stroop effect in 1991, he counted more than 700 published paper on the Stroop effect. Since then, researchers have continued to write articles on the Stroop effect. In PsycInfo, which lists only articles written after 1984, I found 880 articles that contain the word "STROOP".
Why is this effect so interesting? We will talk about this later. First, I want you to have some experience with this task and try to think about the possible explanations to account for it.
STROOP EXPERIMENT IN CLASS AND GROUP ACTIVITY
COMPONENTS OF THE STROOP EFFECT
1) Automatic vs controlled processes
What is the difference between automatic and controlled processes?
Two processes are engaged in this task.
A) COLOR NAMING.The instructions ask the subject to name the color. In order to correctly perform the task we have to process the color, to activate the associated name and to say aloud the name
B) WORD READING. Although the instructions specifically require not to read the word, we are so well practiced in word reading that the presentation of the word AUTOMATICALLY activates the word name.
2) Resolution of a conflict
Even though both color name and word name are activated, we nonetheless are able (almost always) to give the correct response. How do we do this? We have to assume that even if the word is strongly associated to the response of reading the word, there is some mechanisms that allows us to overcome the automatic response and to be able to produce a less automatic and less practiced response.
As you can see, this mechanism has to do with our ability to be flexible and to deal with novel situations, where well learned responses are not appropriate. In some way this capacity to solve conflict is the exact opposite of what behaviorist were proposing. They thought that all behavior could be explained with a complex set of learned association between stimuli and responses.
One model of this conflict resolution mechanism is Norman and Shallice's model of goal oriented behavior. According to Norman and Shallice's model, we have a mechanism that is like our "behaviorist self" and is called "CONTENTION SCHEDULING". We have a big set of well learned schemes of action, that is, organized actions associated to a goal. According to Norman and Shallice, these schemes have different levels of activations, depending on the current external stimulation and other internal states such as intentions or motivations. These schemes compete with each other and the scheme that receives more activation tend to win over the others and be executed. This type of process can be sufficient in very well learned and "standard" situations. There are situations, however, in which the contention scheduling will not work.
Normal and Shallice list five of this situations:
Situations that:
1) require planning or decision making
2) require error correction or troubleshooting
3) require to create novel and not well learned links between the input and the action to execute
4) are difficult or dangerous
5) require a response that competes with a strong habitual response
The last condition is exactly what occurs during the Stroop task. A strong habitual response (reading the word) competes with a weaker response (name the color). In this type of conditions, a second mechanism needs to be activated. Norman and Shallice call it SUPERVISORY ATTENTIONAL SYSTEM. The function of this mechanism is to modify the activations of the action schemes so that the one that is more appropriate for that situation, even when less activated by external factors, will be able to win the competition with other, more activated schemas.
Can you see what is the problem with this model?
One problem with this approach is that the Supervisory Attentional System (SAS) look a lot like the "small man in the head" that Diego was talking about, and it is not very clear how to implement this mechanism in the brain.
A very likely candidate for this function is the anterior cingulate. In neuroimaging studies, the anterior cingulate is activated in in situations similar to the one proposed by Normal and Shallice, and when a strong conflict (as in the case of the Stroop task) is present.
---------------------------------------------------------------------------------------