Chapter 5 - Attention and performance

Chapter Summary


Simulations of Key Experiments

Case Studies

Automatic processes, attention and the emotional Stroop effect

The Stroop effect (Stroop, 1935) occurs when participants are required to name the colour in which a set of words are printed:

  • In the congruent condition, the words are colour words that are printed in the same colour as the word itself (e.g., the word blue printed in blue ink or the word red printed in red ink).
  • In the incongruent condition, the word and the colour in which it is printed differ (e.g., the word blue printed in red ink).

As you might predict, participants are slower to name the ink colours in the incongruent condition than in the congruent condition. The reasoning is that reading the words, as an automatic process, cannot be avoided, and this interferes with naming the colours.

In the emotional Stroop task (e.g., Williams et al., 1996), participants are required to name the ink colour of words, just as in the Stroop task, however now the words are either emotional words (especially threatening words, such as kill) or neutral words. When the participants used in this study had high levels of trait anxiety (that is, they had a long-term disposition towards being anxious, though not necessarily clinically anxious), they were slower to name the colour of the emotional words compared to the neutral words, and slower than those participants with low trait anxiety. Since the slowed responding in the Stroop task is thought to be due to automatic processing of the words, it is not inconsistent to suppose that the slowed responses of high-trait anxiety individuals on the emotional Stroop task are due to the automatic processing of emotional information.

Studies such as this have raised the issue of whether there is an attentional bias towards threat in individuals with raised levels of anxiety, and whether such biases are automatic (in the sense that they occur rapidly and without voluntary intent).

If you are afraid of spiders you may have first-hand experience of this idea. Imagine you are in the attic looking for grandpa’s old photographs and you notice something out of the corner of your eye. Given that it’s an attic, full of old cobwebs, you would be forgiven for assuming that the blob in the corner of your vision is a spider! Someone without a fear of spiders might either (a) not even notice the blob or (b) assume that it was just a piece of dust.

This example demonstrates what has been found in people who are prone to anxiety – they tend to notice potential dangers and threats much more easily and sooner than people who are less prone to anxiety. Clearly, there is a pay-off for the anxious person who notices a threat sooner rather than later: they can take evasive action before being harmed. However, anxious people too often make the error of assuming that something is dangerous or threatening when it is actually harmless. This rapid switching of attention towards threat has been observed in the laboratory in anxious students as well as in people diagnosed with an anxiety disorder.

In many of these experiments, highly anxious individuals have been found to direct their attention towards items related to personal threat (e.g., words related to health, or photographs of mutilation and so on) when these items appear alongside more neutral items. For example, when a smiling face and an angry face are presented at the same time on a computer screen, the anxious individual looks at the angry face first.

Apart from the emotional Stroop task, another much-used task is the “dot probe” method for measuring attention, as devised by MacLeod and Mathews (1988). The idea is that a probe, such as an asterisk (*), is presented on a computer monitor, and the participant is required to press a button as soon as he or she sees it appear. Now, if the person just happens to be looking in the place where the probe appears, then they will respond faster than if they had been looking in a different area of the screen. Therefore we can measure the reaction time of the button press, and it will be related to where the person was looking and where the asterisk appeared. Two words are presented on the screen, one on the left side and the other on the right side – one word is a threatening word (e.g., dagger) and the other word is a non-threatening word (e.g., number). Suppose dagger is presented to the left and number to the right. An anxious person would orient towards the word dagger, while a low-anxiety person would treat each word equally. The probe is then presented on the screen in the middle of the word dagger (e.g., dag*ger). If the participant is looking at dagger, then they will be able to respond sooner than if they had been looking at the word number. In this way it is possible to infer the direction of attention by use of a probe and reaction time.

Specificity of attentional bias

Research has shown that what an anxious person fears most grabs most attention. As examples:

  • People with a spider phobia attend to words such as hairy or creepy.
  • People with a social phobia attend to words related to socialising (e.g., party).
  • People with an eating disorder attend to words related to food (e.g., chocolate).
  • Anxious students tested just before their exams have been shown to attend to words related to success and failure (e.g., error).

The tendency for anxious individuals to attend to threat does not seem to depend on conscious awareness or deliberate intent, as attentional bias has been found to exist even for stimuli that cannot be reported, such as when the words are presented subliminally (e.g., Mogg et al., 1993). It may therefore be a behaviour that is difficult to control or suppress (and this then obviously has implications for therapy).


MacLeod, C. & Mathews, A. (1988). Anxiety and the allocation of attention to threat. The Quarterly Journal of Experimental Psychology,40A: 653–70.

Mogg, K., Bradley, B.P., Williams, R. & Mathews, A. (1993). Subliminal processing of emotional information in anxiety and depression. Journal of Abnormal Psychology, 102(2): 304–11.

Stroop, J.R. (1935). Studies of interference in serial verbal reaction. Journal of Experimental Psychology, 18: 643–62.

Williams, J.M., Mathews, A. & MacLeod, C. (1996). The emotional Stroop task and psychopathology. Psychological Bulletin, 120: 3–24.


Multiple Choice Quiz