Anxiety-Related Stroop Interference in Adolescents

Anxiety-Related Stroop Interference in Adolescents

Anne Richards

ABSTRACT. A group of 16- to 18-year-old students was presented with threat-related and neutral Stroop stimuli on separate cards. Participants were assigned to anxiety groups on the basis of their scores on the Beck Anxiety Inventory (BAI; A. T. Beck & R. A. Steer, 1990). It was found, as predicted, that the high-anxiety group took significantly longer to identify the color of the threat-related word than the neutral words, whereas there was no difference for the low-anxiety group. There was a significant linear relationship between interference on the task and BAI scores, showing that as anxiety increases there is a corresponding increase in interference produced by the threat-related stimuli when compared with the neutral stimuli. This study demonstrates an anxiety-related Stroop interference effect for adolescents consistent with that reported in the adult literature.

IT IS NOW WELL ESTABLISHED that in the adult population there are cognitive biases present that are related to anxiety (see Power & Dalgleish, 1997; Williams, Watts, MacLeod, & Mathews, 1997). Despite this, researchers have largely ignored the relationship between cognition and emotion in adolescent populations. It is of obvious interest to examine the development of cognitive biases in younger populations, because it is not known whether the processes underlying information processing and anxiety are the same for younger people as they are for adults. Martin, Horder, and Jones (1992) used an emotional spider Stroop paradigm in which children between the ages of 6 and 13 were required to name the colors of different types of stimuli including spider-related words. They found a cognitive processing bias in spider-phobic 6- or 7-year-olds; this bias remained constant for the older children and was of a size comparable to that observed in adult populations.

There have been a few other studies examining the relationship between cognition and emotion in children and the findings are generally in line with those obtained from adult populations. Using the probe detection paradigm (MacLeod, Mathews, & Tata, 1986), Vasey, Daleiden, Williams, and Brown (1995) demonstrated an attentional bias toward threatening information in clinically anxious 9-to 14-year-old children, and a similar bias in nonreferred 11- to 14-year-old children with high levels of test anxiety (Vasey, El-Hag, & Daleiden, 1996). Recently, Hadwin, Frost, French, and Richards (1997) developed a pictorial analogue of the threat/neutral homophone task. In the adult version a series of threat/neutral homophones (e.g., bury/berry) are presented auditorily, and the participants are required to spell the words. To avoid the spelling component of the exercise, the researchers presented the words auditorily, and the children were required to choose between two pictures that corresponded to what they had heard . It was found that anxious children exhibited an interpretation bias that caused them to be more likely to choose the picture that corresponded to the threatening rather than the neutral interpretation of the homographs.

Very few researchers have used the emotional Stroop paradigm to examine the relationship between cognition and emotion in adolescents. In a study by Schwartz, Snidman, and Kagan (1996), adolescents who had been classified as behaviorally inhibited or uninhibited during their second year of life were presented with an emotional Stroop task. The authors predicted that adolescents who were inhibited as infants would display threat-related interference compared with those who were uninhibited as infants. This prediction was not supported, although the authors noted that there was a greater number of delayed reactions to threatening words for the inhibited group than for the uninhibited group. However, this latter analysis was performed on a post hoc basis and obviously provides only tentative support for their hypothesis.

Moradi, Taghavi, Doost, Yule, and Dalgleish (1999) obtained interference effects in the predicted direction in posttraumatic stress disordered children and adolescents between the ages of 9 and 17. A study was carried out by Doost, Taghavi, Moradi, Yule, and Dalgleish (1997) in which 9- to 18-year-old children who were classified as clinically depressed, mixed-anxious, depressed, and a control group were presented with differentially valenced Stroop words. The normal controls were faster overall in identifying the color of all words, but there were no differences between the differently valenced words or interactions involving valence. Within the adult literature, it is clear that anxiety is more robustly associated with an encoding bias than is depression (see Wells & Matthews, 1994; Williams, Mathews, & MacLeod, 1996; Williams, Watts, MacLeod, & Mathews, 1997). In particular, mood-congruent emotional Stroop interference has been associated with anxiety using both clinical (e.g., Mathews & MacLeod, 1985) an d nonclinical samples (e.g., MacLeod & Rutherford, 1992; Richards, French, Johnson, Naparstek, & Williams, 1992; Richards & Millwood, 1989).

The present investigation involved a card-based emotional Stoop paradigm to test for an anxiety-related interference effect in a sample of 16- to 18-year-old students. Anxiety was measured using Beck’s Anxiety Inventory (BAI; Beck, Epstein, Brown, & Steer, 1988; Beck & Steer, 1990). This questionnaire was developed as a measure of adult anxiety and has a two-factor structure. These two factors are characterized by physical and cognitive symptoms. Creamer, Foran, and Bell (1995) argue that the BAI is superior to the Spielberger State-Trait Anxiety Inventory (STAI; Spielberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983) in differentiating anxiety from depression. Although the BAI was developed for use with an adult population, it has been used with younger populations. Steer, Kumar, Ranieri, and Beck (1995) administered the BAI to a group of 105 adolescent outpatients between 13 and 17 years of age. They argued that the BAI is a valid tool for investigating self-reported anxiety in outpatient adolescents. Jolly, Aruffo, Wherry, and Livingston (1993) examined the reliability, convergent validity, and discriminant validity of the BAI with a group of 180 adolescents. They found that the BAI was readable for a group of 14-year-old patients, and had high internal consistency (Cronbach’s [alpha] of .94), and high item-total score correlations (M = .66). In addition to this concurrent validity was demonstrated by the significant correlations between ratings by clinicians (r = +.4) and self-report (r = +.58) measures of anxiety. Jolly et al. used a clinical sample, whereas the present experiment examined nonclinical anxiety. However, Creamer et al. (1995) examined the properties of the BAT in a nonclinical sample of 326 undergraduates. They found that the BAI demonstrated a high level of internal consistency, and they observed that the scale appeared to be measuring state anxiety rather than trait anxiety.



Thirty adolescents (13 girls and 17 boys) took part in the experiment. The participants were between 16 and 18 years of age; the average age was 16.45 years. They were part of an opportunist sample taken from students attending a sixth form college (equivalent to a senior high school).


Eight threat-related words and eight matched-neutral words were selected, and two Stoop cards were prepared. The words were matched for length and frequency using Hofland and Johansson (1982) word frequency norms (see Appendix). For the threat-related Stroop card, each threat-related word was presented eight times, twice in each of the four colors: red, green, yellow, and blue. Therefore, there was a total of 96 words on each of the two cards.

The order in which the two Stroop cards were presented to the participants was balanced so that half of the participants were presented with the threat-related card first followed by the neutral card, and the other half of the participants were presented with the neutral card first followed by the threat-related card. All of the participants were presented with the practice card before the two experimental cards.


Participants were tested individually in a quiet room. On arrival, the participants were presented with a sheet instructing them to identify the color of each word on the card as quickly as possible while maintaining accuracy. They were also instructed not to pay attention to the content of the word but to simply identify the color of the ink in which it was written. The time taken to identify the colors of the words on two experimental cards was recorded using a stopwatch.

At the end of the session, the participants completed the BAI.


Participant Characteristics

Participants scoring above the median score of 10.5 on the BAI were assigned to the high-anxiety group, and the remaining participants were assigned to the low-anxiety group. The low-anxiety group had a mean BAI score of 5.4 (SD = 3.0) and comprised 10 boys and 5 girls (mean age of 16.00 years), and the high-anxiety group had a mean BAI score of 16.3 (SD = 5.2) and comprised 7 boys and 8 girls (mean age of 16.9 years). Gillis, Haaga, and Ford (1995) found the average score for a nonclinical sample of adults between 18 and 44 years of age was 7.3 (50th percentile score was 4). Therefore, this sample was more anxious than that reported by Gillis et al.

Emotional Stroop Data

The time taken (in seconds) to name each Stroop card was recorded, and a 2 (Group: high vs. low anxiety) x 2 (Order: threat first vs. neutral first) x 2 (Valence: threat vs. neutral) analysis of variance was performed on the data. This analysis revealed a main effect of Valence, with responses being generally longer to threat than neutral stimuli (means of 71.0 and 65.7 s, respectively), F(l, 26) = 20.2l,p [less than].001, and a main effect of Group, with the high-anxious group overall taking longer to respond than the low-anxious group (means of 72.5 and 64.3 s, respectively), F(1, 26) = 5.24, p [less than].05. However, there was an interaction between Group and Valence, F( 1, 26) = 7.42, p [less than] .01. Two planned t tests, one for each group, were performed to compare the time taken to identify the colors of the threat-related words to the time taken to identify the colors of the neutral words. There was no significant difference in the time taken by the low-anxiety group to respond to the threat-related words (mean latency of 65.3 s, SD = 10.0) in comparison with the neutral words (mean latency of 63.2 s, SD = 8.2), t(14) = 1.23, but the high-anxiety group took significantly longer to identify the colors of the threat-related words (mean of 76.6 s, SD = 11.5) in comparison with the neutral words (mean of 68.3 s, SD = 11.3), t(14) = 5.8, p [less than] .001. There were no main effects or interactions involving the order of presentation factor.

An interference index was calculated by subtracting the time taken to identify the colors of the threat-related words from the time taken to identify the colors of the neutral words. A correlation performed on the whole group between this difference index and the BAI scores revealed a significant linear relationship (r = +.42). Thus, as anxiety increases there is a corresponding increase in the amount of interference produced by the threat-related stimuli compared with the neutral stimuli.


The present investigation reveals anxiety-related Stroop interference effects in line with prediction. High-anxiety adolescents show interference effects comparable to those demonstrated in the adult literature. This is the first study to examine anxiety-related Stroop interference in a nonclinical sample of adolescents. The results are interesting because they show that there is a linear relationship between anxiety and interference in a sample of normal adolescents. These findings suggest that one possible reason for the failure of Doost et al. (1997) to obtain interference effects in their sample is that such interference may be more strongly related to anxiety than to depression (as is typically found in adult populations). This experiment has demonstrated that there is a cognitive bias associated with anxiety in adolescence that is consistent with that obtained in adulthood.


Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory for measuring clinical anxiety-psychometric properties. Journal of Consulting and Clinical Psychology, 56, 893-897.

Beck, A. T., & Steer, R. A. (1990). The Beck Anxiety Inventory. Austin, TX: The Psychological Corporation.

Creamer, M., Foran, J., & Bell, R. (1995). The Beck Anxiety Inventory in a nonclinical sample. Behaviour Research and Therapy, 33, 477-485.

Doost, H. T., Taghavi, M. R., Moradi, A. R., Yule, W., & Dalgleish, T. (1997). The performance of clinically depressed children and adolescents on the modified Stroop paradigm. Personality and Individual Differences, 23, 753-759.

Gillis, M. M., Haaga, D. A. F., & Ford, G. T. (1995). Normative values for the Beck Anxiety Inventory, Fear Questionnaire, Penn State Worry Questionnaire, and Social Phobia and Anxiety Inventory. Psychological Assessment, 17, 450-455.

Hadwin, J., Frost, S., French, C. C., & Richards, A. (1997). Cognitive processing and anxiety in mainstream children: Trait anxiety and interpretation bias. Journal of Abnormal Psychology 106, 486-490.

Hofland, K., & Johansson, 5. (1982). Word frequencies in British & American English. Bergen, Norway: Longman.

Jolly, J. B., Aruffo, J. F., Wherry, J. N., & Livingston, R. (1993). The utility of the Beck Anxiety Inventory with inpatient adolescents. Journal of Anxiety Disorders, 7, 95-106.

MacLeod, C., Mathews, A., & Tata, P. (1986). Attentional biases in emotional disorders. Journal of Abnormal Psychology 95, 15-20.

MacLeod, C., & Rutherford, E. M. (1992). Anxiety and the selective processing of threatening information: Mediating roles of awareness, trait and state variables, and personal relevance of stimulus materials. Behaviour Research and Therapy, 30, 67-68.

Martin, M., Horder, P., & Jones, G. V. (1992). Integral bias in naming of phobia-related words. Cognition and Emotion, 6, 479-486.

Mathews, A., & MacLeod, C. (1985). Selective processing of threat cues in anxiety states. Behaviour Research and Therapy, 23, 563-569.

Moradi, A. R., Taghavi, M. R., Doost, H. T. N., Yule, W., & Dalgleish, T. (1999). Performance of children and adolescents with PTSD on the Stroop color-naming task. Psychological Medicine, 29, 415-419.

Power, M., & Dalgleish, T. (1997). Cognition and emotion: From order to disorder. Hove, UK: Psychology Press.

Richards, A. French, C. C., Johnson, W., Naparstek, J., & Williams, J. (1992). Effects of mood manipulation on performance of an emotional Stroop task using mixed and blocked presentation. British Journal of Psychology, 83, 479-491.

Richards, A., & Millwood, B. (1989). Color-identification of differentially valenced words in anxiety. Cognition and Emotion, 3, 171-176.

Schwartz, C. E., Snidman, N., & Kagan, J. (1996). Early temperamental predictors of Stroop interference to threatening information at adolescence. Journal of Anxiety Disorders, 10, 89-96.

Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the State-Trait Anxiety Inventory (Form Y). Palo Alto, CA: Consulting Psychologists Press.

Steer, R. A., Kumar, G., Ranieri, W. F., & Beck, A. T. (1995). Use of the Beck Anxiety Inventory with adolescent psychiatric outpatients. Psychological Reports, 76,459-465.

Vasey, M. W., Daleiden, E. L., Williams, L. L., & Brown, L. M. (1995). Biased attention in childhood anxiety disorders: A preliminary study. Journal of Abnormal Child Psychology, 23, 267-277.

Vasey, M. W., El-Hag, N., & Daleiden, E. L. (1996). Anxiety and the processing of emotionally threatening stimuli: Distinctive patterns of selective attention among high- and low-test-anxious children. Child Development, 67, 1173-1185.

Wells, A., & Matthews, G. (1994). Attention and emotion: A clinical perspective. Hove, UK: Erlbaum.

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

Williams, J. M. G., Watts, F. N., MacLeod, C., & Mathews, A. (1997). Cognitive psychology and emotional disorders. Chichester, UK: Wiley.


Matched threat and neutral words (frequencies in parentheses; Hofland & Johansson, 1982)

Crazy-Draft (22)

Tense-Sweep (11)

Panic-Ample (15)

Fear-Note (111)

Shaking-Shelves (15)

Nervous-Element (53)

Dying-Breed (21)

Frightened-Provincial (27)

COPYRIGHT 2000 Heldref Publications

COPYRIGHT 2001 Gale Group