Selective Attention in Obsessive–Compulsive Disorder

Selective Attention in Obsessive–Compulsive Disorder
[Short Reports]
Clayton, Ian C.1; Richards, Jeffrey C.2,3; Edwards, Christopher J.1
1Department of Psychology, University of Adelaide, South Australia, Australia.
2School of Behavioral and Social Sciences and Humanities, University of Ballarat, Victoria, Australia.
3Correspondence concerning this article should be addressed to Jeffrey C. Richards, School of Behavioral and Social Sciences and Humanities, University of Ballarat, Mount Helen Campus, University Drive, P.O. Box 663, Ballarat, Victoria 3353, Australia. Electronic mail may be sent to j.richards@ballarat.edu.au.
Received Date: April 18, 1997; Revised Date: June 8, 1998; Accepted Date: June 12, 1998
Abstract
Recent information-processing studies have suggested that a selective attention deficit may be involved in the symptomatology of obsessive–compulsive disorder (OCD). In this study, individuals diagnosed with OCD were distinguished from those with panic disorder and from control participants by their relatively poorer performance on a series of psychometric tasks of selective attention. These results are interpreted as supporting the hypothesis of a diminished ability of people with OCD to selectively ignore competing external (sensory) and internal (cognitive) stimuli, especially intrusive thoughts.
________________________________________


It has been suggested that a global deficit in the ability to selectively attend to relevant stimuli while concurrently screening out unimportant competing environmental and internal stimuli (particularly random thought processes) is central to the etiology of obsessive–compulsive disorder (OCD) because people with OCD appear to give excessive consideration to unimportant environmental details and to somewhat arbitrary cognitive associations (Enright & Beech, 1990, 1993b). Although this overinclusiveness is consistent with impairment in selective attention processes, surprisingly few studies have specifically and directly investigated attentional ability in people with OCD.
Martinot, Allilaire, Mazoyer, and Hantouche (1990) noted impaired performance in those with OCD on tasks sensitive to selective attention deficits. People with OCD performed more poorly than control participants on the Wechsler digit span and on time taken to complete the Trail-Making Test Form B (Reitan, 1958) and the Stroop test (Stroop, 1935), on which they also made more errors when naming the incongruous color of the written words and rarely indicated awareness of mistakes. Because performance on the Stroop test reflects selective attentional capacity to maintain a course of responses despite intrusion by extraneous competing stimuli (Comalli, Wapner, & Werner, 1962), Martinot et al. (1990) suggested that this apparent deficit in selective attention might be responsible for the relative inability of people with OCD to dismiss or selectively disattend to extraneous sensory stimuli and internally generated cognitive associations or random thoughts. However, their interpretation cannot be accepted unequivocally because global memory impairment was also found in the people with OCD.
Evidence for impairment of selective attention specifically related to less effective attentional inhibitory processes in OCD, in comparison to a mixed group of people with other anxiety disorders, was found by Enright and Beech (1990). They used a negative priming paradigm (Tipper, 1985) using the Stroop task where previously ignored information becomes the required response on a following trial (e.g., the word red printed in blue ink is followed immediately by the word green printed in red ink. The task is for the participant to name the ink color on each trial). In control participants, inhibition, occasioned by the previous stimulus, delays the subsequent response (Lowe, 1979; Neill, 1977), the result being a measurable increase in response time relative to other nonnegatively primed trials. Further comparisons with larger groups of people with discrete anxiety disorders (Enright & Beech, 1993a, 1993b; Enright, Beech, & Claridge, 1995) failed to demonstrate negative priming effects among people with OCD, whereas individuals diagnosed with the other anxiety disorders (e.g., social phobia, generalized anxiety, panic disorder, posttraumatic stress disorder, and simple phobia) consistently demonstrated significant negative priming effects. By contrast, Fox (1994) produced data that she interpreted as suggesting that people with anxiety disorders, in general, may not demonstrate negative priming. The applicability of her results to the above studies involving people with clinical anxiety disorders is unclear, however, because her study involved high trait-anxious college students and so awaits replication with clinical populations.
Marcel (1983) suggested that although cognitive inhibition as assessed by negative priming was important to the operation of selective attention, it nevertheless operated at a nonconscious level. Enright et al. (1995) provided support for this selective attention deficit existing at a preattentive level. When compared with people with generalized anxiety disorder, simple phobia, and social phobia, people with OCD demonstrated significantly less negative priming at a stimulus presentation speed of 100 ms but not at slower presentation speeds of 250 ms and 350 ms. They concluded that this reduced negative priming effect in OCD is a result of preattentive cognitive processes because conscious strategies of selective attention operate at longer stimulus presentations. On the basis of the evidence of reduced negative priming associated with OCD, the symptoms of OCD may arise out of a deficit in cognitive inhibition, leaving the person with OCD unable to effectively suppress, at a nonconscious level, intrusive thoughts.
The evidence to date, therefore, suggests that a failure in cognitive inhibition may be responsible for the apparent deficits in selective attention observed in OCD. However, a specific and selective attention deficit in OCD has not yet been identified. The present study directly tested for a failure of selective attention specific to OCD by comparing the performance of people with OCD with those with another common anxiety disorder (panic disorder) and with control participants on a battery of psychometric attention tests known as the Test of Everyday Attention (TEA; Robertson, Ward, Ridgeway, & Nimmo-Smith, 1994). It contains tests sensitive to aspects of selective attention (i.e., the selection of relevant stimuli and processes with simultaneous inhibition of irrelevant ones), attentional switching, vigilance, and divided attention. It was predicted that people with OCD would demonstrate poorer performance relative to panic and control groups on subtests with high factor loadings on selective attention that may indicate a reduced ability to attend to relevant stimuli and inhibit irrelevant stimuli.
Method
Participants
Participants consisted of 17 people diagnosed by a clinical psychologist or psychiatrist as meeting the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; American Psychiatric Association, 1994) criteria for OCD, and 13 people who met the criteria for panic disorder with or without agoraphobia. They were recruited through community media or directly by a consulting clinician. The OCD group consisted of 9 women and 8 men, and the panic group had 10 women and 3 men. Sixteen OCD participants had been prescribed antidepressant medication (fluoxetine), and 3 had also been prescribed benzodiazepines. Five panic participants had been prescribed antidepressant medication, and 3 had been prescribed benzodiazapines.
The control group consisted of 14 participants recruited by word of mouth who, in a brief screening interview with the senior author, reported no significant psychiatric history and who matched the OCD group as closely as possible for age and intellectual level. This group consisted of 10 women and 4 men. Table 1 details age, IQ levels, and measures of anxiety, depression, and OCD symptomatology for each group.

________________________________________

[Help with image viewing]
[Email Jumpstart To Image]
Table 1 One-Way Analysis of Variance Results for Demographic and Psychometric Data From OCD, Panic, and Control Groups
________________________________________
Measures
The Beck Depression Inventory (Revised; BDI; Beck, Rush, Shaw, & Emery, 1979).
The BDI is designed to assess severity of depression in adolescents and adults, is usually self-administered, and consists of 21 groups of statements.
The State Form of the State–Trait Anxiety Inventory (Form Y-1).
The state component of the State–Trait Anxiety Inventory (STAI–S; Spielberger, 1983) consists of 20 statements that address present feelings of apprehension, tension, nervousness, and worry.
The Padua Inventory (Revised; PI-R; van Oppen, Hoekstra, & Emmelkamp, 1995).
The PI-R is a 41-item questionnaire covering a wide range of OCD behaviors and consisting of five subscales: Impulses, Contamination, Checking, Ruminations, and Precision. van Oppen et al. (1995) reported Cronbach internal reliability estimates for both clinical and nonclinical samples as ranging between .89 and .92, with coefficients for individual subscales between .65 and .93.
The National Adult Reading Test (NART; Nelson, 1982).
The NART is a word-recognition test that utilizes the high correlation between reading ability and intelligence in the general population (Crawford, Stewart, & Cochran, 1989). The test is composed of a list of 50 irregular words (i.e., pronunciation does not follow the normal phonetic rules) printed in order of increasing difficulty. A printed list of these words is read aloud, and, on the basis of the number of errors made in pronunciation, an estimation of the Wechsler Adult Intelligence Scale—Revised Verbal, Performance, and Full-Scale IQs can be calculated. Internal reliability estimates based on Cronbach's coefficient are reported as ranging from .90 to .93 (Crawford, Stewart, & Garthwaite, 1988; Nelson, 1982; Nelson & Willison, 1991); interrater reliability estimates range from .96 to .98, and test–retest reliability for a 2-week interval is reported as .98 (Crawford, Parker, Stewart, Besson, & De Lacey, 1989). There do not appear to be any age or gender effects (Nelson, 1982; Nelson & Willison, 1991). Evidence for the construct validity of the NART comes from factor analysis and correlation studies (Crawford, Parker, et al., 1989; Crawford, Stewart, & Cochran, 1989).
The TEA (Robertson et al., 1994).
The TEA provides norm-referenced scores on tests sensitive to selective attention, sustained attention, attentional switching, and divided attention. The battery consists of eight subtests and provides 10 scaled attention scores, adjusted for age.
The test comes in three parallel versions (A, B, & C), with each version standardized on a sample of 154 volunteers ranging from 18 to 80 years who were stratified into four age bands (18–34, 35–49, 50–64, and 65–80) and two estimated IQ score bands, either above or below 100 (on the basis of the NART). Correlations between TEA subtests and estimated verbal IQ are low and not significant. Test–retest reliability based on 1-week interval administrations of alternate versions of the TEA to a normative sample and a sample (N = 74) who had experienced cerebrovascular accidents (CVAs) ranged from .59 to .91 for individual subtests, with an average test–retest correlation of .75 and .78 for the two samples, respectively (Robertson et al., 1994).
The eight subtests in the TEA are composed of the Map Search 1A and 1B, Elevator Counting, Elevator Counting with Distraction, Visual Elevator Accuracy and Visual Elevator Speed, Elevator Counting With Reversal, Telephone Search, Telephone Search While Counting, and Lottery. All subtests except Elevator Counting and Lottery are performed under time pressure. A principal components analysis based on a normative sample yielded a four-factor solution, which explained 62.4% of the sample variance (Robertson et al., 1994). The four factors were Selective Attention, Attentional Switching, Sustained Attention, and Auditory–Verbal Working Memory.
The Selective Attention factor is described by both Map Search (1 and 2 min) and Telephone Search, which are sensitive to visual selective attention deficits reflecting a difficulty in ignoring irrelevant information and picking out targets in complex visual arrays. Elevator Counting With Reversal (Subtest 5), which is an auditory version of the Visual Elevator subtest, loads on both the Selective Attention factor and the Auditory–Verbal Working Memory factor. This subtest appears to be sensitive to the ability to manipulate information in auditory–verbal memory and also to auditory selective attention deficits (Robertson et al., 1994).
The Attentional Switching factor is described by Visual Elevator Accuracy and Speed (Subtest 4), which measures the ability to shift train of thought (shift attention). This test was designed as a measure of cognitive flexibility. Loading on the Sustained Attention factor is Elevator Counting (Subtest 2), which is the easiest of the TEA tests and is used as an entry-level test. All participants in this research obtained a perfect score of 7, and so their data are not included in the results. Lottery (Subtest 8) also loads on this factor. Both Elevator Counting and Lottery are sensitive to the ability to maintain attention on a relatively unchanging task. Initially designed as a test of divided attention, Telephone Search While Counting (Subtest 7) was subsequently found to load on this Sustained Attention factor.
The Auditory–Verbal Working Memory factor is described by Elevator Counting with Distraction (Subtest 3), which appears to detect the ability to manipulate information in auditory–verbal memory. Elevator Counting With Reversal (Subtest 5), as well as loading on Selective Attention, also loads on this Auditory–Verbal Working Memory factor.
Evidence of concurrent validity for the TEA comes from factor analytic studies that indicate that TEA subtests load on the same factors as more established tests of attention such as the d2 visual search, the Stroop test, Trails B, Wisconsin Card Sorting Test, Backward Digit Span, and the Paced Auditory Serial Addition Test (Robertson et al., 1994). In addition, CVA (stroke) patients assessed on the TEA have produced subtest scores that correlated strongly with other commonly used measures of functional status, and TEA subtest scores have been shown to discriminate patients with moderate to severe closed head injury from control participants matched for age and IQ even at 14 months post-injury (Robertson et al., 1994).
Procedure
The senior author assessed participants in their own homes or in a place of their own choosing. Participants completed a demographic questionnaire, the PI-R, the BDI, and the STAI–S scale (Form Y-1) and were then assessed on the NART and the TEA.
Results
Demographics and Psychometrics
Group demographic and psychometric data were analyzed by one-way analyses of variance (ANOVAs) and are presented in Table 1. Post hoc comparisons from Scheffé's test with a 5% significance level showed that both the OCD and panic groups scored significantly higher on measures of depression and anxiety than the control group and that the OCD group scored significantly higher on the PI-R than both other groups (as expected). Therefore, the OCD and panic groups both reported higher levels of anxiety and depression than the control group, and the OCD group also reported higher levels of obsessive–compulsive symptomatology than either. The average age of the panic group was older than that of the control group and the OCD group, but this factor was not of major concern in relation to performance on the TEA because its scores are adjusted for age.
Group Comparisons on the Subtests of the TEA
A series of one-way ANOVAs tested for differences in performance between the OCD, panic, and control groups on the subtests of the TEA. It was predicted that subtests loading on the Selective Attention factor (i.e., Map Search 1 and 2, Telephone Search, and Elevator Counting With Reversal) would discriminate the OCD group from the panic and control groups. We used planned comparisons to test the hypotheses that the pooled mean for the panic and control groups exceeded the mean of the OCD group and that the mean for the OCD group would be less than that of the panic group.
We made post hoc comparisons using the Scheffé's test with a 5% significance level for the remaining subtests (i.e., Visual Elevator Accuracy, Visual Elevator Speed, Elevator Counting With Distraction, Telephone Search While Counting, and Lottery). The results are reported in Table 2.

________________________________________

[Help with image viewing]
[Email Jumpstart To Image]
Table 2 One-Way Analysis of Variance Results Comparing Performance on Subtests of the TEA for OCD, Panic, and Control Groups
________________________________________
There were statistically significant differences between the groups on all subtests except Elevator Counting With Distraction and Lottery. The planned comparisons for the subtests loading on the Selective Attention factor indicated that the mean for the OCD group score was lower on all four subtests relative to the pooled mean of the panic and control groups (Contrast 1) but was lower than the panic group only on Map Search 2, Elevator Counting With Reversal, and Telephone Search (Contrast 2). Post hoc comparisons from Scheffé's test with a 5% significance level for the remaining subtests indicated that the mean for the OCD group was significantly lower only on Visual Elevator Accuracy relative to both the panic and control groups. The performance of the panic group did not differ significantly from that of the control group on this subtest. The post hoc comparisons also revealed that the OCD group performed more poorly than the panic group on Visual Elevator Speed and Telephone Search While Counting but did not differ from the control group on these subtests.
In summary, the results confirmed the hypothesis that the OCD group would exhibit deficits in selective attention as evidenced by lower scores than the panic and control groups on subtests loading on this factor. There was also evidence of deficits in attention switching by the OCD participants, as they scored lower than the panic and control groups on one subtest of the Attention Switching factor (Visual Elevator Accuracy) and lower than the panic group on another loading on this factor (Visual Elevator Speed). Moreover, the OCD participants performed more poorly than the panic group on one subtest of the Sustained Attention factor (Telephone Search While Counting).
Discussion
Evidence was found for a specific deficit in selective attention associated with OCD and the absence of a similar deficit associated with a different anxiety disorder, here panic disorder. That is, people with OCD in this study consistently performed less well compared with panic and control groups on three of the four subtests of the TEA, which loaded highly on selective attention. Moreover, people with OCD may have more general attentional deficits related to capacities to switch attention rapidly as well as to sustain attention. These results suggest that people with OCD have a reduced ability to selectively ignore unimportant external (sensory) and internal (cognitive) stimuli. If obsessional thinking has its origins in normal intrusive cognitions (Rachman & de Silva, 1978), then individuals with a reduced ability to automatically and unconsciously disattend selectively to their own intrusive thoughts may therefore be vulnerable to developing OCD.
There are other possible interpretations of these results, however. The poorer attentional capacities found for the OCD participants may have been due to greater psychopathology, particularly as they seemed to be taking more psychotrophic medication. However, there were no differences between the two clinical groups on the measures of anxiety and depression, which argues against this interpretation. Alternatively, because differences were found only on the timed, but not untimed, tasks, these effects may have been due to excessive caution or slow responding (perhaps associated with medication) on the part of the OCD participants. Future research should address these alternative possibilities as well as utilize a wider range of measures of attentional inhibition.
Although the sample sizes used in this study were small, predicted power values for the analyses were generally large, reflecting large effect sizes. Nevertheless, small sample size represents perhaps the primary methodological vulnerability of this study so that further research should ideally involve larger samples.
References
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders. (4th ed.). Washington, DC: Author. [Context Link]
Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression. New York: Guilford Press. [Context Link]
Comalli, P. E., Wapner, S., & Werner, H. (1962). Interference effects of Stroop Color–Word Test in childhood, adulthood, and aging. Journal of Genetic Psychology, 100, 47–53. [Context Link]
Crawford, J. R., Parker, D. M., Stewart, L. E., Besson, J. A. O., & De Lacey, G. (1989). Prediction of WAIS-IQ with the National Adult Reading Test: Cross-validation and extension. British Journal of Clinical Psychology, 28, 267–273. Bibliographic Links [Context Link]
Crawford, J. R., Stewart, L. E., & Cochran, R. H. B. (1989). Construct validity of the National Adult Reading Test: A factor analytic study. Personality and Individual Differences, 10, 585–587. [Context Link]
Crawford, J. R., Stewart, L. E., & Garthwaite, P. H. (1988). The relationship between demographic variables and NART performance in normal subjects. British Journal of Clinical Psychology, 27, 181–182. Bibliographic Links [Context Link]
Enright, S. J., & Beech, A. R. (1990). Obsessional states: Anxiety disorders or schizotypes? An information processing and personality assessment. Psychological Medicine, 20, 621–627. [Context Link]
Enright, S. J., & Beech, A. R. (1993a). Further evidence of reduced inhibition in obsessive compulsive disorder. Personality and Individual Differences, 14(3), 387–395. [Context Link]
Enright, S. J., & Beech, A. R. (1993b). Reduced cognitive inhibition in obsessive compulsive disorder. British Journal of Clinical Psychology, 32, 67–74. Bibliographic Links [Context Link]
Enright, S. J., Beech, A. R., & Claridge, G. S. (1995). A further investigation of cognitive inhibition in obsessive-compulsive disorder and other anxiety disorders. Personality and Individual Differences, 16(4), 535–542. [Context Link]
Fox, E. (1994). Attentional bias in anxiety: A defective inhibition hypothesis. Cognition and Emotion, 8(2), 165–195. [Context Link]
Lowe, D. G. (1979). Strategies, context and mechanisms of response inhibition. Memory & Cognition, 7, 382–389. [Context Link]
Marcel, A. J. (1983). Conscious and unconscious perception: Experiments on visual masking and word recognition. Cognitive Psychology, 15, 197–237. [Context Link]
Martinot, J. L., Allilaire, J. F., Mazoyer, B. M., & Hantouche, E. (1990). Obsessive–compulsive disorder: A clinical, neuropsychological and positron emission tomography study. Acta-Psychiatrica-Scandinavica, 82(3), 233–242. [Context Link]
Neill, W. T. (1977). Inhibitory and facilitatory processes in selective attention. Journal of Experimental Psychology: Human Perception and Performance, 3, 444–450. [Context Link]
Nelson, H. (1982). The National Adult Reading Test (NART): Test manual. Windsor, England: NFER-Nelson. [Context Link]
Nelson, H. E., & Willison, J. R. (1991). The revised National Adult Reading Test (NART): Test manual. Windsor, England: NFER-Nelson. [Context Link]
Rachman, S., & de Silva, P. (1978). Abnormal and normal obsessions. Behaviour Research and Therapy, 16, 233–248. [Context Link]
Reitan, R. M. (1958). The validity of the Trail-Making Test as an indicator of organic brain damage. Perceptual Motor Skills, 9, 127–130. [Context Link]
Robertson, I. H., Ward, T., Ridgeway, V., & Nimmo-Smith, I. (1994). The Test of Everyday Attention. Bury, St. Edmund, England: Thames Valley Test Company. [Context Link]
Spielberger, C. D. (1983). Manual for the State–Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press. [Context Link]
Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643–662. [Context Link]
Tipper, S. P. (1985). The negative priming effect: Inhibitory priming by ignored objects. Quarterly Journal of Experimental Psychology: Human Experimental Psychology, 37(A), 571–590. Bibliographic Links [Context Link]
van Oppen, P., Hoekstra, R., & Emmelkamp, P. M. G. (1995). The structure of obsessive–compulsive symptoms. Behaviour Research and Therapy, 33, 15–23. [Context Link]
________________________________________

Accession Number: 00004468-199902000-00016
Copyright (c) 2000-2007 Ovid Technologies, Inc.
Version: rel10.5.8, SourceID 1.13281.2.32.1.0.2.197.1.4.1.5