PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by:On: 19 November 2010Access details: Access Details: Free AccessPublisher Psychology PressInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Clinical and Experimental NeuropsychologyPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713657736

Order Effects within the Trail Making and Stroop Tests in Patients withNeurologic DisordersRobert Taylor

Online publication date: 09 August 2010

To cite this Article Taylor, Robert(1998) 'Order Effects within the Trail Making and Stroop Tests in Patients withNeurologic Disorders', Journal of Clinical and Experimental Neuropsychology, 20: 5, 750 754To link to this Article: DOI: 10.1076/jcen.20.5.750.1120URL: http://dx.doi.org/10.1076/jcen.20.5.750.1120

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

* Address correspondence to: R. Taylor, Department of Clinical Neurosciences, Western General Hospital,Edinburgh EH4 2XU, UK.Accepted for publication: July 24, 1998.

Journal of Clinical and Experimental Neuropsychology 1380-3395/98/2005-750$12.001998, Vol. 20, No. 5, pp. 750-754 Swets & Zeitlinger

Order Effects within the Trail Making and Stroop Tests inPatients with Neurologic Disorders*

Robert TaylorDepartment of Clinical Neurosciences, Western General Hospital, Edinburgh, UK

ABSTRACT

Effects of order of presentation within the Trail Making and Stroop tests were investigated by comparing50 consecutive patients receiving detailed neuropsychological assessment given the components of eachtest in standard order with 50 given them in reverse. Groups were well matched on a wide range of demo-graphic and neuropsychological variables. The only significant group difference in mean scores for the testcomponents and derived ratios and differences, a higher Trail Making Test Part B/Part A ratio in the re-verse group than in the standard group, was most easily explained in terms of a slight effect of practicein visual scanning. Greater variance in the Stroop Test Part C score and its derivatives in the standard groupmay have been attributable to factors other than order of presentation. Performance of these tests in patientswith neurological disorders is not crucially dependent upon accumulation of proactive interference fromsimpler to more complex parts; the latter can reasonably be used in isolation.

The Trail Making and Stroop tests are de-servedly popular for clinical and researchpuposes in neuropsychological practice (Lezak,1995). It is not known to what extent perfor-mance on the complex component of each ofthese tests is influenced by or dependent uponthe establishment of a mental set or an accu-mulation of proactive interference from preced-ing, more straightforward, test components and,hence, to what extent it is valid to use the com-plex parts in isolation from their partner testcomponents (as is not uncommon in practice).

The several versions (Lezak, 1995; Spreen &Strauss, 1991) and methodological variants(MacLeod, 1991) of the Stroop test all involvepresentation of the most complex part last.Trenerry, Crosson, DeBoe, and Leber (1989)found the score on the most complex part of theStroop to be more informative than the differ-ence between that and the score on a precedingpart, but indicated that the preceding part shouldalways be given in case of any priming effect

(and because their normative data had been col-lected using such a procedure). Presentation ofthe more complex part (Part B) of the TrailMaking test after the simpler part (Part A) is thestandard procedure, but some clinicians maychoose to omit Part A if this appears redundantin view of other test performances. Conse-quently, order effects have clinical implications,and are of some theoretical interest as regardsunderstanding the sensitivity of these tests toaspects of brain dysfunction.

Effects of order of presentation within eachtest were investigated by comparing the perfor-mances of a group of patients given the testcomponents in standard order with those of agroup given them in reverse. The option of giv-ing the test components in standard and reverseorder to a single group on different occasionswas rejected because of the possibility of prac-tice effects and related complex interactionseven with appropriate counterbalancing of orderof conditions.

Downlo

aded

At:

15:

00 1

9 No

vemb

er 2

010

ORDER EFFECTS IN TRAILS AND STROOP TESTS 751

METHOD

ParticipantsParticipants were 100 consecutive patients referredto clinical neuropsychology within a large regionalclinical neurosciences department who receiveddetailed neuropsychological assessment and hadno gross motor dysfunction in the preferred hand,and no speech disturbance or uncorrected impair-ment of visual acuity.

ProcedureAll participants received the Spreen and Strauss(1991) administration of the Trail Making Test(TMT) in which Part A (TMT-A) requires joiningnumbered circles and Part B (TMT-B) requiresjoining numbered and lettered circles in alternatingsequence. The Comalli, Wapner, and Werner(1962) version of the Stroop Test was adminis-tered, in which Part A requires reading aloud thenames of colours printed in black ink, Part B re-quires naming the colours of small coloured rect-angles, and Part C requires naming the colour ofthe ink in which discordant colour names areprinted. The version used comprises 100 items andthree colour names or colours in each section. Atime penalty of 3 s was added for each uncorrectederror (although such errors were infrequent).

The first 50 participants (32 men, 18 women;mean age 36.3 years, range 1765) completed thetests in the following order: TMT-A, TMT-B,Stroop A, Stroop B, Stroop C; and the remaining50 (28 men, 22 women; mean age 36.8 years,range 1758) in the order: TMT-B, TMT-A,Stroop C, Stroop B, Stroop A. Henceforth, thesegroups will be referred to as the standard and re-verse groups, respectively. Patterns and sources ofreferrals of patients did not change over the courseof the study, and analyses of a wide range of de-mographic and neuropsychological variables de-scribed below suggested that allocating patients togroups consecutively (first 50 and next 50) as op-posed to randomly throughout the study did notintroduce systematic bias.

Other neuropsychological measures used (someof which are unpublished tests in local use) in-cluded the National Adult Reading Test (NART;Nelson, 1991); a selected-item form of the WAIS-R (after Satz & Mogel, 1962; Wechsler, 1981; con-stituent subtests including complete Digit Spanand Digit Symbol, but excluding Comprehension);Ravens (1965) Advanced Progressive MatricesSet I; Mini-Mental State Examination (Folstein,Folstein, & McHugh, 1975); immediate and de-layed recall of a prose paragraph (comparable butnot identical to those of Wechsler, 1987); immedi-

ate and delayed yes-no picture recognition mem-ory; verbal paired-associate learning (Inglis,1959); the Warrington (1984) recognition memorytests for words and faces; copy and immediate re-call of the Rey complex figure (Lezak, 1995);Hooper (1958) Visual Organisation Test items;recognising objects photographed from unusualangles; a collection of very brief visuospatial taskssimilar to those described in Taylor, Gilleard, andMcGuire (1996); Corsi block span (Lezak, 1995);a short version of the Token test (after Spellacy &Spreen, 1969; Coughlan & Warrington, 1978); rep-etition of sentences; naming of pictured objectsand other material; brief simple tests of readingcomprehension, reading aloud, and writing; motorpraxis items (Kertesz & Hooper, 1982); the Weigl(1941) sorting test; the Wisconsin Card SortingTest (Nelson, 1976); paper-and-pencil mazes (fromWechsler, 1974); the Cognitive Estimation Test(Shallice & Evans, 1978; Shoqeirat, Mayes,MacDonald, Meudell, & Pickering, 1990); produc-tion of words in given categories (animal namesand words beginning with the letter F for 1 mineach; Spreen & Strauss, 1991); the Hospital Anxi-ety and Depression scale (HAD; Zigmond &Snaith, 1983); the Wimbledon self-report moodscale (Coughlan & Storey, 1988); a 36-item self-report questionnaire concerning difficulties witheveryday mental functioning (adapted from Taylor,1990); and a parallel rating scale completed by arelative or close acquaintance. The Trail Makingand Stroop tests were given about half waythrough the second of two assessment sessionswith the author.

Final diagnoses (confirmed at least a year afterassessment where necessary) were: head injury (20in the standard group, 19 in the reverse group),intracranial tumour (8, 6), subarachnoid hemor-rhage from aneurysm or arteriovenous malforma-tion (7, 5), cerebral infarction (4, 1), anoxia (2, 0),meningitis (0, 2), encephalitis (0, 1), hydrocepha-lus (1, 0), epilepsy (2, 4), depression or anxiety (4,6), and multiple diagnoses or no definite diagnosis(2, 6). Time since insult or onset of illness variedconsiderably within and between diagnosticgroups.

Statistical analyses were conducted usingStatistica (Statsoft, 1997).

RESULTS AND DISCUSSION

The two groups did not differ significantly bychi-square on gender, consumption of psychoac-tive medication, presence of neuroradiological

Downlo

aded

At:

15:

00 1

9 No

vemb

er 2

010

752 ROBERT TAYLOR

Table 1. Group Differences for Test Component Scores and Relevant Computed Ratios.

Groups

Standard Reverse

M (SD) M (SD) pa

Trail Making Test Part BPart A

B/A

Stroop Test Part CPart BPart A

C/BC/A

2C/(B+A)B/A

75.036.0

2.10

131.571.048.4

1.842.762.201.49

(35.3)(11.2)

(0.70)

(50.4)(19.1)(14.1)

(0.42)(0.85)(0.55)(0.25)

82.634.0

2.58

126.571.749.9

1.772.562.081.45

(28.6)(12.3)

(0.93)

(34.9)(16.6)(11.1)

(0.30)(0.53)(0.36)(0.23)

.239

.388

.005

.562

.841

.545

.317

.162

.216

.375

a Unpaired t test

evidence of brain abnormality, or social class;nor by t test on mean age, years of education,estimated premorbid IQ on the NART (104.3,SD = 14.6 in the standard group; 105.1, SD =13.4 in the reverse group), current Verbal IQ(102.9, SD = 14.4; 99.5, SD = 13.1, respec-tively), Performance IQ (96.6, SD = 13.6; 95.8,SD = 11.0), and Full Scale IQ (100.2, SD = 13.9;97.8, SD = 12.2); or raw scores on any of theother neuropsychological measures listed aboveexcept that the reverse group reported signifi-cantly more anxiety on the HAD anxietysubscale and more difficulties with everydaymental functioning on the 36-item self-reportquestionnaire. These two significant differencesmay be attributable to chance in view of thenumber of comparisons made, and become non-significant if Bonferonni adjustment is applied.

Table 1 shows means, standard deviations,and significance levels (by unpaired t test) of thedifference between groups for test componentscores and relevant computed ratios. The distri-butions of scores were mildly skewed in thesame direction in both groups. The t test isknown to be robust under these circumstancesand was preferred to nonparametric alternativesfor its generally greater statistical power. Repe-tition of analyses using nonparametric (Mann-Whitney U) tests produced the same pattern ofstatistical significance.

The only significant difference was a higherTMT-B/TMT-A ratio in the reverse group, re-flecting slightly slower performance on TMT-Band slightly quicker performance on TMT-A inthat group as compared with the standard group.This difference is opposite to that which wouldbe expected if TMT-B was influenced by proac-tive interference from TMT-A; it is probablymost easily explained by a slight effect of prac-tice in effective visual scanning of the pagegained in the first part given (in either or bothgroups) leading to improved performance on thesecond. The difference between groups is smalland fairly insignificant in clinical terms. Thegroups did not differ significantly on the differ-ence between scores on the two parts (B-A) asopposed to their ratio (B/A). That ratio of scorescorrelated (Pearson r) .59 with Part B itself inall 100 patients, whereas the difference scorecorrelated .93, indicating that the latter is largelya reflection of performance on Part B.

There were no significant differences on themean Stroop measures and derivatives. Thestandard deviations of Part C and its derivativeswere significantly larger in the standard than inthe reverse group, but this was also the case forsome of the other tests listed above and may bea chance difference between groups. The possi-bility that proactive interference from earlierparts of the test affected some (vulnerable) pa-Downlo

aded

At:

15:

00 1

9 No

vemb

er 2

010

ORDER EFFECTS IN TRAILS AND STROOP TESTS 753

tients but not others, thereby increasing thegroup variance, is not particularly plausiblegiven the small difference in group means.There were also no significant differences be-tween group means when difference scores(Stroop C-B, C-A, C-((A+B)/2), B-A) ratherthan ratios were considered. The most crucialrelationship in the Stroop test is considered thatbetween Part C and Part B. The ratio of these(C/B) correlated (Pearson r) .63 with Part C it-self in all 100 patients whereas the differencescore (C-B) correlated .93, indicating that thelatter is largely a reflection of performance onPart C.

Performances in both groups ranged fromnormal to grossly impaired. Women tended toperform various test components more quicklythan did men, but gender was confounded withdiagnosis and, in any case, this tendency doesnot affect the main conclusions that can bedrawn. Correlations with age were negligible,and there were low correlations in expected di-rections with social class, educational level, andestimated premorbid IQ; but again these factorsare not independent of others such as diagnosis.There were no significant differences betweenpatients with head injury and all other patientscombined; diagnostic subgroups were otherwisetoo small to allow worthwhile comparison witheach other.

Not surprisingly, the component parts of theTrail Making and Stroop tests tended to inter-correlate fairly highly; principal componentsanalysis in each group and in both combinedindicated that scores on all five test parts werereducible to only one component with eigen-value greater than 1. Patterns of correlation withother neuropsychological tests appeared broadlysimilar in the two groups.

The results indicate that performance on thecomplex components of the Trail Making andStroop tests in patients with neurological disor-ders is not crucially dependent upon the estab-lishment of a mental set or an accumulation ofproactive interference from preceding simplercomponents, and that these complex test compo-nents can reasonably be used in isolation. This isnot to suggest that the simpler components areof no value, but simply that their omission will

not lead to serious distortion of performance onthe complex components alone.

REFERENCES

Comalli, P.E., Wapner, S., & Werner, H. (1962). In-terference effects of Stroop Color-Word Test inchildhood, adulthood and aging. Journal of Ge-netic Psychology, 100, 47-53.

Coughlan, A.K., & Storey, P. (1988). The Wimbledonself-report scale: Emotional and mood appraisal.Clinical Rehabilitation, 2, 207-213.

Coughlan, A.K., & Warrington, E.K. (1978). Word-comprehension and word-retrieval in patients withlocalized cerebral lesions. Brain, 101, 163-185.

Folstein, M.F., Folstein, S.E., & McHugh, P.R.(1975). Mini-mental state. Journal of PsychiatricResearch, 12, 189-198.

Hooper, H.E. (1958). The Hooper Visual Organi-sation Test. Los Angeles, CA: Western Psycholog-ical Services.

Inglis, J. (1959). A paired-associate learning test foruse with elderly psychiatric patients. Journal ofMental Science, 105, 440-443.

Kertesz, A., & Hooper, P. (1982). Praxis and lan-guage: the extent and variety of apraxia in aphasia.Neuropsychologia, 20, 275-286.

Lezak, M.D. (1995). Neuropsychological assessment(3rd ed.). New York: Oxford University Press.

MacLeod, C.M. (1991). Half a century of research onthe Stroop effect: An integrative review. Psycho-logical Bulletin, 109, 163-203.

Nelson, H.E. (1976). A modified card sorting test sen-sitive to frontal lobe defects. Cortex, 12, 313-324.

Nelson, H.E. (1991). National Adult Reading Test(2nd ed.). Windsor, UK: NFER-Nelson.

Raven, J.C. (1965). Advanced progressive matricessets I and II. London: H.K. Lewis.

Satz, P., & Mogel, S. (1962). An abbreviation of theWAIS for clinical use. Journal of Clinical Psychol-ogy, 18, 77-79.

Shallice, T., & Evans, M.E. (1978). The involvementof the frontal lobes in cognitive estimation. Cortex,14, 294-303.

Shoqeirat, M.A., Mayes, A., MacDonald, C., Meudell,P., & Pickering, A. (1990). Performance on testssensitive to frontal lobe lesions by patients withorganic amnesia: Leng & Parkin revisited. BritishJournal of Clinical Psychology, 29, 401-408.

Spellacy, F.J., & Spreen, 0. (1969). A short form ofthe Token Test. Cortex, 5, 390-397.

Spreen, O., & Strauss, E. (1991). A compendium ofneuropsychological tests. New York: Oxford Uni-versity Press.

Statsoft (1997). Statistica. Tulsa, OK: Statsoft.

Downlo

aded

At:

15:

00 1

9 No

vemb

er 2

010

754 ROBERT TAYLOR

Taylor, R. (1990). Relationships between cognitivetest performance and everyday cognitive difficul-ties in multiple sclerosis. British Journal of Clini-cal Psychology, 29, 251-253.

Taylor, R., Gilleard C.J., & McGuire, R.J. (1996).Patterns of neuropsychological impairment in de-mentia of the Alzheimer type and multi-infarct de-mentia. Archives of Gerontology and Geriatrics,23, 13-26.

Trenerry, M.R., Crosson, B., DeBoe, J., & Leber,W.R. (1989). Stroop Neuropsychological Screen-ing Test Manual. Odessa, FL: Psychological As-sessment Resources.

Warrington, E.K. (1984). Recognition Memory Test.Windsor, UK: NFER-Nelson.

Wechsler, D. (1974). Wechsler Intelligence Scale forChildren-Revised. New York: The PsychologicalCorporation.

Wechsler, D. (1981). Wechsler Adult IntelligenceScale-Revised. New York: The Psychological Cor-poration.

Wechsler, D. (1987). Wechsler Memory Scale-Re-vised. New York: The Psychological Corporation.

Weigl, E. (1941). On the psychology of the so-calledprocess of abstraction. Journal of Abnormal andSocial Psychology, 36, 3-33.

Zigmond, A.S., & Snaith, R.P. (1983). The HospitalAnxiety and Depression scale. Acta PsychiatricaScandanavica, 67, 361-370.

Downlo

aded

At:

15:

00 1

9 No

vemb

er 2

010