This article was downloaded by: [Monash University Library]On: 05 September 2013, At: 06:48Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Clinical NeuropsychologistPublication details, including instructions for authorsand subscription information:http://www.tandfonline.com/loi/ntcn19

Extended norms for the pacedauditory serial addition taskDeborah D. Roman a , Glenace E. Edwall a , Rebecca J.Buchanan a & Jim H. Patton aa Baylor UniversityPublished online: 08 Nov 2007.

To cite this article: Deborah D. Roman , Glenace E. Edwall , Rebecca J. Buchanan & JimH. Patton (1991) Extended norms for the paced auditory serial addition task, ClinicalNeuropsychologist, 5:1, 33-40, DOI: 10.1080/13854049108401840

To link to this article: http://dx.doi.org/10.1080/13854049108401840

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information(the “Content”) contained in the publications on our platform. However, Taylor& Francis, our agents, and our licensors make no representations or warrantieswhatsoever as to the accuracy, completeness, or suitability for any purposeof the Content. Any opinions and views expressed in this publication are theopinions and views of the authors, and are not the views of or endorsed byTaylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor andFrancis shall not be liable for any losses, actions, claims, proceedings, demands,costs, expenses, damages, and other liabilities whatsoever or howsoever causedarising directly or indirectly in connection with, in relation to or arising out ofthe use of the Content.

This article may be used for research, teaching, and private study purposes.Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expresslyforbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

The Clinical Neuropsychologist 1991, Vol. 5, No. 1, pp. 33-40

0920- 163719 1/050 1 -0033$3.00 Q Swets & Zeitlinger

CLINICAL, ISSUES

Extended Norms for the Paced Auditory Serial Addition Task

Deborah D. Roman, Glenace E. Edwall, Rebecca J. Buchanan, and Jim H. Patton Baylor University

ABSTRACT

The Paced Auditory Serial Addition Task (PASAT) and an abbreviated version of the WAIS-R were administered to 143 normal individuals, including 62 persons aged 18 to 27 (58% female), 40 aged 33 to 50 (50% female), and 41 aged 60 to 75 (51% female). PASAT means and standard deviations are presented. Analysis of variance results revealed significant age effects on PASAT performance, with the 60-to 75-year age group scoring significantly lower than the two younger age groups on all four PASAT series. No significant gender effect was found. Small but statistically signifi- cant correlations were found between PASAT performance and estimated IQ.

The PASAT (Gronwall & Sampson, 1974; Gronwall & Wrightson, 1974), con- sidered to be a sensitive measure of information-processing speed and efficiency, concentration skills, and immediate memory, has been used extensively to evalu- ate the effects of head injury. A number of studies have reported defective PASAT performance by subjects who had experienced a concussion or more severe head trauma (Gronwall, 1977; Gronwall & Sampson, 1974; Gronwall & Wrightson, 1974; Levin et al., 1987; O’Shaughnessy, Fowler, &Reid, 1984). Further, PASAT performance has been shown to correlate with subjective complaints and job performance difficulties following head injury (Gronwall & Wrightson, 1974).

Given these results, several investigators (Gronwall, 1977; Levin, Benton, & Grossman, 1982) have recommended the use of the PASAT in clinical practice as a means of measuring postinjury cognitive deficits. Serial testing has been

Communications should be directed to Deborah Roman, now at the University of Minne- sota Hospital and Clinic, Box 390 Mayo Memorial Building, Harvard Street at East River Road, Minneapolis, MN 55455, USA. The authors wish to thank Dr. John Cornell for his assistance with the statistical analysis and Ms. Cindy Reinbold and Ms. Elizabeth Gutz for their help with the data collection. Accepted for publication: March 9, 1990.

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 0

6:48

05

Sept

embe

r 20

13

34 DEBORAH D. ROMAN ET AL.

suggested as a way of determining when head injury victims can be expected to return to their preinjury levels of functioning.

Despite these promising research findings, the PASAT’s utility as a clinical instrument has been limited for several reasons. Adequate age norms are not available for either Gronwall’s (1977) version of the PASAT or for a shorter version of the test, de’veloped at the University of Texas Medical Branch at Galveston, that is in common use in the United States (H. S. Levin, personal communication, September 12, 1988).

Norms for Gronwall’s version are based on the performance of 80 predomi- nantly male, youthful New Zealanders (10 subjects were aged 41 to 55, the rest were younger). Gronwall has cautioned that these norms may not be appropriate for other populations (Gronwall, 1987). Unpublished norms for the Galveston PASAT are based on 30 subjects, aged 21 to 56 (M = 28.6).

Additionally, information regarding the relationship between PASAT per- formance and other cognitive abilities is sketchy. Sampson’s unpublished studies (Gronwall & Sampson, 1974) suggest that the correlation between paced performance and general intelligence is low: .28 using military personnel as subjects. It is not known whether these findings are based on current editions of the PASAT or on earlier paced performance tasks using visually presented stim- uli (Sampson, 1956).

Researchers have endeavored to overcome these obstacles by using paired control groups; however this strategy is not generally feasible in clinical prac- tice. The current study was designed to develop norms for the Galveston version of the PASAT, utilizing approximately equal numbers of male and female sub- jects, aged 18 to 75. The study was further designed to investigate the relation- ship between gender, age, and IQ and PASAT performance.

METHOD

Subjects The subjects were 143 white adults. Based on self-report, no subject had a history of head injury with loss of consciousness, other neurological disorders (such as meningitis, sei- zures, or stroke), substance abuse, or psychiatric disorders. Additionally, no subject was taking psychoactive medication. There were 62 subjects in the 18- to 27-year-old range (58% female), 40 in the 33- to 50-year range (50% female), and 41 in the 60- to 75- year range (51% female).

Subjects were recruited from a variety of sources, including Baylor University under- graduates and employees, students from a local business college, members of service clubs and retired professional groups, employees of several local businesses, attendants at senior citizen centers, and residents from a retirement community. Approximately one fourth of the subjects were paid $5 for their participation; the remainder were not paid.

Apparatus A taped version of the PASAT, provided by Harvey Levin, was used. It consisted of four series, each with 49 pairs of randomly presented single-digit numbers. The length of time

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 0

6:48

05

Sept

embe

r 20

13

PASAT NORMS 35

elapsed between numbers was shorter with each series: The first series was presented 2.4 s apart; the second, 2.0 s apart; the third, 1.6 s apart; and the final series, 1.2 s apart. There was a 30-s pause between each series.

Procedure Subjects provided information regarding their medical/psychiatric history, current health status, and medication regimen. The Vocabulary and Block Design subtests of the WAIS- R were administered in accordance with standardized instructions (Wechsler, 1981). Conversion tables (Silverstein, 1982) were used to convert scaled score totals into esti- mated IQ scores.

Prior to taking the PASAT, subjects were told that they would hear a series of single- digit numbers presented at a fixed speed and that their task would be to add the last digit presented to the digit immediately preceding it. This was demonstrated visually and the first four to five pairs of numbers from the first series (2.4 s speed) were presented for practice. Instructions were repeated as needed to insure comprehension. Subjects were advised to resume their calculations as quickly as possible in the event that they missed a sum or forgot a number. They were told that four separate series of numbers, each sepa- rated by a brief pause, would be presented and that each successive series would be presented at a faster pace. Encouragement was provided during the pauses between each series.

Ten PASAT scores were calculated, including raw and percentage scores (number correct and percent correct, respectively) for each of the four series and for the total of all four series. Means and standard deviations for the PASAT and IQ estimate scores were calculated for the three age groups. Pearson r correlations were computed to estimate the relationship between PASAT performance and IQ. An analysis of variance was used to evaluate the effects of age and gender on PASAT Performance. A profile analysis was conducted to determine if there was an interaction between the speed demands of the task and the age of subjects, and an a priori contrast and trend analysis were performed in order to confirm the findings of the profile analysis (Tabachnick & Fidell, 1989).

RESULTS

Table 1 summarizes the demographic characteristics of the sample. Table 2 provides PASAT means and standard deviations for the three age groups; Table 3 provides frequency distributions of raw scores for the three age groups.

A two-way analysis of variance using the four PASAT series raw scores as a within-group repeated measure and age group and gender as between-group variables revealed a significant age effect on PASAT performance, F (2, 139) = 20.00, p < ,001, but no significant gender effect. Results of the profile analysis indicated that the three age groups’ performances on the PASAT were parallel, F (6,417) = 0.57, n.s., and that the young and middle aged groups differed from the older age group in elevation, F (2, 139) = 20.00, p < .001. Subsequent to the overall analysis, a contrast of PASAT performance for young and middle aged subjects vs. older subjects was tested and found to be significant, F (1, 139) = 37.81, p < .001, with older subjects scoring lower. This result elucidates the age effect found on the profile analysis. Trend analysis indicated a significant in- verse linear trend for PASAT performance as speed of presentation increased, F

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 0

6:48

05

Sept

embe

r 20

13

36 DEBORAH D. ROMAN ET AL.

Table 1: Demographic Data for PASAT Normative Sample

Age Group

Young Adult Middle Aged Older Adult

(n = 62) (n = 40) ( n = 41) Adult

Age Range 18 - 27 33 - 50 60 - 75 (years)

(years)

FemaleMale

M (SD) (years)

M (SD)

A g e M (SO) 19 (1.6) 40 (5.1) 69 (4.1)

Percent 58/42 50150 51/49

Education 12 (77) 15 (2.6) 15 (3.2)

IQ estimate 110 (12.3) 110 (12.3) 107 (11.0)

Table 2: PASAT Norms for Young, Middle Aged, and Older Age Groups

PASAT Series

Presentation 2.4 s 2.0 s 1.6 s 1.2 s Total Speed

Young Adult ( n = 62)

Raw score 4514.3 3917.8 3617.7 2816.7 1 48123.5 MISD

Percent 91.681 79.531 72.791 57.921 7511 2.0 correct MISD 8.9 16.2 15.9 13.8

Middle Aged Adult ( n = 40)

Raw score 4414.9 3817.6 3318.9 2819.3 144127.0 MISO

Percent 90.701 77.501 68.221 57.221 73114.0 correct MISD 10.3 15.3 18.2 18.9

Older Adult ( n = 41)

Raw score 3719.1 3119.2 2718.5 2016.1 11 5129.9 MISD

Percent 75.951 63.041 54.951 41.631 59115.3 correct MISD 18.6 18.9 17.4 12.2

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 0

6:48

05

Sept

embe

r 20

13

PASAT NORMS 37

Table 3: Frequency Distributions of Raw Scores for the Four PASAT Speeds for Three Age Groups.

Raw Score Ranges PASAT

(seconds) 45-49 40-44 35-39 30-34 25-29 224 Age Speed

Young Adult ( n = 62)

Middle Aged Adult ( n = 40)

Older Adult ( n = 41)

2.4 2.0 1.6 1.2

2.4 2.0 1.6 1.2

2.4 2.0 1.6 1.2

39 19 7 1

26 7 2 0

17 4 14 7 15 11 2 12

7 3 14 5 12 6 6 4

13 5 3 6 3 6 0 0

2 13 16 9

0 0 5 4 8 5

30 8

0 0 4 2 5 6 3 18

4 5 11 11 7 18 5 33

(3, 417) = 384.48, p c .001, with no significant non-linear components. This result also confirms the results of the profile analysis (see Figure 1).

Pearson r correlations for the entire sample demonstrated small but signifi- cant correlations between each of the four raw PASAT scores and estimated IQ (series 1: r = .27; series 2: r = .38; series 3: r = .26; series 4: r = .28; all p s < .OOl).

DISCUSSION

The PASAT performances of the young and middle aged groups in the current sample are comparable to Levin’s unpublished norms, which were as follows: series one M = 43, series two M = 38, series three M = 34, and series four M = 28 (H. S. Levin, personal communication, September 12, 1988). The percentage norms for the young and middle aged groups in the current sample were higher than those of Gronwall and Wrightson’s sample (1974), perhaps reflecting edu- cational or cultural differences between the two samples. Alternately, these dif- ferences may be attributable to the greater length of the Gronwall version, which could result in increased fatigue and thus poorer overall performance. Not sur- prisingly, the 60- to 75-year-old age group scored significantly lower than the other two age groups, suggesting a decline in information processing speed oc- curring at some point after the age of 50.

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 0

6:48

05

Sept

embe

r 20

13

38 DEBORAH D. ROMAN ET AL.

- Young Adult - Middle Aged Adult - Older Adult

0 s-1 5-2 s-3 s-4

PASAT Series

Fig. 1. Mean PASAT raw scores and standard error of measurement for young, middle aged, and older adults.

Although statistically significant, the correlations between estimated IQ and PASAT performance are quite small; for practical purposes, IQ appears to bear a minimal relationship to PASAT performance, at least for subjects falling within the normal to superior IQ ranges. Although the current sample approximated a normal distribution of IQs, it should be noted that there were relatively few subjects with IQs less than 90. The low correlations found between IQ and PASAT performance may be due to the somewhat truncated IQ range of the current sample. These norms should therefore be used with caution for individu- als of borderline or lower intelligence. Since it is likely that a minimal level of intellectual functioning is necessary to comprehend the instructions and to per- form the necessary mental functions in a timely fashion, the PASAT may not be an appropriate instrument for measuring acquired deficits in these individuals. Further, it should be noted that many normal subjects find the PASAT to be an aversive, stressful test and become anxious during its administration, despite encouragement and support. The PASAT may not be appropriate for highly

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 0

6:48

05

Sept

embe

r 20

13

PASAT NORMS 39

anxious patients, as excessive levels of anxiety could adversely affect PASAT performance.

There are several additional limitations of these norms. The sample clearly is not large enough to constitute a representative sample of the U.S. population. In particular, it is homogeneous for race and includes only a few subjects with less than a high school education. Therefore, these norms should be used with caution in the evaluation of minority patients and those with minimal formal education. Further, test-retest reliability was not assessed. Previous studies suggest that practice effects occur in this procedure; this is an important consideration in serial testing. Gronwall and Sampson (1974) reported significantly improved performance from the first to the second PASAT administration in their control subjects, but found minimal improvements with subsequent administrations. In contrast, Stuss, Stethem, Hugenholtz, and Richard (1989) found steady improve- ments in the performance of their controls across four administrations with a leveling off of performance by the fifth trial. The mean scores for the fourth administration were approximately one standard deviation higher than for the initial presentation, with most of the gains occurring between the first and sec- ond administration.

Regarding the PASAT’s utility as a clinical instrument, it is interesting to note that consistent differences between head-injury subjects and normal con- trols on PASAT performance have not been found. A number of investigators have found deficits in PASAT performance among head injury subjects (Gron- wall, 1977; Gronwall & Sampson, 1974: Gronwall & Wrightson, 1974; O’Shaughnessy et al., 1984). In contrast, Stuss et al. (1989) did not find signifi- cant differences in the PASAT scores of head-injured vs. control subjects. Levin et al. (1987) found defective PASAT performance in two of the three head- injured samples in their three-center study. The mixed findings in the Levin et al. (1987) study may provide a worthwhile clue to these inconsistent results.

One third of the Bronx sample employed by Levin et al. (1987) received their head injuries as a result of an assault: only 15% had experienced a motor vehicle accident (MVA). In contrast, about half of the subjects in the remaining two samples (from San Diego and Galveston) had suffered MVAs; only one person from these two samples had been assaulted. The Bronx head-injured subjects did not differ from their controls on PASAT performance, while the head-injured subjects in the other two samples did differ. The PASAT performance of the controls across these three settings did not appear to differ significantly.

Compared with assault, MVA typically involves greater acceleration/decel- eration of the brain and more substantial rotation of the head. These factors would be expected to result in increased subcortical involvement due to the effects of stress waves and cerebral strains generated by this type of an event (Gennarelli, 1986; Ommaya & Gennarelli, 1974). The PASAT may prove to be most sensitive to subcortical damage (i.e., temporary or permanent white matter axonal dysfunction): if so, the type of injury would have a bearing on perform- ance on this task. Future head-injury researchers would need to report the types

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 0

6:48

05

Sept

embe

r 20

13

40 DEBORAH D. ROMAN ET AL.

of injuries that characterize their samples i n order t o ascertain whether these speculations are supported.

In conclusion, the PASAT has proven t o be a useful instrument for measuring cognitive deficits associated with head injury. PASAT performance has been shown t o correlate with subjective complaints and functional difficulties. B y providing the clinician with age-appropriate norms, the present findings enable the clinician or researcher t o interpret test results with greater confidence. The results o f this study m a y also facilitate the use o f this instrument i n the evalu- ation of other types o f neurological disorders characterized b y cognitive slowing or concentration deficits (e.g., subcortical lesions, dementia). S ince the PASAT requires minimal materials and takes only 10 min t o administer, it m a y be a worthwhile addition t o the clinician’s assessment battery.

REFERENCES

Gennarelli, T.A. (1 986). Mechanisms and pathophysiology of cerebral concussion. Jour- nal of Head Trauma Rehabilitation, I , 23-29.

Gronwall, D.M.A. (1977). Paced Auditory Serial - Addition Task: A measure of recovery from concussion. Perceptual and Motor Skills, 44, 367-373.

Gronwall, D. (1987). Advances in the assessment of attention and information processing after head injury. In H.S. Levin, J. Grafman, & H.M. Eisenberg (Eds.), Neurobe- havioral recovery from head injury (pp. 355- 371). New York: Oxford University Press.

Gronwall, D.M.A., & Sampson, H. (1 974). The psychological effects of concussion. New Zealand: Aukland University Press/Oxford University Press.

Gronwall, D., & Wrightson, P. (1974). Delayed recovery of intellectual function after minor head injury. The Lancet, 2 , 605-609.

Levin, H.S., Benton, A.L., & Grossman, R.G. (1 982). Neurobehavioral consequences of closed head injury. New York: Oxford University Press.

Levin, H.S., Mattis, S . , Ruff, R.M., Eisenberg, H.M., Marshall, L.F., Tabaddor, K., High, W.M., & Frankowski, R.F. (1987). Neurobehavioral outcome following minor head injury: A three-center study. Journal of Neurosurgery, 66, 234-243.

Ommaya, A.K., & Gennarelli, T.A. (1974). Cerebral concussicn and traumatic uncon- sciousness: Correlation of experimental and clinical observations on blunt head injuries. Brain, 97, 633-654.

O’Shaughnessy, E.J., Fowler, R.S., & Reid, V. (1984). Sequelae of mild closed head injuries. The Journal of Family Practise, 18, 391-394.

Sampson, H. (1956). Pacing and performance on a serial addition task. Journal of Cana- dian Psychology, 10, 219-225.

Silverstein, A.B. (1982). Two- and four-subtest short forms of the Wechsler Adult Intelli- gence Scale-Revised. Journal of Consulting and Clinical Psychology, 50,415-41 8.

Stuss, D.T., Stethem, L.L., Hugenholtz, H., & Richard, M.T. (1989). Traumatic brain injury: A comparison of three clinical tests, and analysis of recovery. The Clini- cal Neuropsychologist, 3 , 145-1 56.

Tabachnick, B., & Fidell, L. (1989). Using multivariate statistics (2nd ed.). New York: Harper and Row.

Wechsler, D. (1981). WAIS-R manw1:Wechsler Adult Intelligence Scale - Revised. New York: Psychological Corporation.

Dow

nloa

ded

by [

Mon

ash

Uni

vers

ity L

ibra

ry]

at 0

6:48

05

Sept

embe

r 20

13