Speed of memory scanning is not affected in early HIV1 infection

This article was downloaded by: [Rush University], [Ms Eileen M. Martin]On: 29 June 2015, At: 15:12Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Clinical and ExperimentalNeuropsychologyPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/ncen19

Speed of memory scanning is notaffected in early HIV-1 infectionEileen M. Martin a , Lynn C. Robertson a , Donna J. Sorensen b ,William J. Jagust a , Kevin F. Mallon b & Valerie A. Chirurgi ca Department of Neurology , University of California , Davisb Research Service, Veterans Affairs Medical Center , Martinezc Department of Medicine , University of California , DavisPublished online: 04 Jan 2008.

To cite this article: Eileen M. Martin , Lynn C. Robertson , Donna J. Sorensen , William J. Jagust ,Kevin F. Mallon & Valerie A. Chirurgi (1993) Speed of memory scanning is not affected in earlyHIV-1 infection, Journal of Clinical and Experimental Neuropsychology, 15:2, 311-320, DOI:10.1080/01688639308402565

To link to this article: http://dx.doi.org/10.1080/01688639308402565

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 warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

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

Journal of Clinical and Experimental Neuropsychology 0 168-8634/93/1502-03 1 1 $25 .OO 1993, Vol. 15, No. 2, pp. 311-320 0 Swets & Zeitlinger

Speed of Memory Scanning is Not Affected in Early HIV- 1 Infection*

Eileen M. Martin’, Lynn C. Robertson’, Donna J. Sorensen2, William J. Jagust’, Kevin F. Mallon’, and Valerie A. Chirurgi3 ‘Department of Neurology, University of California-Davis

2Research Service, Veterans Affairs Medical Center-Martinez 3Department of Medicine, University of California-Davis

ABSTRACT

Thirty-seven nondemented HIV-seropositive and 17 seronegative control subjects were administered the Stemberg speed of memory scanning task, a procedure frequently employed to study mental slowing in patients with subcortical dementing disorders. Experimental and control subjects did not differ in speed of memory scanning, as indexed by the slopes of set size-reaction time functions, nor on mean 0-intercepts for the RT functions, which index stimulus detection and motor re- sponse time. Intercept values were significantly greater for subjects with a positive alcohol abuse history and for subjects with greater self-reported depression, but slopes were not significantly correlated with substance abuse history or psycho- logical distress. Cognitive slowing in early HIV-1 infection is not a nonspecific effect observed across all measures of information processing speed. Underlying component functions measured must be carefully considered when selecting reac- tion time tasks for study with HIV-seropositive subjects. The term “subcortical” dementia may be too general a descriptor, and RT task performance may provide an alternative basis for classification of dementia types.

The AIDS dementia complex (Navia, Jordan, & Price, 1986) has been classified as a “subcortical” dementia, with mental slowing as the most prominent cogni- tive abnormality. In a series of experiments using chronometric measures (Posner, 1986). we have demonstrated that subclinical cognitive slowing is present in nondemented HIV-seropositive subjects (Martin, Sorensen, Edelstein, & Robertson, 1992), as indexed by increased decision-making time, and the discrepancy be-

* Presented at the 20th Annual Meeting of the International Neuropsychological Society. February 1992, San Diego. We thank Howard Edelstein, M.D., Timothy Odell, M.D., KeMeth Shedd, M.D.. Sharon Oster, M.D., and Kent Sack, M.D.. for subject referrals, John Lackey for computer programming, and Robert Rafal, M.D., Mark Kelly, Ph.D.. and two anonymous reviewers for helpful comments on an earlier draft of the manuscript. This research was supported by a Veterans Affairs Merit Review Grant to Dr. Martin. Address correspondence to Eileen M. Martin, Ph.D., Department of Psychiatry (M/C 913), University of Illinois, 912 South Wood St., Chicago, IL 60612, USA. Accepted for publication: May 15, 1992.

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

312 EILEEN M. MARTIN ET AL.

tween simple and choice reaction time (RT). We have also investigated specific component cognitive functions in HIV-positive subjects and demonstrated that cognitive slowing selectively affects attentional processing (Martin, Sorensen, Robertson, Edelstein, & Chirurgi, in 1992~). Controlled attentional processes appear particularly vulnerable to disruption in HIV-1 infection (Martin et al., 1992b). Our primary focus on attentional processes, however, does not exclude the possibility that additional component cognitive functions are disrupted by HIV-1 infection.

The speed of memory scanning paradigm developed by Sternberg (1969) has been employed frequently to study information processing in a variety of neurologic disorders, including Korsakoff syndrome (Cermak, 1977). Friedreich ataxia (Hart, Kwentus, Leshner, & Frazier, 1985), and myotonic dystrophy (Stuss et al., 1987). The task has been of particular interest in the study of cognition in subcortical dementing disorders such as Parkinson disease and multiple sclerosis (Wilson, Kaszniak, Klawans. & Garron, 1980; Rafal, Posner, Walker, & Friedrich, 1984; Rao, St. Aubin-Faubert, & Leo, 1989). Mental slowing or “bradyphrenia” is the primary central cognitive deficit in subcortical dementias, but is often difficult to study experimentally because of the patients’ concurrent motor slowing. The speed of memory scanning procedure permits experimental separation of mental slowing from motor slowing.

Briefly, the speed of memory scanning task consists of a series of two-choice RT trials. On each trial, the subject views a single target digit and must indicate by a button or keypress whether or not the target was present in a set of digits previously memorized. Set size (the number of digits in the memory set) is varied across trials or blocks of trials. The relationship between set size and RT is linear in normal subjects (Sternberg, 1966). The slope of the RT function represents the increment in time required to make a decision as the number of items to be searched in short-term memory increases and thus indexes memory scanning speed; the 0-intercept represents the time required for stimulus detection and response preparation and execution. It is assumed that, regardless of intercept value, the slope of the RT function will be steeper in subjects with cognitive slowing.

In the current study, we investigated memory scanning as an additional po- tential mechanism of cognitive dysfunction in HIV-1 infection. We administered the speed of memory scanning task to groups of symptomatic and asymptomatic HIV-seropositive subjects and matched seronegative controls. We also adminis- tered measures of psychological distress in order to determine the effect of cur- rent depression or anxiety on scanning speed. A finding of increased scanning time (indexed by steeper RT slopes) in the HIV-positive subjects would suggest that speeded memory search, in addition to attentional processes, is disrupted in HIV-1 infection.

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

HlV-1 INFECTION 313

METHODS

Subjects Experimental subjects were 37 HIV-seropositive right-handed males, including 34 gay or bisexual men, 2 subjects with a history of IV drug abuse or transfusion, and 1 subject with unknown risk factors. They were recruited from the Infectious Disease Clinics at the VA Medical Center-Martinez and at Memthew Memorial Hospital, and through local physi- cians. Seventeen subjects were clinically symptomatic. This group included 11 subjects with constitutional symptoms such as fatigue and night sweats (CDC Stage IVA) (Centers for Disease Control, 1987) but without AIDS-defining disorders and 6 subjects with a history of Kaposi sarcoma or pneumocystis carinii pneumonia but no neurologic disease. The asymptomatic group included 14 subjects who were completely clinically asympto- matic and 6 subjects who were asymptomatic except for persistent generalized lymphadenopathy (CDC Stages II/lII). Subjects were staged by their physicians on the basis of the physical examination and laboratory values. All experimental subjects were ambulatory and without serious clinical disease at the time of testing, which was con- ducted on an outpatient basis. All experimental subjects had a normal neurologic exami- nation. Subjects were volunteers and were not referred for study because of neurobehavioral symptoms or signs.

Controls were 16 seronegative gay or bisexual men and 1 heterosexual female (serostatus documented by ELISA in all subjects). Prospective experimental or control subjects with a history of closed-head injury with loss of consciousness greater than 30 min, learning disability, or premorbid psychiatric disorder were excluded from study. The three groups did not differ significantly in mean age, years of education or estimated verbal intelli- gence as measured by WAIS-R Vocabulary scaled scores (F < 1 for each comparison) (Table 1). Eight experimental and 1 control subject had a past history of alcohol abuse and 3 experimental and 1 control subject had a history of drug abuse, but no subject had abused alcohol or drugs for at least 3 years prior to study. Fifteen symptomatic and 7 asymptomatic subjects were taking AZT at the time of testing. Four symptomatic and 3 asymptomatic subjects were taking antidepressant or anxiolytic medication.

Procedure The memory scanning task was administered to all subjects on an AT-compatible compu- ter and VGA monitor. The task consisted of six blocks of 24 two-choice RT trials, pre- ceded by one block of 24 practice trials. We utilized the “varied” version of the memory scanning task, in which a different memory set is presented on each trial, as opposed to the “fixed” version, which employs the same memory set for all trials within a block (Stemberg. 1969). At the start of each trial, an 880-Hz warning tone sounded for 500 ms. A set of

Table 1. Demographic Data for All Subjects. ~

Mean Mean Mean WAIS-R Mean CD4 Subject Group Age (SO) Education (SD) Vocabulary (SD) Count (SD)

HIV+/Symptomatic 38.5 (7.9) 14.6 (2.4) 12.3 (3.8) 355 (248.7)

HIV+/Asymptomatic 36.8 (7.0) 14.9 (2.5) 13.6 (2.3) 596 (310.2)

Control 34.8 (11.0) 15.4 (2.0) 13.4 (3.3) Not obtained

(n=17)

(n=20)

(n=17)

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

314 EILEEN M. MARTIN ET AL.

centered and horizontally arrayed digits appeared on the monitor screen 100 ms after tone offset, and the subject was required to memorize the set. Each digit subtended a visual angle of approximately 2’. The number of digits in the set varied randomly across trials (1,2.4 or 6) and the set remained on the screen for 5 0 0 msldigit. One s after the memory set went off, a single target digit appeared in the center of the display and remained on the screen for 3 s or until the subject responded. If the subject decided that the target had appeared in the memory set just seen, a response button marked “YES” was pressed using the right hand. A separate button marked “NO” was pressed using the left hand if the subject decided the target had not appeared in the set. The target was present in the set on half the trials. RT were recorded to the nearest ms. Error and no-response trials were also recorded.

All subjects completed three self-report measures of psychological distress, including the Beck Depression Inventory (Beck, Ward, Mendelson. & Erbaugh, 1961). the Center for Epidemiological Studies Depression Scale (Radloff, 1977). and the State-Trait Anxi- ety Inventory-State version (Spielberger, Gorsuch, & Lushene, 1970).

RESULTS

Reaction Times Error rates for the three subject groups were computed and analyzed by a Kruskal- Wallis test. Nonparametric analysis was used because error scores were not normally distributed. There were no significant differences in error rates be- tween groups (x2(2)=5.20; p < .12). However, error rates for two asymptomatic and two symptomatic subjects exceeded a 5% cutoff (but were less than lo%), and these subjects’ data were excluded from further analysis. Error rates for the remaining subjects did not exceed 2% for any subject group or stimulus condi- tion. Median RTs were computed and analyzed by a 3 x (2 x 2) mixed-design analysis of variance (ANOVA), with Groups (Symptomatic, Asymptomatic, Con- trol) as the between-subjects factor and Set Size and Trial Type (Target Present vs. Target Absent) as the within-subjects factors. Results indicated an expected significant main effect for Set Size (F(3,141) = 1 8 7 . 4 ; ~ c .OOOl), with longer RT at larger set sizes. There was also a significant main effect for Trial Type (F(1,47) = 12.2; p < .001), with longer RT on Target Absent trials. The main effect and interactions involving the Groups factor, however, were not significant (F c 1 for each comparison). The Group x Trial Type interaction was not significant, but the Set Size x Trial Type interaction was significant (F(3,141) = 5.77; p < .Ol). See Figure 1 for reaction times in the two conditions.

RT-Set Size Functions Because of the significant Set Size x Trial Type interaction, slopes and O-inter- cepts of the RT-set size functions for the three groups were calculated separately for Target Present and Target Absent trials (see Table 2). Slopes and intercepts were calculated using linear regression and analyzed by one-way ANOVA.

There were no significant differences between groups for slopes (Target Present: F(2,47) < 1; Target Absent: F(2,47) = 1.32; p < .30). Intercepts were not signifi- cantly different between groups on Target Present trials (F < 1), but the differ-

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

HlV-I INFECTION

950 -- 900 - - 850 -- 800 -- 750 -- 700 -- 650 -- 600 550 500 450

315

- - -- -- 1-

Target Present Trials

1 050 - 1000 -- 950 -- 900 -- 850 -- 800 - - 750 700 650 600 550 500 450 400

i! i=

-- - - -- 4 - 0 Controls -- m-. HIV+/ASx -- *-A HIV+/Sx - - - -

I I I I I I I I 1 1 I

1000 + I

4-4 Controls B-¤ H IV+/ASx * - * HIV+/Sx

I I I I 400 ! I I I 1 1

I I I

1 2 4 6 Memory Set Size (W of Digits)

Fig. 1. Mean reaction times for all groups plotted by set size and trial type.

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

316 EILEEN M. MARTIN ET AL.

ence between groups for intercepts on Target Absent trials reached borderline significance (F(2,47) = 2.78; p < .07).

Slopes did not differ significantly between experimental subjects on and off AZT (F(1,31) < 1 for both trial types). Similarly, there were no significant differences in intercept values between AZT-treated and untreated experimental subjects (Target Present: (F(1,31) c 1; Target Absent: F(1,31) = 1.75; p < .20). Slopes did not differ between subjects with and without a positive alcohol his- tory (F(1,48) < 1 for both trial types), but intercepts were significantly greater for subjects with a positive alcohol history (Target Present: F(1,48) = 7.95; p < .01; Target Absent: F(1,48)=8.83; p < .005). CD4 lymphocyte counts were es- sentially uncorrelated with memory search variables.

Since overall group differences in memory search variables were not demon- strated between experimental and control subjects, the incidence of abnormal test performance (defined as deviantly high slope or intercept values) in each subject group was evaluated. A total of four standard scores, for slopes and intercepts on Target Present and Target Absent trials, was computed for each subject, using the means and standard deviations obtained for the control group. For each memory search variable, the frequency of standard scores greater than or equal to two SD above the control mean for each group was computed and evaluated by the Fisher Exact Test. There were no significant differences be- tween groups in frequency of deviantly high slopes or intercepts on either target- present or target-absent trials (smallest associated p-value = .24).

Psychological Distress Measures Table 3 shows the mean scores for each of the psychological distress measures. The groups differed significantly on mean Beck Depression Inventory (BDI) scores (F(2,46) = 5.93; p < .005). Post hoc Duncan tests indicated that sympto- matic subjects scored significantly higher than controls (p < .Ol), but means for

Table 2. Mean Slopes and Intercepts of RT-Set Size Functions.

Subject Group Slope* (SD) Intercept+ (SD)

Target Present Symptomatic Asymptomatic Control

Target Absent Symptomatic Asymptomatic Control

68 (29.1) 63 (24.4) 68 (30.4)

74 (40.7) 53 (30.7) 69 (42.3)

479 (175.3) 491 (148.8) 438 (96.5)

480 (123.5) 568 (154.2) 480 (93.6)

* All slopes in ms/digit + All intercepts in ms

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

HIV-I INFECTION 317

asymptomatics and controls did not differ significantly. There was a borderline significant difference between groups on the Center for Epidemiological Stud- ies-Depression Scale (CES-D) (F(2,47) = 3.04; p < .06), with a trend toward higher scores in the symptomatic group compared to controls. Mean State-Trait Anxiety Inventory (STAI) scores for the three groups did not differ significantly (F(2,47) = 1.51; p < .25).

Scores on the psychological distress measures were essentially uncorrelated with slopes of the RT functions for both trial types. Scores on all three measures, however, correlated significantly with intercepts for Target Absent trials (p < .05 for each). Intercepts for Target Present trials correlated significantly with BDI and CES-D scores (p < .05 for both), and the correlation with STAI scores reached borderline significance ( p < .06) (see Table 4).

DISCUSSION

Despite previous studies indicating that subclinical cognitive slowing can be demonstrated in HIV-seropositive subjects using chronometric measures (Martin et al., 1989; Martin et al., 1992a; Martin et al., 1992c; Martin et al., 1992b), experimental subjects in the current study did not show evidence of increased

Table 3. Mean Psychological Distress Scores for All Subjects.

Subject Beck Depression CES Depression State-Trait Group Inventory (SD) Scale (SD) Anxiety (SD)

Symptomatic 11.6** (8.7) 14.0 (10.0) 35.9 (10.5) Asymptomatic 7.6 (7.4) 9.9 (6.4) 32.1 (8.7) Control 3.0 (4.7) 7.1 (7.5) 30.4 (8.6)

** p < .01 (vs. Controls)

Table 4. Correlations of Psychological Distress Scores with Memory Search Variables.

BDI CES-D STAI ______

Target Present Slope -.02 .04 .07 Intercept .31* .28* .22

Target Absent Slope -.12 -.05 -.14 Intercept .32* .23* .30*

* p < .05

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

318 EILEEN M. MARTIN ET AL.

memory scanning time. Although symptomatic subjects showed a nonsignificant trend toward higher error rates compared to controls, analysis of individual sub- jects’ scores on memory search variables did not demonstrate an increased inci- dence of abnormal task performance for the HIV-seropositive subjects.

The lack of differences between groups cannot be attributed to insufficient task difficulty, since we have previously demonstrated group differences be- tween HIV-positive and control subjects on relatively simple RT tasks of lesser difficulty than the memory scanning task (e.g., Martin et al., 1992a). Similarly, the lack of differences between experimental subjects on and off AZT argues against AZT treatment as an explanation for these negative results. In addition, AZT-treated subjects have shown longer RT compared to controls in previous experiments.

Our data are consistent with a previous report of normal scanning speed in asymptomatic subjects (Wilkie, Eisdorfer, Morgan, Loewenstein, & Szapocznik, 1990), but the current study also demonstrated normal scanning speed in sympto- matic subjects. The earlier study by Wilkie et al. reported a significantiy greater intercept value for asymptomatics, but our data indicate that intercepts correlate significantly with psychological distress and with a positive alcohol history rather than with serostatus. This finding is also consistent with an earlier report from our laboratory of a significant correlation between simple RT, an index of motor slowing, and distress scores (Martin et al., 1992a).

Results of RT studies in the HIV-1 literature are inconsistent. Some investi- gators (e.g., Martin et al., 1989; Wilkie et al., 1990) have reported significant differences in RT performance between asymptomatic seropositive subjects and controls. Others, such as Perdices and Cooper (1989) and Miller, Satz, and Visscher (1991) have reported differences between controls and symptomatic subjects only. We propose that these discrepant results are due in part to differences in task selection. Much research on cognition in HIV-1 and other subcortical disor- ders has assumed implicitly that slowing should be apparent across all speeded cognitive tasks. However, our data suggest that cognitive changes in early HIV- 1 infection are more selective than the general term “cognitive slowing” would imply, and that task requirements and the underlying component functions meas- ured must be considered when RT measures are selected for study.

The results of a previous experiment from our laboratory using a reaction time version of the Stroop task (Martin et al., 1992b) support this contention. The Stroop effect has two components: facilitation, measured by the difference in RT to congruent and neutral stimuli, and inhibition, measured by the difference in RT to neutral and incongruent stimuli, and these components are differentially affected by automatic and controlled processes. We found that HIV-positive subjects showed greater Stroop effects than controls, but only one component was affected. The HIV-seropositive subjects showed much greater inhibition, but normal facilitation compared to controls.

Finally, the current data support criticisms of the “cortical/subcortical” clas- sification of dementia as inappropriately nonspecific (e.g.. Whitehouse, 1986). It

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

HIV-1 INFECTION 319

has often been assumed that certain RT tasks such as the Sternberg tap a deficit common to all patients with “subcortical dementias.” However, memory scan- ning time is increased in patients with multiple sclerosis (Rao et al., 1989). variable in patients with Parkinson disease (Rafal et al., 1984; Wilson et al., 1980), and normal in HIV-positive subjects. It is possible that data from RT experiments will be useful in developing a more specific classification of dementias based on profiles of cognitive component functions in addition to primary site of neuropathology .

REFERENCES

Beck, A.T., Ward, C.H., Mendelson, M., & Erbaugh, J.K. (1961). An inventory for meas- uring depression. Archives of General Psychiatry, 4, 561-571.

Centers for Disease Control (1 987). Revision of the CDC surveillance case definition for acquired immunodeficiency syndrome. Morbidity and Mortality Weekly Report, 34,3- 15.

Cermak, L. (1977). The contribution of a “processing” deficit to alcoholic Korsakoff patients’ memory disorder. In I.M. Bimbaum & E.S. Parker (Eds.), Alcohol and hu- man memory (pp. 195-208). Hillsdale, NJ: Erlbaum.

Hart, R.P., Kwentus, J.A., Leshner, R.T., & Frazier, R. (1985). Information processing speed in Friedreich’s ataxia. Annals of Neurology, 17, 612-614.

Martin, A., Kampen, D., Salazar, A.M., Williams, J., Law, W., & Roller, T. (1989). Slowed cognitive processing in HIV+ patients in comparison to psychiatric controls. Fifth International Conference on AIDS: Abstracts, 210.

Martin, E.M., Robertson, L.C.. Edelstein, H.E.. Jagust, W. J., Sorensen, D.J., San Giovanni, D., & Chirurgi, V.A. (1992b). Performance of patients with early HIV-1 infection on the Stroop task. Journal of Clinical and Experimental Neuropsychology, 14,857-868.

Martin, E.M., Sorensen, D.J.. Edelstein H.E.. & Robertson, L.C. (1992a). Decision-mak- ing speed in HIV-1 infection. AIDS, 6, 109-113.

Martin, E.M., Sorensen, D.J., Robertson, L.C., Edelstein, H.E., & Chirurgi, V.A. (1992~). Spatial attention in HIV-1 infection. Journal of Neuropsychiatry and Clinical Neurosciences, 4,288-293.

Miller, E.N., Satz, P., & Visscher, B. (1991). Computerized and conventional neuropsychological assessment of HIV-1 -infected homosexual men. Neurology, 41,

Navia, B.A.. Jordan, B.D.. & Price, R.W. (1986). The AIDS dementia complex: I. Clinical features. Annals of Neurology, 19.5 14-524.

Perdices, M., & Cooper, D.A. (1989). Simple and choice reaction time in patients with human immunodeficiency virus infection. Annals of Neurology, 25, 460-467.

Posner, M.I. (1986). Chronometric explorations of mind. New York: Oxford University Press.

Radloff. L.S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1,385-401.

Rafal, R.D., Posner, M.I., Walker, J.A., & Friedrich. F.J. (1984). Cognition and the basal ganglia: Separating mental and motor components of performance in Parkinson’s dis- ease. Brain, 107, 1083-1094.

Rao, S.M, St. Aubin-Faubert, P., & Leo. G.J. (1989). Information processing speed in patients with multiple sclerosis. Journal of Clinical and Experimental Neuropsychology, 1 I . 47 1-477.

1608-1 61 6.

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5

320 EILEEN M. MARTIN ET AL.

Spielberger, C.D., Gorsuch, R.L., & Lushene, R.E. (1970). State-Trait Anxiety Inventory.

Stemberg, S . (1966). High-speed scanning in human memory. Science, 153,652-654. Stemberg, S . (1969). Memory scanning: Mental processes revealed by reaction time ex-

periments. American Scientist, 57. 421 -457. Stuss, D.T., Kates, M.H., Poirier, C.A., Hylton, D., Humphreys, P., Keene, D., & Lafleche.

G. (1 987). Evaluation of information-processing speed and neurops ychological func- tioning in patients with myotonic dystrophy. Journal of Clinical and Experimental Neuropsychology, 9, 131-146.

Whitehouse, P.J. (1 986). The concept of subcortical and cortical dementia: Another look. Annals of Neurology, 19,l-6.

Wilkie, F.L., Eisdorfer, C., Morgan, R., Loewenstein, D.A., & Szapocznik, J. (1990). Cognition in early human immunodeficiency virus infection. Archives of Neurology.

Wilson, R.S., Kaszniak, A.W., Klawans, H.L., & Garron, D.C. (1980). High speed memory

Palo Alto: Consulting Psychologist Press.

47,433-440.

scanning in parkinsonism. Cortex, 16, 67-72.

Dow

nloa

ded

by [

Rus

h U

nive

rsity

], [

Ms

Eile

en M

. Mar

tin]

at 1

5:12

29

June

201

5