1-s2.0-0028393288900048-main

8/12/2019 1-s2.0-0028393288900048-main

http://slidepdf.com/reader/full/1-s20-0028393288900048-main 1/16

Nr uropsychologra, Vol. 26. No. 5. PP . 685-700, 1988.Pr inted in Great Bntam

0028-3932/88 53.00+0.006’ 1988 Pergamon Press plc

MODELLING DEMENTIA: EFFECTS OF SCOPOLAMINE ONMEMORY AND ATTENTION

P. BROKS,* G. C. PRESTON, M. TRAUB, P. POPPLETON, C. WARD and S. M. STAHL

Neuroscience Research Centre, Merck Sharp Dohme Research Laboratories, Terlings Park, Eastwick Road,Harlow, Essex, CM20 2QR, U.K.

Received 16 August 1987; accepted 18 December 1987)

Abstract-Scopolamine, a muscarinic cholinergic antagonist, is capable of inducing transientmemory impairment in normal subjects. Against the background of the cholinergic hypothesis ofAlzheimer’s disease (AD) the present study was designed to investigate the effects of low oral doses ofscopolamine on a range of cognitive functions known or hypothesized to be affected in AD. Twentyhealthy volunteers (1848 yr) performed a battery of automated cognitive tasks under each of fivetreatments: oral scopolamine 0.3 mg, 0.6 mg, 1.2 mg; oral methylscopolamine 0.6 mg; placebo.Alongside analogous tests of verbal and non-verbal memory, the battery enabled assessments of arange of attentional functions: alerting, sustained attention, selective attention, and covertorientation. A profile of effects was observed within and beyond the realm of memory. While somefunctions were unaffected by the drug (e.g. alerting) and others were impaired at the highest dose (e.g.verbal learning) still others were affected in a linear dose-dependent manner (sustained attention;visual contrast sensitivity). These observations are discussed in the context of the “scopolaminemo el” of AD.

INTRODUCTION

ALZHEIMER’S DISEASE (AD), the most common cause of dementia, is characterized by aprogressive and unremitting deterioration in cognitive function. It has been proposed thatcertain clinical features of the disease may be attributable to degeneration of specificneuronal pathways, particularly those which project diffusely to the cerebral cortex from thebasal forebrain and employ the neurotransmitter acetylcholine [6, 12, 36, 371. The mainbody of evidence linking the intellectual decline of AD with a cholinergic deficit derives fromthe analysis of post-mortem studies. For example, it has been demonstrated that in AD thereis a decline in the cortical levels of choline acetyl transferase (ChAT), an enzyme involved inthe synthesis of acetylcholine and a specific marker for cholinergic neurones. Cholinergicdepletion has been shown to be most severe in hippocampus, parietal and temporal cortexIll, 37,441. Furthermore, the reduction in cortical ChAT has been shown to correlate withthe severity of intellectual decline in life [38].

The “cholinergic hypothesis” of dementia has also received support from the observa-tion that pharmacologically induced blockade of central cholinergic neurones leads tomemory impairments in normal subjects. The hypothesis gains credence to the extent thatsuch effects may be shown to resemble the cognitive disorder of AD, and pharmacologicalmanipulation of cholinergic function in turn provides a valuable strategy for defining the role

*Present address: Department of Psychology, St James’s University Hospital. Beckett Street. Leeds, LS9 7TF.

685

8/12/2019 1-s2.0-0028393288900048-main


686 P. BROKS, G. C. PRESTON, M. TRAU B, P. P OPPL ET ON, . WAK I and S. M. STAHL

ofcholinergic pathophysiology in the dementing process. In this respect, the most extensivelystudied anticholinergic agent is scopolamine (hyoscine). The transient amnestic effects ofscopolamine have long been noted from its application in surgery as a premedicant andearly controlled studies, both clinical [32] and experimental [46], substantially confirmed

clinical observation. In particular, it appeared that the effect of scopolamine was to disruptthe process of memory consolidation. Thus while primary memory as assessed by digitspan, for example, is minimally affected by scopolamine, delayed recall of such material isgrossly impaired [14, 18, 461. Again, impairment is evident on tasks, such as list learningor paired associates, which gauge the efficiency of new learning [S, 10, 14, 16, 18, 19, 39,46, 481. Several studies have supported SAFER and ALLEN’S [46] original observation thatscopolamine does not affect retention of material learned prior to drug administration[19,39] which suggests that the effects of scopolamine may be located at the acquisitionstage.

A comparison of the transient effects of scopolamine with the amnestic symptoms of ADreveals both strengths and weaknesses in the scopolamine model. While it can usefullyaccommodate certain of the learning and recent memory deficits observed clinically andexperimentally [22, 281 in AD patients, its shortcomings become clear if one considers thebroader pattern of memory dysfunction characteristic of true dementia. For example,whereas the disruptive effect of scopolamine appears to be restricted to processes involved insecondary memory, AD is characterized by multiple memory deficits affecting both primaryand secondary systems [24, 281.

The discrepancy is not of itself sufficient to invalidate the model, but important questionsare raised regarding its range of application. It may be that the scopolamine model hasintrinsic limitations; there is now substantial evidence that cholinergic depletion representsonly one facet of the pathophysiology of AD. A number of other transmitters have recently

been implicated [45] which might further account for the general failure of current attemptsat cholinergic replacement therapy [21], and, ofcourse, interest in the neuropathology o D

is not confined to the neurochemical domain [23,35]. Thus the range of application of thescopolamine model would extend only to clinical manifestations specifically associated withcholinergic dysfunction-of which loss of recent memory might be the most prominentexample.

Alternatively, at least some of the limitations may be more apparent than real due toshortcomings in the research methodologies which have been adopted in attempts tosubstantiate the model. For example, in AD there is a chronic loss of cholinergic neuroneswhile experimental studies in man have only investigated the effects of acute or sub-acutedrug administration.

Another possibility for extending and substantiating the scopolamine model is to broadenthe scope ofinquiry beyond memory to incorporate other pertinent aspects ofcognition. Thiswill be a major concern of the present paper. There have been good reasons for thepredominant emphasis on learning and memory, but certain other features of cognitivedysfunction in AD could be explored from the perspective of experimental psychopharma-cology. Most significantly perhaps, few studies have examined attentional processes in anysystematic fashion--either in true dementia or in experimental models. Although attentionaldisorder is rarely explicitly cited as a clinical feature of dementia, it is worth considering thepossibility that some of the commonly noted cognitive dysfunctions associated with AD maycoexist with, and perhaps to some extent be shaped by, more fundamental deficits ofattention.

8/12/2019 1-s2.0-0028393288900048-main


MODELLING DEMENTIA 687

In cognitive psychology the concept of attention has been used to cover a range of separatefunctions. For example, the term “vigilance” has been used synonymous with “alertness” torefer to a subject’s state of optimal preparedness to receive an anticipated stimulus [40]. Itmay also refer to a subject’s capacity to sustain awareness of a series of target stimuli

presented over a relatively protracted period [54]. Another fundamental aspect of attentionis perceptual selectivity; this refers to the process by which a particular perceptual objectemerges from an array of competing sensory inputs [3]. The converse of this may beconstrued as a form of “distractibility” whereby a subject is able to shift the focus of hisattention in response to other significant events in the sensory field [41]. To some extent suchfunctions appear to be neurologically dissociable so that, for example, sustained attentionmay depend on the functional integrity of the frontal lobes [54] whereas the parietal lobesseem to have a special involvement in mediating shifts of attention [42].

For the most part, assessments of attentional deficits in AD patients have relied onrelatively ad hoc clinical procedures such as “letter cancellation”, “serial sevens” or digit-symbol substitution which, while undoubtedly drawing on attentional resources, do so in arather indiscriminate fashion. Digit-span, which has been widely studied in AD patients, isfrequently referred to as an “attentional” task [50] but is probably better construed as ameasure of verbal working memory. The suggestion of attentional impairments in dementiais not new [55] but formal cognitive studies are scarce.

In general, the sporadic reports to be found in the literature on dementia tend to supportthe notion of an intrinsic attentional deficit. For example, patients with dementia have beenshown to be impaired on continuous performance vigilance tasks [2]. Abnormalities insimple reaction time [49] and in electrophysiological response [31] have also been taken toindicate an impairment of the “alerting” (vigilance) response in AD patients. Differencesbetween mild and moderate AD patients in the pattern of performance on a verbal learning

task have also been attributed to differences in the deployment of already depletedattentional resources [30].From studies of normal subjects there is growing evidence that the cholinergic system may

be directly implicated in such attentional processing. Several studies have indicated thatscopolamine impairs signal detection and sustained vigilance [34,5 1,521 and there is someevidence to suggest that the drug impairs mechanisms of selective attention [15].

Given the suggestion of attentional impairments in AD, it is clearly of some interest toestablish the profile and intercorrelation of memory and attentional disturbances induced innormal subjects under scopolamine. The present study employed a cognitive test batteryincorporating both memory and attentional tasks as a preliminary step towards theestablishment of such a profile. In order to exclude as far as possible sedative effects of

anticholinergic treatment, lower doses of scopolamine have been used than in the majority ofprevious studies (for comprehensive reviews see COLLERTON [9] and KOPELMAN [25]). Theinvestigation represents the first phase in a programme of research into dementia which willincorporate neuropsychological studies of patients in parallel with psychopharmacologicalanalyses of cognitive function in healthy volunteers.

Subjects

METHODS

Twenty healthy volunteers (10 male and 10 female), aged between 18 and 48, were paid to take part in the study.All gave informed consent but none were aware of any hypothesis being tested.

8/12/2019 1-s2.0-0028393288900048-main


688 P. BKOKS G. C. PKESTON M. TKAUB P. POPPLET ON . WARD and S. M. STAH L

Apparatus

The experiment was controlled by an IBM PC XT, which also collected data. In addition, each testing stationcontained a Mellordata Digivision 14 in. colour monitor on which graphics displays could be presented to thesubject. These displays were generated by an IO Research Pluto II graphics unit.

Subjects were able to respond using one of the following: (I) by pointing at graphic items on the touch sensitivescreen (Mellordata); (2) by triggering a voice sensitive key (Campden Instruments Ltd); (3) by pressing a hand-heldpush button; (4) by pressing a key on a four-key choice-reaction time keyboard.

One test (the little man test) wasdrawn from the Bexley-Maudsley “BMAPS” battery [l] and administered on anApple IIe computer.

Procrdurr

Subjects attended on five occasions at weekly intervals and received one of the following at each session; oralscopolamine (0.3 mg, 0.6 mg or 1.2 mg), oral methylscopolamine (0.6 mg), or placebo. Treatments wereadministered double blind according to a latin square design. Each experimental session lasted approximately 3 hr.Subjects were given a brief medical examination after which the drug or placebo was administered. During the next45 min the subjects performed automated psychometric tests and rested. Their visual contrast sensitivity was thenmeasured using the Cambridge Low Contrast Gratings test 1531. Next the subjects were given a test of prospectivememory: they were told that the session would be timed and that they must tell the experimenter to stop a timer atthe end of the study period. Then followed approximately 2 hr of cognitive tests, as described below. Finally thesubjects were given a brief medical examination and transported home.

The battery consisted of tests of primary and secondary memory together with tests of a range of attentionalprocesses and perceptual ability. For half the subjects (5 male, 5 female) the battery was given in one order; for theremainder the order was reversed.

(I) S&p/c, wuction tiw: ulrrting finction. This was a reaction time test in which the subject faced the monitorscreen and held a push-button in his/her preferred hand. The task was to press the button as quickly as possible whena large yellow cross was displayed on the screen. The crosseither appeared at the same time as the computer issued adistinctly audible tone (440 Hz. 50 msec) or appeared shortly after the tone such that the tone predicted theappearance of the cross. Stimulus onset asynchronies (SOAs) used were 0,200,400.800, 1600,320O msec. Averageinter-trial intervals (ITIs) were 6 set (range=5-7 set). Three blocks of48 trials were run, with 8 trials at each SOAper block, in a pseudo-random order. The subject was given an opportunity to rest after each block.

(2) Sintpld’hoicr rrmtiorl /ime: unsi~qmrlled. In this test the subject was instructed to place his.‘her finger tips on arubber pad and to moveistrike one of four keys whenever one of four boxes displayed on the monitor lit up in red.The keys were set radially equidistant from the rubber pad and corresponded spatially to the four boxes displayed onthe screen. Seven blocks of20 trials were run. In thesimple reaction time conditions (blocks 2,3,6 and 7) the subjectswcrc informed at the start of each block which of the four keys was the target: throughout the block only the boxcorresponding to that key turned red. Under choice conditions (blocks, I, 4 and 5) the subjects were informed at thebeginning of the block that on each trial any of the four boxes might turn red and that they were simply to hit theappropriate key. Block I was or practice: data from this block were ignored. Average inter-trial intervals were 5 see(range=4 6). After each block of trials the subjects were given an opportunity to rest.

(3) I’isu~~l selrc~rir~ trtrt~rio~~. In this test. which was based on procedures described by ALLIWKT (‘f rri. [3]. thesubject faced the monitor and held ; I microphone connected to a voice-operated switch. He, she was instructed thatsuperimposed pairs of letters would appear on the screen, one orange and one green. The task &as simply to say theorange letter as quickly as possible, ignoring the green one. Three conditions were run. In the control condition therewas a random relationship between the orange and green letters of any given pair: in the constant condition thegreen letter was al ways the same: in the negative priming condition the lirst pair in a block of trials were chosenrandomly; thereafter the orange (target) letter of 21 particular pair was always the green (distractor) letter of theprevious pair.

Nlnc blocks of trials were run. three of each condition, separated by a brief rest. Each block consisted of thepresentation of I I pairs of letters chosen to contain one each of the following: AFHEJINLSTX. Reaction times tothe first pair ofeach block wcrc ignored: thereafter each of the subjects’ utterances ~21s recorded as a reaction-timeand the next pair of lcttcrs uas dlsplayed. Errors of identification were recorded manually hq the experimenter.

(4) C‘owrr oriwrtrrimr. This test makes WC ofa paradigm developed by POSN~K et rl. 1411. Subject5 were requiredIO lixate a central cross on the monitor and press a hand-held push button \I henever a red star was presented m oneof two peripheral grey boxes. Four blocks of 30 trials were run. with a mean IT1 of 6 set (range 5 7 set).

Each trial consisted ofthe dimming to darkness ofone ofthc grcy boxes for 75 msec (the “cue”) followed. at one ofthree intervals. by the appearance of a red star in one of the side boxes (the “target”). Cue target intervals were 325.575 and 1075 msec. On 80”/;1 of trials (the “valid” condition) the target was displayed in the box in which the cue hadbeen given: on the remaining 20% of trials (the “invalid” condition) the target appeared on the uncued side. Therewcrc equal numbers of left and right trials and equal numbers of trials at each SOA in each block.

(5) .S~,\r~r;~r/ (IIIP~I~;O)I. hih teat. based upon ;I procedure dc\cloped h> WLLI M r r 01. [St]. requires the \uhject to

8/12/2019 1-s2.0-0028393288900048-main



listen to trains of auditory “bleeps” from the computer. At the end of each train the subject is asked to indicate (bytouching a grid on the screen) how many bleeps there were in the previous train. Train lengths from 5-14 are given ina random order at three nominal frequencies, 0.25 Hz, 1 Hz and 3 Hz. The bleeps in a train have a duration of 50 msand are separated by a period randomly chosen to be l/Frequency f OS/Frequency, i.e. at a nominal frequency of0.25 Hz tones may be separated from 2 to 6 sec. Two measures are taken in this procedure; the number of errors thesubject makes at each frequency and the actual estimates of train lengths.

(6) Digit span In this test subjects were presented with sequences of digits. At the end of each sequence a gridcontaining the digits l-9 was displayed on the touch screen; the subject was then required to repeat the digitsequence by touching the appropriate items on the grid. Digit sequences were given at a frequency of approxI digit/set. The sequences were chosen randomly with the constraint that no digit could appear more than once in agiven sequence.

The test always started with a sequence of two digits; thereafter the sequence length was increased by one eachtime the subject succeeded twice at a given sequence length. This procedure was adopted until either the subjectfailed twice at a sequence length or succeeded twice at the maximum (9). The span was recorded as the last sequenceat which the subject succeeded twice. The test was then repeated with the exception that the subject was instructed toecho the sequences in reverse order.

(7) Spatial span. This was an analogous test to the digit span test except that the material to be recalled was spatialrather than verbal. An irregular array of rectangles (boxes) was displayed on the screen; these “lit up” (turned red for1 set) in sequence. The subject was required to echo the sequence of boxes by touching them in appropriate order.All other details were as for the digit span test. The test is analogous to the blocks test developed by COW [27].

(8) Verbal free recall This task was based on the Buschke selective reminding test [7]. Subjects were presentedwith 20-word lists on the monitor at a rate of 1 word/3 sec. Using published norms [33], the lists were chosen tocontain equal numbers of high imageabilityihigh frequency, high imageability/low frequency, low imageabilityihighfrequency and low imageability/low frequency words.

After each presentation the subjects were given 90 set to recall as many words as possible. Next the subject wasreminded just of those items which were unrecalled on trial 1. The subject again was required to recall as many itemsas possible from the entire list. Four trials were run using the first list.

Next the subject was given a presentation ofa second (distractor)20_word list, followed by an attempt to recall thesecond list. Finally the subject was asked to recall as many as possible words from the first list. The subjects were toldat the beginning what the pattern of recall/reminding would be.

(9) Spatialfree recall: selecrire reminding This test was devised to be formally similar to the verbal learning test.The lists to be recalled were lists not of words but of pairs of rectangles from the grid used in the spatial span test.Ecight such pairs were presented initially; the subject was then required to make eight unique attempts to recall thepairs. The pairs could be recalled, as in the word learning test, in any order. After the first attempt at recall the subject

was reminded of the unrecalled items. Six trials were run. In this test (unlike the preceding test) there was nodistractor and no final unreminded attempt at recall.

(10) Lirtl e man test This was drawn from the BexleyyMaudsley “BMAPS” battery [l]. The subject was requiredto judge whether a displayed figure was holding an object in it’s right or left hand. The figure was displayed in fourpossible orientations (head up/down, facing toward,‘away).

RESULTS

Data collected throughout were reaction times, errors or scores. These data weretransformed where appropriate and subjected to analysis of variance. The pattern of results issummarized in Table 1. In no case did methylscopolamine produce effects that differedsignificantly from placebo and for clarity the methylscopolamine data will generally not bepresented.

( 1) Simple reaction time: alerting function

As can be seen from Fig. 1, provision of a warning stimulus prior to a target produced afacilitation of reaction time that was optimal at an SOA of approx. 200 msec. This wassupported statistically by a large main effect of SOA [F (5, 95)=47.3, P<O.Ol]. There was,however, no effect of the drug either alone [F (3, 57) = I .88] or in interaction with SOA[F (15, 285) = 1 OO].

(2) Simple/choice reaction time: unsignalled

The subjects’ reaction times in the simple component of this task were considerably fasterthan those in the choice component [F (1, 18)=233.7, PtO.Ol]: see Fig. 2. In this task there

8/12/2019 1-s2.0-0028393288900048-main


690 P. BROKS G. C. PRESTON M. TRALJEI . POPPLETON C. WARD and S. M. STAHL

Table I Summary of results

Test Effect of scopolamine

Simple reaction timeSimple reaction time: unsignalledChoice reaction time: unsignalledVisual selective attentionCovert orientationSustained attentionDigit spanSpatial spanVerbal free recallSpatial free recallLittle man taskProspective memory taskVisual contrast sensitivityVerbal intrusions

No effectlmpaired at 1.2 mg doseImpaired at I 2 mg doseNo effectComplex SOA-dependent effectMonotonic dose-dependent impairmentNo effectNo effectImpaired at 1.2 mg doseNo effectNo effectNo effectMonotonic dose-dependent impairmentMonotonic dose-dependent increase

Placebo Scopolamfne ScoPolamlne SCOpOldRllW03mg 06mg 12mg

M -- t-- e----.

26 12 56

I2 56

1 .

244 ----r- T ~--I I

200 400 800 1600 3200 ’

SOA/ms

FIG. 1. Simple reaction time: alerting function. The effects of scopolamine on the reaction times to a

visual stimulus preceded at varying SOAs by an auditory cue.

was a large main effect of dose [F (3, 54) = 10.36, P< 0.011 but no interaction of dose withcondition [F (3, 54) < 11. This pattern of results was produced by a slowing of reaction timeat the highest dose of scopolamine (1.2 mg) which was independent of the task: both simpleand choice components were affected equally.

(3) Visual selectiue attention

In this test there was no evidence of negative priming (the reaction times in the “negativeprime” condition were no longer than those for the control condition) and so data from the

8/12/2019 1-s2.0-0028393288900048-main


8/12/2019 1-s2.0-0028393288900048-main


692 P. BKOKS, G. C. PRESTON, M. TRAUR, P. POPPLETON, C. WARD and S. M. STAHL

Table 2. Results

Test PlaceboScopolamine

0.3 mgScopolamine

0.6 mgScopolamine

1.2 mg

Selective attention (r.t. in ms)Control 596i 10.1Constant 567& 9.1

Sustained attention (square-root transform)0.25 Hz 1.08 + 0.041.0 Hz 1.08 f 0.043.0 Hz 1.91 kO.12

Digit spanForwards 6.85kO.31Backwards 6.9OiO.30

Spatial spanForwards 5.X0+0.17Backwards 5.35 kO.23

Verbal free recallList learned 5.55kO.71

Spatial free recallList learned 1.70*0.30

Visual contrast sensitivityNon-dominant eye 425i47.8Dominant eye 471 i44.3

594, 9.1568k 8.4

1.12,0&l1.04 & 0.041.95kO.13

602 * 10.6577 f 10.9

1.24 + 0.08I.1 1 kO.081.94kO.13

7.25 +0.256.7OkO.32

5.75kO.205.3OkO.23

7.05+0.386.75 + 0.43

5.75io.175.so+o.17

5.05 f 0.63 5.65kO.74

2.15kO.42

415k41.7435k41.2

2.20*0.41

403 +44.1405 k44.7

597* 8.3577+ 10.3

1.29kO.091.18+0.071.87+0.12

6.8OkO.296.45 kO.33

5.65 + 0.235.10&0.16

3.60i0.44

1.35+0.41

366k31.8344 i 36.0

Placebo Scopolamine SCOpOkllllfle Scopolamine03mg 06mg 12mg

I -r;‘“z’, ,‘“7

500

450 I

V&d lnvalld V&d lnval,d

325 ms SOA 575 ‘lls SOA 1075 ms SOA

FIG. 3. Covert orientatwn. The effects of scopolamine on reaction times to targets preceded by valid(80%) or invalid (20%) cues (means+S.E.).

had no significant effect (mean errors = I. 15 kO.07). This emerged as a linear trend in theanalysis of variance [F(I, 19)=7.07, P<O.O5]. In contrast, at 1 and 3 Hz there were nosignificant effects of dosage [F’s (I, 19) = 1.85 and 0.6 respectively]: at the highest frequencythe drug effect is, if anything, in the opposite direction to that seen at 0.25 Hz.

8/12/2019 1-s2.0-0028393288900048-main



6) Digit span

Performance on this task was not affected by scopolamine (see Table 2). In the forwardcase the average span was 6.8; there were no significant differences between subjects’performance at any of the doses [F (3, 57) = 1.071, nor was there a linear trend across doses

[F(l, 19)=0.19]. In the backward case neither the main effect [F’(3, 57)=0.73] nor thelinear trend of dose [F (1, 19) = 2.531 approached significance.

(7) Spatial span

The pattern of results in the spatial span procedure is analogous to that seen in the verbal,digit span task (see Table 2). There dere no significant effects of dose upon forward spatialspan (all Fs < 1). In the backward case there was a dose effect [F (3, 57) = 3.26, P<O.O5], butthis emerged as a mixture of quadratic and cubic trends [R’s (1, 19) = 4.93,5.06 respectively].This non-systematic effect is not likely to be of interest.

(8/9) Verbal and spati al free r ecall

Performance on these tests can be assessed using a number ofdifferent measures, which aredesigned to reflect either recall from short-term stores, secondary memory, or both [7].

The two indices chosen here are total recalled and “list-learned”. The total number of itemsrecalled on each trial clearly reflects both short- and long-term memory processes. List-learned is calculated as the number of items recalled on a particular trial which are recalled(without reminding) on all subsequent trials; it therefore represents a fairly conservativeestimate of recall from secondary memory.

Verbal : Buschk e sel ecti ve remi ndi ng t ask. The total items recalled on each of the fourlearning trials are shown in Fig. 4. As can be seen, the subjects showed acquisition of theword set. This is supported statistically by a large main effect of trial [F (3, 57)= 117,P<O.OOl] which decomposes into highly significant linear and quadratic trends[E’s (1, 19)=222.8 and 14.3 respectively].

On this measure the main effect of dose approached significance [F (3, 57)=2.69,P=O.O55], but the interaction between the linear trends of dose and trial was highlysignificant [F(l, 19)=9.59, P<O.Ol]. On the “list-learned” measure, which reflectsacquisition into long-term memory, the drug produced a significant impairment (seeTable 2); both the main effect of drug dose [F (3, 57)= 3.51, P~O.051 and the linearcomponent [F (1, 19) = 7.32, P < 0.051 reached significance.

Interestingly, the drug produced an effect on intrusion errors in this task: mean intrusionsfor the placebo, 0.3, 0.6 and 1.2 conditions were 1.295, 1.297, 1.402, 1.524 (square-root

transformed data). This corresponded to a linear dose effect in the analysis of variance[F(l, 19)=4.63, P<O.O5].

Spatial selective reminding. As in the case of the verbal task, the subjects acquired thematerial: there was a large main effect of trial [F (5, 95)=32.7, P~O.051 which consistedmainly of a linear trend [F (1, 19) = 94.8, PC 0.05]. However, in contrast to the verbal casethere was no drug effect on items recalled (the F-ratio for the interaction between linear trendof trial and dose was 0.19: the corresponding ratio for the verbal case was highly significant).Likewise the drug produced no change in spatial long-term memory performance as reflectedin the list-learned measure [F (1, 19) for the linear trend =0.41]. Under these conditionsthen, we could obtain no evidence for a drug effect on spatial memory. Performance on thetwo measures is shown in Fig. 5 and Table 2.

8/12/2019 1-s2.0-0028393288900048-main


694 P. BKOKS, G. C. PRESTON. M. TRAUH. P. POPPLETON. C. WARD and S. M. STAHL

Placebo Scopolamine Scopolamine Scopolamfne03mg 06mg 12mg

- e- t-+ c----+

20

18 I

16i

14 I12

10 1

/

/’/’

.I’

FIG. 4. Verbal free recall. The effects of scopolamine on the total number of words recalled on eachtrial.

Placebo Scopoldmlne Scopolamine Scopolamfne

~-------. A---- A .-__. , 1 ?,03mg 06mg

E 5E

45 -

4 -

35

31- , / I -Ii-

1 2 3 4 5 6

TMi

FIG. 5. Spatial free recall. The effects of scopolamine on the total number of items recalled on eachtrial.

(10) Little man test

In this test the four conditions differed significantly in difficulty [F (3,48)= 19.01,P<O.Ol]. The “head up, facing away” condition produced the fastest reaction times and the“head down, facing away” the slowest. However, despite the comparative complexity of the

8/12/2019 1-s2.0-0028393288900048-main



task, and the ranges of difficulty tested, there were no detectable effects of scopolamine.Neither the main effect of dose [F (3,48) = 0.141; nor the interaction between drug conditionand level of difficulty [F (9, 144) = 0.31 approached significance.

(11) Oth er tests

(a) Prospective memory. There were no significant effects of scopolamine on this measure.The probability of remembering to perform the task was 0.78 under placebo conditions, 0.61at a dose of 0.3 mg scopolamine, 0.89 at 0.6 mg of scopolamine and 0.55 under the highestdose.

(b) Visual contr ast sensit i vit y. The effects of scopolamine on this task are given in Table 2.Five subjects (2 male, 3 female) were left-eye dominant. As can be seen, the drug produced adose-related decline in threshold sensitivity that was independent of the eye being tested(dominant vs non-dominant). These results were confirmed statistically: there was a largemain effect of drug condition [F (3, 51) = 3.81, P C 0.051 that decomposed into a highlysignificant linear component [F (1, 17) = 10.2, P < 0.011 with no significant higher-order

contributions. None of the main effects or interactions involving eye were significant. In thisexperiment all subjects had normal or corrected-to-normal acuity before the drug took effect.We did not, however, have any comparable measures of acuity in the drugged states. In thistask, as in all others, methylscopolamine 0.6 mg p.o. had no significant effect (score averagedacross the two eyes = 436.8 f 35.6).

DISCUSSION

Against the background of the cholinergic hypothesis of Alzheimer’s disease (AD), theprimary objective of this study was to confirm previous findings regarding the differentialeffects of cholinergic blockade on primary and secondary verbal memory and to examinewhether certain features of attention and perception might be affected coextensively. Theopportunity was also taken to study the effects of cholinergic blockade on primary andsecondary spatial memory in a more systematic fashion than has previously been attempted.

In summary, we have been able to demonstrate a profile of effects within the domains ofboth memory and attention (see Table 1). In some cases (e.g. simple alerting) testperformance appears to be resistant to the effects of cholinergic blockade (at least for therange of doses under consideration), whereas the drug has clear effects on the performance ofother tasks. Furthermore, among those tasks which are sensitive to scopolamine we havenoted certain variations in the nature of the dose-response relationship such that some taskswere affected only at the highest dose (e.g. verbal free recall; uncued simple/choice RT) but

others showed linear, dose-dependent effects (sustained attention; visual contrast sensi-tivity). After a more detailed discussion of these results, some general issues concerning thestatus and future development of the “scopolamine model” of dementia will be considered.

With regard to verbal short-term memory, our results confirm the observations of anumber of other investigators in showing that scopolamine causes no impairment [ 14, 18,461. To our knowledge, all previous studies of the effects of scopolamine on such short-termmemory have used aurally-presented verbal material (stimulus items in the presentstudy-digits-were presented visually) and none has examined non-verbal short-termmemory. Experimental studies of both normal subjects and neurological patients indicatethat verbal and visual temporary working memory are dissociable functions [4, 131. It wastherefore of some interest to investigate the effects of scopolamine on analogous measures of

8/12/2019 1-s2.0-0028393288900048-main


696 P. BKOKS, . C. PRESTON, . TKAUB. . POPPLETOK, . WAKII and S. M. ST HL

verbal and non-verbal short-term memory within a single study. We have been able todemonstrate that it is not only verbal working memory that is resistant to the effects of thedrug but that performance in the spatial domain seems to be similarly unaffected.

However, turning to long-term, or secondary, memory there is the suggestion that

scopolamine may produce a verbal/non-verbal dissociation. Again, there is firm evidencethat such processes are neuropsychologically dissociable 1271. By design, the spatial learningtask in the present study bore a formal resemblance to the verbal selective reminding task.Whereas verbal learning was impaired (in line with evidence from previous studies) statisticalanalysis indicated no impairment of spatial learning. It is impossible to conclude from thepresent data whether this dissociation between the effects on the spatial and verbal memorytasks reflects differential involvement of the cholinergic system in linguistic and visuo-spatialfunctions or is merely attributable to unequal sensitivities of these particular tests.(Although, in this connection, it may be relevant to note that the “little man” test, whichplaces heavy demands on visual-spatial function, was insensitive to scopolamine). Certainly,other investigators have observed impairments of spatial memory at higher doses of

scopolamine. For example, LILJEQUIST and MATTILA [26] reported that an intravenous doseof 6 /lg/kg impaired the ability of chess players to recall play positions and CAINE et d. [S],used pattern recall as an intervening distractor task in a verbal fluency test and noted that0.8 mg of scopolamine (intramuscular) caused significant impairment. In pilot studies usingdoses roughly comparable to those employed in the above studies (e.g. 0.4 mg subcutaneous)we have also observed impaired performance on our spatial learning task, but it might bementioned here that sedation is a concomitant effect of such high doses. We have preliminarydata from a second study, involving 18 healthy volunteers, which suggest that an oral dose ofI.2 mg of scopolamine does not cause significant sedation as assessed by scores on the BOND

and LAIER [S] visual analogue subjective rating scales (Preston et al., in preparation). It maybe that. in cases where scopolamine is seen to affect nonverbal memory, sedation is asignificant contributory factor.

Finally. it is worth noting that, in the verbal learning task, scopolamine had the effect ofincreasing the number of non-list intrusions as well as reducing the number of items correctlyrecalled. Such a pattern is consistent with that seen in the performance of AD patients [ 171.

In the realm of attention, we have been able to demonstrate a profile of effects in whichvarious tasks showed different susceptibilities to scopolamine. For example, “vigilance”, inthe sense of preparedness to respond to a single anticipated stimulus, was relativelyimpervious to cholinergic blockade (Table 2). By contrast. “vigilance”, in the sense of thecapacity to sustain attention to a series of target stimuli over a few minutes was vulnerableover the whole dosage range. This impairment of sustained attention is consistent with the

evidence of some previous studies which have adopted more time-consuming continuousperformance or signal detection procedures [34, 51. 521. It is worth emphasizing that thedrug effect here is observed only for the enumeration of stimulus trains presented at thelowest frequency (0.25 Hz). The result cannot therefore be attributed to a disruption of thecapacity to form numerosity judgements as such, nor to an impairment of auditoryperception. Rather, it seems there is a more specific interference with the ability to imposevoluntary attention over the relatively extended periods required for the slow rate of

presentation. As WKKINS c’t rrl. 1541 have noted, traditional vigilance tasks, in which subjectsare required to report infrequent target events at low levels of stimulation, are both lengthyand inefficient (in that for a lapse of attention to be observed, any such failure must coincidewith the rare occurrence of a signal). They are thus generally unsuitable for clinical studies.

8/12/2019 1-s2.0-0028393288900048-main



The present method, being both brief and efficient (any momentary lapse of attention is liableto be detected) might well prove to be useful as a clinical test. In light of the observed sensitivityof this test to experimental cholinergic blockade, it is tempting to speculate that errors on a testof this kind might prove to be an early indicator of disease-related cholinergic depletion.

Although there has been some evidence to suggest that scopolamine may affect auditoryselective attention there was no evidence of any drug effect in our test of visual selectiveattention. In the dichotic-listening study reported by DUNNE and HARTLEY 15] a drug-induced disturbance of selective attention is inferred from the performance of a task whichplaces significant demands on memory (verbal free recall). There, it may be the interaction ofattentional and mnestic processes that is crucial to the observed effect. In the present casethere were no demands on memory. Future studies might explore the effects of scopolamineon visual and auditory selective attention by comparing analogous tasks with and without amemory component.

The pattern of results emerging from the covert orientation task was complex and difficultto interpret. There is the suggestion from the three-way interaction between the factors Dose,SOA and “Validity”, that covert orientation (as inferred from the “validity effect”) wasdisrupted for longer SOAs at the highest dose of scopolamine. The possibility of floor/ceilingeffects presents fundamental difficulties for interpretation of this observation.

In contrast to the cued reaction time task, the simple RT component of the simple/choiceRT test was significantly affected at the highest dose. An obvious difference between this testand the “alerting task” (which was not affected) is, of course, that stimuli in the alerting taskwere signalled whereas no preparatory cues were provided in the simple/choice RT task. Onepossible explanation for the difference, therefore, is that cholinergic blockade induces a kindof response inertia, in the form of a difficulty in initiating unprepared responses, withoutinterfering with the capacity to prepare (and subsequently execute) a response given

appropriate forewarning.Another difference between these two tests is in the mode of responding; the alerting taskrequires fine finger movement whereas the simple/choice RT task requires more grossmovement of the hand and arm. The drug effect discrepancy might therefore be traced todifferences in the motor demands of the two tasks. We are currently investigating thesealternatives. Whatever the reason for the drug effect on the simple/choice RT task, it isinteresting to note that the simple and choice components were not differentially affected. Adecline in choice RTperformance with age is well documented, ifdifficult to interpret [43]. Inthis respect at least, our data would suggest that cholinergic blockade does not provide amodel of normal aging.

Our test of visual contrast sensitivity was incorporated into the battery as a means of

sampling an aspect of visual function known to decline with age [47]. It provided one of themore sensitive measures of anticholinergic influence, showing a linear, dose-dependentresponse similar to that observed for the sustained attention task. This finding has beenconfirmed in a subsequent study in which visual acuity was measured and remainedunchanged (Preston et al., in preparation), and thus the dose related effects on contrastsensitivity may not be attributed to changes in refraction. The evidence on visual contrastsensitivity in AD is somewhat equivocal. SCHLOTTERER t al. [47] report that visual contrastsensitivity may not be disproportionately affected in AD patients compared with age-matched controls. However, NISSEN t al. 1291 found their AD patients to have significantlyelevated contrast sensitivity thresholds compared with controls. It may be, as NISSEN et al.contend, that procedural differences might account for the discrepancy between these two

8/12/2019 1-s2.0-0028393288900048-main


69 P. ROKS,G. . PRESTON, . TRAUB, . OPPLETON,~. ARD and S. . SrAHr

studies. The procedure adopted by SCHLOTTERER and his colleagues required subjects todiscriminate the spatial frequency of a grating once it had been detected (“wide”, “medium”,“fine”) whereas NISSEN et al. simply required their subjects to judge the presence or absence ofa grating. Since, in the present study, spatial frequency was held constant at four cycles per

degree, our procedure may be more akin to that of NIssEN and her colleagues. An impairmentof contrast sensitivity in AD would be consistent with recent evidence from post-mortemstudies that the disease may cause widespread axonal degeneration in the optic nerves [20].An implication of this finding is that visual function tests may ultimately have a valuable roleto play in diagnostic screening and differential diagnosis. In this connection, it is of interest tonote that, from the evidence of the present study, visual contrast sensitivity provides anespecially sensitive index of cholinergic depletion.

As one of the cornerstones of the cholinergic hypothesis of AD, “scopolamine dementia”represents a useful model for preliminary evaluation of putative therapies. Any drug capableof reversing the effects of scopolamine safely and effectively would merit consideration as acandidate for the treatment of AD.

In its crudest conception, the scopolamine model of dementia need not even refer directlyto cognitive function. Any convenient index sensitive to cholinergic manipulation mightprovide a measure of the efficiency of scopolamine reversal. Intuitively, at least, it seems quitelikely that a drug capable of reversing non-cognitive central effects of cholinergic blockade,such as the sedative effects observed at high doses, might also have some success in restoringcognitive function. There are, however, potentially important advantages in seeking to refinethe model by adopting an approach which gives due weight to analyses of psychologicalfunction. In particular, there is the question of separating specific from non-specific effects ofcholinergic blockade.

The notion of specificity here has two, related, aspects: dose specificity and pharmacologi-cal specificity. It is important to determine, for a given dose, which abilities are impaired andwhich are spared. Those abilities which are vulnerable to relatively low doses of the drugmight be considered to provide the most sensitive test parameters in the evaluation of drugshypothesized to reverse the effects of cholinergic blockade. One can expect the relationshipbetween dose and range of effects to be such that the higher the dose the greater the range ofprocesses affected. Certainly, our own experience in pilot studies is that at higher effectivedoses (e.g. 0.4 mg subcutaneous) subjects become sedated and, probably because of this, amuch wider range of functions are impaired. However, since high doses of other centrally-acting drugs, such as the benzodiazepines, are likely to produce similarly widespreadimpairments, the pattern of dysfunction observed with higher doses of scopolamine becomesdifficult to interpret in terms of effects specific to cholinergic deficit. In order to determine

cholinergic specificity at lower doses it therefore becomes necessary to compare the effects ofsuch low doses of scopolamine with comparably low doses of other drugs. Referring again tothe benzodiazepines, attempts to distinguish the amnestic effect of such drugs from that ofscopolamine have so far met with little success [ 161. As long as the cholinergic hypothesis ofAD retains credence, the isolation of specifically “cholinergic” aspects of cognitive functionwill remain a worthwhile goal. Significant progress will depend on the application ofcomprehensive and analytical psychological test batteries in multiple-dose studies of a rangeof centrally-active drugs.

In conclusion, the results of the present study lead us to view the various tasks underconsideration in terms of their relative vulnerabilities to cholinergic blockade. It would be ofconsiderable interest to know whether the patterns of relative vulnerability suggested by the

8/12/2019 1-s2.0-0028393288900048-main



present study find any correspondence in the natural history of cognitive decline in clinicaldementia. On the basis of the present evidence one might hypothesize that subtle, andperhaps therefore unreported, impairments of attention and visual function could antedatethe disturbances of memory which most commonly present as the first sign of dementia. We

propose to examine this question in future studies.

A~knowledyement~We would like to thank Dr Bill Schofield and Dr Arnold Wilkins for invaluable advice anddiscussion, Jean Patman, Kathleen Wets for help in running the study and Jacqueline Tredinnick who typed themanuscript. Thanks also to the anonymous reviewers who took the trouble to make such constructive comments.

1.

2.3.

4.5.

6.

I.

8.

9.

IO.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21

22.

23.

24.

REFERENCESACKER, W. and ACKER, C. Bexley Maudsley Automated Psychological Screening and Bexley MaudsleyCategory Sorting Test. NFER-Nelson, Windsor, 1982.ALEXANDER, D. A. Attention dysfunction in senile dementia. Psycho/. Rep. 32, 229-230, 1973.ALLPORT, D. A., TIPPER, S. P.- and CHMIEL, N. R. Perceptual integration and post-categorical filtering. InMechanisms of’iltrention: Attention and Performance XI, M. I. POSNER and 0. S. MARIN (Editors). Lawrence

Erlbaum, Hilisdale N.J., 1985.BAUDELEY, A. D. Working memory. Phi/. Trans. R. Sot. Land. B 302, 31 l-324, 1983.BOND A. and LADER, M. The use of analogue scales in rating subjective feelings. Br. J. med. Psycho/. 47,211 ~218, 1974.BOWEN, D. M., SMITH, C. B., WHITE, P. and DAVISON, A. N. Neurotransmitter-related enzymes and indices ofhypoxia in senile dementia and other abiotrophies. Brain 99, 459- 496, 1976.BUSCHKE, H. Selective reminding for analysis of memory and learning. J. Verb. Learn. Verb. Behac. 12,543-550.1973.CAINE. E. D., WEINGARTNER, H., LCDLOW, C. L., CUDAHY. E. A. and WEHRY, S. Qualitative analysis ofscopolamine-induced amnesia. Psychopharmacology 74, 74-80, 1981.COLLER~N. D. Cholinergic function and intellectual decline in Alzheimer’s disease. Neurosciencr 19, 1 -28,1986.CROW T. J. and GROVE-WHITE, I. G. An analysis of the learning deficit following hyoscine administration toman. Br. J. Pharwm. 49, 322- 327, 1973.DAVIES, P. Biochemical changes in Alzheimer’s disease-senile dementia: neurotransmitters in senile dementiaof the Alzheimer’s tvoe. In Conaenital and Acauircd Cognitive Disorders, KATZMAN. R. (Editor), pp. 153-166.Raven Press, New York, 1979.”DAVIES, P. and MALONEY, A. J. Selective loss of central cholinergic neurones in Alzheimer’s disease. Lancrt ii1403, 1976.DE RENZI, E. and NICHELLI. P. Verbal and non-verbal short-term memory impairment following hemisphericdamage. Cortex 11, 341~-354, 1975.DRACHMAN, D. A. and LEAVITT, J. Human memory and the cholinergic system. Arch.7 Neural. 30, 113 121.1974.DUNNE, M. P. and HARTLEY, L. R. The effects of scopolamine upon verbal memory: evidence for an attentionalhypothesis. Actu Psycho/. 58 (3), 205 217, 1985.FRITH, C. D., RICHARDSON, J. T., SAMUEL, M., CROW, T. J. and MCKENNA, P. J. The effects of intravenousdiazepam and hyoscine upon human memory. Q, J. exp. Psychol. 36A, 133- 144, 1984.FULD. P. A., KATZMAN, R., DAVIES, P. and TERRY, R. D. Intrusions as a sign of Alzheimer dementia: chemicaland pathological verification. Ann. Neural. 11, 155-159. 1982.GHONEIM, M. M. and MEWALDT, S. P. Effects of diazepam and scopolamine on storage, retrieval andorganizational processes in memory. Psychopharmacologia 44, 257-262, 1975.GHONEIM, M. M. and MEWALDT. S. P. Studies on human memory: the interactions of diazepam, scopolamineand physostigmine. Ps~chopharmucolog~ 52, l--6, 1977.HINTON. D. R., SAIXJN, A., BLANKS, J. and MILLER, C. Optic-nerve degeneration in Alzheimer’s disease. New.Eng. J. Med. 315,485487, 1986.HOLLANDEK, E., MOHS R. C. and DAVIS, K. L. Cholinergic approaches to the treatment of Alzheimer’s disease.Br. Med. Bull. 42, 97-100, 1986.HUPPERT. F. A. and TYM, E. Clinical and neuropsychological assessment of dementia. Br. Med. Bull. 42, 1 l-18,1986.HYMAYX,B. T., VAN HOESEN G. W., DAMASIO, A. R. and BARNES, C. L. Alrheimer’s disease: cell-specificpathology isolates the hippocampal formation. Science 225, 1168&l 170. 1984.KOPELMAN, M. D. Multiple memory deficits in Alzheimer-type dementia: implications for pharmacotherapy.P.Tycho/. Med. 15, 527 541, 1985.

8/12/2019 1-s2.0-0028393288900048-main


25. KOPELMAN, M. D. The cholinergic neurotransmitter system in human memory and dementia: a review. Q. J.cup. Psychof. 38A, 535-573, 1986.

26. LILJEQUIST, R. and MATTILA, M. J. Effect of physostlgmine and scopolamine on the memory functions of chessplayers. Med. Biol. 57, 402405, 1979.

27. MILNEK, B. Interhemispheric differences in the localization of psychological processes in man. Br. Med. Bull. 27,272~277,1911.

28. MORRIS, R. G. and KOPELMAN, M. D. The memory deficits of Alrheimer-type dementia: a review. Q. J. rup.f’sychol. 38A, 575-602, 1986.

29. NISSEN, M. J., CORKI K S. BUONANNO F. S. GROWDON J . H., WKAY, S. H. and BAUEK, J. Spatial vision inAlzhcimer’s disease. Archs Nrurol. 42, 667-671, 1985.

30. OVER B. A., Koss, E., FRIEDLAND . P. and DELIS, D. C. Processes ofverbal memory failure in Alzhcimcr-typedementia. Brain Coynit. 4, 90~-103, 1985.

31. O’CONNOK, K. P. Slow potential correlates of attention dysfunction in senile dementia: I. Biol. Ps~c’hol. II,193-202, 1980.

32. PANIXT, S. K. and DUNUEE, J. W. Pre-operative amnesia. Anuesthesia 25, 4931199, 1970.33. PAIVIO, A., YUILL~.. J. C. and MADIGAN, S. A. Concreteness, imagery, and meaningfulness values for 925 nouns.

_. exp. Psqchol. Mon. Suppl. 76, No. 1 Part 2, I-25, 196X.34. PARROTT, A. C. The effects of transdermal scopolamine and four dose levels of oral scopolamine (0.15,0.3.0.6

and 1.2 mg) upon psychological performance: Psychopharmucnloy~ 89, 347-~354, 1986’.35. PERRY, R. H. Recent advances in neuropathology. Br. Med. Bull. 42, 34 41, 1986.

36. PERKY, E. K., Gmso?r, P. H.. BLESSED, G., PERKY, R. H. and TOMLI?ISON, B. E. Neurotransmitter enzymeabnormalities in senile dementia. J. Neural. Sci. 34, 247- 265, 1977.

31. PERRY, E. K., PERKY, R. H., BLESSED, G. and TOMLINSON, B. E. Necropsy evidence ofcentral cholinergic deficitsin senile dementia. Luncet i, 189, 1977.

38. PERKY, E. K., TOMLINSON, B. E., BLESSED, G., BERGMANN K.. GIBSON P. H. and PEKRY, R. H. Correlation ofcholinergic abnormalities with senile plaques and mental test scores in senile dementia. Br. Med. J. ii,1457-1459,197x.

39. PETEKSEP~, . C. Scopolamine induced learnmg failures in man. Psy~hophurmuco~oy : 52, 283 289, 1977.40. POSNER, M. I. and BOIES. S. J. Components of attention. Psvchol. Rev. 78, 391 408, 1971.41. POSNER, M. I., COHEN, Y. and RAFA;, R. D. Neural systems-control of spatial orienting. Phi/. Truns. R. SK.

Lond. B 298, 187 198, 1982.42. POSNER. M. I.. WALKER, J. A., FKIEDKICH, F. J. and RAFAL, R. D. Eflects of parietal injury on covert orienting of

attention. J. Nruroscr. 4 (7), 1863-1874, 1984.43. RABBIT-~, P. M. A. A fresh look at changes in reaction times in old age. In Thu Psqchohioloyy of’Ayinq: Prohlem,~

and Perspectirrs, STEIN (Editor). Elsevier. Amsterdam, 1980.44. ROSSOR. M. N. Neurotransmittcrs and CNS disease-dementia. Lancet ii, 1200-1204, 1982.45. ROSSOR, M. N. and IV~RSEN, L. L. Non-cholinergic neurotransmitter abnormalities in Alrheimer’s dwzase. Br.

Med. Bull. 42 (1). 70 74, 1986.46. SAFER D. J. and ALLEN. R. P. The central effects of scopolamine in man. Biol. P.$~c’hrarry 3, 347 355, 1971.47. SCHLOTTTKLR, G.. M~S~OW~CH M. and CRAPI’LK-M~LA(.HLAK. D. Visual processing deficits as assessed by

700 P. BROKS, G. C. PRESTON M. TKAUB P. POPPLETOX C. WAKI~ and S. M. STAHL

48.

49.

50.

51.

52.

53.

54.

55.

spatial frequency contrast sensitivity and backward masking in normal agcing and Alrhcimcr’s disease. &tin107, 309-325. 1983.SITAKAM, N., WEINC~AKTN~R, H. and GILLIN. J. C. Human serial learning: enhancement with arechollnc andcholine and impairment with scopolamine. Scirncr 201, 274276, 1978.TECCE, J. J., CATTANACH. L.. BOEHNLK-DAVIS, M. B.. BRAN(‘ONNIFK, R. J. and COLL, J. 0. Etudeneuropsychologique de la haisse de l’attention et traitement mt-dicamenteux chcz les malades atteints dcmaladic d’Alrhcimer. La fresse MCdicule 12, No. 48, 3155. 1983.VITALIANO. P.. BKEEN. A. R.. ALHEKT. M. S., Russo, J. and PKI~Z, P. N. Memory, attention and functional

status in community-residing Alzheimer type dementia patients and optimally healthy aged individuals. .1.Gwmtol. 39 (I ). 58 64. 1984.WESNES. K. and WAKBCKTON, D. M. Effects ofscopolamine on stimulus sensitivity and response bias in a wsualvigdancc task. ,Yrrr,.“~~?.c.hohio/ocl~ 9, I54 157. 1983.Wcsuts. K. and WAKBUKTOY. D. M. E&c& of scopolamine and nicotine on human rapid informationprocessing performance. Ps ~chophu~macol~~~~~2, 147- 150. 1984.WILKIXS. A. J. and ROBSON. J. G. C’cc&widyr Low Contrast Gratings. Clement Clarke lntcrnational Ltd..London.WILKINS. A. J.. SIJALLI(.L. T. and M~.CAKTHY. R. Frontal lesions and sustained attention. Vr,uropc~c,holoyio 25,359 365. 1987.WIL.LIAMS, M. Studies of perception m senile dementia. Br. J. Rfrd. Pswhol. 29, 270 2X7, 1956.

1-s2.0-0028393288900048-main

Documents