Remediation of sustained attention following traumatic brain injury: vigilance task training and the generalization of its effects

REMEDIATION OF SUSTAINED ATTENTION FOLLOWING TRAUM ATIC BR AIN INJURY: V igilance Task Training and the G eneralization o f its

E ffec ts b y

Jon Jay Van Doren

B .S ., U niversity o f Florida, 198 0 M .S c., U niversity o f V ictoria, 1982

A D issertation Subm itted in Partial F u lfillm en t o f the R equirem ents for the D egree o f

' ; P T i ' ? i DOCTOR OF PHILOSOPHY .rV.P L. ' '.'b.i-S

in the Departm ent o f P sy c h o lo g y

’^r \T"X j / Of an W e accept this thesis as conform ing

[ h ' 1' ■e),' to the required standard

Dr. Bram G oldw ater, Supervisor (D epartm ent o f P sy c h o lo g y )

--- ---y— --- ■---1=— ---T---- ---Dr. lLorbn A cker, D epartm ental M em ber (D epartm ent o f P sy c h o lo g y )

Dr. S p ell^ cv . D ^ b a r tm e n ta l/^ e m b e r (D epartm ent o f P sy c h o lo g y )

'dJ . /J o se p h Parsons, O utside M em ber (C o u n sellin g S erv ices)

Dr. R obert K ohlenberg, External E xam iner (D epartm ent o f P sy c h o lo g y , U n iv ersity o f W ashington )

© JON JAY V A N DOREN, 1991 U n iv ersity o f V ictoria

All rig h ts reserv ed . D issertation m ay not be reproduced in w h o le o r in p art, by p h o to co p y in g or other m eans, w ithout the perm ission o f th e author.

11

Supervisor:

Dr.

Bram

G ol dw ate r

A B S T R A C T

S tu d ie s o f su s ta in e d -a tte n tio n retra in in g f o ll o w i n g brain in ju ry are r e v ie w e d , and fo u n d to h a v e p ro d u ced in c o n c lu s iv e r e s u lts . T h e reason fo r th is, it is s u g g e s te d , is that a standard o p e r a tio n a l a n a ly s is o f a tte n tio n has not b een a p p lied , a s e v id e n c e d by c o n s id e r a b le in c o n s is t e n c y in th e d e p e n d e n t m e a s u r e s and tr e a tm e n t m e th o d s u sed fr o m stu d y to stu d y . T h e p r e se n t stu d y a d d r e s s e s th is c o n c e r n by a p p ly in g w e ll e s ta b lish e d p r in c ip le s o f o p era n t c o n d it io n in g to th e a n a ly s is and r e m e d ia tio n o f a tte n tio n d e f ic its . A fte r b r ie fly

r e v ie w in g the v a r ie ty o f ta sk p a ra m eters in th e a tte n tio n ite r a tu re , n o tin g a m b ig u itie s in h eren t in th e v a r io u s c o n c e p t u a liz a t io n s o f a tte n tio n , it is d e c id ed to train v ig ila n c e task p e r fo r m a n c e , a

r e la t iv e ly u n a m b ig u o u s and u n c o n tr o v e r s ia l o p e r a tio n a l d e f in it io n o f s u sta in e d a tten tio n . B o th th e p r in c ip le o f im m e d ia c y o f r e in ­

fo r c e m e n t (fe e d b a c k o f c o r re c t :nd in c o rr e c t on e a c h tria l) and

sh a p in g (g ra d u a l in c re a se o f sp e e d d e m a n d s c o n t in g e n t o n in c r e a s e d p e r fo r m a n c e a c c u r a cy ) are e m p lo y e d . T h e is s u e o f g e n e r a liz a t io n is d e e m e d cen tra l to co n c e rn s o f treatm en t e f f ic a c y , a n d is e x p lo r e d b y a d m in istra tio n o f alternate v e r sio n s o f th e s a m e b a s ic v ig ila n c e ta sk . R e s u lts s h o w that tra in in g w ith im m e d ia te r e in f o r c e m e n t a n d s p e e d - s h a p in g p r o d u c e d better a c q u is itio n o f th e train ed ta sk th an d e la y e d f e e d b a c k and in v a r ia n t s p e e d o f s tim u lu s p r e s e n ta tio n . F u r th e r m o r e , g a in s r e su ltin g fro m tra in in g w e r e e s s e n t ia lly lim ite d to th e ta sk o n

R e m e d ia tio n o l S u sta in ed A tten tion : A b stract_________________________________________ Li] w h ic h tra in in g w a s c o n d u c te d , w ith little e v id e n c e for g e n e r a liza tio n to lik e ta sk s e m p lo y in g d iffe r e n t stim u li. T h e s e r e su lts are d isc u sse d in term s o f the a p p lic a b ility o f th e c o n str u c t o f su sta in e d attention to h ea d in ju ry r e h a b ilita tio n .

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : A b s t r a c t E x a m in e r s : Dr. B ra m O o ld w a te r , S u p er v iso r (D e p a r tm e n t o f P s y c h o lo g y ) D r. L o r e n A c k e r , D ep a rtm en ta l M em b er (D e p a r tm e n t p f P s y c h o lo g y ) D r / F ran k S p e l l a ^ , D ep a rtm en /a l M em b er (D e p a r tm e n t o f P s y c h o lo g y ) D r. J o se p h P a r so n s, O u tsid e M em b er ( C o u n s e llin g S e r v ic e s ) Dr. ^R obert K o h le n b e r g , E xtern al E x a m in e r (D e p t, o f P s y c h o lo g y , U n iv . o f W a sh in g to n )

ABSTRACT

CONTENTS

D efin ing A tten tio n ...2

Focused attention ... 3

Sustained a tten tio n ... ...4

S e lec tiv e attention... 4

Alternating attention... 5

Divided attention...5

Narrowing the Field o f Study...8

Rem ediation o f Sustained A ttention...10 Prevalence o f Attention D e fic its... 1 0 Importance o f Attention to Rehabilitation... 1 2 Evaluating Efficacy o f Attention R em ediation...1 3 R eview o f Attention Retraining Literature 13 Factors U nderlying Inconclu sive R esu lts... 2 4 O perationally D efin ing Sustained A tten tion... 3 2 SECTION 2: METHOD O b je c tiv e ...3 9 D e sig n ...3 9 S u b je c ts ...4 1 Tasks...4 1 D ependent M e a su re s... 4 4 Apparatus...4 6 P r o c e d u r e ... 4 7 SECTION 3: RESULTS and DISCUSSION

Subject 1 P r o c e d u r e ... 5 1 Results...5 1 Training T ask ...5 1 FCAT...5 2 AC AT... 5 3 SCAT... ... 5 5 Conclusions...5 6 Skill A cquisition

Generalization Across Tasks

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : A b s t r a c t

Learning o f the trained ta sk ...5 7 Pattern o f improvement on untrained tasks... 5 7 Subject 2 P r o c e d u r e ... 5 8 Results... 5 9 Training T ask...5 9 FCAT... 5 9 AC AT... 6 1 SCAT... 6 3 Conclusions...6 4 Skill Acquisition... 6 4 Immediate feedback...6 4 Speed shaping... 6 4 Generalization Across T asks... 6 4 D iscrim ination training... 6 4 Speed training... 6 5 Specific vs. N onspecific Treatment E ffects 6^ Learning o f the trained ta sk ... 6 5 Pattern o f improvem ent on untrained tasks... 6 5 Subject 3 P r o c e d u r e ...6 5 Results... 6 7 Training T ask...6 7 FCAT...6 7 AC AT... 6 9 SCAT...7 0 Conclusions...7 2 Skill Acquisition... 7 2 D iscrim ination training... 7 2 Speed T raining... 7 2 Generalization Across T asks... 7 2 D iscrim ination training... 7 2 Speed Training...7 3 Specific vs. N onspecific Treatment E ffec ts 7 3

Learning o f the trained task... 7 3 Pattern o f improvem ent on untrained tasks...7 3 Subject 4 P r o c e d u r e ...7 3 Results... 7 4 Training T ask... 7 4 FCAT... 7 5 AC AT...7 6 SCAT... 7 8

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : A b s t r a c t ___________ v j

Conclusions... 8 0 Skill Acqu-sition... 8 0 Generalization Across T a sk s.. ... 8 0 Specific vs. N onspecific Treatment E ffects 8 0 Discussion o f Results, Subjects 1 - 4 ...8 0

Skill Acquisition...8 1 D iscrim ination Training...8 1 Speed T rain in g ... 8 2 Generalization Across T asks...8 3 D iscrim ination Training ... 8 3 Speed T rain in g...8 4 Specific vs. N onspecific Treatment E ffects...8 4

N on sp ecific Effects Producing Apparent

G en er a liz a tio n ... ... 8 4 N on sp ecific Effects Suppressing Generalization 8 6 General Conclusions, Subjects 1 - 4 ... 8 7

Skill Acquisitior...8 7 G eneralization Across T asks... 8 7 Possible Contribution of Nonspecific Factors 8 7 Subject 5 P r o c e d u r e ... ... ... 8 8 Results...8 9 Training Task... ...8 9 SCAT... ... 8 9 Conclusions...9 1 Skill Acquisition... 9 1 G eneralization Across T asks... 9 1 Specific vs. N onspecific Treatment E ffects 9 1 Subject 6 P r o c e d u r e ... 9 2 Results...9 2 Training T ask ...9 2 SCAT... 9 2 Conclusions...9 4 Subject 7 P r o c e d u r e ... 9 4 Results...9 5 SCAT... 9 5 Conclusions...9 7 D iscussion o f Results, Subjects 5 - 7 ... 9 7 Skill Acquisition...9 7 G eneralization Across T asks... 9 7 Specific vs. N onspecific Treatment E ffects... 9 8 General Conclusions, Subjects 5 - 7 ... 9 8 Subject 8

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : A b s t r a c t ___________________________________________ v i i i Results...9 9 Training Task... 9 9 F C /A ... 1 0 0 ACAT...1 0 1 SCAT... 1 0 2 Conclusions... 1 0 4 Skill Acquisition...1 0 4 Generalization Across T asks... 1 0 4 Subject 9 P r o c e d u r e ...1 0 4 Results... 10 5 Training T ask... 1 0 5 FCAT... 1 0 5 ACAT...1 0 7 SCAT...1 0 8 Conclusions... 1 1 0 Skill Acquisition...1 1 0 G eneralization Across T asks... 1 1 0 SECTION <. GENERAL DISCUSSION

Baseline: Practice Effects vs. Spontaneous R ecovery...1 1 2 Treatment Effects...1 1 3 Generalization o f Treatment... 1 1 5 Functional Utility of Training...1 1 5 Im plications for the Construct o f Sustained A ttention 1 1 7 Training G en eralization... ...1 2 0 Mapping Generalization...1 2 1 Current Results in Light o f Previous Findings... 1 2 2 S u m m a r y ...1 2 3 BIBLIOGRAPHY... 1 2 5 APPENDICES

Dependent M easures in Attention Retraining Studies...1 3 2 Training Tasks Used in Attention Retraining S tu d ies... 1 3 7 SCAT Pilot Data... 1 4 2 Subject Instructions... 1 4 4

LIST OF TABLES

Table 1.1: Operational Definitions o f Attention Subtypes... 0 Table 1.2: Summary o f Attention Retraining Literature... 11 Table 2.1: Subject Profiles ... 4 0 Table 3.1: Current Subjects vs Normative Control Group... 47 Table B .l: Physical Characteristics o f FCAT Stim uli... 114

X

LIST OF FIGURES

3 . 1 . 0 : Subject l ’s Shortest ISI Achieved in T raining 5 1 3 .1 .1 : Subject l ’s Mean FCAT Reaction Time 5 2 3 .1 .2 : Subject 1'.' Standard Deviation o f FCAT Reaction Tim e 5 2

3 .1 .3 : Subject l ’s FCAT Errors 5 3

3 . 1 .4 : Subject l ’s Mean ACAT Reaction T im e... 5 3 3 . 1 . 5 : Subject l ’s Standard Deviation o f ACAT Reaction T im e... 5 4 3 . 1 .6 : Subject l ’s ACAT Errors 5 5 3 .1 .7 : Subject l ’s Mean SCAT Reaction Tim; 5 5 3 .1 .8 : Subject l ’s Standard Deviation o f SCAT Reaction Tim e... 5 6 3 .1 .9 : Subject l ’s SCATErrors ... .5 6 3 . 2 . 0 : Subject 2 ’s Shortest ISI Achieved in T raining 5 9 3 .2 .1 : Subject 2 ’s Mean FCAT Reaction Time 5 9 3 .2 .2 : Subject 2 ’s Standard Deviation o f FCAT Reaction Tim e 6 0

3 .2 .3 : Subject 2 ’s FCAT Errors 6 0

3 . 2 .4 : Subject 2 ’s Mean ACAT Reaction T im e... ... 6 1 3 . 2 . 5 : Subject 2 ’s Standard Deviation o f ACAT Reaction T im e... 6 1 3 . 2 .6 : Subject 2 ’s ACAT Errors 6 2 3 .2 .7 : Subject 2 ’s Mean SCAT Reaction Time 6 3 3 .2 .8 : Subject 2 ’s Standard Deviation o f SCAT Reaction T im e ... 6 3 3 .2 .9 : Subject 2 ’s SCAT Errors...6 3 3 . 3 . 0 : Subject 3 ’s Shortest ISI Achieved in T raining...6 7 3 .3 .1 : Subject 3 ’s Mean FCAT Reaction Time 6 7 3 .3 .2 : Subject 3’s Standard Deviation of FCAT Reaction Tim e 6 8 3 .3 .3 : Subject 3 ’s FCAT Errors... ... 6 8 3 . 3 .4 : Subject 3 ’s Mean ACAT Reaction T im e...6 9 3 . 3 . 5 : Subject 3 ’s Standard Deviation o f ACAT Reaction T im e... 6 9 3 . 3 .6 : Subject 3 ’s ACAT Errors 7 0 3 .3 .7 : Subject 3’s Mean SCAT Reaction Time 7 0 3 .3 .8 : Subject 3 ’s Standard Deviation o f SCAT Reaction Tim e 7 1

3 .3 .9 : Subject 3 ’s SCAT Errors 7 1

3 . 4 . 0 : Subject 4 ’s Shortest ISI Achieved in T raining 7 4 3 .4 .1 : Subject 4 ’s Mean FCAT Reaction Time 7 5 3 .4 .2 : Subject 4 ’s Standard Deviation o f FCAT Reaction Tim e 7 5 3-4.3: Subject 4 ’s FCAT Errors... ... 7 6 3 . 4 .4 : Subject 4 ’s Mean ACAT Reaction T im e...7 6 3 . 4 . 5 : Subject 4 ’s Standard Deviation o f ACAT Reaction T im e... 7 6 3 . 4 .6 : Subject 4 ’s ACAT Errors 7 7

R g m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : L i s t o f F i g u r e s _____________________ x j

3 .4 .7 : Subject 4 ’s Mean SCAT Reaction Time...7 8 3 .4 .8 : Subject 4 ’s Standard Deviation of SCAT Reaction Time. ... 7 8 3 .4,9: Subject 4 ’s SCAT Errors... ... 7 8 3 . 5 . 0 : Subject 5 ’s Shortest ISI Achieved in Training... 8 9 3 .5 .1 : Subject 5 ’s Mean SCAT Reaction Time...8 9 3 .5 .2 : Subject 5 ’s Standard Deviation o f SCAT Reaction Tim e... 9 0 3 .5 .3 : Subject 5 ’s SCAT Errors... 9 0 3 . 6 . 0 : Subject 6 ’s Shortest ISI Achieved in T raining...9 2 3 .6 .1 : Subject 6 ’s Mean SCAT Reaction Time... 9 2 3 .6 .2 : Subject 6 ’s Standard Deviation o f SCAT Reaction Tim e...9 3 3 .6 .3 : Subject 6 ’s SCAT Errors... 9 3 3 .7 .1 : S'ibject 7 ’s Mean SCAT Reaction Time... 9 5 3 .7 .2 - Subject 7 s Standard Deviation o f SCAT Reaction Tim e... 9 5 3.7.3: Subject7’s SCATErrors...9 6 3 . 8 . 0 : Subject 8 ’s Shortest ISI Achieved in T raining...9 9 3 .8 .1 : Subject 8 ’s Mean FCAT Reaction Time...1 0 0 3 .8 .2 : Subject 8 ’s Standard Deviation o f FCAT Reaction Tim e...1 0 0 3.8.3: Subject 8 ’s FCAT Errors... 1 0 0 3 . 8 .4 : Subject 8 ’s Mean ACAT Reaction T im e... 1 0 1 3 . 8 . 5 : Subject 8 ’s Standard Deviation o f ACAT Reaction T im e...1 0 1 3 . 8 .6 : Subject 8 ’s ACAT Errors ... 1 0 2 3 .8 .7 : Subject 8 ’s Mean SCAT Reaction Time...1 0 2 3 .8 .8 : Subject 8 ’s Standard Deviation o f SCAT Reaction Tim e...1 0 3 3.8 .9 : Subject 8 ’s SCAT Errors... 1 0 3 3 . 9 . 0 : Subject 9 ’s Shortest ISI A chieved in T raining...1 0 5 3 .9 .1 : Subject 9 ’s Mean FCA.T Reaction Time...1 0 5 3 .9 .2 : Subject 9 ’s Standard Deviation o f FCAT Reaction Time...1 0 6 3.9.3: Subject 9 ’s FCAT Errors... 1 0 6 3 . 9 .4 : Subject 9 ’s Mean ACAT Reaction T im e... . 1 0 7 3 . 9 . 5 : Subject 9 ’s Standard Deviation o f ACAT Reaction T im e...1 0 7 3 . 9 .6 : Subject 9 ’s ACAT Errors... 1 0 8 3 .9 .7 : Subject 9 ’s Mean SCAT Reaction Time...1 0 8 3 .9 .8 : Subject 9 ’s Standard Deviation o f SCAT Reaction Tim e...1 0 9 3.9 .9 : Subject 9 ’s SCAT Errors... ... 1 0 9

x i i

ACKNOWLEDGEMENTS

This monogrpnh would never have been com pleted without the support and assistance o f many people. Most especially, I would like to acknowledge the efforts o f the subjects o f this study; they have taught me much more than is conveyed within the confines o f these pages.

Furthermore, I must acknowledge the guidance and nurturance o f my supervisor, Dr. Bram Goldwater. His quiet patience must be boundless, else I surely woulo have found its limit.

Finally, thanks to my wife, Ann Van Doren, for her understanding and patience. To say that her support contributed im m ensely to the

Zftiz

to m r ta tto n

iz

to my lot>ma p a te n ts,

C Ja rte s $ . anir J u tia <E. $ att 3&or$tt.

SECTION 1: INTRODUCTION

R a t i o n a le

Accidental trauma is the leading cause o f death for persons under 45 years o f age, and an estimated 65% o f such injuries involve head trauma (Klauber, Barrett-Conner, Marshall, & Bowers, 1981; N augle, 1990). Fortunately the likelihood o f surviving serious head injury has been increasing. Thanks to improved tim eliness o f em ergency m edical response and evacuation, as w ell as to ' ,clinical advances in acute treatment o f trauma victim s, the chances o f surviving such injury are now thought to be in the range o f 7% - 16% (Naugle, 1990), producing an estim ated incidence o f nonlethal traumatic brain injury o f 168-186 per

100.000 cid zens. With a U.S. census o f about 250 m illion, approximately 4 5 0 .0 0 0 new cases are expected nationally each year, continually adding to the numbers o f this chronically disabled population.

Survivors can be faced with a broad range o f lasting physical, em otional, and cognitive challenges. As their numbers have sw elled there has been mounting activism addressing quality o f life issu es, and an attendant rise in demand for rehabilitative services. The present monograph focu ses on treatment o f one o f the more ubiquitous

difficulties plaguing such patients: disorders o f attention. Such disorders can be considered as lim itations in speed or accuracy under certain conditions, potentially impeding performance o f virtually any activity undertaken under those conditions. As w e w ill see, the fact that attention has been quite broadly (and loosely) defined poses serious challenges to its objective study. Before proceeding, therefore, let us confront the task o f identifying various definitions o f attention for the purpose o f distilling their common essence. Doing so at this point w ill more clearly establish the subject matter o f the present paper and w ill ultim ately permit our

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n ■>

settling on an operational definition, and, thus, a particular behavior as the focu s o f treatment.

Defining

Attention

The diversity with which the term "attention" has been used reflects considerable ambiguity in its definition, ambiguity borne largely o f the m entalistic conceptualizations currently in vogue in the field o f attention research. As w ill be discussed below, this has led to failure to develop a universally accepted operational definition o f attention, and has been esp ecia lly noticeable in studies o f attention retraining. Remedial studies have used apparently arbitrary, and w idely varying, m easures to reflect the construct o f attention. The tacit assumption reflected by these

studies is that "attention" is a broad response class that includes

behaviors addressed by the various treatments as v/ell as by the various dependent m easures em ployed.

H ow ever, there has been an emerging trend toward a more operational approach as efforts to study attention retraining have increased. This is no doubt a natural consequence o f the demands inherent in treatment studies, since these tend to be quite practical in nature and are therefore less tolerant o f am biguity regardinfa the

behaviors being trained and require clearly defined indices o f successful treatment. Unfortunately, despite this trend, the conceptualizations o f attention remain som ewhat am biguous and continue to hamper progress in the field. In reviewing the definitions o f attention proposed by two groups o f influential authors, we w ill see the beginning o f a trend toward a m ore operational approach. From there, we w ill establish an

Re m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n

1

Acknowledging the inconsistency with which the construct o f attention lias been applied, Posner and Rafal (1987) identified three man; contexts in which the term has been used: alertness, selection, and vigilance. By their definition, alertness refers to both tonic and phasic arousal, to "preparedness to respond" which corresponds to overall responsiveness to "any and all signals." Selection, on the other hand, refers to "responsiveness to selected information," to "the processing o f specific information, while tuning out other available signals" (p. 184). Finally, in the words o f the authors, vigilance "refers to the ability to sustain conscious attention over long periods" (p. 186), and might therefore be interpreted as consistent m aintenance o f alertness or selective attention over time.

A similar taxonomy, partly stemming from the work o f Posner and his associates, has been proposed by Sohlberg and Mateer (1 9 8 8 ). The

conceptualization p ro p o s^ by these authors differs from the above in that it includes two additional performance characteristics, and is described in more operational terms. "Based on the experim ental attention literature, clinical observation, and patients' subjective complaints," Sohlberg and Mateer identified five "levels" o f attention: focused, sustained, selective, alternating, and divided.

Focused attention refers to responding "discretely to sp ecific visual, auditory, or tactile stimuli," and is comparable to "alertness" o f Posner and Rafal (1987). Both refer to performing a task consistin g o f individual trials with clear forewarning. A focused attention task could, for

example, require subjects to press a button each time green appears among a series o f displayed colors. Each trial w ould be preceded by a warning, such as a countdown, indicating when the next color w as to be displayed; the interval between trials would not necessarily be

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n 4

predetermined, but could be varied according to subject preparedness. D eficit in focused attention would be exhibited as inconsistent

responding to individual trials, even without the requirement of

sustaining performance over time. Such patients are rarely referred for intensive outpatient services, not necessarily because of low incidence, but perhaps because adm itting physicians and/or third party payors see them as so severely impaired that significant benefit from intensive outpatient therapy is deem ed unlikely.

N ext, sustained attention refers to maintenance o f "a consistent behavioral response during continuous or repetitive activity." Both groups o f authors see sustained attention as persistence on a given task over a prolonged period, with responding to discrim inative stim uli that are presented without warning. Such a task could be ides tical the focused task already described, except that it would consist o f ongoing presentation o f colors, requiring performance over tim e and response to targets presented at varying intervals. Patients deficien t in this respect m ight reliably perform discrete trials but show inconsistent performance w hen many trials are presented in series.

S e le c ti.e attention refers to maintenance o f "a cognitive set which requires activation and inhibition o f responses dependent upon

discrimination o f stimuli." Both groups describe this as performance o f a requisite task in the face o f irrelevant stim uli. Although to som e degree inherent in all tasks, the demand for stim ulus discrimination is more explicit and more difficult in selective attention tasks. In such a task, extraneous stim uli would be adm inistered, such as sim ultaneously playing a tape-recording o f spoken colors v hile the subject performs either o f the tw o previously m entioned tasks. Am ong patients with

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n 1

difficulties in this respect, introduction o f stimuli not relevant to the task at hand leads to exaggerated diminution o f performance.

Alterr ung attention refers to "mental flexibility which allow s for m oving between tasks having different cognitive requirements." This requires shifting from one task to another, and may therefore be described as requiring the subject to respond to discrim inative stimuli indicating a change in contingencies such that formerly irrelevant stimuli becom e relevant, and vice versa. In keeping with the above exam ples, alternating attention could involve the experim enter’s occasionally announcing a shift in target, back and forth from the color green to the spoken word “red.”

Finally, divided attention refers to accurate responding to multiple sim ultaneous tasks, and may therefore be considered performance o f tasks consisting o f multiple contingencies, such as pressing a button when green is seen or when "red" is spo'-en. These two contingencies would be operating sim ultaneously, with no requirement to alternate.

In light o f the variety o f senses in which the term attention has been used, the feat o f deriving an overall definition becomes one o f isolating the factor(s) common to all. To assist in identifying such comm on elem ents, let us consider the operational equivalents o f each aspect o f attention, those aspects o f a task which determine the type o f attention to w hich it corresponds (see Table 1.1).

N ote that each attention subtype is defined not by performance o f a specific task, but refers to characteristics o f a potentially broad range o f tasks. A focused attention task, for example, has only to consist o f discrete and forewarned trials. The nature o f the stim uli provided or responses required need not be specified. Furthermore, by varying the task characteristics described above one could develop m ultiple versions

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n _________________________________________ &

o f what would otherwise oe the same task, producing a focused attention version, a sustained attention version, and so on.

Table 1.1

Operational D efinitions o f Attention Subtypes

-SUBTYPE-

-DEFINING

CHARACTERISTICS-F ocused warnings provided, individual trials S u sta in ed no w arnings, m ultiple trials

S e le c tiv e presentation o f irrelevant stim uli A lter n a tin g m ultiple tasks perform ed co n se cu tiv e ly D iv id ed m ultip le tasks perform ed concurrently

Clearly, attention is not presently view ed as a m onolithic construct; rather, various subtypes o f attention have been em erging in the

literature. In general, however, it seems that the basic notion underlying them all is one o f consistency o f task performance across conditions. That is, a patient would be described as suffering an attention deficit only if performance o f a given task was adequate under some conditions and inadequate under others; were performance co n sisten tly inadequate across conditions, the difficulty would not be considered one o f attention but m ight instead be related to basic sensory or motor impairment. Therefore when inferring either deficits or improvem ents in attention, one must demonstrate that elem ents o f the task other than those listed above a e not determining performance, being sure to attribute deficits to the degree o f control exerted by discrim inative stim uli under certain conditions rather than to w holly ineffective discrim inative stim uli or lack o f available reinforcers. An extreme exam ple o f such confounding would be to infer focused attention deficit in a deaf subject who has been asked

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c i i o n ₁

to perforin a task involving auditory warning stim uli. A similar mistake would be to infer improved focused attention in a non-Anglophone subject who had, over time, learned to distinguish the warning stimulus "get ready" from the stimulus "not yet."

The process o f defining subtypes o f attention, it seem s, is sim ply one of identifying those conditions that are associated with variations in "strength" o f the stim ulus-response relationship. Strength o f stim ulus control is reflected in either speed or accuracy o f response. For example, were performance adequate (with respect to speed and accuracy) with invariant contingencies but inadequate (w ith respect to speed or

accuracy) with shifting contingencies, then the deficit w ould be described in terms o f alternating attention. Sim ilarly, should performance that had been good without distractors suddenly fa il with introduction o f

extraneous stimuli, a deficit in selective attention w ould be invoked. Unfortunately it is by no means certain that these subtypes o f attention have any existence o f their own, other than as heuristic constructs. That is, the present author knows o f no em pirical evid en ce indicating that correlations across tasks within subtype are high w hile correlations across subtypes are low.

Assum ing that such subtypes do exist, it seem s unlikely that training to improve one would be expected to generalize to another, given that each subtype consists o f very different behavioral demands. Training a patient to maintain performance in the face o f distraction, for exam ple, would not be expected to improve his shifting from one set o f

contingencies to another when signalled. The question faced by the current monograph w ill be whether, ev en w ithin subtype, training generalizes across tasks (e.g ., whether training on one task with

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n __________________________ _____________

same task with different distractors). Indeed, the degree to which

generalization occurs within, and is lim ited to, individual subtypes would provide empirical support for the existence o f such subtypes. However the scope o f the present study will be limited to exploration of the extent o f generalization within a single subtype, and will not directly evaluate the specificity of generalization to a single subtype. Therefore we must choose one subtype o f attention and devise a research protocol for studying its rem ediation.

Narrowing the Field of Study

W hen considering the five types o f attention, it was apparent that the first two, focused and sustained, are mutually exclusive. The former refers to tasks consisting o f individual trials, the latter to multiple-trial tasks. Note that this dichotom y is comprehensive; any task can be classified as one type or the other. We may therefore consider focused and sustained attention to com prise the entire continuum o f task

durations, "focused" referring to the minimum endpoint o f the continuum tnd "sustained" referring to all tasks consisting o f durations greater than a single trial.

The remaining three types o f attention are, it seem ed, independent of one another. A task's demand for selective attention is manipulated by varying the discrimiriability or ratio o f relevant to non-relevant stim uli, that for divided attention by varying number o f sim ultaneously active contingencies, and alternating attention by varying frequency o f shifts between contingencies. Each o f these task characteristics can be

independently com bined in various permutations, but no task may sim ultaneously include a demand for both focused (single trials) and sustained (m ultiple trials) attention.

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n ; I n t r o d u c t i o n _________________________________________ 9

The latter observation, combined with a desire to study a single attention subtype, immediately narrowed the field o f study to either focused ' sustained attention. A task fitting the defining characteristics o f any o f the other three attentions (e.g ., requiring alternation between sets o f contingencies) would also fit those o f either focused or sustained attention (consisting o f either individual or multiple trials). On the other hand, when designing a task to meet the defining characteristics o f

focused or sustained attention, one could more readily lim it the degree to which other attention-behaviors are involved (i.e ., only one stim ulus- response relationship, no change in contingencies, and elim ination o f extraneous stim uli to the extent possible).

H aving narrowed the focus o f the current investigation to either focused or sustained attention, the field o f study was further narrowed by excluding focused attention. This decision was based on the fact, already m entioned, that subjects with d eficien cies in focu sed attention are rarely referred for intensive outpatient therapy. They were

therefore not expected to be available for study. Furthermore, the focused attention paradigm is rather artificial in that only discrete, forewarned trials are presented, a circum stance rarely encountered in daily experience and therefore inherently less interesting than the sustained attention paradigm.

The object o f the preceding discussion was to provide a clearer understanding, in operational terms, o f the construct o f attention. This led to refinement in definition o f the subject matter o f the current study. That done, we are now prepared to consider the remediation o f sustained attention subsequent to brain trauma.

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n ________________________________________ 1 0

Remediation

of Sustained

Attention

Interest in remediation o f sustained attention derives from three sources: prevalence o f general attention disorders among the head injured, potential importance o f attention remediation to the overall process o f rehabilitation, and inconsistency o f research findings which has led tc controversy regarding the efficacy o f sustained-attention remediation. To firmly establish a rationale for the current study, let us briefly examine the first two before turning to a detailed consideration of the third. The latter w ill lead to identification o f factors that have promoted inconsistency o f results. This, in turn, w ill lead to identification of som e o f the basic characteristics o f a research design chosen to

m inim ize such inconsistencies.

P r e v a le n c e a£

Attention Deficits

The pervasive nature o f attention deficits among the head injured has been noted by many authors. For support they refer to anecdotal clinical experiences common to all who have worked with such patients, and to several surveys o f the enduring complaints o f survivors and their fam ilies. Such surveys have consisted o f symptom ch eck lists including item s thought to relate to attention, such as "concentration difficulty," along with a multitude o f other frequent complaints such as "headaches" and "fatigue." In light o f the fact that an operational definition o f

attention has only recently begun to emerge, we must bear in mind that these pre-operational estim ates o f prevalence may differ from those based upon more recent conceptualizations o f attention.

In a sample o f Korean War veterans, Caveness (1969) found 41% o f these who had suffered head injury to report concentration difficu lties five years after injury, compared to only 14% o f non-head-injured

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n

respondents. As McLean et al (1984) have pointed out, it is important to consider the base rate o f attention complaints in the general population when attempting to draw conclusions regarding the contribution o f head injury in this regard. In a thorough review of the literature, Gronwall (1987) not only referred to the Caveness study but also reviewed the findings o f more recent studies. As indicated by studies em ploying control groups, the base rate o f concentration complaints tends to fall in the 4 - 8 % range (Gronwall, 1987), the one exception being the 14% rate reported by Caveness (1969). M ost studies, including that o f Caveness, found incidence o f concentration com plaints to be much higher among head injury j urvivors.

Unfortunately the validity o f self-report follow ing brain injury is open to question. There are factors w hich could produce overestim ates o f disability, as indicated by M cKinlay et al (1983) who found only a 9% incidence o f attention complaints 12 months post injury among patients without compensation claim s pending, but a 38% incidence among patients with pending claim s. On the other hand brain injured patients are notoriously poor observers o f their ow n behavior, which could produce underestimates o f disability. For exam ple the above-cited M cKinlay et al study found relatives o f brain injured patients without pending compensation claim s to report a 25% incidence, compared to the 9% incidence reported by patients them selves.

D espite shoitcom ings inherent in the self-report o f survivors, and the poorly operationalized nature o f "concentration" in the above studies, it is nevertheless generally accepted that failures o f concentration are among the most common o f posttraumatic symptoms (L idvall et al, 1974; McLean et al, 1984). This may be taken as circumstantial evidence that attention, as currently defined, is frequently impaired fo llo w in g head

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n

injury. This also fits with clinical observations of the present author, who has noted a high incidence o f performance inconsistencies in head

injured patients.

Attentional difficulties are not only common, but can pose significant lim itations on the entire process o f rehabilitation. Let us now br'efly consider the im plications o f such impairment.

Importance

Attention

to

Rehabilitation

A s described by Gronwall (1987), statements thought to reflect attention difficu lties include: '"...couldn't listen and tik e notes at the same time.'; 'Fine for the first part, then I just kind o f drifted off.'; 'Every time som eone near me turned a page or made a noise, I lost the thread.'; and 'I couldn't seem to keep my mind on what I was there for.'" (p. 355 - 3 5 6 ). The variety o f behaviors represented by these statements

highlights the fact that attention has been a rather vague construct. H owever, as described above and as reflected in the exam ples o f

Gronwall, concern with consistency o f task performance across conditions outlined in Table 1.1 unifies the various conceptualizations or subtypes o f attention.

The notion that attention refers to skill acquisition and execution sufficiently attests to the importance of this topic to rehabilitation (e.g., see N ew com be, 1982). A ll aspects of rehabilitation are essentially educational, with patients engaged in the process o f acquiring or

reacquiring skills specific to each therapy. Clients who exhibit deficits in attention am . under som e conditions, lim ited in the performance o f activities that are the focus o f therapy. This can frustrate attempts o f therapists to instill targeted behaviors and, to the extent that different conditions w ill be encountered out o f therapy, can complicate

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n 11

generalization o f these behaviors from the clinic to functional settings. The degree to which conditions contributing to poor performance can be identified, and training to perform under those conditions be provided, the effectiven ess o f instruction in any therapeutic endeavor should be enhanced. Thus, disorders of attention can limit progress in all areas o f therapy (e.g., see Wood, 1984) and efficacious attention retraining would offer promise as a foundation upon which other rehabilitative efforts might build.

The above-mentioned findings with regard to incidence o f attentional impairment (albeit loosely defined), even five years after injury, point to the potentially chronic nature o f such deficits and underscore the fact that they cannot be expected to spontaneously resolve in their entirety. Furthermore if methods for improving general attentional functions exist, there is little question o f the importance o f their application among head injury survivors involved in the arduous process o f rehabilitation. Having established a satisfactory rationale for investigating attention remediation in general, let us now return to consideration o f sustained-attention remediation in particular.

E v a l u a t i n g

To date, there have been few reported studies o f attention

retraining after head injury. In fact the primary body o f this research is comprised of only six studies (see Table 1.2). Sohlberg and Mateer (1987), Ben-Yishay, Piasetsky, and Rattok (1987), and Gray and Robertson (19 3 9 ) obtained encouraging results, w hile Ponsford and Kinsella (1988) and Malec, Jones, Rao, and Stubbs (1984) drew np'^.tive conclusions, and W ood (1986) reported m ixed results. As reflected by

R em ed iation o f S u stain ed A ttention: Introduction___________________________________14

the variety o f conclusions drawn by these authors, consensus regarding the efficacy o f such therapy has been elusive.

T ab le 1.2

S u m m a r y

o f

A tte n tio n

R e t r a in in g

L ite r a tu r e

Authors / Training D epend en t M easures ±

M alec et al (1984) S tr o o p

-v id eo gam e practice C a n c e lla tio n

-Visual RT

-Wood (1986) WAIS D igit Span

-auditory v igilan ce task o f fixed WMS L ogical Memory -d ifficu lty , visu al vigilan ce task Rey AVLT -o f variable d ifficu lty %-time on task + Ben-Y ishay et al (1987) WAIS D igit Span + m u lti-c o m p o n e n t train in g WATS Picture Completion +

p r o g r a m Visual RT +

Sohlberg & Mateer (1987) PASAT +

m u lti-c o m p o n e n t train in g p r o g r a m

Ponsford & K insella (1988) Sym bol D igit M odalities -com puterized tasks o f fixed C a n c e lla tio n

-d iffic u lty le v e l C hoice RT

-Rating S cale

-%-time on task

-Gray & Robertson (1989) WAIS D ig it Span-arithmetic Com posite + com puterized tasks o f fixed

d iffic u lty le v e l

“i ” indicates whether authors drew p ositive (+) or negative (-) conclusions regarding e ffic a c y o f training.

The paucity o f research is surprising, in light o f evidence suggesting a grow ing tendency for clinicians to em ploy "attention remediation"

methods with their head-injured clients. This evidence is largely

circumstantial; the current author know s o f no formal surveys mdicating the frequency with which such methods are applied. H owever, given the high demand for improvem ents in consistency o f performance across the

various "attention conditions" outlined in Table 1.1, and the recent appearance on the market o f considerable materia1 s ostensibly for such remediation, one can assume the practice o f "attention remediation" to be fairly widespread. A lso indicative o f a high prevalence o f attention

R e m e d i a t i o n o f S u s t a i n e d A t t e n t i o n : I n t r o d u c t i o n 1 5

retraining is the widespread use o f personal computers in clinical practice. (A 1984 informal survey o f the readership o f Cognitive

Rehabilitation, a trade magazine, found that 73% o f the 63 respondents used computers in therapy.) The fact that much o f the computer

software marketed for "cognitive rehabilitation" is billed as having been designed for "attention remediation" suggests that these computers are often used in such activities.

The paucity o f research despite the clear need for studies o f efficacy is probably attributable, in part, to ambiguity in the defim'.ion o f attention. In the words o f Ponsford and Kinsella (1988), "One possible reason for the relative lack o f studies o f attention follow ing brain injury is the continuing controversy as to the meaning o f the concept o f

attention and its mechanisms." W hile it is hoped that discussions such as that presented above w ill eventually lead to consensus regarding the definition o f attention, this has not yet been achieved.

In the absence o f a universally accepted operational definition o f attention, it is not surprising that these few studies have not shared com m on approaches to training nor have used comparable dependent measures. It is also not surprising, therefore, that considerable variation in results has obtained. W e w ill now turn to specific consideration o f the major studies in the area, which w ill permit greater understanding o f the factors possibly contributing to disparity in conclusions. This w ill, in turn, permit developm ent o f a protocol for evaluating attention retraining which might lessen the influence o f these factors. For the sake o f brevity, only three o f the six existing studies w ill be reviewed in depth. This should convey an understanding o f this body o f literature sufficient for present purposes.

R em ed ia tio n o f S u stain ed A ttention; Introduction___________________________________

M ost recently, Ponsford and K insella (1988) evaluated the effectiven ess o f such training in a group o f ten survivors o f severe traumatic brain injury (TBI). Subjects were 17 - 38 years o f age, had experienced 10 days - 12 w eeks o f posttraumatic amnesia, and had begun treatment 6 - 3 4 w eeks follow ing onset.

A ll showed deficient performance on at least three o f five measures derived from a total o f three different tasks. Sym bol D igit M odalities (Smith, 1968, 1973) involved written and oral translation o f a series of printed sym bols to numbers, producing a measure for each response m odality consisting o f the number o f correctly translated sym bols in a 90-second period. Two-Letter Cancellation (Ben-Y ishay et al, 1978) involved a sheet o f apparently random letters printed in rows, requiring the subject to mark o ff every instance of two target letters; two

measures, Tim e Taken and Percent Correct, were produced from this task. Choice Reaction Tim e (Van Zomeran, 1981) required the subject to monitor a display consisting o f four lit buttons all equidistant from a single "home" button, to keep the hom e button depressed until one o f the others lit, then to press the lit button as quickly as possible. The time between lighting o f the target button and release o f the hom e button (D ecision Tim e) was the single measure contributed by this task.

In addition, a Rating Scale o f Attentional Behaviours (developed by Ponsford and K insella) was com pleted by each patient's Occupational Therapist, and 30-m inute videotapes o f each patient performing clerical duties were evaluated with regard to percent-tim e spent on-task. This set o f dependent m easures, including the above m entioned five and the two developed by the authors, is described fully in Appendix A and was used as the standard by w hich effectiven ess o f training was judged.

I

R em ed iation o f Sustained A ttention: Introduction___________________________________ LI

Impairment on these measures was defined as performance at least one standard deviation below the mean obtained from a group o f 16 orthopedically injured control subjects. A ll head trauma subjects were judged by their Occupational Therapist as displaying "impaired

attentional behavior" and were judged by neuropsychological evaluation to exhibit "slow information processing" and "poor selective attention." M ost were also said to exhibit "memory problems, as w ell as im pulsivity, poor self-regulation o f behavior, and difficu lties with planning and problem solving." The authors did not elaborate on the objective indices w hich led to their subjects being so characterized.

Training entailed performance o f fiv e different com puterized tasks w hich provided "repeated practice in responding rapidly, but selectiv ely , to information presented visually on a computer screen," and w hich "allowed for measurement o f changes in accuracy as w e ll as speed o f response over time." These five: React, Search, Red Square/Green Square, Spot the Letter, and Evens and Fives are fully described in Appendix B.

The training msks were administered in two contexts. First, for three w eek s patients were required to practice them w ithout ex p licit feedback. Then for another three w eeks the tasks were performed with a therapist providing feedback and graphing results.

H alf o f the subjects underwent three, and the other half six, w eeks o f pretreatment baseline. Follow ing baseline, all were treated identically: three weeks o f practice, follow ed by three w eeks o f practice with

feedback, and three weeks o f return to baseline. The nature o f feedback was not fu lly described, but appeared to consist primarily o f post-session feedback and graphing o f performance. During all phases, the

R em ed iation o f S u stain ed A ttention: Introduction _1&

collected three tim es per week. The rating scale and videotaping were administered at the beginning and end o f each phase.

Performance on training tasks was not reported. As a group, however, the subjects show ed gradual improvem ent on the five psychom etric m easures. This im provem ent occurred primarily during baseline and was therefore not attrilutable to treatment, leading the authors to conclude that their rem edial approach had been in effective. W hen analyzing their data on a case-by-case basis, however, they noted that five o f the ten subjects show ed sign ificant im provem ent on som e m easures when therapist feedback was provided. One subject improved on four o f the five measures, two on three o f the fiv e, and two subjects on one m e a su r e .

Although the group data indicated no significant effect o f training upon the chosen dependent m easures, it nevertheless appeared that, for som e patients, im provem ent in psychom etric performance was related to receivin g therapist feedback regarding training-task perform ance. The authors were unable to find any characteristics which distinguished those who profited from those who did not. N evertheless, given

encouraging results among some o f the subjects, it seem s reasonable to wonder whether refinem ents o f treatment m ethodology m ight have produced more consistency across subjects, despite the authors' negative c o n c lu sio n s.

A lso worthy o f note is that the videotaped behavioral measure was particularly insen sitive to the attentional deficits that had been ascribed to these patients; a pronounced ceilin g effect in percent-time attending to task was noted. This highlights a particularly frustrating feature o f studying attentional dysfunction. It is an elu sive phenomenon in that patients are com m only able to rise to the occasion when asked to

R em ed iation o f Sustained A ttention: Introduction 1 9

perform a familiar task over a prescribed period o f tim e, when they know they are being critically observed. This is often the case even when reports o f both fam ily and staff, as w ell as results o f psychom etric

evaluation, suggest that attentional perform ance is indeed impaired. The other study producing negative findings was that o f M alec et al (1984). In a study involving ten subjects who were within six months o f injury, these authors reported no effect o f performance o f an arcade-like video gam e on three dependent measures derived from the Stroop Tests (G olden, 1978), two paper-and-pencil cancellation tasks, and a visual reaction time task. H owever, the authors were careful to refer to the application o f the video game as "practice" rather than "training," since treatment consisted o f sim ply allow ing subjects to play the standard video gam e without explicitly incorporating aspects o f training (other than the im m ediate feedback inherent in such gam es). Since data regarding performance o f the video gam e itse lf were not reported, and the authors them selves pointed out that "one obvious shortcom ing o f manufactured video gam es is that they offer little opportunity to

individualize performance demands to suit a particular patient's training needs," one m ight question whether there were indeed any effects o f treatment to generalize to the dependent measures.

Sohlberg and Mateer (1987) em ployed a sin g le-ca se experim ental design with m ultiple baselines across tw o dependent m easures, and four subjects ranging in age from 25 to 30 years. T im e since onset ranged from 1 2 - 7 2 months, and coma duration from 24 hours to 7 w eeks. At least three experim ental phases were provided each subject: baseline (pre-treatment assessm en t), attention training, and visu al process training (training designed to remediate deficits in com plex visual stimulus discrim ination). The latter training w as provided as a means o f

R em ed ia tio n o f Su stain ed A ttention: Introduction ₂₀

demonstrating the specific benefits derived from attention training, by show ing absence o f effect o f visual process training on measures o f a tte n tio n .

T w o dependent measures were used. Paced Auditory Serial Addition Task (PASA T) involving taped presentation o f single-digit numbers at a steady rate o f 2.4 seconds per stimulus, requiring the subject to add the current number to the pi«^cuing number and to speak the sum aloud, was intended to represent attentional function. The Spatial Relations subtest o f the W oodcock-Johnson Psychoeducational Battery, involving presentation o f a series o f geom etric figures accompanied by sets o f irregular shapes and requiring the subject to identify those shapes that could be com bined to form the given geometric figure, was intended to represent "visual processing abilities." (Description o f both o f these tasks is included in Appendix A.) Each was administered only occasionally, as a probe.

A ttention retraining consisted o f a m ulticom ponent treatment package, described briefly in Appendix B. Tasks intended to remediate those aspects o f performance described by Sohlberg and Mateer (1987) as sustained, focused, selectiv e, alternating, and divided attention were all administered. Several different tasks thought to address each o f these d eficit areas w ere provided.

A s was the case for Ponsford and K insella, no mention was made o f training task performance. H ow ever, Sohlberg and Mateer reported evidence for both the generalization and the specificity o f their attention retraining procedures, interpreting results as show ing that

improvem ents in PA SAT scores occurred only during attention-

retraining phases and im provem ents in Spatial Relations scores occurred only during visu al process training. Unfortunately, as the authors

R em ed iation o f Sustained A ttention: Introduction 2 1

them selves pointed out, visual inspection o f their graphed results was com prom ised by the infrequency with w hich probe m easures were taken. Adding to difficu lties with interpretation was the fact that probes were not administered at equal intervals, and stable baselines were not always obtained on both measures. Additional concern stems from the fact that the PASAT, unlike standard vigilance tasks, involves varied behavioral demands, not all o f w hich would be interpreted as

"attentional" in character (e.g., requiring rapid numerical addition rather than sim ple target detection). A ssum ing that improved speed o f addition is not synonym ous with improved attention, it is conceivable that

training tasks involving arithmetic could have conferred a benefit to PA SAT perform ance unrelated to generalized im provem ent in sustained attention. We w ill return to this issue below.

B en-Y ishay et al (1 9 8 7 ) described a program o f attention remediation that, like that o f Sohlberg and Mateer, consisted o f several components providing a variety o f training tasks. T hese have been extensively described elsew here (B en-Y ishay et al, 1987; Ben-Y ishay, Rattok, and D iller, 1979) and w ill not be repeated here. These authors found significant im provem ent not only on the training tasks them selves but also, using a pretest-posttest design, on all four psychom etric measures em ployed (visual reaction time, W AIS D igit Span, W AIS Picture

Com pletion, and an in-house Picture Description test). Although the authors reported stability o f these dependent m easures over the months preceding participation in their study, ostensibly m ollifyin g

m ethodological concerns regarding the use o f a pretest-posttest design, this failed to control for nonspecific effects (e.g., a Hawthorne effect) which could have enhanced or em ulated effects specific to treatment.

R em ed ia tio n o f S u stain ed A ttention; Introduction 22

A lso reporting positive findings were Gray and Robertson (19 8 9 ) who, using a sin gle-case design across three subjects, demonstrated

im provem ent across treatment sessions in a com posite score reflecting D igit Span (forward and backward) performance and performance o f a "simple" arithmetic task involving addition with carrying. Unfortunately the interpretability o f these results is compromised by the brief

b aselines obtained, in conjunction with the visual im pression conveyed by their graphed data indicating that the trend o f improvement during treatment may have reflected a continuation o f a trend established during baseline. Bolstering the authors' claim s o f demonstrated

treatment efficacy was the use o f a control measure for each subject. In all three cases the control measure remained level throughout the study, which the authors took as indication that spontaneous recovery (and, presum ably, practice effects) was not underlying im provem ents noted on the attention measure. H ow ever the control measure appeared level by com parison to the attention measure even during the brief baseline periods, and the possibility o f differential effects o f practice on the two m easures was not considered.

W ood (1 9 8 6 ) investigated the effects o f attention retraining among four subjects, ages 19, 25, 30, and 46 years. All had suffered lengthy posttraumatic am nesia, were many years post onset, exhibited "problems o f attentional focu sin g, distractibility, and were unable to sustain

atten tion ."

T w o classes o f outcome measures were em ployed (see Appendix A). One consisted o f "attention-to-task behavior" observed via a tim e-

sam pling procedure im plem ented during standard rehabilitation therapy sessio n s. The other measures were standard tests o f "auditory-verbal memory" (D igit Span, Logical Memory, and Rey Auditory-Verbal