Shaping Attention Span: An Operant
Conditioning Procedure to Improve
Neurocognition and Functioning in
Schizophrenia
by Steven M. Silverstein, Anthony A. htenditto, and Paul Stuve
Abstract
The high prevalence of neurocognitive deficits in schizophrenia, and their association with poorer outcomes, has
created interest in treatments that can improve neurocognitive functioning hi this illness. While a variety of
rehabilitation interventions have been developed, many
are not appropriate for the most severely ill patients,
whose attention spans are so short that they cannot
attend to the material being presented. For this population, the only neurocognitive rehabilitation methods with
demonstrated effectiveness are those that involve the
operant conditioning technique known as shaping. In
this article, we review the rationale for the use of shaping-based methods as neurocognitive retraining techniques for treatment-refractory schizophrenia patients,
review published reports using this intervention, and
offer suggestions for the future development of this
method from both clinical and research perspectives.
Keywords: schizophrenia, attention, cognition, cognitive rehabilitation, psychiatric rehabilitation
Schizophrenia Bulletin, 27(2):247-257, 2001.
Data on the high prevalence of neurocognitive deficits in
schizophrenia (Palmer et al. 1997) and their association with
poorer outcomes (Green 1996; Silverstein et al. 1998«) has
created interest in treatments that can improve neurocognitive
functioning in this illness. Because traditional antipsychotic
medications have had minimal or sometimes deleterious
effects on cognition after the acute phase (Corrigan and Penn
1995; Schwarzkopf et al. 1999), the majority of direct neurocognitive enhancement efforts thus far have focused on
psychological interventions.1 These nonpharmacological
1
More recently, data on the effects of atypical amipsychotic medications on
neurocognition have been repotted (cg^ see reviews in Schizophrenia Bulletin
25[2Tj. While a consensus appears to be building that neurocognitive functioning is better on these medications compered to older drugs, the mechanisms of
this improvement (e.g^ real improvement versus removal of sfrintinn and other
side effects that worsened cognition on "typical" medications) as well as the
magniturir of the changes are still under debate. Therefore, this literature will
not be a focus of this article.
247
treatments are typically grouped together under the rubric
"cognitive rehabilitation." To date, this term has been applied
to treatments such as practicing of cognitive skills in individual (Spaulding et al. 1986; van Der Gaag 1992) or group
(Brenner et al. 1994; Spaulding et al. 1999fc) formats, and
computer-assisted training (Medalia and Revheim 1999).
Because of the early stage of development of neurocognitive rehabilitation technology for schizophrenia,
guidelines have yet to be established regarding which
treatments are appropriate for which patients. Thus, treatments have not necessarily been targeted to specific profiles or severity levels of neurocognitive deficits. This lack
of specificity is most problematic for chronic, severely ill,
and treatment-refractory patients, such as those who are
unable to be discharged from state-hospital settings; these
patients typically have the most severe attentional and
other neurocognitive deficits, and these deficits are related
to particularly poor outcomes. Despite the enormity of this
problem, scattered reports throughout the literature suggest
that one form of neurocognitive rehabilitation, based on
the behavioral principle of shaping, has consistently
demonstrated effectiveness in increasing the attention
spans of such patients. In this article, we discuss issues
related to the use of shaping procedures as neurocognitive
retraining methods. We begin by providing a rationale for
addressing attentional impairment in schizophrenia and for
using behavioral techniques to do so. This is followed by a
rationale for choosing shaping over other behavioral techniques, a review of published reports using shaping methods to improve attention, and a review of future needs.
Rationale for Addressing Attentional
Deficits in Schizophrenia
A growing body of evidence indicates that deficits in
sustained attention and verbal memory are associated
Send reprint requests to Dr. Steven M. Silverstein, Weill Medical
College of Cornell University, 21 Bloomingdale Rd., White Plains, NY
10605; e-mail: [email protected].
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
S.M. Silverstein et al.
with less skill acquisition in treatments such as the
University of California at Los Angeles Social and
Independent Living Skills modules (Mueser et al. 1991;
Kern et al. 1992; Wallace et al. 1992; Bowen et al. 1994;
Corrigan et al. 1994; Silverstein et al. 19986; Silverstein
et al. 1998e). One conclusion that can be drawn from
these studies is that patients who are impaired in their
ability to sustain attention or to remember material presented to them will benefit little from the treatment.
Neurocognitive deficits, including poor attention span,
are also predictive of poorer outcome in other domains,
including community outcomes, work performance, and
social problem solving (reviewed in Green 1996).
Moreover, the relationships between neurocognitive
deficits and outcomes are relatively independent of
symptom effects, and neurocognitive functioning is
more predictive of level of functioning and outcome
than are symptoms (e.g., Mueser et al. 1991; Green
1998). All of these data suggest that finding a method to
improve neurocognitive functioning is an important step
in improving rehabilitation outcomes. When neurocognitive deficits are targeted for direct intervention, it is
thought that gains in functioning may be made and that
such gains will enhance the success of other rehabilitation efforts.
Neurocognitive Remediation for
Schizophrenia
A comprehensive review of the growing field of neurocognitive remediation for schizophrenia is beyond the
scope of this article, and the reader is referred to several
excellent recent reviews and reports of individual studies
(e.g., Kern 1996; Bellack et al. 1999; Spaulding et al.
1999a). The purpose of this brief section, rather, is to
describe the approaches that are currently in use, in order
to distinguish them from the behavioral approach of
shaping that is the focus of this article.
One approach to treating neurocognitive deficits
involves the adaptation of methods from experimental
psychology. For example, while dichotic listening procedures have been used to demonstrate auditory selective
attention deficits in schizophrenia, they have also been
adapted to enable patients to practice attending to relevant stimuli and ignoring irrelevant stimuli (e.g.,
Spaulding et al. 1986). To date, the total number of
patients treated using such techniques, as reported in
published articles, is quite small, and no systematic procedures for using this approach have been developed.
Moreover, as Spaulding et al. (19996) noted, there is little evidence that the improvements generalize to other
areas of functioning or that they enhance response to
other rehabilitation efforts.
248
Another recent approach utilizes computers to
administer tasks based on neuropsychological tests or
exercises developed for remediation of cognitive deficits
in learning disabilities. Data from studies of neuropsychologically oriented computer exercises indicate that
improvement in neurocognitive functioning occurs, as
assessed via laboratory procedures (e.g., Burda et al.
1994). As with the approach discussed above, there is little evidence that the improvements generalize to other
areas of functioning or that they enhance response to
other rehabilitation efforts. Studies using a neuropsychological educational approach to rehabilitation (NEAR)
(e.g., Medalia et al. 1998; Medalia and Revheim 1999)
have shown promise in improving cognitive abilities, but
their effects outside of the laboratory are as yet unknown.
While it is likely that these methods will continue to be
developed and refined, a problem with using them with
severely impaired patients is that they require significant
intrinsic motivation on the part of the patient. This has
been recognized, and one of the strengths of the NEAR
approach is its selection of exercises that patients seem to
enjoy (Medalia and Revheim 1999). However, at this
point, it is not known whether this approach would be
effective with patients who are unmotivated to participate
in rehabilitative treatment.
The most widely reported approach to neurocognitive
remediation of schizophrenia has been group-based therapy. The most popular of these treatments is Integrated
Psychological Therapy (IPT), developed by Brenner et al.
(1994). This intervention targets skills in a hierarchical
fashion, beginning with conceptual differentiation (executive functioning) and moving through social perception,
verbal communication, basic social skills, and interpersonal problem-solving segments. Skills are targeted
through group practice and problem solving using a series
of exercises that increase in complexity over time. Results
from studies of IPT have been mixed (Brenner et al. 1992;
Brenner et al. 1994; Spaulding et al. 19996). In Brenner's
studies, little evidence of generalizability of the effects to
real-world behavior was observed. Spaulding et al.
(19996) reported improvement on a measure of social
cognition; however, behavioral effects outside of laboratory-based assessment procedures were not studied. The
current status of IPT for schizophrenia remains controversial, with some (e.g., Bellack et al. 1999) suggesting that
the effect sizes from published reports are not clinically
significant, and others (e.g., Spaulding et al. 19996) disputing these claims. As discussed below, however, whatever the eventual verdict is regarding IPT, it is clear that it
is not an appropriate treatment for patients with severe
attentional impairment.
Approaches similar to IPT have been developed for
use in individual treatment sessions (e.g., van Der Gaag
1992; Wykes et al. 1999). For example, Wykes and col-
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
Shaping Attention Span
leagues developed an intensive treatment for addressing
executive functioning deficits in schizophrenia patients.
As with the approaches discussed above, some positive
effects were observed on neuropsychological test scores,
but no effects were noted on behavioral functioning. In
addition, the effects of these interventions outside the
treatment context are unknown. Most important, however, as with IPT, it can be argued that treatments targeting high-level cognitive processes are not appropriate
for patients with severe attentional impairment (see
below).
Because of the lack of consistently convincing
effects from prior studies of neurocognitive remediation
in schizophrenia, some researchers have suggested utilizing approaches that focus on helping patients manage
cognitively demanding tasks in the real world (Flesher
1990; Hogarty and Flesher 1999; Velligan and BowThomas 2000). One such approach is Cognitive
Adaptation Training (CAT), which involves the use of
cues and compensating features in the patient's environment (Velligan and Bow-Thomas 2000). As the authors
note, CAT has more in common with case management
than with traditional cognitive rehabilitation, in that it
involves home visits and in vivo supports and is not
viewed as a method for strengthening cognitive functions or their neural correlates. Preliminary data on this
approach are encouraging. The strength of CAT, however, is clearly with outpatient populations who are
already living and working in what is hoped to be a relatively permanent environment for them. For long-stay
inpatients in state hospitals and other residential facilities, interventions are needed that can improve basic
attentional abilities so that these inpatients can participate more fully in psychosocial rehabilitation and eventually be discharged to the community. Flesher's (1990)
cognitive habilitation intervention is more amenable for
use in a hospital setting, although he notes that
"Cognitive habilitation will be most useful for ameliorating the residual deficits in relatively stable remitted
patients" (p. 226). Similarly, the more recently developed Cognitive Enhancement Therapy (Hogarty and
Flesher 1999), while unique and promising, was developed for an outpatient population.
In short, current approaches to treating neurocognitive deficits in schizophrenia do not provide strong evidence of their effectiveness or generalizability, and they
do not seem appropriate for patients with severe attentional impairment (see below). In addition, recent remedial and compensatory approaches are clearly meant for
use with stabilized outpatients. This situation calls for
the development of interventions targeting the severe
attentional impairments of chronic schizophrenia
patients who continue to reside in long-term institutional settings.
Attentional Requirements for
Participation in Traditional
Neurocognitive Rehabilitation
Since all forms of psychosocial treatment require some
degree of attention span, those patients with the most
severe attentional impairment may benefit as little from
group-based (Brenner et al. 1994) or computer-assisted
(Medalia and Revheim 1999) neurocognitive rehabilitation as they do from more traditional skills training procedures. Research data support this view. For example,
Michel et al. (1998) found that failure to benefit from
neurocognitive remediation was related to more severe
deficits in neurocognitive functioning among schizophrenia patients. In addition, Sohlberg et al. (2000), in evaluating the effects of a well-validated form of neurocognitive rehabilitation on persons with acquired brain injury,
found that the treatment was ineffective for individuals
with impaired sustained attention. In contrast, for
patients with relatively intact pretreatment levels of vigilance, the treatment effect was highly significant. Two of
the measures on which pretreatment vigilance level discriminated treatment responders from nonresponders
were measures of executive functioning with demonstrated sensitivity to frontal lobe and cingulate activation
(Perett 1974; Deary et al. 1994). The finding that sustained attention deficits have a rate limiting effect on
cognitive rehabilitation (as they do for skills training) has
implications for the placement of schizophrenia patients
in traditional group-based forms of cognitive treatment.
For example, the most widely used treatment of this type
for schizophrenia patients is Brenner's IPT (described
above, Brenner et al. 1992, 1994). Sohlberg et al.'s
(2000) results imply, however, that patients with severely
impaired sustained attention abilities will be unable to
benefit from even the initial conceptual-differentiation
module.
These data and assertions are consistent with the
attentional and rehabilitation models of Sturm et al.
(1997). Using earlier work by Posner and Boies (1971),
Posner and Rafal (1987), and van Zomeran et al. (1984),
Sturm and colleagues conceptualize attention as involving
a set of distinct processes. They make an important distinction between intensity aspects of attention and aspects
involved in information selection. Subcomponents of the
intensity aspect of attention include (1) phasic alertness,
defined as those processes involved in enhancing a
response to a specific stimulus at a specific time point;
and (2) vigilance, defined as the ability to maintain alertness over a prolonged period in order to respond to relevant but infrequently occurring stimuli. Subcomponents
of the selection aspects of attention include (1) selective
attention, defined as the ability to focus on relevant con-
249
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
S.M. Silverstein et al.
textual stimulus features while suppressing responses to
irrelevant stimuli; and (2) divided attention, defined as the
ability to simultaneously monitor two or more sources of
information that are relevant for behavioral responses.
Sturm et al. studied the effectiveness of a cognitive remediation intervention comprising distinct intervention components targeted separately at alertness, vigilance, selective attention, and divided attention in populations of
patients with focal brain damage of vascular etiology.
One of their key conclusions was that "specific attention
disorders need specific training" (p. 96). This was especially true for the intensity components of alertness and
vigilance, where improvement occurred only after specific training of these functions. Other important findings
from this study were that adequate functioning of selective aspects of attention required a sufficient amount of
alertness, and that among patients with impaired alertness
and vigilance, training on selective aspects of attention
led to clinical deterioration. On the basis of these data,
they stressed that "impairments in a given level can only
be approached by training on the same or a subordinate
level" (p. 100). On the basis of their findings, Sturm et al.
proposed a hierarchical model of attention wherein phasic
alertness and vigilance capabilities are required for selective attention, and selective attention is a prerequisite for
divided attention.
These findings are relevant for cognitive rehabilitation of schizophrenia because the patients who are most
unable to benefit from traditional group-based training
approaches are those with severe impairments in the
intensity aspects of attention such as phasic alertness
(Silverstein et al. 1998e) and vigilance (Nuechterlein
1991; Kern et al. 1992; Silverstein et al. 19986).
Therefore, even though the majority of these patients will
also demonstrate deficits in higher level functions, it is
critical that cognitive remediation efforts begin by targeting basic intensity functions. To date, efforts to improve
cognitive functioning in schizophrenia have focused
almost exclusively on remediation of higher cognitive
functions such as executive functioning and social cognition (e.g., van Der Gaag 1992; Brenner et al. 1994;
Spaulding et al. 19996; Wykes et al. 1999) while ignoring
the more basic aspects of attention. For schizophrenia
patients with severe attentional impairment, placement in
a traditional neurocognitive rehabilitation intervention
such as IPT, which begins with exercises targeted toward
higher level cognitive processes such as executive functioning, is likely to lead to, at best, few gains, and at
worst, further impairment. We have observed this phenomenon clinically among patients who participated in
studies of cognition and skills training outcome
(Silverstein et al. 1998e), where no appreciable change
was observed among those with the most severe atten-
tional impairments when they were placed in IPT (Pierce
et al. 1997). All of this stresses the importance of developing effective interventions for schizophrenia patients
who need to improve their basic attentional abilities as a
precursor to being introduced to other treatments.
The Use of Behavioral Techniques to
Enhance Neurocognitive Functioning
Data from a number of older and more recent studies provide a rationale for using behavioral techniques to
improve cognition. One group of studies used monetary
reinforcement for correct responding to improve performance on laboratory tests of attention (Rosenbaum et al.
1957; Karras 1962, 1968; Wagner 1968; Meiselman 1973;
Kem et al. 1995), or tests of other neurocognitive functions (e.g., Summerfelt et al. 1991). A second group of
studies used operant procedures to increase attention span
during brief, structured social interactions (e.g., Wallace
and Boone 1983; Massel et al. 1991). All of these studies
describe increased attentional functioning after appropriate responding was consistently reinforced with money or
consumables. They therefore can be viewed as precursors
to neurocognitive rehabilitation interventions. These
demonstrations diverge from modern rehabilitation efforts
in that their time span was brief and individual differences in attentional functioning were not taken into
account when delivering the reinforcers. As will be
described below, tailoring reinforcement schedules and
contingencies to patients' baseline levels of functioning,
and proceeding with a systematic and gradual approach,
can achieve real-world results that exceed and are more
meaningful than past laboratory demonstrations. The use
of shaping procedures as neurocognitive rehabilitation
methods allows the clinician/researcher to achieve these
goals.
Shaping
Shaping is the application of several fundamental techniques of learning to bring about new behavior or to modify a certain aspect of an existing behavior. As such, shaping can be viewed as a method to achieve operant
conditioning, with attention being the response that is targeted. The primary technique involved is differential reinforcement of successive approximations. Rather than
waiting for the complete behavior (e.g., a 20-minute
attention span) to occur before offering reinforcement,
reinforcement is provided for successive approximations
or steps toward the final behavior. When the initial step
toward a behavior (e.g., 4 minutes of continuous attention) has been reinforced and occurs fairly regularly, the
250
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
Shaping Attention Span
criterion for reinforcement is advanced to the next step
(e.g., 5 minutes of continuous attention). This sequence
of reinforcing, changing criteria for reinforcement, fading reinforcers for previous "versions" of the behavior,
and limiting reinforcers to behavior meeting the new criterion, is then repeated until the behavior resembles the
final desired response. A strength of shaping is therefore
that it allows for specific learning techniques to be used
to develop and strengthen behavior that does not normally occur, or else occurs at a very low frequency. It is
this feature that makes it suitable for treatment of patients
whose severely impaired attention spans preclude them
from active participation in other forms of psychosocial
treatment, including many forms of neurocognitive remediation.
Shaping procedures can target the form, intensity, or
duration of a behavior. An example of shaping the form
of a behavior was provided by Isaacs et al. (1960), who
shaped verbal behavior in two patients with schizophrenia who had each been mute for over 14 years. Chewing
gum was chosen as a reinforcer and was used to successively reinforce lip movement initially, followed by reinforcement of simple vocalization (initially grunts), then
simple understandable verbalizations ("gum"), and
finally verbal responses to questions. To have reinforced
the intensity of behavior, these investigators could have
provided reinforcement for verbalizations that gradually
approached a specified loudness. For example, they could
have started by offering reinforcement for a whisper and
gradually requiring increased loudness of verbalizations
until they were reinforcing verbalizations of only a reasonable conversational volume. Duration is often a target
of shaping procedures directed toward neurocognitive
functioning, specifically attentional functioning. In such
cases, the behavior of interest is how long a patient can
direct attention to a particular task.
Two remaining issues regarding shaping should be
noted prior to reviewing studies of its use in improving
attention span. One concerns the choice to use shaping
rather than other conditioning techniques. It is clear that
operant techniques, rather than classical conditioning
procedures, are most appropriate for treating attention
span in schizophrenia. This is because on-task behavior
(i.e., attention span) is a voluntary behavior (or operant)
comprising distinct temporal units of engaging with the
environment. Within the field of operant conditioning
itself, many procedures can be employed to modify
behavior. Some of these, such as modeling and vicarious
learning, assume that the target behavior is either already
in the person's behavioral repertoire or can be quickly
learned. Neither of these conditions are met in the case of
severe attentional deficits in schizophrenia. In this case,
the target behavior or criterion (i.e., a period of sustained
attention of sufficient duration to enable learning in other
treatment modalities) is clearly not part of the current
repertoire, and the criterion is unlikely to be reached by
watching others. Similarly, simply making reinforcers
contingent upon meeting the target behavior is unlikely to
be effective, because the target behavior occurs at an
extremely low frequency. That is, if the strategy that is
used is to reinforce patients every time they, for example,
pay attention for 20 minutes, this strategy is unlikely to
be effective because this behavior will occur infrequently,
leading to low levels of reinforcement and therefore little
behavior change. What is needed in schizophrenia, therefore, is an approach that involves a long-term goal of an
adequate attention span but that also aims to reach that
goal gradually over time by reinforcing similar behaviors,
or approximations to the behavior, until the target behavior occurs spontaneously and can be reinforced frequently. It is for this reason that shaping approaches have
been adopted for use with patients with severe attentional
impairment.
A further rationale for using shaping procedures in
schizophrenia is that there is a long history of the successful use of shaping to change behavior in schizophrenia patients. For example, shaping procedures have been
used to increase spontaneous speech and improve conversation skills in severely withdrawn patients (Massel et al.
1991). Moreover, shaping procedures have been shown to
be effective in increasing a wide range of appropriate
behaviors in institutionalized schizophrenia patients (Paul
and Lentz 1977).
Shaping Attention Span as a Form of
Neurocognitive Rehabilitation for
Schizophrenia
Spaulding et al. (1986) examined the effects of shaping
on the continuous work performance of nine inpatients
with severe and persistent schizophrenia. At baseline, all
subjects demonstrated an inability to focus their attention
on a simple work task for more than 5 minutes. Training
consisted of three 30-minute sessions per week in which
subjects performed a simple paper-manipulation task,
such as folding, cutting, stapling, or sorting. A target time
was established for each subject at the beginning of each
session, and verbal feedback, prompting, and praise were
provided by a trainer throughout the session.
Seven of the original nine subjects graduated from
training, having achieved continuous work performance
scores of 30 minutes for five consecutive sessions.
Considerable individual differences were noted in the
patients' response to treatment, with the time to achieve
graduation ranging from a low of 12 sessions for one
251
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
S.M. Silverstein et al.
patient to a high of over 250 sessions (nearly 2 years) for
another. Similar individual differences were found for
these seven patients in overall psychiatric status, as
improved continuous work performance scores did not
appear to be systematically related to scores on the
Nurses' Observation Scale for Inpatient Evaluation, 30
items (Honigfeld, Gillis, and Klett 1966).
Menditto et al. (1991) used shaping procedures to
increase the attention span of seven forensic inpatients
with severe and persistent schizophrenia or schizoaffective disorder. Average length of stay for these patients was
10.4 years, and they were considered to be among the
most ill and least responsive patients in the hospital.
Shaping classes were held three times daily during weekdays. Assignments consisted of paper and pencil tasks
focusing on practical language and mathematics skills.
Target times for on-task behavior were initially quite brief
(30-60 seconds), and 2-3 trials were typically required
per session. Patients received prompting and encouragement as necessary throughout the session, and upon successful completion of each trial they received specific verbal praise, a shaping chip, a small food snack, and a
prompt specifying the requirements for the next trial.
After successful completion of the last trial, a participation token was awarded. Tokens were used to "purchase"
a variety of goods or privileges, such as snacks, coffee,
grounds passes, and TV time, from the ward "token
store." As patients demonstrated success with each target
for several sessions, targets were gradually increased, typically in increments of 30-60 seconds, until the patient
consistently completed two consecutive 10-minute trials.
After 12 months of training, six of the seven patients had
demonstrated substantial improvements in attentional
functioning, with four of these graduating from shaping
classes and progressing to more traditional academic
classes on the ward. They continued to perform quite well
in those classes, with 1-year followup showing successful
completion of academic class assignments an average of
84 percent of the time.
Silverstein et al. (1998<f) replicated the findings of
Spaulding et al. (1986) and Menditto et al. (1991) using a
four-session-per-week shaping intervention with four
patients who had been unable to tolerate any form of
group treatment. All patients demonstrated improvements
over the course of 50-55 training sessions, with average
on-task behavior increasing to 45-55 minutes. One patient
had an IQ within the mentally retarded range and another
patient had borderline intellectual functioning. Consistent
with the findings of Spaulding et al. (1986), no changes in
Brief Psychiatric Rating Scale scores were observed for
any of the patients while they were in shaping classes.
These data support accumulating evidence on the relative
independence of cognitive deficits and symptoms in
schizophrenia (Green 1998).
252
Bellus et al. (1999) compared the academic skill performance of seven lower functioning patients in shaping
classes to a group of seven higher functioning patients in
traditional academic classes over a 9-month period. Most,
but not all, patients in the Bellus et al. study had chronic
psychotic disorders. In shaping classes, feedback was
given at varying temporal intervals across patients, consistent with their individualized goals, while in academic
classes, feedback was given at fixed 10-minute intervals.
Additionally, in shaping classes, shaping chips were
paired with a choice of a consumable reinforcer (e.g.,
penny candies, 1-ounce cups of juice), while no consumable reinforcers were used in traditional classes. Both
shaping and traditional academic classes were generally
50 minutes in duration. Patients in shaping classes
increased their reading and mathematics performance
close to one and two grade levels, respectively. In contrast, patients in traditional academic classes did not show
significant performance improvement in these subjects.
Silverstein et al. (1999) sought to determine the
effectiveness of integrating shaping and skills training
procedures. It was thought that such an integration could
potentially increase the efficiency of treatment and reduce
length of hospital stay. This is because patients would no
longer need to wait until months of shaping classes were
completed before they entered several more months of
skills training. The method used in this shaping-skills
training integration was to identify inattentive behaviors
characteristic of each patient and then use shaping techniques to improve these behaviors and facilitate acquisition of new knowledge and skills during group sessions.
For each patient in the Silverstein et al. (1999) study,
the most problematic verbal and nonverbal inattentive
behaviors were identified. Nonverbal behaviors (e.g., eyes
open, head up, eye contact with speaker) were rated each
minute using interval-sampling procedures, while verbal
behaviors (e.g., responding within 5 seconds, making
spontaneous relevant comments) were rated using eventsampling procedures. Two noninteractive observers
recorded the individualized target behaviors and reported
their frequency at 15-minute intervals. Patient goals initially reflected an average of 4 weeks baseline, preshaping
performance. After initiation of shaping procedures, for
each 15-minute review period, patients who met or
exceeded their goal (e.g., 60% of that period with their
head up) received a token. Patients turned in tokens at the
end of the group and received 25 cents for each token
earned. As patients began to exceed their goals consistently, the criteria were increased to facilitate continued
progress. Two findings were noteworthy from this project.
First, all patients demonstrated significant increases in
attentive behavior using this procedure. Second, for one
patient who did not respond initially to the 15-minute
reinforcement schedule, a continuous reinforcement
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
Shaping Attention Span
schedule was implemented wherein he was given 5 cents
and a piece of candy each time he opened his eyes. This
eventually led to increases from 10 percent to over 80 percent of the time in keeping his eyes open, with subsequent
greater spontaneity and participation, and responses that
were more relevant to the group.
A second limitation of past studies is that they were
not controlled. In essence, all past reports, with the
exception of Bellus et al. (1999), were a series of case
studies. This raises the possibility that nonspecific factors
may be responsible for all or part of the gains made by
patients receiving shaping interventions. Controlled studies that vary critical intervention components (e.g.,
amount, nature, and frequency of reinforcement) are necessary to rule out this possibility. Such studies will also
lead to greater efficiency and effectiveness of shaping
treatment by identifying its most critical parameters.
Another design that may be useful is a multiple-baseline
design. This was used by Massel et al. (1991) in their
study of shaping to improve conversation skills. Multiple
baseline designs use patients as their own controls and
allow for assumptions to be made regarding relationships
between time of treatment initiation and treatment
effects.
Two of the biggest unresolved issues regarding shaping involve posttreatment maintenance of gains and generalizability. Specifically, do the performance improvements from shaping classes generalize to other
environments, and do they persist after shaping treatment
is discontinued (and if so, for how long)? Limited data
are available on these issues. Spaulding et al. (1986)
reported that seven of nine patients reached the 30minute criterion and then continued to progress in subsequent stages of vocational rehabilitation. Menditto et al.
(1991) found that after shaping groups ended and patients
were transferred to regular academic classes, patients
successfully completed assignments and earned participation tokens in these classes an average of 84 percent of
the time during a 12-month followup period. In addition,
the four patients in that project who met the most stringent improvement criterion subsequently earned promotions to higher step levels in the social-learning program
following their graduation from shaping classes.
While these data are encouraging, they are no substitute for controlled longitudinal studies and systematic
assessments of generalization to other environments. An
important issue for such future studies involves the nature
of the outcome variables. While generalizability to other
forms of treatment is an important index, it will be critical
to evaluate factors such as spontaneous participation in
voluntary activities, length of conversations, ability to
gain and maintain employment, and other real-world outcomes.
Interpretation of Effects
An important issue that is rarely addressed in the literature on neurocognitive rehabilitation in schizophrenia
concerns the nature of the treatment effects. Simply
stated, the issue is this: Do the observed improvements in
neurocognitive performance represent maximization of
functioning or do they reflect actual regaining of lost
functions? In the case of shaping, it could be argued that
by making the pairing between the stimulus (i.e., the
task), the response (i.e., on-task behavior), and the reinforcement highly salient, competing stimuli (internal and
external) are made relatively less salient, and the likelihood of responding to other internal or external stimuli
(i.e., the response of "distractibility") is reduced. On the
other hand, it could be argued that the shaping procedures
are having the effect of strengthening those neural circuits
involved in sustained attention. The data available do not
allow for a resolution of these two positions, and both
effects may be operating. Thus, determining the relative
contributions of these effects is an important goal for better understanding the mechanisms of shaping and other
forms of neurocognitive rehabilitation. We would caution,
however, that since the schizophrenia literature is replete
with examples of environmental stimuli influencing
behavior (e.g., Zarlock 1966; Salzinger 1984), the contribution of the stimulus field should not be ignored in
designing rehabilitation interventions. Indeed, it represents a fertile area for further development of these treatment techniques. This is because, in theory, the combining
of optimal training tasks (i.e., those that activate relevant
neural structures and circuits) with optimal learning conditions should lead to the greatest gains.
Limits of the Literature
Enough reports have now appeared in the literature to indicate that shaping is an effective method to increase attention
span and work performance among chronic, severely ill
schizophrenia patients. Despite these encouraging data, however, there are important limitations to those reports that have
appeared thus far. One limitation involves sample size. All
published reports of shaping as a form of neurocognitive
rehabilitation have included small numbers of patients. What
is now needed is a study that utilizes a standardized variant
of the shaping technique with a larger number of patients.
Future Directions
One of the most important tasks in the continued study of
shaping is to better characterize those patients who
require shaping interventions as well as those patients
253
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
S.M. Silverstein et al.
who benefit most from the treatment. Regarding the first
issue, patients have typically been chosen based on clinical judgment For example, patients have been chosen for
shaping on the basis of being the most "regressed"
patients in a treatment program (Menditto et al. 1991) or
being "low functioning" and having made little or no
gains in prior efforts at more traditional skills training
approaches (Silverstein et al. 199&/)- It is possible that
efficiency of treatment would be improved if there were
methods to identify relevant candidates and place them in
shaping interventions prior to extended periods of treatment failures. One potential mediod to identify appropriate candidates is via neurocognitive testing on admission.
Data now exist indicating that specific neurocognitive
deficits are associated with poorer psychiatric rehabilitation outcomes (Green 1996; Silverstein et al. 1998e),
thereby highlighting the need for development and dissemination of a neurocognitive screening battery. Efforts
at the development of both test batteries (Ashbrook et al.
1988; Silverstein et al. 1998a; Silverstein et al. 1998e)
and new neurocognitive measures (Silverstein et al.
1998*) that may be effective in the screening of candidates for different levels of psychiatric rehabilitation
interventions have already begun. Much more work needs
to be done, however, before treatment decisions based on
test data can be made with high levels of confidence.
Another important issue is the integration of shaping
procedures with other forms of neurocognitive rehabilitation. The benefits of this would be to systematize the delivery of reinforcement that patients receive as they are
engaged in the training. As noted at the outset, data from
laboratory studies of neurocognitive functioning under
conditions of reinforcement indicate that performance
improves relative to typical test-taking conditions. Thus,
there is reason to believe that the systematization of reinforcement delivery, in addition to the usual focus on task
content and difficulty level, would have a synergistic
effect. One area in which this integration would appear to
be relatively simple is with the training technique known
as errorless learning. Errorless learning involves beginning
training on tasks for which there is a high expectation of
success and proceeding through a graded series of tasks
that become increasingly more complex. The goal of this
procedure is to minimize the commission of errors while at
the same time achieving performance mastery. Once performance mastery without errors is achieved at a given
level, tasks at the next level of complexity are introduced.
Errorless learning has demonstrated effectiveness in the
treatment of developmentally disabled and neurologically
impaired individuals (Baddeley 1992; Kern 1996).
Moreover, it has been used as a technique to improve die
performance of people with schizophrenia on neuropsychological tests (Summerfelt et al. 1991; Benedict et al.
254
1994; Stratta et al. 1994; Vollema et al. 1995; Bellack et al.
1996; Wexler et al. 1997) and to treat executive functioning impairments widiin the context of a larger neurocognitive rehabilitation program (Wykes et al. 1999). The success of errorless learning strategies in the cognitive
rehabilitation of schizophrenia raises the issue of whether
systematically pairing reinforcement with errorless learning procedures might lead to increased gains in neurocognitive rehabilitation. Similar pairings can be tried with
computer-assisted forms of neurocognitive training. A
related but unexplored issue is the extent to which shaping
attentive group behaviors, as demonstrated by Silverstein
et al. (1999), within die context of group-based forms of
neurocognitive rehabilitation such as IPT (Brenner et al.
1994) would lead to increased treatment effects.
References
Ashbrook, R.M.; Spaulding, W.; and Cromwell, R.L.
Computer-assessed cognition in psychiatric settings: An
examination of the issues. In: Mancuso, J., and Shaw, M.,
eds. Cognition and Personal Structure: Computer Access
and Analysis. New York, NY: Praeger, 1988. pp. 253-278.
Baddeley, A.D. Implicit memory and errorless learning: A
link between cognitive theory and neuropsychological
rehabilitation? In: Squire, L.R., and Butters, N., eds.
Neuropsychology of Memory. 2nd ed. New York, NY:
Guilford, 1992. pp. 309-314.
Bellack, A.; Blanchard, J.; Murphy, P.; and Podell, K.
Generalization effects of training on the Wisconsin Card
Sorting Test for schizophrenia patients. Schizophrenia
Research, 19:189-194, 19%.
Bellack, A.S.; Gold, J.M.; and Buchanan, R.W. Cognitive
rehabilitation for schizophrenia: Problems, prospects, and
strategies. Schizophrenia Bulletin, 25(2):257-274, 1999.
Bellus, S.B.; Kost, P.P.; Vergo, J.G.; Gramse, R.; and
Weiss, B.A. Academic skills, self-care skills, and on-ward
behavior with cognitively impaired, chronic psychiatric
inpatients. Psychiatric Rehabilitation Skills, 3:23-40,
1999.
Benedict, R.; Harris, A.; Markow, T.; McCormick, J.;
Nuechterlein, K.; and Asamow, R. Effects of training in
information processing in schizophrenia. Schizophrenia
Bulletin, 20(3):537-546, 1994.
Bowen, L.; Wallace, C.J.; Glynn, S.M.; Nuechterlein,
K.H.; Lutzker, J.R.; and Kuehnel, T.G. Schizophrenics'
cognitive functioning and performance in interpersonal
interactions and skills training procedures. Journal of
Psychiatric Research, 28:289-301, 1994.
Brenner, H.; Hodel, B.; Roder, V.; and Corrigan, P.
Treatment of cognitive dysfunctions and behavioral
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
Shaping Attention Span
Kern, R.S.; Green, M.F.; and Satz, P. Neuropsychological
predictors of skills training for chronic psychiatric
patients. Psychiatry Research, 43:223-230, 1992.
deficits in schizophrenia. Schizophrenia
Bulletin,
18(l):21-26, 1992.
Brenner, H.; Roder, V.; Hodel, B.; Kienzle, N.; Reed, D.;
and Liberman, R. Integrated Psychological Therapy for
Schizophrenic Patients. Toronto, Canada: Hogrefe and
Huber, 1994.
Kern, R.S.; Green, M.F.; and Goldstein, M.J.
Modification of performance on the span of apprehension,
a putative vulnerability marker to schizophrenia. Journal
of Abnormal Psychology, 104:385-389, 1995.
Burda, P.; Starkey, T.; Dominguez, F.; and Vera, V.
Computer-assisted cognitive rehabilitation of chronic psychiatric inpatients. Computers in Human Behavior,
10:359-368, 1994.
Massel, H.K.; Corrigan, P.W.; Liberman, R.P.; and Milan,
M.A. Conversation skills training of thought-disordered
schizophrenic patients through attention focusing.
Psychiatry Research, 38:51-61,1991.
Corrigan, P.W., and Penn, D.L. The effects of antipsychotic and antiparkinsonian medication on psychosocial
skill learning. Clinical Psychology Science and Practice,
2:251-262, 1995.
Medalia, A.; Aluma, M.; Tryon, W.; and Merriam, A.E.
Effectiveness of attention training in schizophrenia.
Schizophrenia Bulletin, 24(1): 147-152, 1998.
Corrigan, P.W.; Wallace, C.J.; Schade, M.L.; and Green,
M.F. Learning medication self-management skills in
schizophrenia: Relationships with cognitive deficits and
psychiatric symptoms. Behavior Therapy, 25:5-15,1994.
Medalia, A., and Revheim, N. Computer assisted learning
in psychiatric rehabilitation. Psychiatric Rehabilitation
Skilb, 3:77-98, 1999.
Meiselman, K.C. Broadening dual modality cue utilization in chronic nonparanoid schizophrenia. Journal of
Consulting and Clinical Psychology, 41:447^153, 1973.
Deary, IJ.; Ebmeier, K.P.; MacLeod, K.M.; Dougall, N.;
Hepburn, D.A.; Frier, B.M.; and Godwin, G.M. PASAT
performance and the pattern of uptake of -super(99m)Tcexametazine in brain estimated with single photon emission tomography. Biological Psychiatry, 38:1-18, 1994.
Menditto, A.A.; Baldwin, L.J.; O'Neal, L.G.; and Beck,
N.C. Social learning procedures for increasing attention
and improving basic skills in severely regressed institutionalized patients. Journal of Behavior Therapy and
Experimental Psychiatry, 22:265-269, 1991.
Flesher, S. Cognitive habilitation in schizophrenia: A theoretical review and model of treatment Neuropsychology
Review, 1:223-246, 1990.
Michel, L.; Danion, J-M.; Grange\ D.; and Sandner, G.
Cognitive skill learning and schizophrenia: Implications
for cognitive remediation. Neuropsychology, 12:590-599,
1998.
Green, M.F. What are the functional consequences of neurocognitive deficits in schizophrenia? American Journal
of Psychiatry, 153:321-330, 1996.
Mueser, K.T.; Bellack, A.S.; Douglas, M.S.; and Wade,
J.H. Prediction of social skill acquisition in schizophrenic
and major affective disorder patients from memory and
symptomatology. Psychiatry Research, 37:281-296, 1991.
Green, M.F. Schizophrenia From a Neurocognitive
Perspective. Boston, MA: Allyn and Bacon, 1998.
Hogarty, G.E., and Flesher, S. Practice principles of cognitive enhancement therapy for schizophrenia.
Schizophrenia Bulletin, 25(4):693-708, 1999.
Nuechterlein, K.H. Vigilance in schizophrenia and related
disorders. In: Steinhauer, S.R.; Gruzelier, J.H.; and Zubin,
J., eds. Handbook of Schizophrenia: Neuropsychology,
Psychophysiology, and Information Processing. Vol. 5.
Amsterdam, The Netherlands: Elsevier, 1991. pp. 397^133.
Honigfeld, G.; Gillis, R.; and Klett, J. NOSIE-30: A treatment-sensitive ward behavior scale. Psychological
Reports, 19:180-182, 1966.
Isaacs, W.; Thomas, J.; and Goldiamond, I. Application of
operant conditioning to reinstate verbal behavior in psychotics. Journal of Speech and Hearing Disorders,
25:8-12, 1960.
Karras, A. The effects of reinforcement and arousal on the
psychomotor performance of chronic schizophrenics. Journal
ofAbnormal and Social Psychology, 65:104-111,1962.
Palmer, B.W.; Heaton, R.K.; Paulsen, J.S.; Kuck, J.;
Braff, D.; Harris, M.J.; Zisook, S.; and Jeste, D.V. Is it
possible to be schizophrenic yet neuropsychologically
normal? Neuropsychology, 11:437^446, 1997.
Paul, G. L., and Lentz, R. J. Psychosocial Treatment of
Chronic Mental Patients: Milieu vs. Social Learning
Programs. Cambridge, MA: Harvard University Press, 1977.
Karras, A. Choice reaction time of chronic acute psychiatric patients under primary or secondary aversive stimulation. British Journal of Social and Clinical Psychology,
7:270-279, 1968.
Perett, E. The left frontal lobe of man and the suppression
of habitual responses in verbal categorical behavior.
Neuropsychologia, 12:323-330, 1974.
Kem, R.S. Cognitive rehabilitation of people with mental
illness. Psychiatric Rehabilitation Skills, 1:69-77, 1996.
Pierce, D.; Silverstein, S.M.; Dyjak, M.; DeCarlis, L.; and
Schenkel, L. "Cognitive Rehabilitation of Chronically
255
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
S.M. Silverstein et al.
Psychotic Patients." Presented at the New York State
Office of Mental Health Research Conference, Albany,
NY, December 1997.
Posner, M.I., and Boies, S.W. Components of attention.
Psychological Review, 78:391-408, 1971.
Spaulding, W.D.; Fleming, S.K.; Reed, D.; Sullivan, M.;
Storzbach, D.; and Lam, M. Cognitive functioning in
schizophrenia: Implications for psychiatric rehabilitation.
Schizophrenia Bulletin, 25(2):275-289, 1999a.
Spaulding, W.D.; Reed, D.; Sullivan, M.; Richardson, C ;
and Weiler, M. Effects of cognitive treatment in psychi-
Posner, M.I., and Rafal, R.D. Cognitive theories of attention and the rehabilitation of attentional deficits. In:
Meier, M.J.; Benton, A.L.; and Diller, L., eds.
Neuropsychological Rehabilitation. Edinburgh, Scotland:
Churchill Livingstone, 1987. pp. 182-201.
atric
rehabilitation.
Schizophrenia
Bulletin,
25(4):657-676, 1999ft.
Spaulding, W.D.; Storms, L.; Goodrich, V.; and Sullivan,
M. Applications of experimental psychopathology in psychiatric rehabilitation. Schizophrenia
Bulletin,
12(4):560-577, 1986.
Rosenbaum, G.; MacKavey, W.R.; and Grisell, J.L.
Effects of biological and social motivation on schizophrenic reaction times. Journal of Abnormal and Social
Psychology, 54:364-368, 1957.
Stratta, P.; Mancini, F.; Mattei, M.; Casachia, M.; and
Rossi, A. Information processing strategy to remediate
Wisconsin Card Sorting Test performance in schizophrenia: A pilot study. American Journal of Psychiatry,
151:915-918, 1994.
Salzinger, K. The immediacy hypothesis in a theory of
schizophrenia. In: Spaulding, W.D., and Cole, J.C., eds.
Nebraska Symposium on Motivation, 1983: Theories of
Schizophrenia and Psychosis. Lincoln, NE: University of
Nebraska Press, 1984. pp. 231-282.
Sturm, W.; Willmes, K.; Orgass, B.; and Hartje, W. Do
specific attention deficits need specific training?
Neuropsychological Rehabilitation, 7:81-103, 1997.
Summerfelt, A.T.; Alphs, L.D.; Wagman, A.M.I.;
Funderburk, F.R.; Hierholzer, R.M.; and Strauss, M.E.
Reduction of perseverative errors in patients with schizophrenia using monetary feedback. Journal of Abnormal
Psychology, 100:613-616, 1991.
Schwarzkopf, S.B.; Crilly, J.F.; and Silverstein, S.M.
Therapeutic synergism: Optimal pharmacotherapy and
psychiatric rehabilitation to enhance functional outcome
in schizophrenia. Psychiatric Rehabilitation Skills,
3:124-147, 1999.
Silverstein, S.M.; Hitzel, H.; and Schenkel, L. Cognitive barriers to rehabilitation readiness: Strategies for identification
and intervention. Psychiatric Services, 49:34-36,1998a.
van Der Gaag, M. The Results of Cognitive Training in
Schizophrenic Patients. Delft, The Netherlands: Eburon
Publishers, 1992.
Silverstein, S.M.; Light, G.A.; and Palumbo, D.R. The
Sustained Attention Test: A measure of cognitive dysfunction. Computers in Human Behavior, 14:463-475, 1998ft.
van Zomeran, A.H.; Brouwer, W.H.; and Deelman, B.G.
Attentional deficits: The riddles of selectivity, speed, and
alertness. In: Brooks, D.N., ed. Psychological Deficits
after Head Injury. London, UK: Oxford University Press,
1984. pp. 74-107.
Silverstein, S.M.; Osborn, L.M.; and Palumbo, D.R. ReyOsterreith Complex Figure Test performance in acute,
chronic, and outpatient schizophrenia patients. Journal of
Clinical Psychology, 54:985-994, 1998c.
Velligan, D.I., and Bow-Thomas, C.C. Two case studies
of cognitive adaptation training for outpatients with schizophrenia. Psychiatric Services, 51:25-29, 2000.
Ventura, J.; Lukoff, D.; Nuechterlein, K.H.; Liberman,
R.P.; Green, M.F.; and Shaner, A. Manual for the
Expanded Brief Psychiatric Rating Scale, version 4.0.
International Journal of Methods in Psychiatric Research,
3:221-224, 1993.
Silverstein, S.M.; Pierce, D.L.; Saytes, M.; Hems, L.;
Schenkel, L.; and Streaker, N. Behavioral treatment of
attentional dysfunction in chronic, treatment-refractory
schizophrenia. Psychiatric Quarterly, 69:95-105, 1998a1.
Silverstein, S.M.; Schenkel, L.S.; Valone, C ; and
Nuernberger, S. Cognitive deficits and psychiatric rehabilitation outcomes in schizophrenia. Psychiatric Quarterly,
69:169-191, 1998e.
Vollema, M.; Guertsen, G.; and Van Voorst, A. Durable
improvements in Wisconsin Card Sort Test performance
in schizophrenic patients. Schizophrenia
Research,
16:209-215, 1995.
Silverstein, S.M.; Valone, C ; Jewell, T.C.; Corry, R.;
Nghie'm, K.; Saytes, M.; and Potrude, S. Integrating shaping and skills training techniques in the treatment of
chronic, treatment-refractory individuals with schizophrenia. Psychiatric Rehabilitation Skills, 3:41-58, 1999.
Wagner, B.R. The training of attending and abstracting
responses in chronic schizophrenia. Journal of
Experimental Research in Personality, 3:77-88, 1968.
Sohlberg, M.M.; McLaughlin, K.A.; Pavese, A.; Heidrich,
A.; and Posner, M.I. Evaluation of attention process training
in persons with acquired brain injury. Journal of Clinical
and Experimental Neuropsychology, 22:656-676, 2000.
Wallace, C.J., and Boone, S.E. Cognitive factors in the
social skills of schizophrenic patients: Implications for
treatment. In: Spaulding, W.D., and Cole, J.C., eds.
256
Shaping Attention Span
Schizophrenia Bulletin, Vol. 27, No. 2, 2001
Nebraska Symposium on Motivation, 1983: Theories of
Schizophrenia and Psychosis. Lincoln, NE: University
of Nebraska Press, 1983. pp. 283-318.
Wallace, C.J.; Liberman, R.P.; MacKain, S.J.;
Blackwell, G.; and Eckman, T.A. Effectiveness and
replicability of modules for teaching social and instrumental skills to the severely mentally ill. American
Journal of Psychiatry, 149:654-658, 1992.
Wexler, B.; Hawkins, K.; Rounsaville, B.; Anderson,
M.; Semyak, M.; and Green, M. Normal neurocognitive
performance after extended practice in patients with
schizophrenia. Schizophrenia Research, 26:173-180,
1997.
Wykes, T.; Reeder, C ; Corner, J.; Williams, C ; and
Everitt, B. The effects of neurocognitive remediation on
executive processing in patients with schizophrenia.
Schizophrenia Bulletin, 25(2):291-307, 1999.
Zarlock, S.P. Social expectations, language, and schizo-
257
phrenia. Journal of Humanistic Psychology, 6:68-74,
1966.
The Authors
Steven M. Silverstein, Ph.D., is Associate Professor of
Psychology in Psychiatry, Weill College of Medicine of
Cornell University, New York, NY, and Director, Second
Chance Program and the Schizophrenia Neurocognitiion
Laboratory at New York Presbyterian Hospital, White
Plains, NY. Anthony A. Menditto, Ph.D., is Director of
Psychosocial Rehabilitation at Fulton State Hospital,
Fulton, MO, and Clinical Assistant Professor of
Psychiatry at the University of Missouri School of
Medicine, Columbia, MO. Paul Stuve, Ph.D., is
Psychosocial Rehabilitation Program Coordinator at
Fulton State Hospital and Clinical Assistant Professor of
Psychiatry at the University of Missouri School of
Medicine, Fulton, MO.