Schizophrenia Bulletin vol. 32 no. 2 pp.

238–245, 2006
doi:10.1093/schbul/sbj013
Advance Access publication on October 27, 2005

The Structure of Negative Symptoms Within Schizophrenia:

Implications for Assessment

Jack J. Blanchard1,2 and Alex S. Cohen2 toms in schizophrenia. These instruments include the
2
Department of Psychology, University of Maryland, Scale for the Assessment of Negative Symptoms
College Park (SANS),1, 2 the Positive and Negative Syndrome Scale
(PANSS),3 and, for the assessment of primary and endur-

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at UCSF Library on March 14, 2015

ing negative symptoms, the Schedule for the Deficit Syn-
This review examines the structural validity of negative drome (SDS).4 Given the last 15–20 years of research on
symptoms focusing on 2 questions: (1) Do negative symp- these instruments, it is an appropriate time to evaluate
toms represent a domain separate from other symptoms in our understanding of the measurement of negative symp-
schizophrenia? and (2) Within negative symptoms, is there toms. In particular, the current review will focus on
a structure that suggests multidimensionality? Results from the structural validity of negative symptoms.5 Structural
exploratory and confirmatory factor analytic studies are validity can be examined using factor analysis in order
examined to address these questions. Across studies and to address important theoretical questions regarding
symptom instruments, negative symptoms appear to consis- the intercorrelations of schizophrenic symptomatology.
tently emerge as a factor separate from other dimensions Specifically, 2 structural questions arise: (1) Do nega-
of the illness in schizophrenia. Whether 2-, 3-, or 5-factor tive symptoms represent a domain separate from other
models are identified, negative symptoms consistently load symptoms in schizophrenia? and (2) Within negative
on a factor separate from positive symptoms, affective symptoms, is there a structure that suggests multidimen-
symptoms of depression or anxiety, and symptoms of dis- sionality (ie, are there different components of negative
organization. Focusing on negative symptoms themselves, symptoms that may require separate assessment so as
factor analytic findings suggest that this construct is mul- to provide adequate measurement of the broader con-
tidimensional with at least 2 factors (involving diminished struct)? The following review will address these questions
expression and anhedonia-asociality). Although these fac- in turn.
tors were replicable, serious limitations were noted in this
literature. Thus, 2- (or even 3- or 5-) factor models of neg- The Overall Structure of Schizophrenic Symptomatology
ative symptoms should not be considered definitive, but A large number of studies have now examined the valid-
rather all converge to support the general conclusion of ity of the distinction between positive and negative symp-
the multidimensionality of negative symptoms. The later toms (for a review, see Peralta and Cuesta 20016). These
findings indicate the importance of employing assessments have included both exploratory factor analytic (EFA)
that provide adequate coverage of the broad domain of studies, as well as more rigorous testing of competing
negative symptoms. Importantly, caution is noted in the models utilizing confirmatory factor analysis (CFA), in-
interpretability of findings based on existing instruments, cluding a meta-analytic CFA.7 A number of symptom
and implications for future assessment are discussed. measures have been studied, including the SANS, the
Scale for the Assessment of Positive Symptoms (SAPS),8
Key words: negative symptoms/factor analysis/ the Brief Psychiatric Rating Scale (BPRS),9 and the
SANS/PANSS PANSS,3 among others. The following general summary
will provide an examination of results across instruments
Although there is compelling evidence regarding the clin- that include items or scales tapping negative symptoms.
ical and theoretical importance of negative symptoms, Attention to a variety of instruments allows for an ap-
a critical issue concerns how best to measure this domain preciation of how characteristics of input variables (the
of phenomenology within schizophrenia. A variety of type and number of symptoms) have a direct impact
instruments have become available to rate negative symp- on the number of factors that emerge from studies of
schizophrenic symptomatology. Relatedly, the examina-
1
To whom correspondence should be addressed; e-mail: tion of factor structure across instruments provides an
jblanchard@psyc.umd.edu. informative test of the robustness of a negative symptom
Ó The Author 2005. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved.
For permissions, please email: journals.permissions@oxfordjournals.org.
238
 The Structure of Negative Symptoms Within Schizophrenia

factor (ie, is this factor replicable despite differences in ful, as the SAPS and SANS do not include evaluation of
the type and number of symptoms assessed in different important clinical symptoms such as depression and anx-
instruments?). iety. The BPRS is a frequently utilized clinical rating in-
Early exploratory factor analyses of symptom assess- strument that assesses a range of symptomatology. Using
ments tapping positive and negative symptoms (eg, relying CFA techniques, Mueser, Curran, and McHugo31 exam-
on the combined SANS and SAPS) initially suggested a 2- ined the factor structure of the BPRS in schizophrenia.
dimensional model composed of independent and largely After trimming 2 items (disorientation and excitement),
uncorrelated positive and negative symptom factors.10–22 results indicated support for a 4-factor model involving
However, subsequent studies have found the 2-factor thought disturbance (positive symptoms), anergia (nega-
model inadequate. Three-factor models,10–12, 18,23–27 tive symptoms of blunted affect, emotional withdrawal,
have consisted of positive and negative symptom factors motor retardation), affect (depression and anxiety), and
with a third factor that, while differing in precise item con- disorganization (conceptual disorganization). Longitudi-
tent across studies, is often described as a disorganiza- nal analyses of individuals with schizophrenia diagnoses
tion factor (typically involving thought disturbance and have demonstrated that this 4-factor model is stable over

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at UCSF Library on March 14, 2015

bizarre behavior but at times including poverty of content time.32 As with the 3-factor solutions for the SAPS/
of speech and attentional impairmenteg,12,18). In a meta- SANS, the 4-factor models derived from the BPRS again
analysis of the SANS and positive symptom ratings, provide support for independent factors of positive and
Grube, Bilder, and Goldman28 found that data across negative symptoms. Importantly, negative symptoms
10 empirical studies fit the 3-factor model involving posi- were also independent of symptoms of depression and
tive, negative, and conceptual disorganization factors. anxiety. Similar to the 3-factor model based on the
Other 3-factor models have found the third factor to in- SANS and SAPS, a disorganization factor was also rep-
volve disordered social adjustment.29 licated. The limitation of BPRS-based factor analytic
Results from these studies provide compelling support studies is that the BPRS fails to assess the full range of
for the distinctiveness of negative symptoms, but they negative symptoms (eg, it lacks items relating to anhedo-
have also indicated that some symptoms originally con- nia or asociality). Thus, the BPRS-based factor structure
sidered to be related to negative symptoms may fit poorly may not adequately represent the structure of symptoms
with the negative symptom factor. Potentially problem- within schizophrenia.
atic symptoms include inappropriate affect, poverty The PANSS3 was developed to tap both positive and
of content of speech, and attentional impairment.10,23 negative symptoms, as well as general psychopathology
Moreover, scale-level analyses have found alogia and at- (adapting 18 items from the BPRS). Over 17 published
tention to load on the negative, as well as on the disor- studies have examined the factor structure of the PANSS
ganization factor.eg,28 Item-level analyses focusing on (see review by Emsley, Rabinowitz, and Torreman33) with
negative symptoms may clarify these issues and will be a 5-factor solution involving negative, positive, activation,
addressed below. dysphoric mood, and autistic preoccupation. The 5-factor
Although the nature of the 3-factor solutions may dif- solution has been supported by both EFA and CFA in a
fer across studies, negative symptoms consistently emerge large multisite study.34 The 5-factor model appears to be
as one of the factors, separate from positive symptoms applicable across both acute and chronic phases of schizo-
and symptoms of disorganization or thought disorder. phrenia35 and has also emerged in both male and female
However, despite the replicability of the 3-factor model, schizophrenia patients.36 Despite the apparent consistency
some data have suggested a more complex factorial struc- with the 5-factor solution, the composition of the negative
ture. Peralta and Cuesta30 conducted an item-level EFA symptom factor within the PANSS pentagonal model has
of the SANS and SAPS within a large sample of individ- varied. Emsley, Rabinowitz, and Torreman33 found that
uals with psychotic disorders, including schizophrenia. in the forced 5-factor solution, the negative symptom fac-
The authors identified 11 factors, 2 of which reflected tor included blunted affect, emotional withdrawal, poor
negative symptoms (poverty of affect/speech and social rapport, passive/apathetic social withdrawal, lack of spon-
dysfunction). Second-order factor analysis did identify taneity and flow of conversation, as well as items from the
higher-order factors that fit the 3-factor solution PANSS ‘‘general’’ scale, including active social avoidance,
described above, but as noted by Peralta and Cuesta,30 motor retardation, and disturbance in volition. Similar
this 3-dimensional model may not adequately account item loadings on the negative symptom factor were
for variance in symptoms assessed. It is interesting that reported by White et al.34 (after trimming 5 items from
even in the multifactorial model obtained by Peralta the PANSS), with the addition of mannerisms and second-
and Cuesta,30 negative symptoms were separate from ary loadings with uncooperativeness and impulsivity. A
other symptoms, with blunted affect and asociality- more restricted solution was found by Nakaya, Suwa,
anhedonia comprising separate (but correlated) factors. and Ohmori,35 using CFA with the negative symptom fac-
Although these findings are informative, consideration tor consisting of blunted affect, poor rapport, and lack of
of broader symptom assessment instruments may be use- spontaneity and flow of conversation.
239
J. J. Blanchard & A. S. Cohen

The pentagonal model of symptoms derived from the then arises as to whether negative symptoms are a unitary
PANSS appears robust; however, other results raise construct or whether this domain is itself best described
questions about the adequacy of this model. Although as multidimensional. As this larger issue is considered, it
Emsley, Rabinowitz, and Torreman33 replicated a forced is important to note that the various negative symptom
5-factor solution that corresponded to earlier factor scales differ in regard to precise content and number of
structures, these investigators also noted that an EFA items included. Excluding global items, the SANS1,2 cur-
yielded a 7-factor structure (with motor, depression, rently consists of 19 items (prior versions included an
and anxiety symptoms emerging as separate factors). additional item of inappropriate affect2 that was sub-
These authors concluded that the apparent uniformity sequently dropped39), representing 5 rationally derived
in 5-factor findings is somewhat misleading.33 Relatedly, scales: Affective Flattening or Blunting, Alogia, Avolition-
two CFA studies of the PANSS have found that the pen- Apathy, Anhedonia-Asociality, and Inattention. The
tagonal model had inadequate goodness of fit.37,38 negative symptom scale from the PANSS3 includes 7
items tapping blunted affect, emotional withdrawal,
poor rapport, passive/apathetic social withdrawal, diffi-
Summary

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at UCSF Library on March 14, 2015

culty in abstract thinking, lack of spontaneity and flow
In examining the literature on the overall structure of of conversation, and stereotyped thinking. The SDS4
symptoms within schizophrenia, it appears that a good consists of 6 negative symptoms involving restricted af-
deal of support has been marshaled for 3- and 5- factor fect, diminished emotional range, poverty of speech,
models. One evident pattern is that broader symptom curbing of interests, diminished sense of purpose, and
assessments yield a larger number of factors. However, diminished social drive.
critical reviews of this literature have noted that these mod- One issue to consider in looking at these different items
els may continue to be inadequate. In a meta-analysis of or subscales is whether they truly represent different
CFA studies (including the SANS, SAPS, BPRS, and (though correlated) aspects of the negative symptom con-
PANSS), Smith, Mar, and Turoff noted that ‘‘even three struct. If negative symptoms are indeed multidimen-
factors do not capture the structure among schizophrenic sional, this would indicate the need to ensure broader
symptoms. Either more latent variables underlie the inter- assessment of these multiple domains and may suggest
correlations among schizophrenic symptoms, or the set of that these aspects of negative symptoms may have unique
symptoms most often studied is incomplete.’’7(p.67) Peralta etiological, functional, or treatment correlates. Alterna-
and Cuesta6 have also cogently summarized the limitations tively, if a single factor were to emerge, this could indicate
of the extant literature, including the need to consider var- an opportunity to simplify assessment with a focus on
iability in data reduction techniques, the method used for a smaller set of representative items.
selecting and rotating factors, assessment instruments As has been noted, a variety of instruments are avail-
used, whether items or scale scores are employed in anal- able to assess negative symptoms. However, factor ana-
yses, and the characteristics of the clinical sample studied. lytic work that exclusively examines the structure of
Despite the utilization of different instruments across negative symptoms has largely focused on the SANS.
different patient populations and the application of a The attention given to this measure likely arises from
variety of factor analytic methods, the one replicable the fact that this was the earliest instrument to measure
finding has been that negative symptoms emerge as a negative symptoms. Additionally, the SANS represents
factor separate from other dimensions of the illness in a larger number of items (20 originally, 19 in the current
schizophrenia. Whether 2-, 3-, or 5-factor models are version) versus some other instruments (eg, the PANSS
identified, negative symptoms consistently load on a fac- includes 7 negative symptom items), and thus the SANS
tor separate from positive symptoms, affective symptoms may provide more interpretable factor findings. Factor
of depression or anxiety, and symptoms of disorganiza- analyses on the SDS4 are not yet available, as this instru-
tion. Even when more complicated 11-factor models have ment typically focuses on the generation of dichotomous
been identified,30 negative symptoms stand apart from classification into deficit and nondeficit subtypes rather
other symptoms (though blunted affect and asociality- than dimensional ratings of symptoms. The current re-
anhedonia separate into correlated factors). With evi- view will focus on studies examining the structure of
dence that negative symptoms form a distinct factor the SANS.
within broad assessments of schizophrenic symptomatol- In the first reported CFA examining the structure of
ogy, we can next focus our attention on understanding the SANS within schizophrenia, Keefe et al.40 found sup-
the underlying structure of negative symptoms. port for a 3-factor model involving diminished expres-
sion, social dysfunction, and disorganization. While
suggestive of a multidimensional nature of negative
The Structure of Negative Symptoms.
symptoms, this study is limited in that only 13 of the
Given that negative symptoms represent a separate di- original 20 SANS items were included, so that a previous
mension of the illness in schizophrenia, the question 2-factor model (obtained using principal component
240
 The Structure of Negative Symptoms Within Schizophrenia

analysis) identified by Liddle18 could be directly tested. ication. Using the 20-item SANS, 2 factors emerged
Additionally, the items used in the CFA included ‘‘inap- across assessments: Affective Flattening (involving items
propriate affect,’’ which has been dropped from later from the Affective Flattening or Blunting scale, as well as
versions of the SANS41 because it is conceptually incon- the ‘‘poverty of speech’’ item) and Diminished Motiva-
sistent with the construct of negative symptoms. Thus, tion (including items from the Avolition-Apathy and
while suggestive, the findings of Keefe et al.40 are limited Anhedonia-Asociality scales). Other factors that emerged
in understanding the structure of the full range of symp- inconsistently across assessments were related to what
toms assessed by the SANS. the authors described as disorganization items, includ-
Using CFA, Peralta and Cuesta42 examined the full 20- ing social inattentiveness, blocking, latency of response,
item version of the SANS within a sample of schizophre- poverty of content, inattentiveness during mental testing,
nia patients. They concluded that a model based on the and inappropriate affect. Interestingly, the use of inatten-
original 5, empirically derived scales fit ‘‘reasonably well’’ tiveness, poverty of content, and inappropriate affect
and that the best-fitting model involved trimming the ‘‘in- have raised questions in other factor studies.eg,43
appropriate affect’’ item. Although support for 5 factors In summary, an accumulation of factor analytic studies

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at UCSF Library on March 14, 2015

was obtained, results also indicated high intercorrelations have demonstrated that the structure of negative symp-
among some of the factors, with the highest correlations toms is not unidimensional. Multidimensional models
obtained between Affective Flattening and Alogia (.76) have been indicated in both EFA and CFA. The precise
and between Anhedonia-Asociality and Avolition- number of factors that best represents negative symptoms
Apathy (.69). Importantly, Sayers, Curran, and Mueser43 (as measured by the SANS) is somewhat unclear; how-
observed that none of the models tested by Peralta and ever, the most reliable domains to emerge appear to relate
Cuesta42 fit well, with even the best models having to diminished expression and a combined anhedonia-
high v2:df ratios and fit indices less than the recommen- asociality factor. Factors reflecting diminished expres-
ded cutoff of .90.44,45 Additionally, Sayers, Curran, and sion and anhedonia-asociality have consistently emerged
Mueser43 noted that the 5-factor model involved a factor in CFA analyses42,43,49 and EFA analyses of the SANS.48
with 2 indicators, and models with 2 indicators tend to be These 2 negative symptom factors also emerged in an
unstable.46,47 EFA of the combined SANS/SAPS.30
In the most comprehensive study of the SANS to date, Data further suggest that a 3-factor model might be
Sayers, Curran, and Mueser43 examined the structure of more parsimonious than the rationally derived 5 factors
negative symptoms within a large sample of schizophre- of the SANS.43 However, the Inattention-Alogia factor
nia patients (N = 457) over 2 assessments. They utilized obtained by Sayers, Curran, and Mueser43 was highly cor-
CFA and EFA to test various models. An examination of related with the Diminished Expression factor (r = .83).
the original 5-factor structure using CFA yielded poor fit. The reasons for this high intercorrelation are unclear,
A subsequent EFA on the data yielded 3 factors (repli- but items such as poverty of speech (which had dual load-
cating an earlier EFA on a subset of these subjects ing with Diminished Expression and Inattention-Alogia)
reported by Mueser et al.41) corresponding to Diminished and increased latency of response (with a somewhat ele-
Expression (including items from the Affective Flattening vated loading on Diminished Expression) may relate to
or Blunting scale, as well as the ‘‘poverty of speech’’ general decreased expressivity. The inclusion of inatten-
item), Inattention-Alogia (which included items from tion in ratings of negative symptoms is also somewhat
the Inattention and Alogia scales, as well as the ‘‘poor problematic, as questions have been raised as to whether
eye contact’’ item), and Social Amotivation (reflecting attentional problems are indeed conceptually related to
items from the Anhedonia-Asociality and Avolition- the negative symptom construct.
Apathy subscales). Subsequent CFA was used to cross- It is interesting to speculate about the implications for
validate the EFA within the same sample at a second the 2-factor model of negative symptoms with regard to
assessment. A 3-factor model was supported with 3 intervention studies. Since the Diminished Expression
correlated factors involving Diminished Expression, In- factor reflects behaviors evident during direct observa-
attention-Alogia, and Social Amotivation. Factor inter- tion within a clinical interview, one might expect these
correlations were .83 for Diminished Expression and behaviors to show a different time course of response
Inattention-Alogia factors, .56 between Inattention- compared with the anhedonia-asociality symptoms that
Alogia and Social Amotivation, and .56 between the Di- reflect more global social engagement occurring in the
minished Expression and Social Amotivation factors. community. Specifically, changes in expressivity ob-
Validity analyses indicated that the Social Amotivation served during an interview might be more quickly evident
factor had unique correlates with independent ratings compared with improvement in the number and quality
of social functioning and treatment outcome.43 of social relationships developing within the community
In an EFA of the SANS in schizophrenia, Kelley, van (which presumably would improve more slowly as the de-
Kammen, and Allen48 examined the factor structure of velopment of relationships depends on complex changes
negative symptoms while patients were on and off med- in long-standing social networks). Along similar lines, it
241
J. J. Blanchard & A. S. Cohen

is possible to parse the Anhedonia-Asociality factor to and more extensive assessments (involving family or the
determine if changes in interest, drive, or anticipated observations of other treatment providers55) might have
pleasure in social interactions appear earlier during an on the obtained structure is unclear but may be relevant
intervention than actual social success. in considering the interpretability and generalizability of
these findings.
Conceptual issues regarding assessment are also impor-
Dimensions and Subtypes tant in the interpretation of findings obtained with current
As one considers the structure of negative symptoms, an instruments. In looking at the negative symptom factors
inevitable question that arises is if this domain is best con- of diminished expression and anhedonia-asociality, a vari-
sidered a pure dimension with individuals varying in de- ety of interpretations are available to explain what may un-
gree of severity, or if there is a distinct subtype of negative derlie these domains. One might conjecture that these
symptom schizophrenia. This latter categorical model is factors reflect underlying processes associated with the
reflected in typological assignments that can be obtained illness.43 Thus, the factor of diminished expression may
with clinical rating scales2,3 and in the ‘‘Deficit Syndrome,’’ emerge from the deficits in emotional expression that

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at UCSF Library on March 14, 2015

which explicitly adopts a model that enduring, primary have been identified in laboratory paradigms,eg,56–59 while
negative symptoms (or deficit symptoms) reflect a dis- the anhedonia-asociality domain may reflect the separate
tinct subtype of schizophrenia.50,51 The reviewed factor domain of the experiential aspects of emotion in schizo-
analytic studies do not address whether a categorical dis- phrenia, including the reduced capacity to experience plea-
tinction may characterize negative symptoms. Blanchard, sure.60,61 An alternative interpretation is that these 2
Horan, and Collins52 utilized taxometric statistical pro- factors reflect measurement influences unrelated to dis-
cedures53 to determine if negative symptoms are purely tinct processes that are related to the illness or more
dimensional or if a latent class (or taxon) was also a fea- broadly to emotion. Items reflecting diminished expres-
ture of these symptoms. Taxometric analyses yielded sion or blunted affect may cohere because they are all
findings consistent with a latent class of negative symp- related to behavioral observations of interpersonal expres-
tom schizophrenia with a base rate of approximately sion observed during the interview (decreased eye contact,
28% to 36%. Compared with the other schizophrenia paucity of gestures, lack of facial expression or vocal into-
patients, members of this latent class or taxon were nation). With regard to anhedonia-asociality items,
more likely to be male and to have poorer social func- as summarized in this issue by Horan, Kring, and
tioning. The results of this study suggest that negative Blanchard,62 it may be problematic to interpret these
symptoms may be informative in delimiting the pheno- items as indicators of emotional experience. Rather,
typic heterogeneity of schizophrenia and may allow for anhedonia-asociality items may aggregate as a separate
the identification of a subtype for further study regarding factor because they tap general social activity outside of
etiological mechanisms and targets for treatment. the interview that is largely based on patients’ self-reports
(recreational interests, sexual interest and activity, ability
to feel intimacy, and relationships with peers and family).
Summary A critical issue in considering this literature is the mun-
As with factor analyses conducted on broader domains of dane but important fact that factor analytic solutions
symptoms, the interpretability of results examining the depend on the variables used in analyses.6,7 Results
structure of the SANS is limited by a variety of method- will relate to the number of symptoms assessed, as well
ological and conceptual issues. Studies have varied as basic psychometric properties of input variables,
greatly in the samples studied, whether items or subscale including the reliability of individual symptoms that
scores are analyzed, which items are excluded or later comprise the scales. The above factor solutions are
trimmed (with studies examining 13, 18, 19, and 20 constrained by reliance on the SANS. As noted by
SANS items), and how SANS assessments were con- Smith, Mar, and Turoff,7 although there may be consen-
ducted. Investigatorseg43,54 have emphasized these limita- sus about what symptoms should be studied in the assess-
tions in addition to noting that important issues, such as ment of schizophrenic symptomatology, consensus does
the distinction between primary and secondary negative not ensure comprehensiveness. For example, Horan,
symptoms,50 were not addressed when utilizing the Kring, and Blanchard62 make the case that our current
SANS. Other important issues concern the source of understanding of anhedonia-asociality may be severely
information and time frame sampled. As noted by Sayers, constrained by instrument limitations, including a poten-
Curran, and Mueser,43 SANS ratings in their study fo- tial failure to adequately address experiential deficits in-
cused only on the week prior to the interview and relied dependent from behavioral achievements. While the
exclusively on patient reports and observations during available evidence clearly supports a multidimensional
the interview (deviating from Andreasen’s39 recommen- structure of negative symptoms, the development of
dation that ratings should ideally be based on multiple more extensive assessment instruments may very well
sources of information). What role broader time frames yield a different latent structure.
242
 The Structure of Negative Symptoms Within Schizophrenia

Based on current findings, it will be informative to ei- References

ther modify current negative symptom scales or develop
1. Andreasen NC. Negative symptoms in schizophrenia:
entirely new items and assess the impact of these modi- definition and reliability. Arch Gen Psychiatry. 1982;39:
fications on structural findings and external correlates. 784–788.
One strategy would be to disaggregate complex items 2. Andreasen NC. Scale for the Assessment of Negative Symp-
or scales into component parts so that precision of mea- toms. Iowa City: University of Iowa Press; 1983.
surement can be enhanced. For example, items can be 3. Kay SR, Fiszbein A, Opler LA. The Positive and Negative
modified to more accurately assess the potentially distinct Syndrome Scale (PANSS) for schizophrenia. Schizophr Bull.
domains of anhedonia (as this relates to actual emotional 1987;13:261–276.
experience) and asociality (as relating to behavioral indi- 4. Kirkpatrick B, Buchanan RW, McKenney PD, Alphs LD,
cators of social activity).62 As content is reviewed for the Carpenter WT Jr. The Schedule for the Deficit Syndrome:
an instrument for research in schizophrenia. Psychiatry Res.
development of new items, it may be particularly fruitful 1989;30:119–123.
to consider the literature from basic research on emotion 5. Loevinger J. Objective tests as instruments of psychological
and social behavior. The performance of modified assess- theory. Psychol Rep. 1957;3:635–694.

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at UCSF Library on March 14, 2015

ment scales can then be examined with regard to their 6. Peralta V, Cuesta MJ. How many and which are the psycho-
structural and external validity.5 pathological dimensions in schizophrenia? issues influencing
their ascertainment. Schizophr Res. 2001;49:269–285.
7. Smith DA, Mar CM, Turoff BK. The structure of schizo-
phrenic symptoms: a meta-analytic confirmatory factor anal-
Conclusions ysis. Schizophr Res. 1998;31:57–70.
An accumulation of evidence supports the structural 8. Andreasen NC. Scale for the Assessment of Positive Symp-
toms. Iowa City: University of Iowa Press; 1984.
validity of negative symptoms. In broad assessments
9. Overall JE, Gorham DR. The Brief Psychiatric Rating Scale.
of schizophrenic symptomatology, negative symptoms Psychol Rep. 1962;10:799–812.
have repeatedly emerged as a separate factor, indepen- 10. Andreasen NC, Arndt S, Miller D, Flaum M, Nopoulos P.
dent of positive symptoms, disorganization, and affective Correlational studies of the Scale for the Assessment of Neg-
symptoms, including depression and anxiety. Additional ative Symptoms and the Scale for the Assessment of Positive
research focusing on the latent structure of negative Symptoms: an overview and update. Psychopathology. 1995;
symptoms themselves suggests that this symptom domain 28:7–17.
is not unidimensional. The most reliable factors to 11. Arndt S, Alliger RJ, Andreasen NC. The distinction of posi-
emerge within negative symptoms include diminished ex- tive and negative symptoms: the failure of a two-dimensional
model. Br J Psychiatry. 1991;158:317–322.
pression (typically involving symptoms of reduced facial
12. Bilder RM, Mukherjee S, Rieder RO, Pandurangi AK.
and vocal expressivity and reduced verbal output) and
Symptomatic and neuropsychological components of defect
a factor tapping anhedonia and asociality (composed states. Schizophr Bull. 1985;11:409–419.
of symptoms of anhedonia, diminished interest, and de- 13. Green M, Walker E. Neuropsychological performance and
creased social engagement). Other factors may involve positive and negative symptoms in schizophrenia. J Abnorm
aspects of alogia and inattention. Although these factors Psychol. 1985;94:460–469.
were replicable, serious limitations were noted in this lit- 14. Johnstone EC, Owens DG, Gold A, Crow TJ, MacMillan JF.
erature. Thus, 2- (or even 3- or 5-) factor models should Institutionalization and the defects of schizophrenia. Br J
not be considered definitive, but rather all converge to Psychiatry. 1981;139:195–203.
support the general conclusion of the multidimensional- 15. Kay SR, Opler LA, Fiszbein A. Significance of positive and
ity of negative symptoms. The later findings indicate the negative syndromes in chronic schizophrenia. Br J Psychiatry.
1986;149:439–448.
importance of employing assessments that provide
16. Kay SR, Opler LA, Lindenmayer JP. Reliability and validity
adequate coverage of the broad domain of negative of the Positive and Negative Syndrome Scale for schizo-
symptoms. Importantly, a number of limitations in our phrenics. Psychiatry Res. 1988;23:99–110.
current understanding of the structure of negative symp- 17. Lewine RR, Fogg L, Meltzer HY. Assessment of negative and
toms were noted, including potentially critical issues in positive symptoms in schizophrenia. Schizophr Bull. 1983;
how symptoms are assessed. The current findings may 9:368–376.
help to inform the development of future instruments 18. Liddle PF. The symptoms of chronic schizophrenia: a re-
for the assessment of negative symptoms. examination of the positive-negative dichotomy. Br J Psychi-
atry. 1987;151:145–151.
19. Lindenmayer JP, Kay SR, Friedman C. Negative and positive
schizophrenic syndromes after the acute phase: a prospective
Acknowledgments follow-up. Compr Psychiatry. 1986;27:276–286.
20. Losonczy MF, Song IS, Mohs RC, et al. Correlates of lateral
Preparation of this article was supported by Grant ventricular size in chronic schizophrenia: I. behavioral and
MH-51240 from the National Institute of Mental treatment response measures. Am J Psychiatry. 1986;143:
Health (Dr. Blanchard). 976–981.

243
J. J. Blanchard & A. S. Cohen

21. Pogue-Geile MF, Harrow M. Negative and positive symp- in schizophrenia: negative symptoms. Psychiatry Res. 1992;
toms in schizophrenia and depression: a follow-up. Schizophr 44:153–165.
Bull. 1984;10:371–387. 41. Mueser KT, Sayers SL, Schooler NR, Mance RM, Haas GL.
22. Rosen WG, Mohs RC, Johns CA, et al. Positive and negative A multisite investigation of the reliability of the Scale for
symptoms in schizophrenia. Psychiatry Res. 1984;13:277–284. the Assessment of Negative Symptoms. Am J Psychiatry.
23. Andreasen NC, Arndt S, Alliger R, Miller D, Flaum M. 1994;151:1453–1462.
Symptoms of schizophrenia: methods, meanings, and mecha- 42. Peralta V, Cuesta MJ. Negative symptoms in schizophrenia:
nisms. Arch Gen Psychiatry. 1995;52:341–351. a confirmatory factor analysis of competing models. Am J
24. Arndt S, Andreasen NC, Flaum M, Miller D, Nopoulos P. A Psychiatry. 1995;152:1450–1457.
longitudinal study of symptom dimensions in schizophrenia: 43. Sayers SL, Curran PJ, Mueser KT. Factor structure and con-
prediction and patterns of change. Arch Gen Psychiatry. 1995; struct validity of the Scale for the Assessment of Negative
52:352–360. Symptoms. Psychol Assess. 1996;8:269–289.
25. Malla AK, Norman RM, Williamson P, Cortese L, Diaz F. 44. Bentler PM, Bonett DG. Significance tests and goodness of fit
Three syndrome concept of schizophrenia: a factor analytic in the analysis of covariance structures. Psychol Bull. 1980;
study. Schizophr Res. 1993;10:143–150. 88:588–606.
26. Kulhara P, Chandiramani K. Positive and negative subtypes 45. Bentler PM, Chou C. Practical issues in structural modeling.

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at UCSF Library on March 14, 2015

of schizophrenia: a follow-up study from India. Schizophr In: Long, JS, ed. Common Problems/Proper Solutions: Avoid-
Res. 1990;3:107–116. ing Error in Quantitative Research. Newbury Park, Calif:
27. Thompson PA, Meltzer HY. Positive, negative, and disorga- Sage; 1988:78–117.
nisation factors from the Schedule for Affective Disorders 46. Boomsma A. The robustness of LISREL against small sam-
and Schizophrenia and the Present State Examination: ple sizes in factor analysis models. In: Joreskog KG, Wold H,
a three-factor solution. Br J Psychiatry. 1993;163:344–351. ed. Systems Under Indirect Observation: Causality, Structure,
28. Grube BS, Bilder RM, Goldman RS. Meta-analysis of symp- Prediction. Vol 1. Amsterdam: North Holland; 1982:78–117.
tom factors in schizophrenia. Schizophr Res. 1998;31:113–120. 47. McDonald R. Factor Analysis and Related Methods. Hill-
29. Lenzenweger MF, Dworkin RH, Wethington E. Examining sdale, NJ: Erlbaum; 1985.
the underlying structure of schizophrenic phenomenology: 48. Kelley ME, van Kammen DP, Allen DN. Empirical validation
evidence for a three-process model. Schizophr Bull. 1991; of primary negative symptoms: independence from effects of
17:515–524. medication and psychosis. Am J Psychiatry. 1999;156:406–411.
30. Peralta V, Cuesta MJ. Dimensional structure of psychotic 49. Keefe RS, Lobel DS, Mohs RC, et al. Diagnostic issues in
symptoms: an item-level analysis of SAPS and SANS symp- chronic schizophrenia: kraepelinian schizophrenia, undiffer-
toms in psychotic disorders. Schizophr Res. 1999;38:13–26. entiated schizophrenia, and state-independent negative symp-
31. Mueser KT, Curran PJ, McHugo GJ. Factor structure of toms. Schizophr Res. 1991;4:71–79.
the Brief Psychiatric Rating Scale in schizophrenia. Psychol 50. Carpenter WT Jr., Heinrichs DW, Wagman AM. Deficit
Assess. 1997;9:196–204. and nondeficit forms of schizophrenia: the concept. Am J
32. Long JD, Brekke JS. Longitudinal factor structure of Psychiatry. 1988;145:578–583.
the Brief Psychiatric Rating Scale in schizophrenia. Psychol 51. Kirkpatrick B, Buchanan RW, Ross DE, Carpenter WT Jr.
Assess. 1999;11:498–506. A separate disease within the syndrome of schizophrenia.
33. Emsley R, Rabinowitz J, Torreman M. The factor structure Arch Gen Psychiatry. 2001;58:165–171.
for the Positive and Negative Syndrome Scale (PANSS) in 52. Blanchard JJ, Horan WP, Collins LM. Examining the latent
recent-onset psychosis. Schizophr Res. 2003;61:47–57. structure of negative symptoms: is there a distinct subtype of neg-
34. White L, Harvey PD, Opler L, Lindenmayer JP. Empirical ative symptom schizophrenia? Schizophr Res. 2005;77:151–165.
assessment of the factorial structure of clinical symptoms in 53. Waller NG, Meehl PE. Multivariate Taxometric Procedures:
schizophrenia: a multisite, multimodel evaluation of the fac- Distinguishing Types From Continua. Thousand Oaks, Calif:
torial structure of the Positive and Negative Syndrome Scale. Sage; 1998.
The PANSS Study Group. Psychopathology. 1997;30:263–274. 54. Peralta V, Cuesta MJ, de Leon J. Positive and negative
35. Nakaya M, Suwa H, Ohmori K. Latent structures underlying symptoms/syndromes in schizophrenia: reliability and validity
schizophrenic symptoms: a five-dimensional model. Schizophr of different diagnostic systems. Psychol Med. 1995;25:
Res. 1999;39:39–50. 43–50.
36. Hayashi N, Igarashi Y, Yamashina M, Suda K. Is there a 55. Ho BC, Flaum M, Hubbard W, Arndt S, Andreasen NC.
gender difference in a factorial structure of the Positive and Validity of symptom assessment in psychotic disorders: infor-
Negative Syndrome Scale? a test by structural equation mation variance across different sources of history. Schizophr
modeling. Psychopathology. 2002;35:28–35. Res. 2004;68:299–307.
37. Fitzgerald PB, de Castella AR, Brewer K, et al. A confirma- 56. Aghevli MA, Blanchard JJ, Horan WP. The expression and
tory factor analytic evaluation of the pentagonal PANSS experience of emotion in schizophrenia: a study of social
model. Schizophr Res. 2003;61:97–104. interactions. Psychiatry Res. 2003;119:261–270.
38. Lykouras L, Oulis P, Psarros K, et al. Five-factor model of 57. Berenbaum H, Oltmanns TF. Emotional experience and ex-
schizophrenic psychopathology: how valid is it? Eur Arch pression in schizophrenia and depression. J Abnorm Psychol.
Psychiatry Clin Neurosci. 2000;250:93–100. 1992;101:37–44.
39. Andreasen NC. The Scale for the Assessment of Negative 58. Kring AM, Kerr SL, Smith DA, Neale JM. Flat affect in
Symptoms (SANS): conceptual and theoretical foundations. schizophrenia does not reflect diminished subjective experi-
Br J Psychiatry Suppl. 1989;49–58. ence of emotion. J Abnorm Psychol. 1993;102:507–517.
40. Keefe RS, Harvey PD, Lenzenweger MF, et al. Empirical 59. Neale JM, Blanchard JJ, Kerr S, et al. Flat affect in
assessment of the factorial structure of clinical symptoms schizophrenia. In: Flack WF, Laird JD, eds. Emotions in

244
 The Structure of Negative Symptoms Within Schizophrenia

Psychopathology: Theory and Research. New York: Oxford Flack WF, Laird JD, eds. Emotions in Psychopathology: The-
University Press; 1998:353–364. ory and Research. New York: Oxford University Press;
60. Meehl PE. Schizotaxia, schizotypy, schizophrenia. Am Psy- 1998:336–352.
chol. 1962;17:827–838. 62. Horan WP, Kring AM, Blanchard JJ. Anhedonia in schizo-
61. Blanchard JJ. Hedonic capacity: implications for understand- phrenia: a review of assessment strategies. Schizophr Bull.
ing emotional and social functioning in schizophrenia. In: This issue.

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at UCSF Library on March 14, 2015

245