Personality and Individual Differences 31 (2001) 383±400

www.elsevier.com/locate/paid

What's the use of neuroticism?$

G. Claridge a,*, C. Davis b
a
Department of Experimental Psychology, University of Oxford, 1 South Parks Road, Oxford OX1 3UD, UK
b
Kinesiology and Health Science, York University, Canada

Received 21 January 2000; received in revised form 11 June 2000; accepted 19 July 2000

Abstract
In this paper we examine two aspects of neuroticism (N): its status as a personality descriptor and its role
in the personality dynamics of abnormal states. We ®rst suggest that high N is such a universal accom-
paniment of abnormal functioning (both psychological and biological) that by itself it has little descriptive
or explanatory value. Then, acknowledging that N has more utility when used alongside other personality
variables, we argue that here the most informative are disorder-speci®c characteristics that have unique
variance, while also correlating with N. We propose that N's role in aetiology is that of a moderator vari-
able, in¯uencing the expression of these disorder-speci®c characteristics to produce (or not) the clinical
conditions to which they relate. By way of illustration, examples are taken from our joint studies of the
eating disorders and of schizotypy and schizophrenia. # 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Neuroticism; Personality theory; Multiple regression interactions; Moderator variables; Zone analysis;
Biological basis; Genetics

1. Preamble

The origins of this paper lie in discussions between the authors about neuroticism, prompted
by our collaboration on various clinically based studies that re¯ect our interests in two research
areas: eating disorders and schizotypy/schizophrenia. In all of these investigations we used a
variety of personality scales, relevant to the particular problem of interest; but common to all of
them was the inclusion in the test battery of a measure of neuroticism (N). In the course of ana-
lysing these sets of data we were both drawn to a similar question Ð summed up in the title of
this paper.

* Corresponding author. Tel.: +44-1865-515320; fax: +44-1865-3100447.

E-mail address: gordon.claridge@psy.ox.ac.uk (G. Claridge).
$
This paper is based on a joint presentation of the same title at the 9th Biennial Meeting of the International
Society for the Study of Individual Di€erences held in Vancouver, Canada, July 1999.

0191-8869/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0191-8869(00)00144-6
384 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

Interestingly, in the course of our discussion we realised that, although we had asked the same
question, we had each come to it from a di€erent perspective. Consequently, our conclusions
were also rather di€erent. Indeed they were, to an extent, contradictory, representing two opi-
nions that can be described informally as, on the one hand, the pessimistic view and, on the other
hand, the optimistic view of N. That comparison will form the basis of the structure of the present
paper, the purpose of which is two-fold: ®rst, to articulate a number of observations about neu-
roticism as a personality construct (These seemed worth airing since, while individually not new,
they appear not to have been brought together in one place before); and, secondly, to present two
contrasting, but we believe complementary, answers to the question we have posed.

2. The ubiquity of N

Despite the continuing debate about what constitutes the optimal dimensional structure of
personality, one point of agreement remains. It is the consensus that a signi®cant part of per-
sonality variation can be ascribed to traits concerning an individual's emotional reactivity, ten-
dency to worry, susceptibility to negative mood, proneness to psychopathology, or Ð as it is
most often called Ð neuroticism. This cluster of traits reliably appears in factor analyses of per-
sonality data. It does so whatever else emerges and irrespective of what motivates the research or
which statistical methodology is used; viz in the form of a higher-order composite of more pri-
mary traits (Gorsuch & Cattell, 1967; Krug & Johns, 1986); alongside a varying number of other
largish factors (Cloninger, 1998; Costa & McCrae, 1992); or as one of an exclusive set of con-
structs considered necessary (and sucient) to explain personality (Eysenck & Eysenck, 1985).
Because of its strong face validity and ease of measurement, it is not surprising that a massive
data base has accumulated on N. For years it has been common practice to include N in the test
battery whenever some quantitative measures of personality are seen to be required, or are of
interest. In the broadly clinical domain alone there is a huge variety of phenomena found to be
associated with raised neuroticism. N correlates positively with susceptibility to most sources of
pain (Bru, Myletun & Svebak, 1993; Costa, 1987); all manner of psychosomatic complaints
(Kentle, 1989; Sainsbury, 1960; Yadav, Jain, Bahre & Gupta,1990); the symptoms of manifestly
physical illness, such as the common cold (Carr, 1981); the premenstrual syndrome (van den
Akker, Eves, Stein & Murray, 1995); eating disorders of both the bulimic and anorectic type
(Davis, 1997); the predisposition to (Saklofske, Kelly & Janzen, 1995), and failure to recover
from (Duggan, Lee & Murray, 1990), depression; suicidality (Statham et al., 1998); and with the
abuse of drugs as di€erent as nicotine (Breslau, Kilbey & Andreski, 1993), alcohol (Prescott,
Neale, Corey & Kendler, 1997), and cocaine (Kilbey Breslau & Andreski, 1992). Increased N is
found, predictably, not only in `dysthymic' reactions such as anxiety neurosis (Claridge, 1967;
Eysenck & Eysenck, 1975) and obsessive-compulsive disorder (Slade, 1974), but also in most
personality disorders (Trull, 1992), as well as in acute schizophrenia (Claridge, 1967; McGuire,
Mowbray & Vallance, 1963). Indeed, ®nding high N to be a correlate of deviance is so unre-
markable that it comes as quite a shock to discover exceptions: such as the abnormally low N
scores reported early on by Kissen (1964) in lung cancer patients, and the relatively reduced
neuroticism that can be observed in some individuals su€ering from dissociative (hysterical con-
version) reactions (Ingham & Robinson, 1964). Yet even there a convincing argument can be
 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400 385

made that the reduced N is part of a temporary coping response to stress, thus preserving the idea
that the underlying personality structure of people susceptible to such conditions is also highly
neurotic.
It is clear then, that N is a very potent general indicator of abnormal psychobiological func-
tioning. At the same time, its speci®city is extremely low: it is only when the population of
interest Ð e.g. schizophrenics or middle aged men at risk for heart disease Ð has been de®ned in
advance that N has any use, either as a statistical predictor, or as a guide to understanding causal
in¯uences.
Even appealing to biological accounts of personality does not help to rescue N from its posi-
tion as an over general construct. Ever since it emerged as a major personality dimension, di-
culties have been encountered in trying to conceptualise and demonstrate a biological basis for N.
For example, in Eysenck's (1967) theory Ð the most prominent player in this regard Ð the causal
status of N has always been ambiguous, a problem that, over the years, authors have continued
to articulate (Claridge, 1967; Stelmack, 1981; Fahrenberg, 1992).
The point we are making here can be simply illustrated with some early work by one of us
(G.C.) on di€erential drug e€ects as they relate to the Eysenck dimensions of introversion±
extraversion and neuroticism (Fig. 1). Drug response was measured by the `sedation threshold'
procedure, which involves determining the dose of an intravenously administered barbiturate
required to bring the individual to a pre-de®ned level of drowsiness. Although, for ethical rea-
sons, no longer usable, in its day the technique proved to be extremely powerful in demonstrating
strong di€erences in central nervous arousability, in relation to both normal personality and
psychopathology (Claridge, 1967; Claridge & Herrington, 1960; Shagass, 1954). As the ®gure
illustrates, sedation threshold ®ndings for the Eysenck dimensions were quite clear-cut: neurotic

Fig 1. Diagram illustrating how increasing levels of neuroticism (N) can be associated with both very high and very
low barbiturate sensitivity (sedation threshold).
386 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

introverts had very signi®cantly greater tolerance than neurotic extraverts (true whether assessed
in healthy or in clinical groups). This meant that neuroticism itself was associated with both
extremes of drug tolerance; so that both very high and very low resistance to sedation (arousa-
bility) could be found in high N people. The result is especially problematic when interpreted for
the clinical population, where two types of patient of demonstrably quite di€erent behavioural
pro®le and presumed biological status Ð the neurotically anxious and the impulsively personality
disordered Ð appeared to share the same level of neuroticism.
The more modern literature on the biology of N o€ers little to enlighten us. Indeed there is a
surprising dearth of studies Ð as though psychologists have lately given up on the topic. It is
instructive to note, for example, that Matthews & Gilliland (1999), in their recent comprehensive
comparative review of the Eysenck and Gray theories, make almost no reference to N per se, and
none directly to its biology. The reasons for this neglect of N as a theoretical construct are not
hard to ®nd: as discussed below, where laboratory studies of N have been conducted, they have
used disparate experimental procedures that are dicult to connect to one another. And often the
®ndings have been equivocal, even contradictory, so that no coherent picture has emerged.
Measurements of autonomic function have continued to disappoint, as exempli®ed by a recent
study demonstrating that there was no relationship between N and cardiovascular and catecho-
lamine response in a large sample of normal subjects (Miller, Cohen, Rabin, Skoner & Doyle,
1999). The picture is equally bleak at the other end of the continuum of methodological
sophistication: neuroimaging. PET scan studies, while showing some e€ects for extraversion,
have mostly drawn a blank for neuroticism (Fischer, Wik & Fredrikson, 1997; Haier, Sokolski,
Katz & Buchsbaum, 1987). In discussing their own results, the latter authors note the complete
absence of an association with activity in the limbic brain areas, a devastating failure of the bio-
logical theory of N proposed by Eysenck (1967), whose questionnaire they used. As for EEG
studies, Pritchard (1989) found a signi®cant correlation between N and P300; but only in males
and even then only with what was identi®ed as a `trait anger' component of N, an e€ect thought
to be secondary to a correlation with extraversion. Lateralised EEG activation has also been of
interest, with a recent claim that negative a€ect (N) is associated with right frontal brain hyper-
activity (Pauli, Wiedemann & Nickola, 1999). Other authors (Hagemann, Naumann, Luerken,
Becker, Maier & Bartussek, 1999) have suggested that the connection is to activity in the left
cerebral hemisphere!
Studies of the genetics of N have enjoyed more continuity and a more concerted e€ort, using
both conventional behaviour genetics strategies and, recently, the methods of molecular genetics.
On the former there is a consensus that N shows signi®cant heritability (e.g. Jang, Livesley &
Vernon, 1996; Loehlin, McCrae, Costa & John, 1998; Viken, Rose, Kaprio & Koskenvuo, 1994);
even though, compared with extraversion, there is less consistency across studies and rather
poor replicability at the item level (Loehlin, 1992). Of course, such investigations basically
draw their conclusions about the strength of inheritance of traits from comparative kinship
data on self-report scales very distant from any underlying biology. High heritablities on super-
ordinate dimensions can seem persuasive, but they do beg the question about what is actually
inherited; a serious limitation, especially where Ð as in the case of N Ð there are no
agreed biological phenotypes for the dimension. For this reason alone, therefore, studies of the
possible molecular genetics of N have, so far, been less impressive than their behavioural genetics
counterparts.
 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400 387

A favoured target in the search for the genetic basis of N has been the transporter gene (5-
HTTL-PR) regulating the uptake of the neurotransmitter, serotonin. In two studies, functional
polymorphisms in the gene were shown to be positively related to N (Goldman, 1996; Lesch et al.,
1996). The result has been partially replicated by Mazzanti et al. (1998), and not at all by Ball et
al. (1997); Deary, Battersby, Whiteman, Connor, Fowkes and Harmar (1999); Flory, Manuck,
Ferrell, Dent, Peters and Muldoon (1999); Gelertner, Kranzler, Coccaro, Siever and New (1998);
Gustavsson et al. (1999); Jorm, Henderson, Jacomb, Croft and Easteal (1997); and Jorm et al.
(1998). One research group (Sirota, Greenberg, Murphy & Hamer, 1999) has reported that the
relationship between 5-HTTL-PR and neuroticism is non-linear; though the signi®cance of this is
unclear.
Examining the putative biology Ð including the genetics Ð of N (conceptualised as a single
homogeneous dimension) has clearly not so far solved the problem of its non-speci®city, or
identi®ed any unique, unitary property that can explain what it has in common across so many
di€erent forms of abnormal behaviour. For the moment, therefore, it seems as though we need to
look elsewhere if we are to ®nd a utility for N.

3. Going beyond N

Because of the ubiquitous quality of N, it is generally necessary Ð in exploring individual dif-

ferences further Ð to seek to specify some additional personality feature or features peculiar to
the population of interest. This is true whether the aim is to explore personality structure or to
explain the causes (`aetiology') of the di€erences observed. There are several ways to do this.
One is by `pro®ling', which is the way most dimensional personality theories proceed. Here, N
is simply considered alongside other personality dimensions, with which, in a factor analytic
sense, it has equal status. Individuals Ð or types of individual Ð are then described according to
their relative scores on the di€erent dimensions contained in the theory. Fig. 2 shows an example

Fig. 2. Diagram comparing the pro®les of Avoidant Personality Disorder and Borderline Personality Disorder on the
®ve dimensions of the Five Factor Model (see text for source of data).
388 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

of this approach, illustrated by the use of Five Factor theory to characterise two of the DSM Axis
II personality disorders: Avoidant Personality Disorder and Borderline Personality Disorder. The
data in the ®gure are derived from results reported by Widiger and Costa (1994) and represent
averaged correlations between the `Big Five' factors and measures of the two forms of personality
disorder, as found in various samples. It can be seen that, although Avoidant and Borderline are
almost identical on N, there are indeed some di€erences on the other factors. So by looking at the
latter it is possible, in principle, to go beyond the similarity in N and ®nd patterns of traits that
di€erentiate forms of personality disorder. However, it could be questioned whether these other
dimensions are very informative. Does it really help to know that Borderline patients are
disagreeable and not very conscientious; or that Avoidant patients are very introverted? Nor
does resorting to the more detailed facets that make up the broader dimensions of the Five Factor
system seem likely to improve matters. For clinical and clinical research purposes Ð which
are solely our concern here Ð it could be concluded that descriptors derived from normal
personality theory are not a particularly good way of trying to `go beyond N' in the search for
more speci®city.
Another approach to the problem is to get rid of N altogether, an example of this being Gray's
modi®cation of Eysenck's theory (Gray, 1981; and see Fig. 3). Here N is simply absorbed into
two alternative dimensions: Anxiety (an introverted form of neuroticism) and Impulsivity (an
extraverted form of neuroticism). Individual di€erences are then accounted for by the relative
weighting on these two new dimensions, with no reference to N being necessary. In Gray's case
this arrangement is very much driven by a biological theory about the di€erential sensitivity to
reward and punishment along the two dimensions. This gives it some power over both purely
descriptive accounts and the original Eysenckian theory from which it was derived. But it is still

Fig. 3. Gray's suggested modi®cation of Eysenck's personality theory, substituting Anxiety and Impulsivity for
Introversion±Extraversion (I±E) and Neuroticism (N).
 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400 389

fairly restricted in its descriptive and explanatory value, being con®ned to disorders (and features
of those disorders) that come within the rather narrow terms of reference of the Gray model.
A third way to go beyond N Ð that advocated here Ð would start, not from the trait descriptors
found in these highly formalised theories of normal personality, but from disorder-speci®c dimensions.
By this is meant dimensions which, descriptively and conceptually, are derived from and stay
close to the clinical states themselves; primarily chosen, in the ®rst instance, because, judged
against the disorders to which they relate, they seem to have an obvious face validity. Virtually by
de®nition such dimensions correlate with neuroticism. But they do have variance unique to
themselves and it is this variance Ð the `news', as it were, in the data Ð that is crucial in
exploring the speci®cs of various expressions of N. This arrangement is shown diagrammatically
in Fig. 4, using examples from the personality constructs that have informed some of our own
work.
From a statistical viewpoint, Fig. 4 of course represents the logic of multiple regression analysis,
and our thinking about N can be further illustrated with two examples using that technique; one
taken from our schizotypy research and the other from our work on the eating disorders. Here we
were interested in disentangling the relative in¯uence of N and other personality variables on an
associated, theoretically relevant, behaviour. N in both cases was measured with the neuroticism
scale from the Eysenck Personality Questionnaire (EPQ) (Eysenck and Eysenck, 1975).
The ®rst example comes from a study of the relationship between schizotypy and nightmares
(Claridge, Clark & Davis, 1997). Nightmare experience was measured with Belicki's (1992)
nightmare distress scale and schizotypy with the `schizotypy personality scale' (STA; Claridge,
1997). STA is known to correlate moderately highly with neuroticism, while the latter also proved
in the study in question to correlate signi®cantly with nightmare distress. Using multiple regres-
sion analysis to separate these e€ects we found that it was schizotypy that signi®cantly predicted
the nightmare experience; N explained no unique variance beyond that.

Fig. 4. Venn diagram to illustrate the idea of `disorder-speci®c' personality constructs that overlap with neuroticism
(N) but have unique variance corresponding to di€erent forms of abnormality.
390 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

In the second example Ð taken from the eating disorders part of our work (Davis, Claridge &
Cerullo, 1997) Ð the dependent variable in the multiple regression analysis was weight pre-
occupation. As independent variables, in addition to N, we entered a range of disorder-speci®c
personality measures, which included narcissism, borderline personality, and obsessiveness-com-
pulsiveness. Again it was these more speci®c clinical dimensions that signi®cantly de®ned weight
preoccupation, N having no additional e€ect, over and above that explained by its correlation
with the other personality traits.1
It seems, therefore, that in trying to go beyond N, we would be wise to concentrate on per-
sonality characteristics which, although enclosed by N, are relevant to, and conceptually derived
from, the disorder or disorders that are under investigation.
It is important to note that the point being made here is not just about description and classi-
®cation of personality and the clinical disorders; or narrowly about statistical prediction in mul-
tiple regression. It also has to do with explanation and aetiology. Take, for example, the positive
simple correlation between neuroticism and nightmare distress referred to earlier. Knowing that
neurotic people su€er more from their nightmares takes us very little further in understanding
why that is so. True, it is presumably because they are more emotional, or anxious, or depressed.
But that is self-evident, almost tautological. In contrast, the demonstrable connection to schizo-
typy opens up a wealth of psychological and biological theorising about schizophrenia and the
dream process that could help to explain the nightmare data.
The same is true of the eating disorders ®ndings. Constructs like narcissism provide an easy
and obvious theoretical interface to the understanding of the obsessive concerns about the self
involved in weight preoccupation. Whereas it is of very little help, by itself, to know that women
who are preoccupied with their bodies also rate themselves as more emotionally unstable.

4. N as a moderator variable

The view of N promulgated so far in this paper has been relatively pessimistic: N relates to too
many things of a deviant kind to be informative and usually fades into generality as a (frequently)
signi®cant, though rather uninteresting, correlate of abnormal behaviour. However, this does not
mean that N can thereby be totally excluded from further consideration. N represents a major
source of individual variation in emotionality and it is inconceivable that it does not enter into
the expression of clinical dysfunction Ð either as a personality trait or as a state variable,
re¯ecting negative a€ect. This was brought home to us when we began to construe N in a more
`dynamic' sense; examining N, not as a variable presumed simply to have unitary linear e€ects,
but through its contribution as a possible moderator variable, one that modulates the the role of
other personality in¯uences in behaviour.
The idea of moderator variable e€ects goes back a long way in personality psychology
(Ghiselli, 1963) and surfaced, in relation to Eysenck's theory, as ``zone analysis''; the
latter referred to a way of examining how each of the four combinations of neuroticism and

1
Note that, in both examples of multiple regression analysis cited here, other non-personality variables were
included, and found to have signi®cant e€ects, as predictors: sex in the case of nightmare distress and, in the case of
weight preoccupation, Body Mass Index.
 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400 391

introversion±extraversion relate to experimental variables (Eysenck, 1967; Furneaux, 1961).

Using zone analysis it could be shown, for example, that the failure of a particular experimental
measure to correlate with extraversion (E) was in some cases due to a moderating e€ect of N;
resulting in High N/Low E subjects behaving quite di€erently from Low N/Low E subjects, even
though they had low E in common. This e€ect can be illustrated in an extension to the drug tol-
erance data referred to earlier. As shown in Fig. 5, although it is true Ð as previously discussed Ð
that neuroticism appears to have no direct in¯uence on drug response, N evidently is important in
the way it combines with E to produce di€erential drug sensitivity. Thus, the greater barbiturate
tolerance of introverts is only true of those also high in N: the reverse is true at the opposite end
of the neuroticism dimension (Claridge, Donald & Birchall, 1981; Claridge & Ross, 1973). Similar
cross-over e€ects involving E and N have been shown for other depressant drugs Ð e.g. nitrous
oxide (Rodnight and Gooch, 1963) Ð and occasionally for other psychophysiological measures.
These have included electrodermal activity (Sadler, Me€erd & Houck, 1971). and, less con-
sistently, the EEG. In the latter case, Irmis (1994), who indexed the EEG with the alpha rhythm,
reported null results. On the other hand, Robinson (1996), using what he regards as a more valid
measure of EEG Ð its `natural frequency' Ð has described associations with N; but he also
points out that these e€ects can only be interpreted in conjunction with simultaneous variations
that also relate to introversion±extraversion.
Other than these scattered references in the experimental literature, and despite its long history
and evident importance in personality research, one nowadays hears of the moderator variable
e€ect only intermittently, and even then, mostly in the context of complex meta-analyses of psy-
chological data (Viswesvaran & Sanchez, 1998). This is perhaps because most researchers are
looking for a simple life; trying to ®nd an uncomplicated relationship between two variables,

Fig. 5. `Zone analysis' of barbiturate sensitivity (sedation threshold), illustrating how the association with high
neuroticism (N) (shown earlier in Fig. 1) is reversed in low N subjects, in both cases as a function of introversion±
extraversion (I±E).
392 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

without looking at the possibility that a third variable might interact in that relationship. The
neglect is unfortunate because in our experience such e€ects seem to be quite common, and N, in
particular, seems to act as a powerful and reliable moderator variable. We can illustrate this with
three examples from our recent work on body image and the eating disorders. In these cases we
have looked for moderator e€ects as interaction terms in multiple regression, the dependent
variable always being a measure taken from the Body Esteem Scale of Franzoi and Shields
(1984). The statistical methodology is therefore di€erent from `zone analysis'; but they are
equivalent ideas and one can easily be translated into the other. In all three studies N was mea-
sured with the neuroticism scale from the EPQ(R) (Eysenck and Eysenck, 1991).
The ®rst two studies to be cited concern commonly held views about the correlates of body
esteem or body satisfaction among women. Here an important personality variable is narcissism.
It is generally agreed that there are both adaptive and maladaptive aspects of narcissism. Healthy
narcissism constitutes part of normal development, especially during adolescence, as the child
attempts to establish a separate identity through activities and psychological mechanisms inten-
ded to enhance self-esteem and foster positive self-regard (Cramer, 1995). But failures in or dis-
tortions of this process can lead to unhealthy narcissism, in the form of excessive self-absorption,
egocentricity, and other signs of a chronic de®cit in the feeling of self-worth (Miller, 1992). It
should be noted that di€erent scales tend to vary in how closely they relate to one or other of
these two aspects of narcissism. Furthermore, those studies reporting a positive association to
body satisfaction (e.g. Jackson, Ervin & Hodge, 1992) have mostly used the Narcissistic Person-
ality Inventory (NPI; Raskin & Hall, 1979), which tends to load on the healthy aspects of the
trait.

Fig. 6. Diagram showing how the association between narcissism and body esteem varies as a function of neuroticism
(N) . Five values of N are shown, increasing in a downward direction. For interpreting numerical values along the two
axes of the ®gure normative data for the scales used (see text) are as follows: Body Esteem Ð mean 110.10, S.D. 6.73;
Narcissism Ð mean 14.72, S.D. 6.35.
 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400 393

In our own study, we too, used the NPI, but did so alongside a measure of N (Davis, Claridge
& Brewer, 1996). We found that the picture was more complicated than had hitherto been
claimed. As shown in Fig. 6, there was a signi®cant interaction of the NPI with N, such that the
positive e€ect of high narcissism on body esteem only obtained when individuals also fell at low
to average levels of N. As N increased beyond average levels, the in¯uence of narcissism was
systematically eroded, so that at higher levels of N the relationship did not exist at all; irrespective
of the degree of narcissism, body satisfaction was low.
A second study illustrating the same point has to do with sex-role orientation (Davis, Dionne &
Lazarus, 1996). Evidence is regularly quoted claiming that women who possess characteristics
conventionally attributed to the male Ð e.g. con®dence, assertiveness, strength of convictions Ð
report greater body satisfaction (Jackson, Sullivan & Rostker, 1988; Kimlicka, Cross & Tarnai,
1983). Yet again our data demonstrate that this relationship is modi®ed by the individual's posi-
tion on the N dimension. The result is summarised in Fig. 7, the measure of masculinity referred
to there being taken from the Bem Sex-Role Inventory (Bem, 1973). As the plot of the interaction
shows, the positive relationship between masculinity and body satisfaction was only present when
N was at low to average levels. At high levels of N, possessing conventionally masculine char-
acteristics conferred no bene®t on body-image perception.
Our ®nal illustration of a moderating e€ect by N comes from a set of data on eating disordered
patients (both anorexic and bulimic) (Davis, 1997); in this case examining self-oriented perfec-
tionism, as indexed by the scales of Hewitt and Flett (1989). Echoing the claims referred to in our
two earlier examples, it is repeatedly stated that perfectionism is positively associated with eating

Fig. 7. Diagram showing how the association between body satisfaction and masculinity varies as a function of neu-
roticism (N). Five values of N are shown increasing in a downward direction. For interpreting numerical values along
the two axes of the ®gure normative data for the scales used (see text) are as follows: Body Satisfaction Ð mean 46.91,
S.D. 6.30; Masculinity Ð 95.80, S.D. 13.21 (raw BEM score not divided by the number of items, as in test manual).
394 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

Fig. 8. Diagram showing how the association between body esteem and perfectionism varies as a function of neuroti-
cism (N) (four values of N are shown increasing in a downward direction). For interpreting numerical values along the
two axes of the ®gure normative data for the scales used (see text) are as follows: Body Esteem Ð mean 110.10, S.D.
6.73; Perfectionism Ð mean 67.52, S.D. 15.52.

disorder symptomatology, including the poor body image that forms part of the syndrome.
However, our own results again suggest a relationship very much in¯uenced by neuroticism (Fig.
8). At high levels of N, the commonly identi®ed simple relationship does exist; viz highly perfec-
tionist women do report low levels of body esteem. But as N decreases, this e€ect does not merely
disappear: it actually reverses into a positive relationship between perfectionism and body esteem,
even in this clinical sample.

5. Conclusions

We indicated at the beginning of this paper that a single answer to the question posed in the
title was not to be anticipated. Both of what we referred to as, respectively, the `pessimistic' and
the `optimistic' views of N turn out to be partly true. Certainly, taken by itself N tells us very little
about personality; and, crucially, given its label and content, it fails dismally to discriminate
between di€erent forms of abnormality, over a wide range of dysfunctions.
N fares better when used in combination with other dimensions or personality constructs. Here
it can help in pro®ling individuals; though which personality constructs Ð additional to N Ð one
decides to work with depends very much on the purpose of the exercise. Conventional personality
dimensions like extraversion, agreeableness, and so on are probably satisfactory enough
when applied to, say, occupational or other forms of selection among normal samples. But in
clinical psychology the conventional personality theories have signally failed, in our view, to `go
(usefully) beyond N'. For that purpose, dimensions that are ®rmly grounded in the psychological
 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400 395

disorders themselves are likely to be more informative. These are what we called `disorder-speci®c
constructs', like schizotypy, narcissism, obsessive-compulsiveness, and so on. Here, incidentally,
we should further emphasise that, although deliberately chosen because they actually correlate
with N and N-related pathologies, such constructs also describe continuous variations in the
healthy population. As our narcissism example illustrates, they can, therefore, be viewed as
simultaneously subsumed by N and moderated by it. Which of these features we attend to will
depend on whether the question of interest is the structure of personality, or its dynamics. In this
regard it could be said that, in the structural sense, N has the appearance of being (merely) a
statistically derived collecting point for disparate sources of abnormality. Dynamically, its func-
tion seems to be that of determining the extent to which the personality variables it encompasses
are expressed (or, alternatively, not expressed) as abnormal behaviours.
Intriguingly then, both the strengths and the weaknesses of N Ð and here it seems to be unique
among personality variables Ð derive from its very generality; from its overarching quality as a
personality and clinical descriptor that captures the essence of deviance in most potential forms of
abnormality, yet retains sucient speci®city of its own to help determine whether the latter are
manifested and, if so, to what degree. This raises the issue of how N achieves its e€ect; and,
¯owing from that, what sort of individual di€erences variable it really is.
Attempts to answer these questions in the past have often been shackled by a determined
search for an orthogonality for N, a view that we have explicitly abandoned here, as inappropri-
ate to the understanding of personality variations which, by de®nition, must at their extremes
correlate with N. In practice, N's orthogonality from other personality dimensions Ð like intro-
version±extraversion Ð has often failed to appear, in which case, researchers have usually resor-
ted to the habit of equating neuroticism with anxiety. This has the advantage of tidily narrowing
down what has to be explained, but is clearly unsatisfactory: as Gray's handling of the Eysenck
dimensions makes explicit, N has a broader scope, according to him taking in even strong posi-
tive emotion. But even negative a€ect can stem from, or be associated with, a wider range of
temperamental reactions than the dysthymia or neurotic introversion into which N has often, for
convenience, been strait-jacketed. Quite apart from the social fear of the overanxious, emotional
dysfunction can include the ennui of the frustrated sensation seeker, the impatience of the
impulsive, the anger of the borderline personality, the existentialist panic of the schizoid or schi-
zotype, and the depression that can accompany all of these.
Perhaps high N is therefore simply something that goes along with extreme deviations along
any dimension that has the capacity to act as a disposition to disordered functioning. Its role as a
moderator variable then, conceivably comes about because the `bad mood' that constitutes high
neuroticism potentiates the negative elements in co-existing features of the individual, trans-
forming them from adaptive, to unhealthy, behaviours. The examples we gave from our eating
disorders research illustrate the point quite well. There, it is easy to see how moderate degrees of
self-regard, stemming from healthy narcissism, could promote adaptive attitudes and behaviour
towards the physique, leading to high body satisfaction. However, carried to an extreme the very
same tendencies might occupy so much of the individual's emotional and mental `space' that they
become irksome, transforming them into maladaptive (highly neurotic) preoccupations about the
body, and consequent psychopathology. It is at this stage that N could be said to reveal another
of its evident qualities Ð as an `emotional ampli®er', exaggerating existing personality tendencies
to the point where unhealthy behaviour takes over from healthy functioning.
396 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

An important suggestion in the above analysis is that it does not presuppose that the beha-
viours associated with high N are driven by any particular component or type of `bad mood'. In
the example given it might indeed be the case that anxiety Ð the most commonly quoted
expression of neuroticism Ð drives the individual from health to illness. But in other cases it
might be depression, anger, or the pathological need for novelty that serves to translate normality
into abnormality. In this sense, N could be regarded as a truly heterogeneous construct, with
multiple connections across the personality matrix.
Some of the questions about N that we have raised here ought eventually to be resolvable at
the biological level. For example, providing a more sophisticated view is taken of the construct, it
seems virtually certain that continuing research in molecular genetics will elucidate how heritable
factors contribute to what currently we label neuroticism. Not through the discovery of a `gene
for N' Ð a mildly ridiculous notion Ð but rather by deconstructing the genetic architecture of
this very heterogeneous psychological dimension. Be®tting the continuous nature of most, if not
all, of the component parts of N, that is most likely to be achieved by Quantitative Trait Loci
genetics, a methodology well suited to analyses in the personality domain (Plomin, Owen &
McGun, 1994).
There are still issues, however, of `what to look for the genetics of' and how to measure it (or
them). On the latter point, the availability of reliable, relevant biological markers Ð whether de®ned
directly or bootstrapped from the genetics research itself Ð would certainly help. As for the broader
question Ð What is N? Ð our hunch is that, insofar as answers will come from biology, they will
often do so incidentally: from understanding the nature of temperamental variations contributing to
N or the disorder-speci®c characteristics with which it correlates; and also sometimes from unex-
pected directions, such as discoveries about the genetics of schizophrenia and other illnesses.
Finally, although itself part of the biological remit, there is an aspect of N that other sorts of
enquiry should also help to elucidate. We are referring to the state/trait issue, which continues to
be an important concern in individual di€erences research (Cooper, 1998; Matthews & Deary,
1998). The background in personality and temperament theory from which N emerged is very
much predicated on the notion of relatively ®xed characteristics, a view now increasingly rein-
forced by talk of `biological markers' and `gene loci'. In such a climate, it is easy to forget that
such knowledge only constitutes the beginning, the sketching out of constitutional dispositions
for behaviours and psychological states, as reactions to environmental events and, indeed, to
other sources of variation within the individual. N is a particularly relevant example because of
its strong association with mood. It is perhaps that particular function of neuroticism that
accounts for some of the phenomena we have referred to here Ð such as N's moderating e€ects
on other variables and its possible role as an ampli®er of co-existing temperamental tendencies to
di€erent forms of psychopathology.

References

Ball, D., Hill, L., Freeman, B., Eley, T. C., Strelau, J., Riemann, R., Spinath, F. M., Angleitner, A., & Plomin, R.
(1997). The serotonin transporter gene and peer-rated neuroticism. Neuroreport, 8, 1301±1304.
Belicki, K. (1992). The relationship of nightmare frequency to nightmare su€ering with implications for treatment and
research. Dreaming, 2, 143±148.
 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400 397

Bem, S. L. (1973). The Bem Sex Role Inventory: professional manual. Palo Alto: Consulting Psychologists Press.
Breslau, N., Kilbey, M. M., & Andreski, P. (1993). Vulnerability to psychopathology in nicotine-dependent smokers:
an epidemiologic study of young adults. American Journal of Psychiatry, 150, 941±946.
Bru, E., Myletun, R. J., & Svebak, S. (1993). Neuroticism, extraversion, anxiety and Type A behaviour as med-
iators of neck, shoulder and lower back pain in female hospital sta€. Personality and Individual Di€erences, 15,
485±492.
Carr, A. B. (1981). Vitamin C and the common cold: a second MZ cotwin study. Acta Genetica Medicae et Gemello-
logia: Twin Research, 30, 249±255.
Claridge, G. (1967). Personality and Arousal. Oxford: Pergamon.
Claridge, G. (1997). Schizotypy: Implications for illness and health. Oxford: Oxford University Press.
Claridge, G., & Herrington, R. N. (1960). Sedation threshold, personality and the theory of neurosis. Journal of Mental
Science, 106, 1568±1583.
Claridge, G., & Ross, E. (1973). Sedative drug tolerance in twins. In G. Claridge, S. Canter, & W. I. Hume, Personality
di€erences and biological variations: a study of twins (pp. 115±131). Oxford: Pergamon Press.
Claridge, G., Donald, J. R., & Birchall, P. M. A. (1981). Drug tolerance and personality: some implications for
Eysenck's theory. Personality and Individual Di€erences, 2, 153±166.
Claridge, G., Clark, K., & Davis, C. (1997) Nightmares, dreams, and schizotypy. British Journal of Clinical Psychol-
ogy, 36, 377-386.
Cloninger, C. R. (1988). The genetics and psychobiology of the seven-factor model of personality. In K. R. Silk,
Biology of Personality Disorders (pp. 68±92). Washington: American Psychiatric Press.
Cooper, C. (1998). Individual di€erences. London: Arnold.
Costa, P. J. (1987). In¯uence of the normal personality dimension of neuroticism on chest pain symptoms and
coronary artery disease. American Journal of Cardiology, 60, 20J-26J.
Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO PI-R) and NEO Five-Factor Inventory
(NEO-FFI) Professional Manual. Odessa, FL: Psychological Assessment Resources.
Cramer, P. (1995). Identity, narcissism, and defence mechanisms in late adolescence. Journal of Research in Personality,
29, 341±361.
Davis, C. (1997). Normal and neurotic perfectionism in eating disorders: an interactive model. International Journal of
Eating Disorders, 22, 421±426.
Davis, C., Claridge, G., & Brewer, H. (1996). The two faces of narcissism: Personality dynamics of body esteem.
Journal of Social and Clinical Psychology, 15, 153±166.
Davis, C., Claridge, G., & Cerullo, D. (1996). Re¯ections on narcissism: con¯icts about body image perceptions in
women. Personality and Individual Di€erences, 22, 309±316.
Davis, C., Claridge, G., & Cerullo, D. (1997). Personality factors and weight preoccupation: a continuum approach
to the association between eating disorders and personality disorders. Journal of Psychiatric Research, 31, 467±
480.
Davis, C., Dionne, M., & Lazarus, L. (1996). Gender-role orientation and body image in women and men: the
moderating in¯uence of neuroticism. Sex Roles, 34, 493±505.
Deary, I. J., Battersby, S., Whiteman, M. C., Connor, J. M., Fowkes, F. G., & Harmar, A. (1999). Neuroticism and
polymorphisms in the serotonin transporter gene. Psychological Medicine, 29, 735±739.
Duggan, C. F., Lee, A. S., & Murray, R. M. (1990). Does personality predict long-term outcome in depression? British
Journal of Psychiatry, 157, 19±24.
Eysenck, H. J. (1967). The biological basis of personality. Spring®eld: Charles C. Thomas.
Eysenck, H. J., & Eysenck, M. W. (1985). Personality and individual di€erences. A natural science approach. New York:
Plenum Press.
Eysenck, H. J., & Eysenck, S. B. G. (1975). Eysenck Personality Questionnaire. London: Hodder and Stoughton.
ysenck, H. J., & Eysenck, S. B. G. (1991). The Eysenck Personality scales. Sevenoaks, Kent: Hodder & Stoughton.
Fahrenberg, J. (1992). Psychophysiology of neuroticism and anxiety. In A. Gale, & M. W. Eysenck, Handbook of
individual di€erences. Biological Perspectives (pp. 179±226). Chichester: John Wiley.
Fischer, H., Wik, G., & Fredrikson, M. (1997). Extraversion, neuroticism and brain function: a PET study of person-
ality. Personality and Individual Di€erences, 23, 345±352.
398 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

Flory, J. D., Manuck, S. B., Ferrell, R. E., Dent, K. M., Peters, D. G., & Muldoon, M. F. (1999). Neuroticism is not
associated with the serotonin transporter (5-HTTLPR) polymorphism. Molecular Psychiatry, 4, 93±96.
Franzoi, S. L., & Shields, S. A. (1984). The Body Esteem Scale: multidimensional structure and sex di€erences in a
college population. Journal of Personality Assessment, 48, 173±178.
Furneaux, W. D. (1961). Neuroticism, extraversion, drive and suggestibility. International Journal of Clinical and
Experimental Hypnosis, 9, 195±214.
Gelertner, J., Kranzler, H., Coccaro, E. F., Siever, L. J., & New, A. S. (1998). Serotonin transporter protein gene
polymorphism and personality measures in African American and European American subjects. American Journal of
Psychiatry, 155, 1332±1338.
Ghiselli, E. E. (1963). Moderating e€ects and di€erential reliability and validity. Journal of Applied Psychology,
47, 81±86.
Goldman, D. (1996) High anxiety. Science, 274, 1483.
Gorsuch, R., & Cattell, R. B. (1967). Second stratum personality factors de®ned in the questionnaire realm by the
16PF. Multivariate Behavioural Research, 2, 211±224.
Gray, J. A. (1981). A critique of Eysenck's theory of personality. In J. Eysenck, A Model for Personality (pp. 246±276).
New York: Springer-Verlag.
Gustavsson, J. P., Nothen, M. M., Jonsson, E. G., Neidt, H., Forslund, K., Rylander, G., Mattila, E. M., Sedvall, G.
C., Propping, P., & Asberg, M. (1999). No association between serotonin transporter gene polymorphisms and per-
sonality traits. American Journal of Medical Genetics, 88, 430±436.
Hagemann, D., Naumann, E., Luerken, A., Becker, G., Maier, S., & Bartussek, D. (1999). EEG asymmetry, disposi-
tional mood and personality. Personality and Individual Di€erences, 27, 541±568.
Haier, R. J., Sokolski, K., Katz, M., & Buchsbaum, M. S. (1987). The study of personality with Positron Emission
Tomography. In J. Strelau, & H. J. Eysenck, Personality Dimensions and Arousal (pp. 251±267). New York: Plenum
Press.
Hewitt, P. L., & Flett, G. L. (1989). The Multidimensional Perfectionism Scale: development and validation. Canadian
Psychology, 30, 339.
Ingham, J. G., & Robinson, J. O. (1964). Personality in the diagnosis of hysteria. British Journal of Psychology, 55,
276±284.
Irmis, F. (1994). Relation of extraversion and neuroticism to EEG alpha indicators: Four extreme groups. Studia
Psychologica, 36, 382±384.
Jackson, L. A., Ervin, K. S., & Hodge, C. N. (1992). Narcissism and body image. Journal of Research in Personality,
26, 357±370.
Jackson, L. A., Sullivan, L. A., & Rostker, R. (1988). Gender, gender role, and body image. Sex Roles, 19, 429±443.
Jang, K. L., Livesley, W. J., & Vernon, P. A. (1996). Heritability of the big ®ve personality dimensions and their facets:
a twin study. Journal of Personality, 64, 577±591.
Jorm, A. F., Henderson, A. S., Jacomb, P. A., Christensen, H., Korten, A. E., Rodgers, B., Tan, X., & Easteal, S.
(1998). An association study of a functional polymorphism of the serotonin transporter gene with personality.
Molecular Psychiatry, 3, 449±451.
Jorm, A. F., Henderson, A. S., Jacomb, P. A., Croft, L., & Easteal, S. (1997). Quantitative trait loci for neuroticism: an
allelic association study with the serotonin receptor (HTR2) and monoamine oxidase A (MAOA) genes. Personality
and Individual Di€erences, 22, 287±290.
Kentle, R. L. (1989). Headache symptomatology and neuroticism in a college sample. Psychological Reports, 65, 976±
978.
Kilbey, M. M., Breslau, N., & Andreski, P. (1992). Cocaine use and dependence in young adults: Associated psychia-
tric disorders and personality traits. Drug and Alcohol Dependence, 29, 283±290.
Kimlicka, T., Cross, H., & Tarnai, J. (1983). A comparison of androgynous, feminine, and undi€erentiated women on
self-esteem, body satisfaction, and sexual satisfaction. Psychology of Women Quarterly, 7, 291±294.
Kissen, D. M. (1964). Relationship between lung cancer, cigarette smoking inhalation and personality. British Journal
of Medical Psychology, 37, 203±216.
Krug, S. E., & Johns, E. F. (1986). A large scale cross-validation of second-order personality structure de®ned by the
16PF. Psychological Reports, 59, 683±693.
 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400 399

Lesch, K. P., Bengel, D., HeilsA., Sabol, S. J., Greenberg, B. D., Petri, S., Benjamin, J., Muller, C. R., Hamer, D. H.,
& Murphy, D. L. (1996). Association of anxiety-related traits with a polymorphism in the serotonin transporter gene
regulatory region. Science, 274, 1527±1531.
Loehlin, J. C. (1992). Genes and environment in personality development. London: Sage.
Loehlin, J. C., McCrae, R. R., Costa, P. T., & John, O. P. (1998). Heritabilities of common and measure-speci®c
components of the Big Five personality factors. Journal of Research in Personality, 32, 431±453.
Matthews, G., & Deary, I. J. (1998). Personality traits. Cambridge: Cambridge University Press.
Matthews, G., & Gilliland, K. (1999). The personality theories of H.J. Eysenck and J.A. Gray: comparative review.
Personality and Individual Di€erences, 26, 583±626.
Mazzanti, C. M., Lappalainen, J., Long, J. C., Bengel, D., Naukkarinen, H., Eggert, M., Virkkunen, M., & Linnoila,
M. (1998). Role of the serotonin transporter promoter polymorphism in anxiety-related traits. Archives of General
Psychiatry, 55, 936±940.
McGuire, R. J., Mowbray, R. M., & Vallance, R. C. (1963). The Maudsley Personality Inventory used with psychiatric
inpatients. British Journal of Psychology, 54, 157±166.
Miller, G. E., Cohen, S., Rabin, B. S., Skoner, D. P., & Doyle, W. J. (1999). Personality and tonic cardiovascular,
neuroendocrine, and immune parameters. Brain, Behavior and Immunity, 13, 109±123.
Miller, I. J. (1992). Interpersonal vulnerability and narcissism: A conceptual continuum for understanding and treating
narcissistic psychopathology. Psychotherapy, 29, 216±224.
Pauli, P., Wiedemann, G., & Nickola, M. (1999). Pain sensitivity, cerebral laterality, and negative a€ect. Pain, 80, 359±364.
Plomin, R., Owen, M. J., & McGun, P. (1994). The genetic basis of complex human behaviours. Science, 264, 1733±
1739.
Prescott, C. A., Neale, M. C., Corey, L. A., & Kendler, K. S. (1997). Predictors of problem drinking and alcohol
dependence in a population-based sample of female twins. Journal of Studies on Alcohol, 58, 167±181.
Pritchard, W. S. (1989). P300 and EPQ/STPI personality traits. Personality and Individual Di€erences, 10, 15±24.
Raskin, R., & Hall, C. S. (1979). A narcissistic personality inventory. Psychological Reports, 45, 590.
Robinson, D. L. (1996). Brain, mind, and behavior. Westport, CT: Praeger.
Rodnight, E., & Gooch, R. N. (1963). A new method for the determination of individual di€erences in susceptibility to
a depressant drug. In H. J. Eysenck, Experiments with drugs (pp. 169±193). Westport, CT: Praeger.
Sadler, T. G., Me€erd, R. B., & Houck, R. L. (1971). The interaction of extraversion and neuroticism in orienting
response habituation. Psychophysiology, 8, 312±318.
Sainsbury, P. (1960). Psychosomatic disorders and neurosis in outpatients attending a general hospital. Journal of
Psychosomatic Research, 4, 261±273.
Saklofske, D. F., Kelly, I. W., & Janzen, B. L. (1995). Neuroticism, depression and depression-proneness. Personality
and Individual Di€erences, 18, 27±31.
Shagass, C. (1954). The sedation threshold. A method for estimating tension in psychiatric patients. Electro-
encephalography and Clinical Neurophysiology, 6, 221±233.
Sirota, L. A., Greenberg, B. D., Murphy, D. L., & Hamer, D. H. (1999). Non-linear association between the serotonin
transporter promoter polymorphism and neuroticism: A caution against using extreme samples to identify quanti-
tative trait loci. Psychiatric Genetics, 9, 35±38.
Slade, P. D. (1974). Psychometric studies of obsessional illness and obsessional personality. In H. R. Beech, Obsessional
states (pp. 95±109). London: Methuen.
Statham, D. J., Heath, A. C., Madden, P. A., Bucholz, K. K., Bierut, L., Dinwiddie, S. H., Slutske, W. S., Dunne,
M. P., & Martin, N. G. (1998). Suicidal behaviour: an epidemiological and genetic study. Psychological Medicine, 28,
839±855.
Stelmack, R. M. (1981). The psychophysiology of extraversion and neuroticism. In H. J. Eysenck, A model for per-
sonality (pp. 38±64). London: Methuen.
Trull, T. J. (1992). DSM-III-R personality disorders and the ®ve-factor model of personality: an empirical comparison.
Journal of Abnormal Psychology, 101, 553±560.
van den Akker, O. B. A., Eves, F. F., Stein, G. S., & Murray, R. M. (1995). Genetic and environmental factors in
premenstrual symptom reporting and its relationship to depression and a general neuroticism trait. Journal of Psy-
chosomatic Research, 39, 477±487.
400 G. Claridge, C. Davis / Personality and Individual Di€erences 31 (2001) 383±400

Viken, R. J., Rose, R. J., Kaprio, J., & Koskenvuo, M. (1994). A developmental genetic analysis of adult personality:
Extraversion and neuroticism from 18 to 59 years of age. Journal of Personality and Social Psychology, 66, 722±730.
Viswesvaran, C., & Sanchez, J. I. (1998). Moderator search in meta-analysis: a review and cautionary note on existing
approaches. Educational and Psychological Measurement, 58, 77±87.
Widiger, T. A., & Costa, P. T. (1994). Personality and personality disorders. Journal of Abnormal Psychology, 103, 78±
91.
Yadav, S. K., Jain, A. K., Bahre, P. B., & Gupta, J. P. (1990). Neuroticism in patients with irritable bowel syndrome.
Indian Journal of Gastroenterology, 9, 29±31.