Personality and Individual Differences 30 (2001) 995±1007

www.elsevier.com/locate/paid

Sensational interests and sensation seeking in mentally

disordered o€enders
Vincent Egan a,*, Philip Charlesworth a, Cathryn Richardson b,
Marie Blair c, Mary McMurran d
a
Department of Psychology, Arnold Lodge, Cordelia Close, Leicester LE5 0LE, UK
b
Clinical Psychology Doctorate, School of Health and Social Sciences, Coventry University, Coventry CV1 5BH, UK
c
Department of Psychology, University of Plymouth, Drake Circus, Plymouth PL4 8AA, UK
d
Centre for Applied Psychology (Forensic Section), University of Leicester, University Road, Leicester LE1 7RH, UK

Received 27 November 1999; received in revised form 10 March 2000

Abstract
Violent and sensational interests are reported in the background history of some mentally disordered
o€enders. As sensation-seeking accounts for the drive for intellectual and physical excitement, it was
thought that this trait would also underlie an interest in sensational material. We examined the di€erences
in sensational interests and SS between patients detained under the Mental Health Act (1983) classi®ca-
tions of psychopathic disorder (PD) and mental illness (MI), the degree to which sensational interests
related to sensation-seeking, and the relationship between these factors and self-reported personality dis-
orders. The Sensational Interests Questionnaire (SIQ), Zuckerman's Sensation-Seeking Scale (SSS), the
Standard Progressive Matrices (SPM) and the MCMI-II were given to 42 patients (32 MI, 10 PD). There
were no signi®cant di€erences between the MI and PD groups for scores on the SIQ or the SSS. The SIQ
was not related to scores on the SPM. As predicted, the SIQ's subscales of militarism and the violent-occult
were signi®cantly and positively associated with sensation-seeking, especially the SSS subscales of Disin-
hibition and Thrill and Adventure Seeking. Sensation-seeking mediated the relationship between sensa-
tional interests and measures of personality disorder. # 2001 Elsevier Science Ltd. All rights reserved.
Keywords: Sensation-seeking; Personality; Intelligence; Violence; Occultism; Interests; Forensic; O€enders

1. Introduction

It is popularly assumed that those who commit apparently senseless and horri®c crimes are
fascinated (and perhaps even driven to o€end) by topics of a lurid and morbid nature. For

* Corresponding author. Fax: +44-0116-225-6127.

E-mail address: vince.egan@arnoldl.cnhc- tr.trent.nhs.uk (V. Egan).

0191-8869/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0191-8869(00)00088-X
996 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007

example, a recent case of a schizophrenic who killed schoolgirls was said to have shown a parti-
cular interest in accounts of murder and the occult (Wainwright, 1995). Brittain's (1970) paper
`The Sadistic Murderer' crystallised this stereotype when he suggested that those who commit
such crimes are often withdrawn, introverted, socially isolated, over-controlled and timid, that
such individuals compensate for their social isolation by having a rich fantasy life and that they
are fascinated with atrocities and cruelty such as those committed by the Nazis. Others were said
to have an interest in werewolves, vampires, black magic, torture, and `escapology' (p. 203). He
suggested such o€enders have an inordinate interest in weapons, often having a large collection
which may be `lovingly handled', to the extent of `endowing some with pet names' (p. 201 and
202).
Brittain's description came from clinical anecdote and experience rather than formal research,
and some ostensibly deviant interests have a high base-rate of interest in the population. While it
is tempting to conclude that investigation of lurid interests is unhelpful for understanding the
mostly commonplace nature of even severe o€ending, the presence of such interests may indeed
di€erentiate the deviant from the more stable o€ender. Prins (1988; 1990a,b; 1998) suggests that
assessing an individual's interests (e.g. books, videotapes) may provide evidence of their fantasy
life and as such it is important to recognise such material. McCullough, Snowden, Wood and
Mills (1983) have also provided evidence that deviant fantasies are one stage in a chain of beha-
viours approximating ever more closely to the ®nal sadistic act. Deviant thoughts and interests
may therefore warrant exploration.
Egan, Auty Miller, Amadi, Richardson and Gargan (1999) developed a Sensational Interests
Questionnaire (SIQ), based on an item analysis of the overall items identi®ed by Brittain along
with others proposed by forensic clinical psychologists. Table 1 presents a summary of the items
extracted from an analysis of the SIQ items proposed by Brittain and shows that 11 (50%) of the
original Brittain `sadistic interests' items do not exist as coherent latent constructs (Egan, 1999).
For example, an interest in serial killers and Hitler/fascism did not load signi®cantly on the main
sensational interest factors of the newly developed questionnaire. The pilot SIQ study found that
general sensational interests encompass two main, but highly correlated, sub-components; mili-
tarism and violent occultism, both of which share an interest in weaponry. The `violent occultism'
factor re¯ected subcultural lifestyles, violence, and a preoccupation with the supernatural,
whereas the `militarism' factor re¯ected a general interest in aggression, dominance and inde-
pendence.
Egan et al. (1999) found forensic patients signi®cantly more willing than control subjects to
express a positive interest in militarism and violent-occultism. In the same study, scores on the
SIQ were examined in relation to general personality traits using the NEO-FFI (Costa &
McCrae, 1992) and a demographic-based estimate of verbal IQ (Egan, Crawford, Goodwin &
Brettle, 1990). Militaristic interests were independently associated with higher extraversion and
lower estimated verbal IQ, whereas an interest in the violent-occult was independently associated
with lower Agreeableness and lower Conscientiousness; both factors were associated with
younger age. The associations between sensational interests and general personality traits were
modest, suggesting the relationship between these factors was insucient to provide a compre-
hensive explanation of why people are interested in sensational topics. We recommended that
future studies should utilise more speci®c scales of psychopathology and better validated mea-
sures of personality, for example, sensation-seeking (Zuckerman, 1979).
 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007 997

Table 1
Items of possible forensic interest extracted from pilot research into the SIQa

Original `sadistic' interest Identi®ed in factor Substitute item which Identi®ed factor
proposed by Brittain analysis of pilot SIQ? loaded in pilot SIQ

Vampires and werewolves Yes 1st PC, F2

Pets No The Paranormal 1st PC
Martial arts Yes 1st PC, F1
Dinosaurs No Motorbikes 1st PC, F1
Crossbows, knives and swords Yes 1st PC, F1, F2
Serial killers No Flying Saucers 1st PC
Attending religious services No Body-building 1st PC, F1
Philosophy Yes 1st PC
Psychology and Psychiatry Yes F3
Drugs Yes 1st PC, F2
Black magic Yes 1st PC, F2
Mercenaries and the SAS Yes 1st PC, F1
Guns and shooting Yes 1st PC, F1, F2
Science ®ction and fantasy No Alternative medicine F3, F4
Hitler and fascism No Tattoos and body-piercing 1st PC, F2
True crime No Astrology F4
Horror ®lms and stories No Survivalism 1st PC, F1
Detective ®lms and stories No
Paganism Yes 1st PC, F2, F4
Thinking about spirituality No
The armed forces Yes 1st PC, F1
Funerals and death No
a
1st PC=loaded on 1st principal component; F1=loaded on ®rst varimax factor; F2=loaded on second varimax
factor; F3=loaded on third varimax factor; F4=loaded on fourth varimax factor.

Sensation-seeking, as measured by the Sensation-Seeking Scale (SSS; Zuckerman, 1984a) is a

useful explanatory construct for a range of pathological states and behaviours including psycho-
pathy (Daderman, 1999); antisocial personality (Montag & Birenbaum, 1986); abuse of alcohol
and illegal drugs, and anti-social behaviour (Zuckerman, 1983; Newcomb & McGee, 1991; Perez
& Torrubia, 1985). Sensation-seeking is also associated with a range of interests and preferences,
including adrenaline sports, an interest in torture, execution, and corpses (Zaleski, 1984), and the
representation of the latter in the mass media (Potts, Dedmon & Halford, 1996).
The relationship between sensation-seeking and horror-®lm viewing is well-established, and
emphasises the sensation-seeker's general inclination to witness or read about arousing and novel
experiences `whether they are sexual, morbid, or anything else' (Zuckerman & Little, 1986). Thus
Edwards (1984) revealed that the SSS total score and all four SSS subscales were signi®cantly
correlated with higher levels of horror ®lm viewing, with the correlations been highest for the SSS
total and SSS Disinhibition (Dis) scales. Zuckerman and Little (1986), examining the relationship
between personality and `curiosity about morbid and sexual events' found attending horror and
X-rated erotic movies positively related to sensation-seeking and Psychoticism. They found total
SSS score and the SSS Dis subscale showed the strongest, most consistent pattern of correlations
with curiosity and horror and sex ®lm attendance in both females and males. The SSS Thrill and
998 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007

Adventure Seeking (TAS) subscale correlated mostly with violence curiosity, whereas the SSS
Experience Seeking (ES) scale correlated primarily with broad attitudes. Johnston (1995) identi-
®ed two horror-®lm viewing types characterised by high levels of TAS; the `gore' and the `thrill'
watcher. She suggests gore-watching motivations re¯ect a curiosity about physical violence, an
interest in killing, and an attraction to the grotesque, whereas the thrill-watching motivations
re¯ect the fun of being startled and scared, so indicating di€erent motivations may underlie
viewing horror ®lms. Fabregat and Beltri (1998) replicated the Zuckerman and Little's result, also
®nding the SSS (especially Dis and TAS) had a marked relationship with interests in sex and
horror-related topics, and in actually going to see horror and sex ®lms.
Several mechanisms to explain why sexual, horrifying or violent stimuli are intrinsically more
rewarding for individuals higher in sensation-seeking (and thus why they may be drawn to more
sensational interests) have been advanced:

1. `Excitation-transfer' theory posits that exposure to stimuli generates excitational states that
intensify post-exposure emotional responses (Zillmann, 1978, 1980). The physiological
arousal generated by exposure to such stimuli is transferred (or labelled) by individuals in an
e€ort to ascribe meaning to their experiences. Thus viewing-induced arousal to graphic
horror images may be converted into feelings of distress or delight (Sparks, 1991).
2. `Optimal level of arousal' (OLA) theories explain an interest in experiential stimuli that can
elicit anxiety, anger and even disgust (Zuckerman, 1979). According to this model, the sen-
sation-seeker's need for novel and arousing experiences drives a desire to witness or read
about `sensational' events. The arousal of ostensibly negative emotions becomes positively
reinforcing if it brings an individual to an optimal level of cortical arousal (or just beyond).
High sensation seekers seek intense and/or novel stimulation because they are generally
under-aroused relative to their higher OLA (Zuckerman, 1994, 1997).
3. Learning processes may also be important, as beyond arousability high sensation-seekers
may habituate more rapidly to repeated exposure to arousing stimuli. The arousing potential
of any stimulus is reduced by repetition, and high sensation seekers seek novel experiences
and activities in order to avoid the inevitable decline in arousal produced by familiarity.
4. Lastly, high sensation-seeking may re¯ect neurochemical states. Fabregat and Beltri (1998)
associate sensation seeking levels with di€ering levels of catecholamines (catecholamines
being monoamine neurotransmitters in the central nervous system such as noradrenaline,
dopamine, and serotonin, all of which are associated with brain mechanisms underlying
behaviour in general). By modulating these neurotransmitters an individual can arguably
sustain their particular OLA (Zuckerman, 1993).

The present study examines the broader construct of sensational interests (rather than horror
®lms alone) in relation to sensation seeking in mentally disordered o€enders using the SSS, the
SIQ and the Millon Clinical Multiaxial Inventory-II (MCMI-II; Millon, 1987). Given the litera-
ture, we expected sensational interests to be positively related to sensation seeking, particularly
the SSS's Dis and TAS subscales. The MCMI-II enables one to examine the relationship between
sensational interests and personality disorder, and determine the role of sensation seeking in such
a relationship. We hypothesised that SS mediates the relationship between sensational interests
and the MCMI-II personality disorders.
 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007 999

2. Method

2.1. Participants

Forty-two patients (40 males and 2 females) participated in this study, all of whom were resi-
dent in a secure psychiatric unit for mentally disordered o€enders and were detained under the
Mental Health Act (MHA) 1983 (HMSO, 1986). Thirty-two (76.2%) were classi®ed as su€ering
from Mental Illness (MI), 10 (23.8%) as Psychopathic Disorder (PD). The mean age of the MI
participants was 32.6 years (SD=10.2), for the PDs it was 27.4 years (SD=5.7). In light of the
relatively small number of females (N=2), statistical comparisons between the two legal group-
ings was performed without division by gender. It should be emphasised that the groupings of MI
and PD used accorded with legal rather than clinical labels. Broadly, those in the category of
psychopathic disorder show some form of personality disorder, while those falling in the mental
illness category have at some time exhibited symptoms of psychiatric disturbance, such as psy-
chosis, with personality disorders in some cases being a secondary diagnosis (Blackburn, 1987;
Blackburn & Maybury, 1985).

2.2. Design

The present study used both between subjects and correlational approaches. This design was
used to see whether those detained under the MHA (1983) classi®cations di€ered in their interest
for `sensational' topics, as well as di€erences in general personality traits. We sought to examine
the correlation between the SIQ and the SSS-V, and secondarily the association between these
measures and various personality disorders, as measured by the MCMI-II. To minimise con-
founding associations, a measure of non-verbal intelligence was taken, as well as age, allowing us
to control for such factors within the statistical analyses.

2.3. Materials

The following inventories were administered to all participants:

2.3.1. The Sensational Interests Questionnaire (SIQ)

The SIQ, (Egan et al., 1999), formally measures potentially pathological interests of a lurid,
morbid, or violent kind (Appendix 1). The overall sensational interests score consists of 19 items
of potentially forensic interest which loaded on the ®rst principal component during the original
validation study (items 3±5, 7±10, 13, 14, 16±19, 22±25, 27 and 28). The ®rst principal component
breaks down into two main Varimax subscales: `Militarism' (items 4, 7, 9, 13, 14, 16, 17, 22, 23
and 25) and `Violent Occultism' (items 3, 7, 8, 13, 18, 24, and 28). There are three other subscales:
`Intellectual Activities' (items 1, 11, 15, 19, 20, 21 and 26), `Credulousness' (items 1±3, 10, 18, 27
and 28) and `Wholesome Interests' (items 5, 6, 9, 12 and 26). These are subscales derived from
®ller items used to embed the forensically oriented items. The subject gives one of ®ve responses
which are verbally anchored, and which ranges from `Great Dislike' to `Great Interest', corre-
sponding to scores between ÿ2 and +2. The scales have alpha coecients between 0.68 and 0.84
(Egan et al., 1999).
1000 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007

2.3.2. The Sensation Seeking Scale±form V (SSS)

Form V of the Sensation Seeking Scale has 40 items, (Zuckerman, 1994), which breaks-down
into four subscales of 10 items: Thrill and Adventure Seeking (TAS), Experience Seeking (ES),
Disinhibition (Dis), and Boredom Susceptibility (BS). The format of the scale is forced-choice for
each item. The scale has good psychometric properties.

2.3.3. Raven's Standard Progressive Matrices (SPM)

The SPM, (Raven, Court & Raven 1996), is a non-verbal intelligence test that measures general
intelligence. Its psychometric properties are exceptional and it is widely used. The SPM was
converted to an IQ score using a transformation equation based on data from the SPM manual
(Egan, 1994).

2.3.4. The Millon Clinical Multiaxial Inventory±II (MCMI-II)

The MCMI-II (Millon, 1987), is a 175-item, true-false, self-report inventory which produces a
diagnostic pro®le consisting of 22 clinical scales divided into four main sections: basic personality
pattern scales; severe or pathological personality disorder scales; moderate clinical syndromes;
and severe clinical syndrome scales. The MCMI-II is normalised to provide a typology of per-
sonality disorders in categories that roughly correspond to disorders of the third, revised edition
of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R; American Psychiatric
Association, 1988). According to Millon (1987) the MCMI-II has good psychometric properties
and validation data with a variety of populations suggest that it can be used with `a reasonable
level of con®dence in most clinical settings' (1987, p. 5).

2.4. Procedure

The study utilised stored data on 42 patients resident at Arnold Lodge, Leicester. These data
were collected for an ongoing, ethically approved project aimed at describing the characteristics
of the patient population, and determining who bene®ts most from the services provided at this
facility. Information from the psychometric measures used in this study was extracted and stored
within an SPSS data ®le. Procedurally, after informed consent was secured patients were inter-
viewed individually in a quiet room adjacent to their main ward. Patients were informed that they
may withdraw at any time and that participation was not compulsory. No individuals were
interviewed when ¯oridly psychotic or in any other acutely disordered state. Patients who had
diculties reading had the questionnaires read out to them and their responses recorded by the
author. No time limits were imposed for these interviews.

2.5. Statistical analysis of the data

An independent samples t-test (two-tailed) were used to test the signi®cance of any di€erences
between the MI and PD samples on their responses to the SSS and the SIQ, the SPM, and age.
Pearson product-moment correlations (r) were used to ascertain the statistical association
between the SIQ scales and the scales on the SSS; the SIQ and the MCMI-II scales (Clinical
personality pattern and Severe personality pathology scales only); and the SSS and the MCMI-II
scales. When statistical associations were found between the SIQ scales and the MCMI-II scales,
 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007 1001

partial correlations (r) were utilised to further examine the association whilst controlling for the
scales on the SSS-V. This test was only adopted when scales on both measures (SIQ & SSS-V)
correlated with a scale on the MCMI-II. Computer programmes from the SPSS package were
used for the computations.

3. Results

To examine the di€erences between the MI and PD participants, independent t-tests were used
to compare mean scores for each test measure. There were no signi®cant di€erences between the
MI and PD participants in relation to age, SPM, SIQ or SSS scores. Linear transformation of the
SPM scores suggested that the mean IQ for the cohort was 93.7 (SD=18.9). As there was no
di€erence between MI and PD participants with respect to their legal classi®cation they were
deemed a homogenous sample, and data were pooled.

3.1. Correlations between the SIQ, SPM and the SSS

Pearson's r correlation coecients were calculated between the SIQ, the SPM and the SSS for
the entire sample (Table 2). General sensational interests and the SIQ sub-components of violent
occultism and militaristic topics were signi®cantly positively correlated with general sensation
seeking and the SSS subscales of Dis and TAS. No signi®cant association was found between
general SIQ or the SIQ subscales and the ES or BS components of the SSS, or with the SPM. The
SPM was associated with the ES subscale of the SSS. These results clearly demonstrate that the
SIQ and its sub-components of militarism and the violent-occult are signi®cantly and quite highly
positively associated with sensation seeking, and in particular with the sub-scales of Dis and TAS,
but not IQ. The lack of association between IQ and the SIQ cannot be accounted for by restricted
IQ variance, as the range of scores was not attentuated.

3.2. Correlations between the SIQ and the MCMI-II personality disorder subscales (Table 3)

Pearson's r correlations were calculated between the SIQ scales and the MCMI-II personality
disorder sub-scales. This revealed that total SIQ was signi®cantly positively correlated with the

Table 2
Correlations between the SIQ, SSS and SPM for the entire samplea

SSS Total P< Dis P< ES P< TAS P< BS P< SPM P<

SIQ total 0.55 *** 0.56 *** 0.29 n.s. 0.46 ** 0.12 n.s. 0.02 n.s.
Militarism 0.44 ** 0.43 ** 0.16 n.s. 0.47 ** 0.05 n.s. ÿ0.14 n.s.
Violent occultism 0.50 *** 0.52 *** 0.20 n.s. 0.43 ** 0.15 n.s. 0.05 n.s.
SPM 0.29 n.s. 0.16 n.s. 0.37 * 0.13 n.s. 0.20 n.s.
a
n.s.=not signi®cant; *=P<0.05; ** P<0.01; *** P<0.001.
1002 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007

Anti-Social and Borderline personality disorder scales, and the violent occultism subscale was
positively associated with the Avoidant and Borderline personality disorder scales. However, no
other signi®cant associations were found between the SIQ scale and the remaining MCMI-II
scales. This indicates that individuals with a greater interest in `sensational' topics are perhaps
also characterised by higher scores on both the Antisocial and Borderline personality disorder
scales, as measured on the MCMI-II.

Table 3
Correlations between the SIQ and MCMI-II self-rated personality disorder subscalesa

SIQ total score P< SIQ militarism P< SIQ violent-occultism P<

Schizoid 0.11 n.s. 0.17 n.s. 0.14 n.s.

Avoidant 0.28 n.s. 0.23 n.s. 0.33 *
Dependent ÿ0.21 n.s. ÿ0.07 n.s. ÿ0.14 n.s.
Histrionic 0.05 n.s. 0.03 n.s. ÿ0.08 n.s.
Narcissistic 0.18 n.s. 0.14 n.s. 0.04 n.s.
Antisocial 0.37 * 0.30 n.s. 0.29 n.s.
Aggressive/Sadistic 0.19 n.s. 0.12 n.s. 0.13 n.s.
Compulsive ÿ0.27 n.s. ÿ0.07 n.s. ÿ0.36 n.s.
Passive/Aggressive 0.15 n.s. 0.06 n.s. 0.17 n.s.
Self-Defeating 0.15 n.s. 0.13 n.s. 0.19 n.s.
Schizotypal 0.26 n.s. 0.22 n.s. 0.31 n.s.
Borderline 0.38 * * 0.28 n.s. 0.41 **
Paranoid 0.19 n.s. 0.20 n.s. 0.13 n.s.
a
n.s.=not signi®cant; *P<0.05; ** P<0.01.

Table 4
Signi®cant correlations between the SSS and the MCMI-II personality disorder scalesa

SSS total P< Dis P< ES P< TAS P< BS P<

Avoidant 0.23 n.s. 0.34 * 0.05 n.s. 0.02 n.s. 0.26 n.s.
Dependent ÿ0.31 n.s. ÿ0.20 n.s. ÿ0.34 * ÿ0.18 n.s. ÿ0.12 n.s.
Narcissistic 0.39 * 0.47 ** 0.29 n.s. 0.06 n.s. 0.29 n.s.
Antisocial 0.34 * 0.58 *** 0.12 n.s. ÿ0.02 n.s. 0.32 *
Agg/Sad 0.32 * 0.47 ** 0.16 n.s. 0.02 n.s. 0.26 n.s.
Compulsive ÿ0.26 n.s. ÿ0.17 n.s. ÿ0.33 * ÿ0.09 n.s. ÿ0.17 n.s.
Passive/Agg 0.22 n.s. 0.44 ** 0.09 n.s. ÿ0.14 n.s. 0.32 *
Schizotypal 0.15 n.s. 0.40 ** ÿ0.05 n.s. ÿ0.10 n.s. 0.24 n.s.
Borderline 0.26 n.s. 0.54 *** 0.09 n.s. ÿ0.03 n.s. 0.15 n.s.
Paranoid 0.24 n.s. 0.48 ** 0.01 n.s. ÿ0.07 n.s. 0.31 *
a
Agg/Sad=Aggressive/Sadistic; Passive/Agg=Passive/Aggressive; Total=Sensation seeking total score; Dis=-
Disinhibition; ES=Experience Seeking; TAS=Thrill and Adventure Seeking; BS=Boredom Susceptibility; n.s.=not
signi®cant; * P<0.05; ** P<0.01; *** P<0.001.
 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007 1003

3.3. Correlations between the SSS-V scales and selected MCMI-II scales

In order to study the relationship between the SSS and the personality disorder subscales scales
of the MCMI-II, Pearson's r correlations were also calculated (Table 4). This revealed that the
total SSS score was signi®cantly and positively correlated with the Narcissistic, Antisocial, and
Aggressive/Sadistic personality disorder scales. No signi®cant associations were found with the
remaining MCMI-II scales. The Dis subscale showed signi®cant positive correlations with the
Avoidant, Narcissistic, Antisocial, Aggressive/Sadistic, Passive/Aggressive, Schizotypal, Borderline
and Paranoid personality disorder scales. The ES subscale of the SSS-V was signi®cantly negatively
associated with the Dependent and Compulsive personality scales of the MCMI-II. For the BS sub-
scale correlation coecients were signi®cantly positively associated with the Antisocial, Passive/
Aggressive, and the Paranoid personality scales on the MCMI-II. No signi®cant correlations were
found between the TAS subscale of the SSS and the MCMI-II personality disorder scales.
To examine if the association between the SIQ and the personality disorder subscales of the
MCMI-II was mediated by sensation seeking, partial correlations controlling for sensation-seek-
ing were conducted for those MCMI-II scales that were signi®cantly associated with both sensa-
tion seeking scales and sensational interest scales to start with. After controlling for total SSS
score there was no signi®cant association between the total SIQ scale and the Antisocial person-
ality disorder scale on the MCMI-II (r(37)=0.23). This suggests that the relationship between
`sensational' interests and certain personality disorders is mediated by general sensation seeking.

4. Discussion

Our study found that a greater interest in general sensational topics was associated with higher
scores on sensation seeking, in particular the SSS's sub-components of Dis and TAS. Neither IQ,
BS nor ES were related to sensational interest scores. The correlations between the SIQ and the
MCMI-II personality disorders con®rm the mediating in¯uence of normal personality traits on
expressed personality disorders, as the association between SIQ and measures of Antisocial and
Borderline personality disorder breaks down once sensation seeking is controlled for. Our results
show that the relationship between sensational interests and psychopathology is essentially
mediated by the trait of sensation-seeking. The lack of signi®cant di€erence on responses to the
SIQ and SSS-V between MI and PD participants reiterates doubts about the validity of using
Mental Health Act (1983) classi®cations to di€erentiate groups of o€enders. Parallel studies by us
using slightly larger cohorts demonstrate that legal classi®cation does not di€erentiate MI and
PD patients on the NEO-FFI, measures of social-problem solving, or criminal cognitions (Egan,
McMurran, Richardson & Blair, 2000; McMurran, Egan, Blair & Richardson, 2000).
The Dis scale, which shows the strongest association with general sensation interests and the
violent occultism subscale, is said to represent the more non-conforming or less socially accep-
table kinds of sensation seeking (Zuckerman, 1978). The Dis scale also shows the strongest rela-
tionship of the SSS to psychopathy (Horvath & Zuckerman, 1993) and relates to Eysenck's
Psychoticism (P) scale, which is also linked to seeking new sensations and stimuli (Eysenck &
Eysenck, 1975). Although the TAS scale re¯ects a type of more socially acceptable sensation
seeking (Perez & Torrubia, 1985), its association with sensational interests can be explained due
1004 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007

to the nature of such interests. The failure to ®nd associations between the BS and ES scales and
sensational interests can be explained by way of the scale characteristics. For example, despite ES
re¯ecting the need for novel situations and experiences through the mind and senses, Zuckerman
and Little (1986) argue that it is most relevant to broad tastes (e.g. conventional art or intense
rhythmic music) and not violent and morbid curiosity. Given the nature of the topics on the SIQ
it would be surprising if such an association existed, although it may for the other (not investi-
gated) SIQ components (e.g. intellectual activities).
Is the association we identify of any use? Excitation-transfer theory posits that exposure to
stimuli may generate excitational states that intensify post-exposure emotional responses (Zill-
mann, 1980). Viewers who seek graphic horror for excitement report positive a€ect both before
and after viewing, whereas viewers who have di€erent motivations (e.g. who watch as a result of
anger) report negative a€ect (Johnston, 1995). Sensation-seekers may display an increased inter-
est in sensational topics to heighten arousal, and depending on motivation, label such arousal
euphoric or dysphoric. Sensation-seekers may tend to activate catecholamine systems in the brain
from their initially low levels found in unstimulated conditions. At optimal levels of activity these
systems may serve as the biological substrate of pleasurable arousal or `intrinsic reward' (Zuck-
erman & Little, 1986). One might hypothesise that an interest in sensational topics may have a
similar activation on central catecholamine activity, with habituation to the arousal generated by
the topic leading to more obvious behavioural try-outs, as suggested by MacCullough, Snowden,
Wood and Mills (1983). This cascade of mechanisms provides a top-down model of arousal in
which psychological drives in¯uence biological states. One task of clinicians and carers of this
population would be to divert the person from arousal via sadistic material to arousal from more
wholesome material to prevent escalation.
The rating scheme adopted on the SIQ is not ideal. In principle, the scale anchors responses of
individuals verbally to reduce the ambiguity and uncertainty in individual ratings and so increase
the reliability (Egan, Miller & MacLennan, 1998). However, negative responses to certain interest
preferences may cancel out positive responses to others, resulting in arti®cially low scores. Future
research could adopt alternative rating schemes that could be self-report or observer rated. For
example, using a rating scheme similar to that used by the revised Psychopathy Checklist (Hare,
1991) the presence of such interests could be rated from `0' (not present), to `1' (present in some
respect) and ®nally `2' (de®nitely present). Another possibility is that the SIQ re¯ects o€ence
types rather than psychopathology itself. Individuals who have a greater interest in such topics,
regardless of MHA (1983) classi®cation, could share some o€ence characteristic, such as some
form of sadism that emphasises power over others. This argument gains support from the fact
that the interest topics contained in the SIQ were originally developed from the interest pre-
ferences outlined by Brittain (1970) concerning sadistic sexual murderers.
Overall, however, the present study unambiguously demonstrates that the SIQ and its sub-
components of militarism and the violent-occult are associated with sensation seeking, and in
particular the SSS subscales of Dis and TAS. Sensation-seeking mediates the relationship between
such interests and the expression of personality disorders, and is independently associated with a
number of more pathological personality disorders. These results clarify the relationship between
personality characteristics and interest preferences, and provide a theoretical underpinning for
this area of research. The present study also further con®rms the mediating in¯uence of normal
personality traits on expressed personality disorder.
 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007 1005

Appendix

The sensational interests questionnaire (SIQ)

Great Slight No Slight Great

dislike dislike opinion interest interest
1. Alternative medicine GD SD NO SI GI
2. Astrology GD SD NO SI GI
3. Black magic GD SD NO SI GI
4. Body-building GD SD NO SI GI
5. Camping GD SD NO SI GI
6. Country and hill-walking GD SD NO SI GI
7. Crossbows, knives & swords GD SD NO SI GI
8. Drugs GD SD NO SI GI
9. Fishing GD SD NO SI GI
10. Flying Saucers GD SD NO SI GI
11. Foreign travel GD SD NO SI GI
12. Gardening GD SD NO SI GI
13. Guns and shooting GD SD NO SI GI
14. Martial Arts GD SD NO SI GI
15. Medicine GD SD NO SI GI
16. Mercenaries and the SAS GD SD NO SI GI
17. Motorbikes GD SD NO SI GI
18. Paganism GD SD NO SI GI
19. Philosophy GD SD NO SI GI
20. Psychology and psychiatry GD SD NO SI GI
21. Singing and making music GD SD NO SI GI
22. Sporting activities GD SD NO SI GI
23. Survivalism GD SD NO SI GI
24. Tattoos and body-piercing GD SD NO SI GI
25. The Armed Forces GD SD NO SI GI
26. The environment GD SD NO SI GI
27. The paranormal GD SD NO SI GI
28. Vampires and werewolves GD SD NO SI GI

References

American Psychiatric Association (1988). Diagnostic and statistical manual of mental disorders (Revised ed.).
Washington, DC: APA.
Blackburn, R. (1987). Two scales for the assessment of personality disorder in antisocial populations. Personality and
Individual Di€erences, 8, 81±93.
Blackburn, M., & Maybury, C. (1985). Identifying the psychopath: the relation of Cleckley's criteria to the inter-
personal domain. Personality and Individual Di€erences, 6, 375±386.
1006 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007

Brittain, R. P. (1970). The sadistic murderer. Medicine, Science, and the Law, 10, 198±207.
Costa, P. T., & McCrae, R. R. (1992). The NEO PI-R professional manual. Odessa, Florida: Psychological Assessment
Resources.
Daderman, A. (1999). Di€erences between severely conduct-disordered juvenile males and normal juvenile males: the
study of personality traits. Personality and Individual Di€erences, 26, 827±845.
Edwards, E. (1984). The relationship between sensation seeking and horror movie interest and attendance. Unpublished
doctoral dissertation, University of Tennessee, Knoxville.
Egan, V. (1994). Intelligence, inspection time and cognitive strategies. British Journal of Psychology, 85, 305±315.
Egan, V. (1999). Sensational and lurid interests in mentally disordered o€enders. Presentation to the ISSID conference,
July 1999, Vancouver, Canada.
Egan, V., Auty, J., Miller, R., Amadi, S., Richardson, C., & Gargan, I. (1999). Sensational interests and general per-
sonality traits. Journal of Forensic Psychiatry, 10, 563±578.
Egan, V. G., Crawford, J. R., Goodwin, G. M., & Brettle, R. P. (1990). The Edinburgh cohort of HIV-positive drug
users: current intellectual function is impaired, but not due to early AIDS dementia complex. AIDS, 4, 651±656.
Egan, V., McMurran, M., Richardson, C., & Blair, M. (2000). Criminal cognitions and personality: what does the
PICTS really measure? Submitted. Criminal Behaviour and Mental Health. (in press).
Egan, V., Miller, E., & MacLennan, I. (1998). Does the personality questionnaire provide a more sensitive measure of
cardiac surgery related anxiety than a standard pencil-and-paper checklist? Personality and Individual Di€erences, 24,
465±473.
Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the Eysenck personality questionnaire. London, UK: Hodder and
Stoughton.
Fabregat, A. A., & Beltri, R. T. (1998). Viewing of mass media violence, perception of violence, personality and aca-
demic achievement. Personality and Individual Di€erences, 25, 973±989.
Hare, R. D. (1991). The Hare psychopathy checklist±revised. Toronto: Multi-Health Systems.
HMSO (1986). Mental Health Act 1983. London: HMSO.
Horvath, P., & Zuckerman, M. (1993). Sensation seeking, risk appraisal, and risky behaviour. Personality and Indivi-
dual Di€erences, 14, 41±52.
Johnston, D. D. (1995). Adolescents' motivations for viewing graphic horror. Human Communication Research, 21,
522±552.
MacCullough, M. J., Snowden, P. R., Wood, P. J., & Mills, H. E. (1983). Sadistic fantasy, sadistic behaviour and
o€ending. British Journal of Psychiatry, 143, 20±29.
McMurran, M., Egan, V., Blair, M., & Richardson, C. (2000). The relationship between social problem-solving and
personality in mentally disordered o€enders. Personality and Individual Di€erences. (in press).
Millon, T. (1987). Millon Clinical Multiaxial Inventory-II: Manual for the MCMI-II (2nd ed.). Minneapolis, MN:
National Computer Systems.
Montag, I., & Birenbaum, M. (1986). Psychopathological factors and sensation seeking. Journal of Research in Per-
sonality, 20, 338±348.
Newcomb, M. D., & McGee, L. (1991). In¯uence of sensation seeking on general deviance and speci®c problem
behaviours from adolescence to young adulthood. Journal of Personality and Social Psychology, 61, 614±624.
Perez, J., & Torrubia, R. (1985). Sensation seeking and antisocial behaviour in a student sample. Personality and
Individual Di€erences, 6, 401±403.
Potts, R., Dedmon, A., & Halford, J. (1996). Sensation seeking, television viewing motives, and home television view-
ing patterns. Personality and Individual Di€erences, 21, 1081±1084.
Prins, H. (1988). Dangerous Clients: Further observations on the limitation of mayhem. British Journal of Social Work,
18, 593±609.
Prins, H. (1990a). Some observations on the supervision of dangerous o€ender patients. British Journal of Psychiatry,
156, 157±162.
Prins, H. (1990b). Bizarre behaviours: boundaries of psychiatric disorder. London, UK: Routledge.
Prins, H. (1998). Dangerous o€enders: some problems of management. International Review of Law Computers &
Technology, 12, 299±313.
 V. Egan et al. / Personality and Individual Di€erences 30 (2001) 995±1007 1007

Raven, J. C., Court, J. H., & Raven, J. (1996). Standard progressive matrices: manual for Raven's progressive matrices
and vocabulary scales. Oxford Psychology Press.
Sparks, G. (1991). The relationship between distress and delight in males' and females' reactions to frightening ®lms.
Human Communication Research, 17, 625±637.
Wainwright, M. (1995). Killer wrote about murder and occult. The Guardian, 13/12/95, 7.
Zaleski, Z. (1984). Sensation seeking and preference for emotional visual stimuli. Personality and Individual Di€erences,
5, 609±611.
Zillman, D. (1978). Attribution and misattribution of excitatory reactions. In J. H. Harvey, W. J. Ickes, & R. F. Kidd,
New directions in attribution research, Vol. 2. Hillsdale, NJ: Lawrence Erlbaum Associates.
Zillman, D. (1980). Anatomy of suspense. In P. H. Tannenbaum, The entertainment functions of television (pp. 133±
163). Hillsdale, NJ: Lawrence Erlbaum.
Zuckerman, M. (1978). Sensation seeking and psychopathy. In R. D. Hare, & D. Schalling, Psychopathic behaviour:
approaches to research (pp. 165±186). London: Wiley.
Zuckerman, M. (1979). Sensation-seeking: beyond the optimal level of arousal. Hillsdale, NJ: Erlbaum.
Zuckerman, M. (1983). A biological theory of sensation seeking. In M. Zuckerman, Biological Bases of Sensation
Seeking, Impulsivity and Anxiety (pp. 37±76). Hillsdale NJ: Erlbaum.
Zuckerman, M. (1984a). Sensation seeking: a comparative approach to a human trait. Behavioural and Brain Science, 7,
413±471.
Zuckerman, M. (1993). P-impulsive sensation seeking and its behavioural, psychophysiological biochemical correlates.
Neuropsychobiology, 28, 30±36.
Zuckerman, M. (1994). Behavioural expressions and biosocial bases of sensation seeking. Cambridge University Press.
Zuckerman, M. (1997). The psychobiological basis of personality. In H. Nyborg, The scienti®c study of human nature:
tribute to Hans J. Eysenck at 80. Pergamon.
Zuckerman, M., & Little, P. (1986). Personality and curiosity about morbid and sexual events. Personality and Indivi-
dual Di€erences, 7, 49±56.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.