Delusion, the overvalued idea and religious beliefs: a tive act and is underpinned by a thought

process, it does not usually involve intro

spection and judgement, unless the object
comparative analysisof their characteristics perceived is of such an idiosyncratic nature
that it calls into question well-established
EDGAR JONES and J. [Link]
beliefs. The sighting of an alien spaceship,
for example, would cause many people to
disbelieve the evidence of their eyes and to
re-examine their existing beliefs about the
nature of the universe. Churchland (1979)
suggested that observational beliefs are
natural rather than artificial, stable rather
Background The studysoughtto In recent years the literature of descriptive than fluid, manifest rather than speculative,
investigate the form ofthe delusion in psychopathology has been enriched by ideas and autonomous rather than parasitic.
drawn from philosophy. Hemsley & Garety Derived or non-observational beliefs involve
schizophreniaandthe overvalued ideain
(1986) proposed Bayes's decision-making introspection and focused thought, and are
anorexia, employing a range of belief theorem as a theoretical basis for examining subject to the will. The belief that democracy
characteristicsto discoverwhether differ reasoning biases in deluded subjects, and this is a superior form of political organisation to
ences could be detected between them, hypothesis was subsequently tested in clinical oligarchy is an example of this type. It relies
studies (Huq et a!, 1988; Garety et a!, 1991). upon a complex collection of other initial and
and how they stood in general relationship
Delusion has been defined in a variety of derived beliefs about the capacities and rights
to the religiousbeliefsof normals.
ways, and to date no consensus has been of individuals and the means of dividing
reached on either its nature, taxonomy or power between them. This abstract proposi
Method A beliefratingscalewas
origins (Winters & Neale, 1983; Oltmans & tion could not be asserted with conviction
devised with 2 characteristics, and Maher, 1988; Butler & Braff, 1991; Roberts, solely on the basis of perception.
completed by 20 schizophrenics, 1992). Although a number of investigators It was hypothesised that delusions and
20 anorectics, and 20 normal controls. have dissented from the notion that it other abnormal beliefs exhibit certain char
Comparisons were drawn between popu constitutes a form of belief (Fulford, 1989; acteristics that may be used to draw
Spitzer, 1990; Berrios, 1991), many continue distinctions between them or to define a
lations using the Mann—Whitney test, and
“¿to
treat delusions as generally describable as group of subtypes, and that the schizo
differenttypes of beliefwere contrasted beliefs―(Garety & Hemsley, 1994). Belief phrenic delusion could be differentiated
within diagnosticgroups usingeachsubject itself has been the subject of much inquiry by from the overvalued idea in anorexia solely
astheir own control by repeated-measures philosophers of mind, and some oftheir ideas in terms of their phenomenal qualities. It
MANOVA. have been used to frame this investigation. was further proposed that a mismatch may
However, a distinction has to be drawn arise between the form and content of
Results The schizophrenic
delusionwas between those who argue that beliefs are abnormal beliefs. The content of most
simply the result of complex organic schizophrenic delusions suggest that they
differentiated from the overvalued idea in
processes and therefore require a physical should be treated as derived beliefs, as they
anorexia by a number ofvariables, which account, and a functional school which are commonly of a persecutory nature such
also served to distinguish both phenomena proposes that subjective phenomenal proper that the subject regards himself as the victim
from religious beliefs held by [Link] ties may serve to define specific mental states of an external attack or conspiracy. To
schizophrenic delusion exhibited many of and help to explain why individuals may reach a similar conclusion, a normal would
treat their beliefs in different ways. Some have needed to gather evidence, work out
the qualitiesofan initial (or observational)
functionalist philosophers have suggested motives and consider the probabilities of
belief, when its content suggested that it that belief may be interpreted in two distinct this situation happening. The delusion
shouldmanifestthose ofa derived belief. ways: those that a subject reports as being would, therefore, qualify as a derived
The anorectic overvalued idea,although based principally on observation (called belief involving introspection and judge
occasionally an initial belief in terms of its ‘¿initial')
and a second group that have been ment. Given that most delusions are said
describedvariouslyas ‘¿theoretical'
(Church to arise suddenly and fully-formed, it was
content, was typically held in the form of a
land, 1979), ‘¿derived'
(Smith & Jones, 1986) hypothesised that they would, in fact,
derived belief. or ‘¿non-observational'(Table 1). exhibit the qualities of initial beliefs, while
What, then, are the respective character the overvalued idea in anorexia would take
Conclusions A wider rangeof istics of the two forms of belief? Firstly, the form of a derived belief.
characteristics is required to define all the initial or observational beliefs (defined by
differences between delusion and the Churchland (1979) as “¿acquired as sponta METHOD
overvalued idea, and these have implica neous non-inferential responses to sensory
states of the perceiver―)possess something of Subjects
tions for beliefmodification programmes.
the ‘¿I
know' about them; the product of Subjects were in three groups: 20 paranoid
perception, they gain a sense of certainty. schizophrenics selected by ICD—1O(World
Although observation is commonly a posi Health Organization, 1992) criteria (F20.0),

381
TableI Thecharacteristics
of initial(observational)
andderived(non-observational)
beliefs and the role of focused thought in its
formation. The 12 characteristics were
Initial
beliefsIbeliefsDerived
represented by the following pole opposites:
(a) Conviction: believe absolutely —¿
believe
.Primary andimmediateI .Secondary and forming by
[Link]-inferential not at all.
(b) Influence on behaviour: makes a big
difference to what I do —¿
makes no
[Link] [Link] introspection and difference to what I do.
to I know―statements [Link]
assenta with varying levels ofconviction or
(c) Influence on cognition: has a major
[Link].
high degree of
impact on my thinking —¿
has no impact
on my thoughts.
(d) Truthfulness: is absolutely true —¿
is not
true at all.
who held delusions with conviction; 20 No significant differences were observed
(e) Importance: is very important to me —¿
is
anorectics (FS0.0), who were at least 10% between groups for age and, in the case of
not at all important.
below their ideal body weight; and 20 schizophrenics and normals, they were
normal controls drawn principally from matched by gender. It proved impossible to (f) Frequency: think about it all the time —¿
NHS employees, who held religious beliefs match the anorectics in this way, as all of never think about it at all.
in the mainstream Christian tradition and those who agreed to take part in the study (g) Acceptability: acceptable to everyone —¿
regularly attended church. The schizo were female. The educational status of all acceptable to no-one.
phrenics were recruited consecutively from subjects was recorded (both academic and
(h) Use of imagination: requires great imagi
four acute adult wards at Guy's Hospital, vocational qualifications) and no significant
nation —¿
requires no imagination.
which served an inner-city area of southeast differences were observed between the three
London, and of those approached six groups using a Friedman two-way ANOVA. (i) Speed of formation: forms very quickly —¿
refused to participate. They comprised 12 forms very gradually.
males and six female patients with a mean (j) Perceptualevidence:is based largelyon
age of 32.72 years (s.d.=11.08). Subjects Design of the rating scales perception —¿
is not based on perception
with organic brain disease or a history of It had been shown that deluded subjects are at all.
drug or alcohol misuse were excluded from capable of rating the characteristics of their (k) Focused thought: requires a lot of
the study. The anorectic population were belief in clinical investigations. Garety thought —¿
requires no thought at all.
out-patients attending an eating disorders (1985) tested two patients over time,
unit at the Royal Free Hospital in north asking them to rate the fixity and intensity (1) Affective content: evokes strong feelings
—¿involves no feelings at all.
London, and had a mean age of 28.07 years of their delusions on a five-point scale. Brett
(s.d.=8.03). They were recruited consecu Jones et a! (1987) used a similar rating scale Most deluded patients were interviewed
tively, three declining to take part in the to assess conviction, preoccupation and on two occasions; first to explain the nature
study. The anorectics were interviewed to interference (in terms of having an identifi of the inquiry and to elicit the nature of their
discover the key beliefs that influenced their able influence on behaviour) for nine single beliefs, and second to gather the data, an
excessive concern with low weight. These case studies in which deluded subjects were exercise that took between 40 minutes and
proved to be of considerable variety, ranging interviewed on average 12 times over six one hour. All subjects were volunteers and
from statements such as “¿I am overweight― months. Garety & Hemsley (1987) then confidentiality was guaranteed. Each was
or “¿I
look fat―,to “¿calories
are dangerous― widened the number of belief characteristics shown a diagram of the rating scale and how
and “¿eatingis a form of moral weakness―. to 11, adding resistance, dismissibility, this related to particular characteristics.
The 20 normals, with a mean age of 31.99 perceived absurdity, the extent to which They were then asked to rate on a five
years (s.d.=8.21), belonged to the following the belief is thought self-evident, reassur point scale the degree to which the char
faiths: Church of England (six), Roman ance-seeking, the extent to which the belief acteristic represented his or her experience
Catholic (five) and non-conformist Christian causes worry, and unhappiness, together (one at its greatest and five for its absence or
sects (nine). They too were recruited with pervasiveness, and asked subjects to opposite). Every effort was made to explain
consecutively (with no refusals) and all had rate them on a scale of 1 to 10 (Garety et a!, the meaning fully, and ample time was given
an established faith, none being the result of 1988). It was decided to adopt five of these so that participants could make an unpres
a conversion experience. Members of idio variables: conviction; capacity to influence sured assessment, particularly in the case of
syncratic minority groups were excluded. behaviour; frequency of consideration; the more complex variables, such as use of
This sampling bias was introduced to cultural acceptability (absurdity); and affec imagination. If the subject encountered
increase the severity of the comparison, tive content. A further seven characteristics, difficulties, the investigator would return
since some of the beliefs held by religious some drawn from the philosophy of belief, on a later occasion. The rating exercise for the
extremists may be considered delusional. A were added: capacity to influence other schizophrenic group had to be abandoned in
Christian population was selected in order thoughts; objective truthfulness of the three cases, either because the subject suffered
to contrast the characteristics of faith and belief; importance to the subject; speed at from intrusive hallucinations or because
delusion, given that the latter commonly which the belief formed; role of imagination thought disorder prevented clear comprehen
have a religious theme. in its genesis; role of perceptual evidence; sion of their meaning. All the anorectics and

382
the controls were able to complete the a genuine attempt to record a subjective could have influenced their rating. Since the
exercise at a single interview. experience. variables are not truly independent and
The idea of including control beliefs for because they have been analysed by non
deluded subjects had been explored by Statistical analysis parametric methods, a Bonferroni correc
Garety (198S), who devised statements tion was applied, and P values of < 0.008
Because the data collected were ordinal in
that would not be expected to fluctuate, were considered significant.
nature, and pilot studies suggested that the
such as the belief that “¿the sun will rise
scoring would not be normally distributed,
tomorrow―, and that “¿God exists―.In view Delusion versus the overvalued idea
the Mann—Whitney test was selected as a
of the distinction drawn by philosophers and religious belief
way of analysing any differences that might
between initial and derived beliefs, it was Delusion in schizophrenia was differentiated
arise between populations by specific belief
decided to include representatives of both from the overvalued idea in anorexia for
characteristics. Three comparisons were
types as controls for each subject. For the seven of the 12 belief variables, namely:
performed:
former, a simple observational belief, such conviction, truthfulness, acceptability,
as “¿thisis a chair―, was selected. For the (a) schizophrenic delusion v. anorectic over
imagination, speed, use of perception, and
latter a more complex proposition of valued ideas;
use of thought (Table 2). Significant differ
particular meaning to the subject was (b) schizophrenic delusion v. normal reli ences were found between delusion and
chosen, examples including a belief in “¿the gious beliefs; religious belief (that God exists) for four
value of law and order―or “¿equal rights for variables (frequency, imagination, speed and
(c) anorectic overvalued ideas v. normal
all―.While only two control observational perception), while the results for thought
religious beliefs.
beliefs were employed for all subjects, it and affect may indicate a trend. The
proved impossible to find a single derived In addition, to eliminate the effect of
comparison between overvalued ideas in
belief to which all subjects assented. With subjective scoring, each individual was used
anorexia and religious beliefs in normals
this intention, “¿theexistence of gravity―was as their own control in order to draw intra
demonstrated that they appeared to have
included in the rating scale, but several group comparisons between specific belief
much in common, although the latter were
subjects dissented from the proposition. As a types. Taking the schizophrenic population
rated as being significantly more truthful
result, a derived belief of personal impor as an example, the ratings for delusion
and acceptable, while having a lower
tance was chosen as the appropriate recorded by the first patient were compared
affective component.
comparison, given that both the delusion with their respective control derived beliefs,
and the overvalued idea are commonly and then the procedure repeated for all
invested with great significance. It is not members of the sample in order to build up Delusion versus its control beliefs
claimed that a homogeneity existed between composite results for each characteristic. In The first repeated-measures MANOVA
all these control derived beliefs, but they the absence of a non-parametric test which compared delusion with its control initial
were selected because, like their corre can perform these calculations with a beliefs and showed that the two could not be
sponding abnormal beliefs, they were held reasonable level of power and reliability, differentiated by the following variables:
with conviction, and unlike them there the repeated-measures MANOVA was conviction, truthfulness, use of imagination
could be no dispute about their content. employed in a cautious spirit of inquiry, in and perceptual input (Table 3). In other
Hence, each individual was asked to rate a the knowledge that the variable scores do words, delusion had the impact and
minimum of three beliefs: one abnormal or not exhibit a normal distribution and the certainty of an observed event without any
religious, and two control beliefs, consisting ratings are not interval in nature. Within the element of fantasy. Yet significant differ
of one initial and one derived. schizophrenic group, delusion was ences were detected between the two that
In assessing delusion according to five contrasted with both control initial and probably related more to the circumstances
belief characteristics, Brett Jones et a! (1987) derived beliefs, the same procedure being in which delusion arises than to its inherent
also explored the stability of these subjective followed for the overvalued ideas of the properties. Delusion was rated as being
ratings. For the variables ‘¿conviction' and anorectics. more important, considered more
‘¿interference',they recorded high coeffi frequently, and its capacity to influence
cients of reliability. Pilot studies undertaken RESULTS behaviour and cognition were far greater,
for this investigation confirmed these find while it had a more powerful affective
ings and, in order to measure any changes to Although a number of significant P values component. Although delusion formed
beliefs, the rating scale was repeated after an were generated by the Mann—Whitney test, quickly, it was typically less sudden than
interval of one or two weeks with two of the these results had to be qualified because the an observation, required more thought in its
schizophrenics and two of the anorectics in characteristics themselves were not truly genesis and was considered less acceptable.
the study. A comparison by Mann—Whimey independent. For example, a belief which If delusion assumes some of the features of
test, adjusted by the Bonferroni correction, is held with great conviction is often an initial belief, then it may also be
revealed that there were no significant regarded as important and commonly distinguished from control derived beliefs.
differences between the scores of the same thought to be truthful. Such a belief is The repeated-measures MANOVA did
individuals for all three types of belief. This likely to have the capacity to determine both discriminate between these two types of
suggested that, although a subject's ratings action and thought. While these features can belief for five variables. Delusion was rated
for delusion or overvalued ideas may alter as be distinguished semantically, in practice as being more important than control
a result of treatment, scores given at a there were elements that overlapped or had derived beliefs and, forming more rapidly,
particular time are not random but represent implications for other characteristics and it was considered more frequently but

383
Table2 Comparison
byMann—Whitney
test (Pvalues) anorectic overvalued ideas and their
control initial beliefs for all characteristics
Characteristic Schizophrenic
[Link] v. Schizophrenic delusion v. Anorectic idea (Table 4). The comparison between over
valued ideas and their corresponding control
beliefConviction idea normal religious belief normal religious
derived beliefs showed that, for five char
0.020lnf. 0.006 0.637 acteristics, the two groups could not be
[Link]
behaviour 0.839 0.070 differentiated: conviction, importance, use
0.154Truthfulness 0.431 0.548 of imagination, perceptual evidence and use
0.001Importance <0.00 I 0.035
of thought. Hence, overvalued ideas in
anorexia exhibited many of the features of
0.780Frequency 0.938 0.709
derived beliefs, being differentiated by their
0.054Cult. 0.377 0.006
greater ability to influence both behaviour
0.002Imagination
acceptability 0.005 0.955
and cognition, the frequency with which
0.618Speed <0.001 <0.001 they are considered, their doubtful objective
0.039Perception <0.00I <0.00I truthfulness, a low level of acceptability,
0.380Thought 0.00 I <0.00 I and their powerful affective component.
0.043Affect <0.00 I 0.0 I I
0.003Table 0.773 0.0I8 DISCUSSION
The pattern of schizophrenic beliefs
MANOVACharacteristic
3 Comparison of beliefswithin the schizophrenicgroup by repeated-measures One of the investigation's hypotheses was
that a mismatch occurs between the appro
priate form and content in delusion; an
Delusion
belief95%Cl
v. control initial belief Delusion v. control derived abnormal belief which by virtue of its
content should exhibit the form of a
PConviction P 95%Cl
derived belief, takes on the characteristics
0.577lnf. —¿0.10—0.17 0.577 —¿0.09—0.17 of an initial belief. Delusion could not be
0.096lnf.
behaviour 0.38—0.90 <0.00 I —¿0.03—0.39 distinguished from a control observation by
0.234Truthfulness
cognition 0.76—1.86 <0.00 I —¿0.22—0.86
the four variables of conviction, truthful
ness, use of imagination and perceptual
0.088Importance —¿0.15—0.23 0.083 0.05—0.61
evidence. Forming quickly with little or no
0.004Frequency 0.69-1.78 0.00 I 0.15—0.61
focused thought, in the nature of an
<0.001Cult. 0.61—1.22 <0.001 0.34-1.00
inspirational idea, delusion appears to be
0.126Imagination
acceptability —¿
l.72——-0.76 <0.00 I —¿0.90—0.12 invested with great importance and is
ISpeed —¿0.
I7-0.3 I 0.54I —¿2.43—--
I.48 <0.00 regularly considered. It is emotionally
0.001Perception —¿0.72--—0.060.024 0.43—1.48 charged, and, by the subject's own assess
0.367Thought —¿0.59—0.16 0.249 0.31-0.81 ment, has great capacity to influence
0.102Affect 0.67—1.73 <0.00 I —¿0.86——0.08 behaviour and cognition. It manifests many
1.26—2.35 <0.00I 0.14-0.85 0.009 of the qualities of ‘¿I know' statements.
While perception appears to play an impor
tant part in schizophrenic delusion when
compared with its control initial beliefs and,
required less imagination. It had a higher gorised as abnormal. It may be the case, indeed, with the overvalued ideas of anor
affective component. Nevertheless, both therefore, that belief formation in schizo ectics, this may, in part, be related to the fact
types of belief were held with similar levels phrenia is underpinned by a general cogni that 16 of the 20 patients in this group had
of conviction and their ability to influence tive bias, and that delusion is not the isolated experienced or continued to suffer from
cognition and action were rated as of symptom of a specific defect. Schizophrenics auditory hallucinations. To guard against
equivalent strength. They were scored as might have a tendency to rate perceptual this effect, the observational aspect of
having a similar level of acceptability, evidence far higher than normals or, rather, perceptual evidence was stressed by the
implying a lack of insight into the bizarre to temper it less by introspective thought. investigator according to the definition of
quality of many of the abnormal beliefs in Within the inter-group comparisons, obser an initial belief. In addition, it had been
the study. Although delusion had been vation played a lesser part in the overvalued planned to contrast the results from the
scored as relying significantly more upon ideas of the anorectics and religious beliefs schizophrenic population with subjects diag
thought, but no less on perception, than a of normals than in the delusions held by the nosed as having persistent delusional
control observational belief, it was not schizophremcs. disorder (F22.0) and, therefore, not influ
differentiated significantly from control enced by perceptual abnormalities. They
derived beliefs by these two characteristics The overvalued idea versus its proved particularly difficult to find and
of evidence. This implies that delusion is control beliefs only seven subjects were tested, too few to
held in a similar way to other beliefs whose The repeated-measures MANOVA test provide a significant sample. However, their
content would not lead them to be cate discriminated significantly between the ratings for perception are recorded as they

384
 Table4 Comparisonof beliefswithinthe anorecticgroupbyrepeated-measures
MANOVA deficit related to the processing of informa
tion (Kaney & Bentall, 1989; Bentall et a!,
CharacteristicOvervalued idea v. control idea v. control
1994). In order to attribute events in this
fashion, it may be necessary to abandon
belief95%Cl
initial beliefOvervalued derived
normal inferential processes such that the
PConviction—0.52.-—0.l8 P95%Cl resulting delusions acquire the immediacy,
high levels of conviction and assumed
0.055lnf. <0.00 I—0.43—0.0 I truthfulness of perceptually-based beliefs.
0.004lnf.
behaviour0.45—1.04 <0.00 I0.15—0.70 In addition, delusions are commonly
0.015Truthfulness—
cognition1.34—2.33 <0.00 I0.10—0.82 supported by auditory hallucinations,
<0.001Importance.40—2.56 l.65——0.76 <0.00 I— I.42——0.56 thereby reinforcing their basis in percep
0.163Frequency0.50—1.06 <0.00I—0.09—0.52 tion.
0.002Cult. <0.00I0.23—0.90 These results must be treated with
caution, since the investigation has a
acceptability—2.59—— 1.79
<0.001Imagination0.71 <0.00 I— l.53——0.8l
number of limitations. The first concerns
0.273Speed— —¿
I .76 <0.00 I—0.21 —¿0.7
I
the nature of the population under scrutiny.
0.061Perception— 1.94—— 1.24 <0.00 I—0.80—0.02
None of the schizophrenics rated was
0.368Thought1.96—2.64 l.l8——0.52 <0.00 I—0.69—0.27
grossly thought-disordered. Although some
0.204AffectI <0.00I—0.10—0.46 subjects exhibited mild derailment, most in
.92—2.54 <0.00 I0.03—0.82 0.036 this group were able to concentrate upon a
diverse range of topics and score specific
characteristics within a single interview.
Those schizophrenics who were deluded
scored this characteristic highly: four scored able use of imagination, together with a and difficult to comprehend, because of
1, two scored 2, and one scored 3. These lesser input from perception. Also highly desultory thought or drivelling, were not
results provide anecdotal evidence that emotionally charged, they too are rated as included in the inquiry. Others, who as a
observation is implicated in individuals' having a considerable capacity to influence result of preoccupying thoughts were unable
interpretations of their delusions. The thought and action. to discuss unrelated subjects, were also
general finding is that delusion is phenom excluded. Both of these populations hold
enologically more complex than proposed, CONCLUSIONS delusions, and it may be the case that if a
manifesting characteristics significant in method could be devised for measuring
terms of both initial and derived beliefs. The underlying assumption in this investiga them, they would reveal a different pattern
tion was that delusion, like the overvalued from the abnormal beliefs of schizophrenics
idea, may be regarded as akin to a normal who were comprehensible. Although all of
The pattern of anorectic beliefs belief in that it can be defined according to a those tested were English-speaking and had
It was further hypothesised that a discont variety of characteristics and responsive to received all or part of their education in the
. inuity might arise between form and content external influence. These phenomena can be UK, some had been born overseas or were
@ in the case of the overvalued idea: that what measured in terms other than simply convic the children of immigrant parents and were,
I are often expressed as initial beliefs assume tion. This confirms earlier findings that therefore, subject to external cultures. This
I the characteristics of derived beliefs. deluded subjects can reflect upon their applied in all three groups, which had each
I However, it was found that many of the beliefs and give apparently valid ratings for been recruited from inner-city London. The
C key beliefs that led to extreme dieting were their experience (Garety & Hemsley, 1994). impact of culture upon the content of
C derived in character, and those which were It also appears that some of the qualities delusion is well established, and it remains
t based upon perception (such as “¿Iam employed by philosophers of mind to an open question as to whether it may also
C overweight―) also involved an element of distinguish between initial (observational) have an effect on the way that beliefs are
r interpretation. Not surprisingly, therefore, and derived (non-observational) beliefs may held.
C the anorectic overvalued idea was sharply be applied with value to the descriptive What, then, are the implications for
c differentiated from control initial beliefs and psychopathology of delusion. clinical practice that emerge from these
5. appeared akin to control derived beliefs. The finding that schizophrenic delu findings? Firstly, the fact that delusion is
a Hence, overvalued ideas in anorexia may sions have much in common with percep held to be objectively true, and supported
h occasionally be initial in content, but all tually-based beliefs may have implications by perception in the manner of ‘¿I know'
. . appear to exhibit the form of derived beliefs. for their origins and formation. It has been statements, reinforces the conclusions of
it They may be summarised as being held with argued that persecutory delusions reflect various belief-modification studies. These
St conviction but less forcefully than a delu abnormal inferences about the causes of demonstrated that direct confrontation
:@:sion.
Although
considered
important
and threatening events, which have the function had markedly less success than
g repeatedly examined, their objective truth is of maintaining a positive sense of self. programmes which taught patients to
se open to doubt and they are not regarded as Subjects ascribe negative outcomes to question and re-evaluate the evidence on
@ acceptable to the general population. external causes, suggesting the existence which they based their delusions (Watts et
@ 0 Arrived at more gradually, they are the of an established cognitive bias, possibly a!, 1973; Milton et a!, 1978). This
L product of rumination and involve consider underpinned by a specific neurological investigation shows that the former

ne

385
 approach is like telling a normal person
that their belief that grass is green is
CLINICAL IMPLICATIONS
mistaken, and that it is actually red.
Because conscious introspection and the I Delusions are multidimensional phenomena and can be measured in terms other than
weighing-up of data seem to be limited in strength of conviction.
the process of delusion formation, their
deliberate introduction should help to U The importance ofperception as a source ofevidence for delusions may have implica
reduce levels of conviction and place a tions for the designof beliefmodification programmes.
brake on the ‘¿jumpingto conclusions' style
. The emphasison perceptionindelusionmayalsohaveimplicationsfor a reasoningbias
manifested by many schizophrenics
(Garety et a!, 1991). in the formation of delusions.
The nature of the overvalued idea in LIMITATIONS
anorexia suggests that a different strategy
should be followed in attempting to modify . The subjectivenatureofscoring,howevercarefullythe ratingscaleisdefined,
beliefs. Supported by imagination and produces ordinal data which reducesthe reliability ofthe results.
subject to repeated thought, it seems that
I The difficulty oftesting thought-disordered or very aroused schizophrenics has led to
these beliefs may be more complex than they
a comprehensible subpopulation being examined.
appear. If the schizophrenic delusion repre
sents the crystallising of imagination into . Althoughallthosetestedwere English-speaking andreceivedallor part oftheir
reality, then the overvalued idea comes into education in the UK, some had been influenced by other cultures and this may have
being when reality is distorted and expanded
affected the way thatthey held their beliefs.
by fantasy. The anorectic, for example, who
believes that she is overweight or that
calories are dangerous has taken a factual
notion and imbued it with associations,
consequences and emotions which it may
EDGARJONES,PhD,[Link],FRCPsych,UMDS Divisionof Psychiatry,Guy's [Link]
not fully warrant. A programme of modifi
cation should, perhaps, concentrate upon Correspondence: Dr Edgar Jones,Division of [Link]'s Hospital, London SEI9RT
why this has happened and explore the
implications of what began as a relatively (First received 12May 1995,final revision 28 August 1996,acceptedS September 996)
straightforward statement.

ACKNOWLEDGEMENTS Fulford, K. M.W (1989)Moral Theoryand MedicalPractice. Kaney. S. a Bentail, R. P.(1989) Persecutorydelusionsand
Cambridge:CambridgeUniversity Press. attributionalstyle.&itishJournalofMedicalPsychology,
62,
This paper was based upon a PhD thesis supervised 191—
198.
byProfessor
[Link] Gaiety, P. A. (I9@5) Delusions: problems in definition and
[Link].
58, 25—34. Milton, F., Patwa,V. K. a Hafn.r, R. J. (19fl) Confrontation
text, the authors thank Dr A. j. D. Macdonald of vs. beliefmodificationin [Link]
@
Hither Green Hospital. and for their help with the —¿ a a R. (1987) Characteristics of delusional JournalofMedical Psychology. SI. 27—130.
statistical analysis. Cohn [Link] ofthe University [Link] Psychiatryand Neurological
of Westminster,Richard Neville and Dr Sube Baner Sciences,236, 294—298. OItmanns,T. F.a Mahsr, B. A. (.ds) (1980) Delusional
jee ofthe Institute of Psychiatry. Belief@.
New York:JohnWiley & Sore.
—¿ a (1994) Delusions, Investigations into the Psychology of

DelusionalReasoning.
Oxford: Oxford Univeristy Press. [Link], C. (1992)The [Link]
REFERENCES Psychiatry.161,293—308.
—¿ . [Link], B. S. a Hem@ie@, D. R. (I@8$) The character

isticsofdelusions:a cluster analysisofdeluded subjects. Smith, P.a Jones,0. R. (1986) ThePhilosophyofMind. An

@ [Link]. R. P.,[Link]*n, P.a S. (1994) The self,
EuropeanArchivesof Psychiatryand NeurologicalSciences, 237, [Link]:CambridgeUniversity Press.
attributionai processesandabrormal beliefs:towards a model
112—114.
of persecutory [Link] Therapy.
Spitzer, M. (1990) The phenomenologyof delusions.
32. 331—¿341.
—¿ , [Link].y, D. R. a Weueiy, S. (1991) Reasoning in Comprehensive Psychiatry,31,377—397.
B@rr1os,G. (1991)Delusionsas‘¿wrong beliefs':a conceptual deludedschizophrenicand paranoidpatients:biasesin
[Link] Psychiatry.159(suppi.14).6—13. performance
ona probabilistic
inference
[Link]
of Watts, F. N., PoweII,G. E. a A@t1n, 5.V. (1973) The modi
Nervousand Mental Disease,179,l94—2l0. ficationof [Link] Medical
[Link], J., Garety, P. a [Link], D. (1987) Measuring Psychology.46, 359—363.
delusionalexperiences:a method and its [Link] [Link].y, D. R. a Garety, P.A. (1986)The formation and
Journalof ClinicalPsychology,
26, 257—265. maintenanceofdelusions:a [Link] [Link], K. C. a r@i.&., j. N. (1983) Delusions and delu
ButI.r, R.W a Braff, D. L (I9@I) Delusions: a review and Psychiatry.149,SI—¿56. sionalthinking in psychotics:a review ofthe [Link]
[Link]
Bulletin,17.633—647. Psychology Review,3 227—253.
Huq, S. F.,[Link], R A. a Hemiley. D. R. (19U) Probabil
churchland, P.M. (1979)ScientifIcRealismandthe Plasticityof isticjudgementsin deludedand ron-deluded [Link]. World Health OrganIzation (1992) TheICO-lO Classifica
Mnd. Cambridge:CambridgeUniversity Press. terlyJournalofExperimentalPsychology,
40A, 801—812. tion of Mental and Behavioural Disoiders. Geneva: WHO.

386
Delusion, the overvalued idea and religious beliefs: a comparative
analysis of their characteristics.
E Jones and J P Watson
BJP 1997, 170:381-386.
Access the most recent version at DOI: 10.1192/bjp.170.4.381

References This article cites 0 articles, 0 of which you can access for free at:
[Link]
Reprints/ To obtain reprints or permission to reproduce material from this paper, please write
permissions to permissions@[Link]

You can respond /letters/submit/bjprcpsych;170/4/381

to this article at
Downloaded [Link] on March 9, 2017
from Published by The Royal College of Psychiatrists

To subscribe to The British Journal of Psychiatry go to:

[Link]