In this article a critical evaluation is given of Eysenck's Psychoticism model. It is argued that in this model two sub‐models can be distinguished, which, contrary to Eysenck's presentation, cannot be regarded as true extensions of each other. With respect to one of these models, the ‚genotypic’︁ P‐model, the theory is criticized in that both schizophrenia and affective psychosis are determined by a common genetic predisposition which can phenotypically manifest itself in variations of P. Instead of this theory, the likelihood is put forward that a high EPQ‐P score, albeit in combination with a high N and a low E score (and notwithstanding the fact that criminals or psychopaths can also obtain high P scores), is (only) related to the schizoid state, and hence, that P seems to be relevant either as a predisposing factor contributing to the development of schizophrenic psychosis, or as a factor on which biological relatives of schizophrenics obtain higher scores on average than normals do. In this respect, Eysenck's theory that the non‐schizoid form of psychopathy can also be found among first‐degree relatives of schizophrenics, and hence, that psychopathy and schizoidia are genetically related, is also criticized. Furthermore, it is argued that Eysenck's EPQ‐P scale is not optimal for measuring those traits of the schizoid personality which are independent of N and/or E. Both arguments regarding the contents of this scale and arguments with respect to the demonstrated lack of invariance of the EPQ‐P factor are adduced to support this statement. Thus, an alternative scale for measuring ‚P’︁ (labelled S or Insensitivity) was designed by us. The S‐scale is based on literature concerning the schizoid state and reflects the results of a series of principal components analyses of (potential) S items, together with N and E items, put into execution with the intention of investigating the invariance of the S factor (and of E and N) with respect to six sample and other parameters. These investigations were carried out on a large, representative sample of the Dutch population. Additional investigations were carried out concerning the reliability and validity of the three newly formed scales. The results of these investigations turned out to be very satisfactory or, in some respects, at least promising. Finally, in this article, comments are made on the nature of the S factor, comparing this dimension with both Eysenck's P factor and the dimensions Agreeableness and Conscientiousness, as proposed, for instance, in McCrae and Costa's version of Nor‐man's 5‐factor model. As against P, the S or Insensitivity factor seems to be only (negatively) related to Agreeableness and not to Conscientiousness. It is also argued that this finding seems to be in accord with the supposed schizoid nature of S and the criticisms levelled at Eysenck's EPQ‐P scale.

In this reply to Eysenck, we attempt to clarify why we have criticized Eysenck's Psychoticism model and have found it necessary to introduce our own alternative model. It k concluded that the validity of the P scale as a measure of Psychoticism has not been demonstrated and that the P construct itsew is untenable.In 'How valid is the psychoticism scale? A comment on the Van Kampen critique ', Eysenck (1995) states that 'To summarize, Van Kampen (1993) has criticized the P scale, and suggested the validity of his S scale, with complete disregard for the universally recognized need for construct validity. His account of the P scale never mentions the large body of evidence supporting its construct validity, and his claim for the validity of his S scale simply refers to low correlations of his scale with other measures of similar concepts. Without evidence of construct validity, based on a properly formulated theory from which testable deductions can be made, personality traits and their corresponding questionnaire type measuring scales are merely correlational artifacts without a firm hold on reality' (p. 107).Of course, these statements of Eysenck seem to point to a very significant omission. For if the P scale is indeed a valid measure of the Psychoticism construct, there would have been no need to write such a critical article about Eysenck's P model, nor to replace the P scale with a scale for the measurement of S or Insensitivity. But in fact, things are not as simple as Eysenck suggests, and in our original publication many criticisms are levelled at P that Eysenck apparently passed over in silence. So, perhaps it makes sense to reiterate some of the main points addressed in our previous article to see why we have criticized Eysenck's P model and found it necessary to introduce the S scale.According to Eysenck (1992a), the theory of Psychoticism rests on two major presuppositions. 'The first is that all functional psychoses (schizophrenia, manicdepressive illness, schizoaffective disorders, monopolar depression) are related and do not form independent categories' (p. 670). Elsewhere, this presupposition is worded somewhat differently by stating that there exists 'some degree of generality

Postulating that the predisposition to illness in Claridge's disease model of schizophrenia can be equated with the personality dimensions S or Insensitivity, (low) E or Extraversion, and N or Neuroticism, as measured by Van Kampen's 3DPT, and assuming that the mode of transmission of schizophrenia is basically polygenic, the genetic and environmental etiology of S, E, and N was assessed in a sample of 52 MZ and 76 DZ twin pairs and their parents by means of LISREL. Besides, in a sample of 2118 subjects MAXCOV±HITMAX analyses were conducted for these factors as well as for the personality dimension G or Orderliness, but now assessed by the 4DPT, in order to ®nd out whether a discrete or quasi-discrete variable might also underlie these dimensions, giving support to the possibility of dominance or epistasis. The results obtained in these investigations favoured a model for all three dimensions, allowing for both additive and non-additive genetic eects in combination with non-shared environmental in¯uences. It was not possible to choose between a model involving dominance and a model involving epistatic genetic eects. With the use of scores corrected for sex and age, which were converted to normal scores, the proportion of variance explained by additive genetic factors was 20% for S, 40±41% for E, and 26±29% for N. Dominance or multiple-gene epistasis accounted for 37±38% (S), 19±20% (E), and 30±31% (N), and unshared environmental in¯uences for 42±43% (S), 41% (E), and 42±43% (N) respectively.