Dimensionality of schizotypy in young people according to sex and age

https://doi.org/10.1016/j.paid.2008.03.011Get rights and content

Abstract

The dimensionality of schizotypy and differences, according to sex and age, were studied with a new instrument called The Thinking and Perceptual Style Questionnaire (TPSQ). Participants were 789 young people with a mean age of 19.7 years (SD = 1.65). The results indicated that the TPSQ has adequate psychometric properties, with a Cronbach’s α coefficient of 0.91, ranging from 0.73 to 0.87 for the subscales. The construct validity gives a four-dimensional solution of the scales. Furthermore, statistically significant differences in schizotypy were found as a function of sex and age. Men scored higher on Anhedonia and Illusion subscales, whilst women scored higher on Negative Evaluation, Social Paranoia and Hallucination subscales. By age, the youngest participants tend to score higher on the Physical Anhedonia and Hallucination subscales. Thus, the TPSQ would appear to be a useful instrument for measuring schizotypal traits in young people, although future research will be needed to explore its relationship with neurocognitive, clinical and social functioning measures.

Introduction

Clinical and research experience shows that psychotic symptoms are not specific to certain disorders; rather, they are expressed on a continuum, and can be found with less intensity but in a persistent way in normal persons (Beauchaine et al., 2008, Claridge, 1997). However, unusual experiences have been considered as indicating proneness to psychotic disorders, and based on this assumption, “high risk” psychometric studies have been carried out to determine risk of psychosis, often using schizotypy scales which, although measuring personality traits (Verdoux & van Os, 2002), include many items which, in their most extreme form, can resemble psychotic symptoms.

The first attempts to measure schizotypy were made by Golden and Meehl (1979), whose tradition has been absorbed into what are generally referred to as “psychosis proneness” scales (Chapman et al., 1995, Mason and Claridge, 2006, Raine, 1991, Venables et al., 1990).

One of the contributions to the debate around the concept of schizotypy and psychosis proneness is the empirical evidence that the construct is not one-dimensional, but rather multi-structural, with components that may actually reflect different (dys)functions. Since the 1980s, numerous psychometric studies of schizotypy based on factor analysis with normal population have described schizotypy as a two-dimensional category made up of a positive factor (perceptual-cognitive deficit) and a negative factor (social-interpersonal deficit, of an anhedonic type) (Kwapil, Barrantes Vidal, & Silvia, in press), clearly similar to the dimensions observed in schizophrenia; however, the majority of recent solutions based on factor analysis give three or four schizotypy factors (Fonseca Pedrero et al., 2007). The three-factor versions usually add a disorganization, social decline or affective factor to the traditional positive and negative ones (Lewandowski et al., 2006, Venables and Rector, 2000). Moreover, in some three-factor solutions, the resulting schizotypy factors appear to be invariant across sex and age (Badcock and Dragovic, 2006, Fossati et al., 2003).

As regards the four-factor schizotypy structure (Linscott and Knight, 2004, Mason and Claridge, 2006, Rawlings et al., 2001, Stefanis et al., 2004, Suhr and Spitznagel, 2001, Venables and Bailes, 1994), this adds, to the three-factor solution, one of impulsive non-conformity, covering disinhibited characteristics guided by impulses, which may be violent or extreme (Claridge et al., 1996). Finally, some studies have found five-factor (DiDuca & Joseph, 1999) or even six-factor solutions (Mass et al., 2007).

The accumulated empirical evidence also suggests that schizotypal traits, as occurs with the symptoms of schizophrenia, vary according to certain demographic characteristics, and may have different pathognomonic meanings. Thus, for example, research has shown that schizotypal traits (except anhedonia) tend to diminish with age (Mason and Claridge, 2006, Mata et al., 2005, Rawlings et al., 2001), and that women display more positive symptoms and men more asocial aspects of an impulsive or anhedonic nature (negative symptomatology) (Fonseca Pedrero et al., 2007). Comparison between the different factorial studies of schizotypy becomes complicated, therefore, due to three elements which vary: the model of analysis, the scales employed for its measurement, and the type of sample.

The purpose of the present work was to study the dimensionality of schizotypy in a sample of young people, through a factor analysis of the scales and items of a recent instrument for the measurement of schizotypy (Linscott & Knight, 2004). Furthermore, given the scarcity of consistent data on differences according to sex and age in factorial studies on schizotypy (Fonseca-Pedrero, Lemos-Giráldez, Muñiz, García-Cueto, & Campillo-Álvarez, 2008), we set out to explore the relationship between the schizotypal dimensions, sex, and age.

Section snippets

Participants

A total of 789 young people (72.9% women), made up of college (social sciences) students (n = 639) and secondary school (technical courses) students (n = 150). Mean age of the sample was 19.7 years (SD = 1.65), with a range of 18 to 24. Participants were recruited from schools, selected to cover rural and urban areas, as well as different socio-economic statutes. Participants’ cooperation was voluntary, and no incentive was offered for participation.

Measurement instruments

Thinking and Perceptual Style Questionnaire (TPSQ) (

Descriptive statistics of the scales

The descriptive study of the Thinking and Perceptual Style Questionnaire scales reveals that the indices of asymmetry and kurtosis range from +1 to −1, except for the Perceptual Illusion scale, which presents an asymmetry of 1.54 and a kurtosis value of 2.57. Correlations between the scales of the questionnaire are shown in Table 1, and the following results are the most noteworthy: (1) the majority of the correlations are statistically significant, except for that of Physical Anhedonia and

Discussion and conclusions

The first objective of the present work was to study the dimensionality of the schizotypy construct in young people, through a factor analysis of the scales and items of the TPSQ; the second objective was to identify possible differences in schizotypal traits according to sex and age. The results support the TPSQ as a satisfactory measure for the assessment of schizotypy in populations of young people. The psychometric qualities of the questionnaire, as regards its internal consistency and

Acknowledgements

This work was supported by grants provided from the Spanish Ministry of Science and Technology (BES2006-12797, SEJ2005-08924, SEJ2005-08357) and the Principality of Asturias (IB05-02, COF05-005).

References (36)

Cited by (24)

  • Functional hemispheric asymmetry and nicotine dependency as variables mediating neurobiological vulnerability to schizotypy in a non-clinical population of college students

    2019, Personality and Individual Differences
    Citation Excerpt :

    It should be noted that so far only two studies have employed both a RHD and a LHD laterality task and have found no significant relation between PS and brain laterality (Herzig et al., 2010; Schofield & Mohr, 2014). Furthermore, findings are likely prone to gender differences: females have been found to score higher on PS, whereas males commonly score higher on NS (Miettunen & Jaaskelainen, 2010; Paíno-Piñeiro, Fonseca-Pedrero, Lemos-Giráldez, & Muñiz, 2008). Another possible explanation of the inconsistent findings reported in the literature pertains to the elevated consumption of psychotropic substances (e.g. nicotine) commonly observed in schizophrenia populations.

  • Psychometric properties of two measures for the assessment of schizotypy in adolescents

    2010, Psychiatry Research
    Citation Excerpt :

    These factors are quite similar to those found in the construct validity study at the level of the TPSQ subscales, even though an Illusion factor is added. Recently, Paino et al. (2008), carrying out a second-order factor analysis at the item level of the TPSQ in a sample of college students, found the same number and content of second-order factors, albeit in a different order. Fifth, construct validity analysis of the MSTQ-R found that the three-dimensional solution based on the factors PSs, NSs and INC presented an adequate fit to the data.

View all citing articles on Scopus
View full text