Finn publikasjon

Helge Andreas Hoff

Thinking About Symptoms of Psychopathy in Norway

Avhandling for graden philosophiae dpctor (ph.d.), University of Bergen. | 2015



The aim of this thesis is to study the construct of psychopathy, in particular how Norwegians think and talk about the features of psychopathy. This was explored by examining the content validity of a recently developed concept map of psychopathy named the Comprehensive Assessment of Psychopathic Personality Disorder (CAPP). The CAPP is based on the lexical approach to personality, and consists of 33 symptoms from six broad domains of personality functioning.

Method and Results

The first study obtained prototypicality ratings from 796 community residents, forensic mental health workers and corrections professionals in Norway. Participants rated the typicality of the 33 CAPP symptoms and nine symptoms of other personality disorders both for psychopathy and for a person without psychopathy. Confirmatory Factor Analysis was used to examine if the symptoms seem to represent the domains of the model as proposed by the CAPP authors. The vast majority of the CAPP symptoms were rated moderately or highly typical of a psychopath as well as untypical of a non-psychopath. There was a high degree of agreement across samples in terms of which symptoms were viewed as central and what symptoms were viewed as peripheral to the prototype of psychopathy. CFA indicated that unidimensional measurement of each of the six domains could be achieved after removal of a few symptoms from the analyses.

The second paper compared prototypicality ratings in Norwegian from the first paper to prototypicality ratings made using the original English CAPP in order to examine the degree of consistency of the CAPP across the two language versions. The latter ratings came from as study by Kreis and colleagues (2012). Multi-group confirmatory factor analyses (CFA) and Item Response Theory (IRT) methods were used to test for structural and measurement invariance across ratings based on the two language versions. A strong similarity of prototypicality ratings across the English and Norwegian language versions of the CAPP was found. CFA indicated structural invariance and IRT analyses indicated measurement consistency, the latter with a few possible exceptions.

The third paper used card sort methodology to examine if the CAPP authors’ grouping of symptoms into functional domains make sense to people in Norway. Groups of students and mental health workers (26 and 22 groups respectively) physically sorted the CAPP symptoms into the six domains proposed by the CAPP authors. Both students and mental health workers alike were able to categorize the CAPP symptoms speedily and intuitively according to the CAPP model, although some CAPP symptoms seemed more troublesome to sort according to the a-priori suggested concept map.


The three studies suggest that Norwegians share a lexical concept of psychopathy in the form of a prototype. The large majority of CAPP symptoms were rated as close to this prototype, though a few were not. Furthermore, Norwegians’ concept of psychopathy seems to resemble that of people who have made use of the English CAPP version when evaluating the CAPP symptoms. Finally, the CFA, IRT and the Card Sort all suggest that the CAPP symptoms represent the domains as proposed by the CAPP authors, with some possible exceptions Together this thesis supports the overall content validity of the CAPP as lexically based concept map of psychopathy in Norway.