Correspondences between theory of mind, jumping to conclusions, neuropsychological measures and the symptoms of schizophrenia
Introduction
Research in the field of cognitive neuropsychiatry in schizophrenia typically employs a univariate group-study approach, whereby measures from one experimental task are compared between patients and controls, and/or patient subgroups. In this fashion the cognitive neuropsychiatry literature has gradually evolved to a state where multivariate associations are implied by the pattern of univariate results. Frequently these interrelationships involve associations with the symptomatology of schizophrenia, such that several cognitive/neuropsychological tasks can be linked to the symptoms of schizophrenia in complementary patterns, potentially providing insight into the nature of the underlying neurocognitive systems if empirically supported.
For example, tasks measuring social cognition and reasoning biases have been applied to the study of schizophrenia in order to shed light on the cognitive underpinnings of positive symptoms (Blackwood et al., 2001, Garety and Freeman, 1999). With respect to social cognition, it was initially speculated that impairments in understanding the mental states of others would lead to a misinterpretation of social cues, possibly resulting in paranoia and other delusions (Corcoran et al., 1995, Frith and Corcoran, 1996). However, the empirical evidence for these associations is often not well founded. Some schizophrenia studies have shown associations between measures of social cognition and delusions (Corcoran et al., 2007, Pickup and Frith, 2001, Pousa et al., 2008), but others have shown associations with negative symptoms (Bell and Mishara, 2006, Brüne, 2005, Harrington et al., 2005, Mazza et al., 2001, Shamay-Tsoory et al., 2007) or disorganization (Sarfati et al., 1997, Sprong et al., 2007). With respect to reasoning biases, it was hypothesized that impairments in reasoning, such as early termination of information gathering, could also result in paranoia and other delusions (Garety et al., 1991, Huq et al., 1988); however, the empirical evidence for whether or not reasoning biases correlate with positive symptoms is mixed (Corcoran et al., 2007, Garety and Freeman, 1999, McKay et al., 2007, Menon et al., 2006, Moritz and Woodward, 2005, Young and Bentall, 1997).
Negative symptoms are typically found to be associated with standardized neuropsychological measures such as verbal fluency, memory and executive functioning (Liddle and Morris, 1991, Moritz et al., 2001b, Norman et al., 1997, Woodward et al., 2003, Woodward et al., 2004), and this has been largely confirmed in meta-analytic reviews (Aleman et al., 1999, Johnson-Selfridge and Zalewski, 2001). However, associations between positive symptoms and executive measures have also emerged in meta-analytic reviews (Johnson-Selfridge and Zalewski, 2001). Moreover, these associations are sometimes not replicated (Green and Walker, 1985, Liddle, 1987, Norman et al., 1997).
Thus, while reasoning biases and social cognition are theorized to associate with positive but not negative symptoms, and memory, executive and fluency performance are theorized to associate with negative but not positive symptoms, this pattern of association and dissociation is not particularly well supported by the empirical evidence from univariate studies. Advancement of these theoretical accounts would benefit from multivariate studies, whereby measures of reasoning biases, social cognition, verbal fluency, memory and executive functioning could be administered to a single sample, and factor analytic techniques used to explore how these measures share variance. Shared variance between these measures suggests that they may be underpinned by a common cognitive operation. Factor or component scores can then be computed and saved, and correlated with positive and negative symptoms.
In the current study we were primarily interested in two of the most commonly studied reasoning and social cognition paradigms: jumping to conclusions (JTC; Garety et al., 1991) and theory of mind (ToM; Frith and Corcoran, 1996). JTC paradigms typically involve the beads task, where the subject is presented with jars containing beads of two colours (e.g. red and white beads divided 60–40 in one jar and 40–60 in the other) and asked from which jar beads are being drawn when the jars have been hidden from view. Schizophrenia patients tend to request fewer beads before deciding on which one of the jars is the source of the beads. The JTC response pattern on probabilistic reasoning studies is thought to be caused by a “data gathering bias” (Garety and Freeman, 1999) whereby patients with delusions arrive at hasty conclusions using less information than healthy controls. This JTC response pattern has been widely replicated in patient groups (Garety et al., 1991, Huq et al., 1988, Menon et al., 2008, Menon et al., 2006, Moritz and Woodward, 2004, Woodward et al., 2008), and some studies have concluded that this response pattern might be associated with positive symptoms as a whole (Menon et al., 2008).
ToM is perhaps the most commonly investigated aspect of social cognition in schizophrenia, for which consideration of situational variables is required in order to determine the perspectives and intentions of other people. This has been of interest for the study of positive symptoms, because delusions can involve a misunderstanding of what is on the mind of others (e.g., paranoid delusions). The Hinting task (Corcoran et al., 1995) is a widely used ToM task, and measures the ability to infer the real intentions behind indirect speech (e.g., what did the child really mean when they said “Mom I'm hungry” when passing the candy aisle). The need to relate ToM measures to other aspects of cognition has been mentioned in reviews, as this has been identified as an important future direction for schizophrenia research (Harrington et al., 2005, p. 282). These relationships have been investigated in other fields of research; for example, in developmental psychology associations have been observed between ToM and inhibitory control (Carlson et al., 2002) and in autism associations have been observed between ToM and set shifting (Ozonoff et al., 1991).
Based on theory, and the original motivation for applying JTC and ToM cognitive challenge tasks to the study of schizophrenia, JTC and ToM may be expected to share variance with each other and with positive symptoms, whereas verbal fluency, memory and executive functions may be expected to share variance with each other and with negative symptoms. The objective of the current project was to use a multivariate approach to assess whether this pattern of associations and dissociations would emerge from an empirical multivariate analysis. This involved two steps: a data-driven factor analytic approach to investigate the pattern of inter-relationships among the cognitive and clinical neuropsychological measures, followed by correlating the resulting factors with the positive and negative aspects of schizophrenia.
Section snippets
Participants
Forty-six patients were recruited from the inpatient and outpatient services of the Centre for Addiction and Mental Health (CAMH), Toronto. The study and the recruitment procedures were approved by the Research Ethics Board of CAMH and the University of Toronto, and all subjects provided written, informed consent prior to their participation. Thirty-five patients had a current diagnosis of schizophrenia, nine of schizoaffective disorder, and two of schizophreniform disorder. All clinical
Results
Using Statistical Package for the Social Sciences (SPSS) 16.0 for Windows, the following measures were submitted to a principal component analysis with varimax rotation: Acquisition (M = 10.52, S.D. = 2.93), Free recall (M = 4.41, S.D. = 2.02) Recognition (M = 7.02, S.D. = 1.72), Verbal Fluency (M = 8.72, S.D. = 2.59), Category Fluency (M = 7.17, S.D. = 2.08), Rule Extraction (M = 48.35, S.D. = 9.39), ToM (M = 16.76, S.D. = 3.40), and JTC (M = 7.04, S.D. = 5.35). The eigenvalue-greater-than-one rule and scree plot converged
Discussion
Based on theory, and the original motivation for applying cognitive challenge tasks to the study of schizophrenia, JTC and ToM may be expected to share variance with each other and with positive symptoms, whereas verbal fluency, memory and executive functions may be expected to share variance with each other and with negative symptoms. The objective of the current project was to use a multivariate approach to assess whether this pattern of associations and dissociations would emerge from an
Acknowledgements
The authors thank Sandy Richards, RN, CMHPN(c), CCRC and Ivana Furimsky, RN, MN, CPMHN(c), CCRC for their recruitment efforts, and Pablo Rusjan, PhD for software development. TSW is supported by a Scholar award from the Michael Smith Foundation for Health Research, and a New Investigator Award from the Canadian Institutes of Health Research. RM is supported by a New Investigator Award from the Canadian Institutes of Health Research.
References (59)
- et al.
Does negative symptom change relate to neurocognitive change in schizophrenia? Implications for targeted treatments
Schizophrenia Research
(2006) - et al.
Explaining delusions: a cognitive perspective
Trends in Cognitiv0e Sciences
(2006) - et al.
Social functioning, theory of mind and neurocognition in outpatients with schizophrenia; mental state decoding may be a better predictor of social functioning than mental state reasoning
Psychiatry Research
(2006) Theory of mind and the role of IQ in chronic disorganized schizophrenia
Schizophrenia Research
(2003)Emotion recognition, ‘theory of mind,’ and social behavior in schizophrenia
Psychiatry Research
(2005)- et al.
Schizophrenia, symptomatology and social inference: investigating “Theory of mind” in people with schizophrenia
Schizophrenia Research
(1995) - et al.
Selective impairments of theory of mind in people with schizophrenia
Schizophrenia Research
(2001) - et al.
‘Jumping to conclusions’ and delusions in psychosis: relationship and response to treatment
Schizophrenia Research
(2008) - et al.
The selective effect of antipsychotics on the different dimensions of the experience of psychosis in schizophrenia spectrum disorders
Schizophrenia Research
(2006) - et al.
Neuropsychological correlates of schizophrenic syndromes in patients treated with atypical neuroleptics
European Psychiatry
(2001)