Metacognition in schizophrenia spectrum disorders: Methods of assessment and associations with neurocognition and function

Background and Objectives: Research has confirmed that many with schizophrenia experience deficits in metacognitive capacity defined as impairments in the ability to think about thinking, both with regards to their own thinking and the thinking of others. These difficulties are related to, but not reducible to symptoms. One question posed here regards how these deficits are linked to other forms of cognitive deficits, including deficits in neurocognition, and how they and other forms of cognitive deficits are related to the ability to function. As neurocognition is degraded in schizophrenia, does the ability to think about one’s own thinking diminish? Do deficits in metacognition affect function in a manner semi-independent of deficits in neurocognition? Methods: To explore these possibilities, this paper reviews recent studies of metacognition as assessed within personal narratives of self and illness spontaneously generated by adults with schizophrenia spectrum disorders. Results: Studies are reviewed which suggest that impairments in verbal memory and executive function may interfere with the ability to form and sustain representation of one’s own internal state as well as the internal states of others. Additionally, results are deMETACOGNITION IN SCHIZOPHRENIA SPECTRUM DISORDERS... 221 The term “Metacognition” refers to a general ability to think about one’s own thinking. This capacity is thought to reflect a wide range of semi-independent faculties which allow individuals to form representations of their own mental states and the mental states of others; these representations enable one to form, challenge, and revise ideas of what is believed, felt, dreamt of, or feared in a number of rapidly evolving contexts1. Over the last two decades, research has suggested that deficits in metacognitive capacity are common in schizophrenia. Relative to those without psychosis, persons with schizophrenia have difficulties distinguishing the source of internal experiences, perceiving themselves as agents in the world, and detecting the intention and emotions of others from verbal and visual cues2-7. These deficits appear to be relatively stable over time and, while correlated with severity of psychopathology, are not simply a straightforward consequence of symptoms or other clinical features of schizophrenia8-11. The discovery of metacognitive deficits has led to interest in their relationships with other aspects of the disorder and with their potentially significant role in determining long term outcomes. In this paper we will first describe the development and psychometric properties of a novel method for assessing metacognition from personal narratives of self and illness. Next we will review recent studies exploring the link between metacognition, assessed using these procedures, with neurocognition, as well as its unique contribution to functional outcome, semiindependent of neurocognitive capacity. Assessing metacognition with personal narratives To date research on metacognition in schizophrenia has been hampered by a number of methodological issues. For one, most studies assess metacognition by measuring performance on a laboratory task which simulates social interaction or a self-reflective task, and so performance on these tests may not adequately capture the individual’s ability to engage in metacognitive acts in day-to-day situations where the content may be emotion-laden and personally salient. Second, since laboratory-based tasks cue or call for specific metacognitive acts at specific times, these tasks may not measure how persons think and act spontaneously. Third, metacognition is often studied as a one-dimensional phenomenon while metacognitive acts may involve capacities that tailed which suggest that metacognitive deficits directly affect function prospectively and may mediate the impact of neurocognitive deficits on functioning. Conclusions: Results are consistent with the possibility that a certain level of neurocognition is needed to think about thinking in a complex manner and that the ability to think about thinking is intimately related to the ability to work and relate to others among persons with schizophrenia. Received: 30 June 2010 Accepted: 21 July 2010 are conceptually distinct and may involve semi-independent neurocognitive functions12-14. Accordingly, it might be overlooked that the capacity to perform different metacognitive acts are linked with separable neurocognitive functions and that as a result some persons may experience deficits in certain domains of metacognition but not others. To measure metacognitive function as it might exist in emotionally salient tasks which do not cue for responses, we have proposed a method to rate metacognitive ability from a spontaneously generated speech sample about a person’s own life story. That speech sample is obtained through a semistructured interview called the Indiana Psychiatry Illness Interview (IPII). The IPII elicits a narrative about one’s self and illness and typically lasts 30 to 60 minutes15. The IPII differs from other psychiatric interviews in that it produces a self-narrative in which specific metacognitive acts may appear spontaneously with minimal scaffolding by the interview’s structure. To quantify metacognitive capacity within IPII narratives, we have used a modified version of the Metacognition Assessment Scale (MAS)16. Originally designed to detect changes in metacognition within psychotherapy sessions for persons with severe personality disorders -, the MAS has been adapted for the study of IPII transcripts as part of a collaborative effort with the original authors17. The MAS contains four scales which pertain to different foci of metacognitive acts: “Self Reflectivity”, or the comprehension of one’s own mental states, “Understanding of others’ minds”, or the comprehension of others’ mental states, “Decentration”, or the ability to see others as having independent motives, and “Mastery”, or the ability to use one’s mental states to accomplish cognitive tasks or cope with psychological distress. Metacognitive capacity can vary along a continuum, and each of the four MAS subscales are accordingly broken down into an ordinal scale of increasing complexity. Thus, persons given higher rating on a certain MAS sub-scale are thought to be able to perform more complex acts of metacognition within that domain. Evidence of good interrater reliability and internal consistency among the four MAS scales has been presented18. Concerning validity, MAS scores have been linked with assessments of insight, social cognition, coping preferences and self-reflectivity17,19,20. In one ongoing study, preliminary observations suggest that persons with schizophrenia experience graver deficits in metacognition as assessed within narratives compared to others with chronic medical conditions in the absence of psychosis. Four studies: Neurocognition, metacognitive capacities and functional outcome To study the relationship between neurocognition and metacognition, we first sought to determine whether ratings on the MAS were linked with performance on a range of neurocognitive tests17. Participants were 61 men with DSM-IV diagnoses of a schizophrenia spectrum disorder, recruited from an outpatient treatment setting. All were in a non-acute phase of illness and were not substance dependent. Participants were administered the IPII along with a neurocognitive test battery which included the Wisconsin Card Sorting Test (WCST)21, a test of abstract flexibility; the Hopkins Verbal Learning Test (HVLT)22, an auditory verbal memory test; the Vocabulary subtest from the Wechsler Adult Intelligence Scale III (WAIS III)23, a test of verbal intellectual function; the Digit Symbol sub222 PAUL H. LYSAKER ET AL.