The evaluation of insufficient cognitive effort in schizophrenia in light of low IQ scores

Highlights

• A total of 16% of schizophrenia patients failed 2 effort tests and 32.1% failed 1.

• Low effort accounts for a large proportion of cognitive test performance variance.

• In schizophrenia patients, negative symptoms and low IQ predicted effort test scores.

Abstract

Low IQ has recently been shown to predict neuropsychological effort test failure in healthy and neurological populations. Although low IQ is common in schizophrenia (SZ), its effect on effort test performance remains unclear in this population. The current study examined the role of IQ in effort test performance in a sample of 60 outpatients with SZ and 30 demographically matched healthy controls (CN). Participants were administered a battery of neuropsychological tests, and insufficient effort was calculated using two embedded effort indices: the Reliable Digit Span Effort Index and the Finger Tapping Effort Index. Results indicated that 16.1% of SZ patients and 0% CN failed both effort measures and that 32.1% of SZ and 3.3% of CN failed one measure. In SZ, IQ in the <70 or 70–79 range was associated with the highest rates of falling below the effort cut-off scores; however, patients with IQs in the low-average or higher range (>80) did not fall below effort cut-offs. Findings suggest that low IQ is a significant predictor of insufficient effort during neuropsychological test performance in schizophrenia, calling into question the validity of neuropsychological effort testing in SZ patients with low IQ.

Introduction

Early clinical conceptualizations of schizophrenia (SZ) considered cognitive impairment to be a core feature of the disease (Kraepelin, 1919). Modern empirical studies have confirmed these observations, with meta-analyses of neuropsychological test performance indicating that SZ patients perform on the order of 1.0 standard deviation below that of healthy controls on standardized tests (Dickinson et al., 2007, Fioravanti et al., 2005). However, studies have yet to uncover a neurocognitive profile or pattern of differential deficits that characterizes the majority of people with SZ (Reichenberg and Harvey, 2007). Rather, individuals with SZ have been found to display deficits of similar magnitude across a range of cognitive domains (e.g., attention, working memory, executive functioning), with moderate interrelationships among tests (Dickinson, 2008, Dickinson et al., 2004, Dickinson et al., 2008). These findings have lead some to suggest that SZ is characterized by a “generalized neurocognitive deficit” (Dickinson and Harvey, 2009, Dickinson et al., 2008).

Several biological accounts have been proposed to explain the generalized neurocognitive deficit in SZ, including central nervous system (CNS) (e.g., impaired signal integration across neural networks, gray and white matter abnormalities, NMDA and GABA interneuron receptor dysfunction) (Dickinson and Harvey, 2009) and “general systems” abnormalities (e.g., inflammation, oxidative stress, metabolic dysfunction) (Asevedo et al., 2013, Friedman et al., 2010, Lindenmayer et al., 2012, Ribeiro-Santos et al., 2014). However, psychological explanations of the generalized neurocognitive deficit have received relatively little attention despite the fact that clinicians have long suspected that symptoms of avolition and apathy lead some patients to put forth insufficient effort during neuropsychological testing. The field of clinical neuropsychology has developed and validated several measures designed to detect insufficient effort during neuropsychological testing (Sollman and Berry, 2011, Vickery et al., 2001). When these measures have been applied to study insufficient effort in SZ, inconsistent rates of effort test failure have been observed. The majority of studies have indicated that approximately 15–25% of SZ patients fall below standard clinical cut-offs for low effort (Avery et al., 2009, Back et al., 1996, Duncan, 2005, Egeland et al., 2003, Gierok et al., 2005, Hunt et al., 2014, Moore et al., 2013, Pivovarova et al., 2009, Schroeder and Marshall, 2011) although a few studies report failure rates as high as 60%–72% (Gorissen et al., 2005, Hunt et al., 2014). Several factors may contribute to discrepant rates of effort test failure observed across studies, including inpatient versus outpatient status, level of personal and parental education, symptom severity, the sensitivity and specificity of effort tests administered, and whether the effort test used was embedded or free-standing (see Gorissen et al., 2005, Hunt et al., 2014, Morra et al., 2015, Strauss et al., 2015). Although the proportion of SZ patients failing effort tests may be inconsistent across studies, the ability of effort tests to predict neurocognitive impairment is reliable, with low effort accounting for approximately 25–35% of variance in global neuropsychological test scores (Avery et al., 2009, Gorissen et al., 2005, Morra et al., 2015, Strauss et al., 2015). Thus, there is some evidence that insufficient effort contributes at least to some extent to the generalized neurocognitive deficit that characterizes SZ.

It should also be noted that although falling below clinically derived effort cut-offs is taken as evidence for malingering (i.e., intentionally feigning cognitive impairment) in most clinical contexts (Bush et al., 2005) this is not the most plausible explanation for low effort in SZ. In the majority of studies published to date, neuropsychological evaluations were conducted solely for the purposes of research, not for clinical service or disability determination. Participants therefore had no readily identifiable incentive to put forth suboptimal performance during testing. Instead, the most viable explanation seems to be that individuals with SZ fail effort tests due to motivational impairments that are core to the disease and affect approximately 1/3 of the SZ population (Ahmed et al., 2014, Foussias and Remington, 2010, Strauss et al., 2013)Several studies now support this “motivational” interpretation, reporting that negative symptoms and psychological processes associated with negative symptoms (e.g., defeatist performance beliefs) are associated with effort test failure in SZ (Avery et al., 2009, Gorissen et al., 2005, Morra et al., 2015, Strauss et al., 2015).

In addition to negative symptoms, there are several other plausible moderators of insufficient effort that have received surprisingly little attention to date. Global intellectual functioning, as indexed via intelligence quotient (IQ) scores on standardized tests, is one particularly underexplored factor. Effort test performance has traditionally been thought to be immune to the effects of low IQ (Flaro et al., 2007); however, there is some evidence that IQ may in fact be highly predictive of insufficient effort in neuropsychiatric populations. For example, Dean et al. (2008) found that IQs in the borderline or lower range of intellectual functioning (<80) were associated with elevated rates of effort test failure in a sample of epilepsy patients. IQs in the <80 range are not uncommon in SZ (Aylward et al., 1984); however, it is unclear whether SZ patients with low IQ are at increased likelihood of failing effort testing. Only one study that we know of has evaluated the role of low IQ in effort test performance in SZ, with a correlation of r = −0.35 reported between the RBANS Effort Index and Full-Scale IQ (Morra et al., 2015). Thus, there is some preliminary evidence that low IQ may indeed be an important moderator of insufficient effort in SZ; however, more systematic investigations are needed to test this hypothesis fully.

The aims of the current study were threefold: 1) to evaluate the proportion of SZ patients falling below clinically derived cut-scores on well-validated embedded effort-tests; 2) to determine whether clinically rated positive, negative, and disorganized symptoms of SZ are significant moderators of effort-test performance; 3) to examine the role of low IQ in effort test performance in SZ. Based on prior studies, it was hypothesized that a minority of SZ patients (approximately 15–25%) would fall below established effort cut-off scores (Duncan, 2005, Egeland et al., 2003, Gierok et al., 2005, Hunt et al., 2014, Moore et al., 2013, Schroeder and Marshall, 2011)and that insufficient effort would be predicted by low IQ and greater severity of negative symptoms (Avery et al., 2009, Gorissen et al., 2005, Morra et al., 2015, Strauss et al., 2015). Additionally, consistent with the notion that low effort contributes to the generalized neurocognitive deficit, we predicted that insufficient effort would account for a significant proportion of variance in neuropsychological performance on tests not used to index insufficient effort.