This research assessed the neuropsychological function of FES patients after initial stabilization with atypical antipsychotic medications. Patients showed impaired performance on tasks requiring self-regulation (MSET) and EF (FAB, M-WCST) as compared to healthy controls. This study also used the DEX and BDSI questionnaires to characterize the DB that affects everyday life in FES. This is the first time that the latter questionnaire has been applied in a FES population. Importantly, we found that DEX and BDSI scores were correlated with aspects of EF, including self-regulation. Patients did not appear to lack awareness of their deficits (measured as the difference between DEX-S and DEX-SO results).
Strategic self-regulation
EF impairment is a cognitive deficit central to the symptomatology of schizophrenia. However, traditional neuropsychological tests of EF may not be sensitive enough to capture everyday dysexecutive problems in FES. These tests were originally designed to evaluate neurological patients and may therefore miss psychiatric-related cognitive deficits.
Qualitative information on EF is also important for diagnosis and treatment in psychiatric populations (26). The MSET, which requires patients to plan, organize, and monitor behavior over a brief period while carrying out a simple task, provides some qualitative data and is often used to measure EF in neurological patients (27). Our FES group performed more poorly than controls on MSET tasks. Specifically, patients completed fewer tasks and broke more rules; in other words, they showed impaired self-regulation. Shallice and Burgess emphasize that few neuropsychological tests require patients to organize or plan their behavior over long stretches of time or prioritize among competing tasks (28) even though this type of executive ability is an important component of many real-life activities. Our results indicate that patients already show impairment in this area after a single episode of schizophrenia despite stabilization with medication.
Our results are consistent with studies in chronic, treated FES patients (26,29,30,31) and one study of unmedicated FES patients (32) that report impaired performance on the MSET. This study demonstrates for the first time that schizophrenia slows cognitive processing even during its earliest stages. We propose that clinicians adopt this simple test as part of routine evaluation in FES. The literature is inconsistent in terms of the longitudinal evolution of self-regulation problems, and the specific cognitive deficit underlying the impairment remains unclear. Patients may lack the ability to perform the number of tasks required, for example, or they may perseverate on a single task, break the stated rules, or deviate from optimal distribution of time spent per task. Therefore, it is crucial that we continue to explore EF in schizophrenia. A prior study evaluated EF over the course of disease progression, applying the MSET in medication-naïve patients after a first episode of schizophrenia and then following the patients for several years. Consistent with our results, FES patients demonstrated impairment as compared to controls on the MSET. Importantly, this impairment persisted from the medication-naïve state to clinical stabilization and throughout the three years following the first psychotic episode, despite improved performance on a conventional executive test (M-WCST). MSET performance was not related to intelligence, education level, changes in symptoms, age of onset, or duration of untreated psychosis. Furthermore, better MSET performance during the medication-naïve state predicted improvements in negative and positive symptoms over the three-year study period. These findings suggest that impaired self-regulation, as measured by the MSET, is a primary deficit in schizophrenia that begins early in the course of the illness and remains present irrespective of clinical status for at least three years following the first episode (33). The Chinese version of the task has been adapted for use in patients with first-episode and chronic schizophrenia and has been tested in healthy adults in Hong Kong, with results indicating adequate sensitivity to deficits in attention allocation and planning (32,34,35). In one study, impaired attention allocation and planning at illness onset (FES), as measured by the MSET, were associated with risk of residual semantic disorganization after one year (36).
Dysexecutive behavior
Problems in DB are common clinical observations in FES, and we believe that standardized testing for this syndrome should be a routine part of clinical psychiatric practice. Various questionnaires have been designed to measure the impact of dysexecutive syndrome on daily life and overcome the low sensitivity of standard neuropsychological tests. The Dysexecutive Questionnaire (DEX) from the Behavioral Assessment of the Dysexecutive Syndrome (23,24) and the Behavioral Dysexecutive Syndrome Inventory (BDSI) from the Groupe de Réflexion sur l'Évaluation des Fonctions Exécutives (25) probe for symptoms that reflect DB in everyday life. In designing this study, we hypothesized that FES patients would receive scores indicative of DB on such questionnaires.
To test this hypothesis, we applied the DEX and BDSI questionnaires to a group of patients with FES and to a control group. We found that there were significant differences between the patient and control groups in terms of total score for both instruments. This result supports our hypothesis; patients with FES showed evidence of DB.
The BDSI has excellent diagnostic accuracy for executive disorders in mild-to-moderate Alzheimer's disease, stroke, and traumatic brain injury (37,38,39). This study marks the first application of this instrument in FES. We found statistically significant differences between patients and controls for most items on the instrument, suggesting that it is a sensitive test for DB in FES. Only three questions showed no significant differences between patients and controls. The first pertained to emotional symptoms (irritability-impulsivity-aggressiveness) that are clinically more typical of bipolar disorder than schizophrenia. The second, environmental autonomy is associated with catatonic symptoms, which are rare in FES. The third items address severe behavioral problems, which are also rare in this population.
In terms of the DEX, this study is the first to analyze not only total score but also item and factor scores in a FES population. The main behavioral problems reported by informants on the DEX-SO were related to the intentionality factor (including planning problems, poor decision-making, and lack of insight); executive memory factor (temporal sequencing problems); negative affect factor (apathy); and restlessness. On the DEX-S, the main behavioral problems recognized by the patients were related to the intentionality and executive memory factors and the impaired abstract reasoning item. These problems are common clinical complaints in FES and may partially explain the problems with school, work, and general social participation that this population experiences. As the above results indicate, there was significant overlap between symptoms detected by caregivers and self-reported problems. Studies using the DEX with chronic schizophrenia populations have reported that the total DEX-S score is typically lower than the total DEX-SO score (indicating incomplete awareness of impairment), and that only the latter score differs significantly from results for controls. Our study, in contrast, found that early-stage FES patients were generally aware of their deficits, suggesting that insight may deteriorate later in disease progression (29). This finding also separates FES from numerous neurological disorders associated with significant anosognosia.
Correlations between strategic self-regulation and dysexecutive behavior
Ecological validity, that is, the applicability of test results to real-life function, has become a major focus of neuropsychological research. Characterizing EF deficits is critical to predicting functional capacity in FES (40,41); if we assume that the MSET is a valid measure of EF and that the DEX and BDSI questionnaires are valid measures of the dysexecutive problems encountered in everyday life, we would expect strong correlation among the measures. In designing this trial, we hypothesized that self-regulation would correlate with more quantitative measures of DB in a sample of FES patients. However, we did not find any correlations between total MSET and total dysexecutive questionnaire scores. This finding is consistent with results in chronic FES and distinguishes this population from neurological patients, in whom the two measures are typically linked (29).
Total BDSI and DEX scores provide an overall measure of DB, including the numerous dimensions that might reflect underlying cognitive processing deficits. Burgess et al. identified at least 5 behavioral dimensions addressed by the DEX, such as inhibition and intentionality (24). The negative result here suggests that a more precise analysis of cognition in FES might require teasing out the relationships among strategic self-regulation and the specific dimensions and items on the dysexecutive questionnaire. In fact, we found a negative correlation between MSET: Subtasks Completed and DEX-SO: Negative Affect Factor; that is, lower strategic self-regulation scores were associated with more marked symptoms of this DB. This DEX dimension includes generalized apathy, which could affect successful execution of the MSET task. A study of chronic schizophrenia found a negative correlation between MSET: Categories Completed and DEX-SO total score, but that study did not analyze the dimensions separately (27). We also found a negative correlation between total MSET and BDSI: Euphoria, Lability, and Moria; that is, lower strategic self-regulation scores were associated with more marked symptoms of this DB. This BDSI item involves problems with attention and impulsivity, which could impede planning and behavioral monitoring during the MSET task.
One possible explanation for the scarce correlations here is that while many persons diagnosed with FES might indeed have EF deficits, some of their behavioral symptoms may be unrelated to executive deficits. Alternatively, the symptoms may be associated with executive deficits not addressed by the MSET. These symptoms would in any case cause problems for the patient in everyday life and therefore be detectable by the DEX and BDSI. It is clear that this topic merits further exploration. It is also possible that FES patients might have EF impairments but no obvious symptoms of dysexecutive syndrome. By the same token, many FES patients demonstrate moderate or marked memory impairments on standardized tests, yet family members and caregivers rarely report this symptom (29).
Executive function
FES patients have been shown to suffer EF deficits such as difficulty with rule shifts, planning, and coordination of two competing tasks. These deficits are consistent across patients, suggesting that the impairment is intrinsic to the disease (9,41,42). Given the above, it would be reasonable to imagine that FES patients would score lower on EF tests than healthy controls. Accordingly, we found a detectable deficit even with a brief test such as the FAB. This study marks the first application of this test in a sample of FES patients. Patients showed difficulty with conceptualization, mental flexibility, motor programming, resistance to interference, and inhibitory control items on the FAB. These findings are consistent with results from studies using other tests, which have reported abnormal verbal fluency scores in FES and chronic schizophrenia (43). This result suggests that the FAB might have utility as a routine clinical test in this population, with the exception of the environmental autonomy subtest. Patients with neurological disorders such as Alzheimer's disease, frontotemporal dementia, and amyotrophic lateral sclerosis have also been found to have abnormal FAB scores (21,44). However, studies with greater numbers of patients and in various stages of disease evolution are necessary to confirm the utility of the FAB in cognitive evaluation of FES.
EF involves such abilities as abstract reasoning, concept formation, decision-making, and planning of behavior. Based on a rule-learning paradigm, which invokes these abilities, the M-WCST is one of the most widely applied neuropsychological measures of EF (45). The M-WCST is particularly sensitive to lesions of the dorsolateral prefrontal cortex (DLPFC) and upper medial regions of the prefrontal cortex. Importantly, reductions in DLPFC gray matter volume are significantly more pronounced in schizophrenia patients with greater executive dysfunction as measured by the M-WCST (17). However, the M-WCST should be used with caution as a frontal measure because retro-Rolandic cortex lesions, such as hippocampal lesions, have also been associated with impairments, especially perseverative errors. Chronic and FES patients show difficulty with inhibiting previously-learned responses and shifting attention towards relevant stimuli; that is, they perseverate on an answer already noted to be incorrect. The poor performance of patients with schizophrenia may reflect a difficulty in inhibiting inappropriate responses (17).
We also explored the executive dysfunction associated with FES using the M-WCST; as widely described in the literature, patients demonstrated numerous problems in carrying out the study tasks. In our sample, the FES patients committed more total errors, failures to maintain set, and perseverative errors than controls. Patients also used more cards to complete the M-WCST than the controls.
Correlations between executive function and dysexecutive behavior
The literature indicates that many EF tests show only moderate ecological validity when used to predict individual functional capacity (40). Therefore, it is important to assess whether our EF tests, the FAB and M-WCST, have ecological relevance in our sample.
a) FAB
EF is primarily associated with the frontal lobes. Scores for all six FAB subtests are significantly correlated with frontal metabolism in patients with frontal lobe damage according to PET studies (17). Our study demonstrates for the first time in schizophrenia that the EF deficit assessed by the FAB also correlates with dysexecutive questionnaire results. Specifically, we found a statistically significant correlation between total BDSI and FAB inhibitory control scores. Difficulty inhibiting inappropriate responses and controlling impulsiveness are common clinical observations in FES, and this impairment likely underlies some of the behavioral problems that these patients experience. We found that motor programming performance as measured by the FAB (in which the patient must remain attentive to the examiner's movements for several minutes) was statistically correlated with BDSI: Disinterest and Indifference to One's Own Concern and other items. Performance on the FAB: Resistance to Interference task (in which the patient must use self-control to ignore conflicting instructions) may also predict behavioral problems in FES associated with difficulties in anticipation, planning, and initiation as measured by the BDSI.
Performance on the total FAB and FAB: Conceptualization (which requires abstract reasoning) may predict behavioral problems related to DEX-SO: Temporal Sequencing Problems (patient mixing events with each other and confusing the order in which they occurred), and DEX-SO: Executive Memory factor (which includes confabulation and temporal sequencing problems). FAB: Motor Programming, which requires motor control, was significantly correlated with DEX-S: Planning Problems. FAB: Resistance to Interference (in which the patient must apply self-control to ignore conflicting instructions) may also predict behavioral problems in FES related to DEX-S: Planning Problems and the DEX-S: Intentionality factor, which includes planning problems, poor decision-making, lack of insight and social awareness, distractibility, knowing-doing dissociation, and perseveration. All of these behaviors depend on self-control.
These results suggest that the FAB is ecologically relevant and might have utility as a routine clinical test in this population, with the exception of the environmental autonomy subtest (FAB: Prehension Behavior).
b) M-WCST
Our study is the first to explore correlations between the deficits identified using the M-WCST and the results of dysexecutive questionnaires in FES. One study in the literature reports a correlation between M-WCST total errors and DEX total score in chronic FES, but the authors did not specify the type of DEX used (46).
In our study, correlation analysis indicated statistically significant relationships between various aspects of M-WCST performance and DB. Specifically, in terms of the subtest scores, we found statistically significant correlations between M-WCST: Categories Completed and M-WCST total errors with DEX-SO: Temporal Sequencing Problem, with patients confusing events with one another as well as the order in which they occurred. Moreover, M-WCST: Categories Completed, M-WCST: Total Errors, M-WCST: Failures to Maintain Set, and M-WCST: Perseverative Errors may predict behavioral problems in FES related to the DEX-SO: Executive Memory factor, which includes confabulation and temporal sequencing problems. These findings would seem to be related to the role of EF in memory, especially given that markedly perseverative patients tend to be those who confabulate (24).
In terms of the DEX-S, there was a significant correlation between M-WCST: Perseverative Errors and total DEX-S. This result likely reflects the fact that the FES assesses many of the executive dysfunctions commonly associated with frontal injuries, such as perseverative deficits, incoherent actions, and unstructured behaviors. Again, as already discussed, M-WCST: Perseverative Errors may also predict behavioral problems in FES related to the DEX-S: Executive Memory factor. Finally, there was a significant correlation between M-WCST: Categories Completed and DEX-SL Impaired Abstract Reasoning (“she or he has problems understanding what other people mean unless they keep things simple and straightforward”).
We found strong correlations between M-WCST performance and DB. In contrast, the associations between psychiatric symptoms (positive, negative, etc.) and cognitive performance on this executive test were typically weak, suggesting relative independence of these disease processes (45).
Our study revealed various significant correlations between EF tests and DB. However, the cognitive tests explained only a small percentage of the variance in the behavioral assessments. In considering what other factors might explain this variance, it is important to keep in mind that DB may be linked to many factors not evaluated in this study:
1. Performing activities of daily living in the context of FES requires self-motivation and persistence, which is supplied by the examiner in the study setting;
2. Multiple and idiosyncratic real-life environmental demands, as well as the compensatory strategies used to address them, can affect the validity of neuropsychological tests in evaluating EF as related to activities of daily living (40);
3. EF tests do not require patients to organize their behavior over long periods of time or prioritize competing tasks to the degree that real-life tasks demand;
4. The tests are performed in a highly structured environment, unlike real-life situations in disorganized environments.
Moreover, reducing complex human behaviors to a test score or questionnaire item limits our insight into the nature of the deficit. EF implies integration and monitoring of cognitive functions, meaning that qualitative as well as quantitative measures (such as a description of how the patient performs the task) are needed to comprehensively characterize the executive deficit. Therefore, directly measuring those processes would significantly improve EF evaluation (40,46). Our results support our hypothesis; however, future studies are needed to identify the role of the various factors in everyday function and to confirm the ecological validity of the neuropsychological tests used to explore EF.
Antipsychotics
While our sample of FES patients were treated with different atypical antipsychotics, studies involving cognitive testing with currently-treated, previously-treated, and never-treated patients have reported similar results regardless of medication status, suggesting that antipsychotics have a relatively minor effect on most neuropsychological functions, although there are some conflicting results in the literature (8,47,48,49). We believe that the disordered behavior observed in this study is likely to reflect dysfunction attributable to the FES itself rather than side effects of the medications.
Neuropsychological heterogeneity
The pathophysiology of FES seems to involve many different degrees and types of global and/or specific cognitive deficits, including varying levels of problems with attention, EF, and memory. In general, the cognitive function of patients with FES is better than that of chronic patients. This cognitive heterogeneity was observed in our study for performance on the MSET and global cognition tests. For instance, while the MSET is fairly sensitive to deficits in abilities traditionally categorized as EF (29), some of the FES patients in our sample (approximately 27%) performed within normal limits. These results are consistent with other studies of FES that also reported cognitive heterogeneity (50,51).
Cognitive deficits can be considered central to schizophrenia, as they are present from the onset of the first episode —ruling out the possibility that the symptoms are completely attributable to illness duration, aging, psychosocial deprivation, cardiometabolic disease, institutionalization, or intensive treatments (electroconvulsive therapy)— and persist throughout the duration of the disease. The confounding cognitive effects of these other factors are a major problem in most neuropsychological studies of schizophrenia and may partially explain the heterogeneity of findings for this population. Therefore, we and numerous other investigators suggest that future research should focus on the population of first-episode patients, as their condition is more likely to reflect the true pathophysiology of this severe brain disorder.