• No results found

Dopamine in schizophrenia : what is the role for cognition?

N/A
N/A
Protected

Academic year: 2021

Share "Dopamine in schizophrenia : what is the role for cognition?"

Copied!
34
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

Dopamine in Schizophrenia:

What is the Role for Cognition?

Abstract

The association of dopamine with schizophrenia is broadly accepted (Kapur, Mizrahi & Li, 2005). Because the precise role of dopamine in schizophrenia is unknown, this reflects the ‘explanatory gap’ between small (molecular) and big (symptom) levels of description that forms a barrier for using neurobiological knowledge in clinical practice (Montague, Dolan, Friston & Dayan, 2012). Cognition, as defined here, is affected in 70-80% of schizophrenia patients (Holthausen et al., 2002). The main question of this thesis is how cognition is related to the association of dopamine with schizophrenia. It is hypothesized that cognition is a mediator. The first part reviews the associations between schizophrenia symptoms, dopamine and cognition. In the second part the ‘network perspective of psychopathologies’ is described and the possibilities of an adjusted version of this approach for schizophrenia research are discussed. A research is proposed in part three where the hypothesis is investigated using the network approach.

Bachelor thesis by Linda Kooiman Student number: 10020187

Address: Carolina MacGillavrylaan 2954, 1098XK, Amsterdam Mentor: Tobias Donner

(2)

Contents

Introduction 3-6

PART I

Schizophrenia and dopamine 6-7

Symptoms and cognition 7-11

Cognition and dopamine 11-14

Conclusion and Discussion Part I 14-15

PART II

Network perspective 15-19

Conclusion and Discussion Part II 19-21

PART III

Research Proposal 21-27

(3)

Schizophrenia, derived from ancient Greek, and meaning ‘split brain’, is because of its name often confused with multiple personality disorder (Comer, 2011). But the real reason why Bleuler named it schizophrenia was because he thought the disease was characterized by split psychological functions. And this, he thought, caused ideas to lose their coherence (Kessels, Eling, Ponds, Spikman & van Zandvoort, 2012). Now schizophrenia symptoms are described as positive or negative symptoms, respectively phenomenon’s that are ‘gained’ because of the disorder, such as hallucinations or ‘lost’, such as diminished emotional expression (American Psychiatric Association, 2013). The positive symptoms are hallucinations, delusions,

disorganized speech and disorganized or catatonic behavior. The negative symptoms include diminished emotional expression, avolition (a decrease in taking initiative), alogia (decreased speech output), anhedonia (a decreased experience of pleasure), and asociality (a decreased interest in social interactions).

For a person to get the diagnosis of ‘schizophrenia’ according to the DSM-5, he or she must have at least one of the following symptoms: hallucinations, delusions and disorganized speech. Furthermore, at least one other positive symptom or negative symptoms should be present (APA, 2013). These must be present for at least one month, but in total there must be signs of disturbances for half a year. Furthermore, there must be a decline in social

functioning (APA, 2013). Note that these diagnostic criteria allow two people with a different set of symptoms to be both diagnosed with schizophrenia. In comparison to the DSM-IV the symptom threshold for getting the diagnosis has gone up (from at least one specified symptom to at least two). Furthermore, the subtypes have been omitted. Identifying subtypes were not helpful because they were not stable (APA, 2013). This means that the symptoms of the patients often changed so that, for example, they would fit in one subtype description at the first encounter with the clinic, but in another subtype one year later.

Even though cognitive deficits are not part of the diagnostic criteria (APA, 2013) they are seen in 70-80% of schizophrenia patients (Holthausen et al., 2002). The American

neuropsychological group MATRICS found eight cognitive domains in which schizophrenic patients scored aberrant compared to the normal population: speed of processing,

attention/vigilance, working memory, verbal learning, visual learning, reasoning and problem solving, verbal comprehension and social cognition (Nuechterlein et al., 2004). Apart from these specific domains, a decline in overall cognitive functioning seems to be robustly found in schizophrenia (Kessels et al., 2012). In this thesis, the terms ‘cognition’ and ‘cognitive deficits’ are defined as cognitive functioning as measured by neuropsychological tests. The importance of finding ways to improve the quality of life of the people suffering

(4)

from schizophrenia is self-evident. But to make this point even more clearly: the World Health Organization (2001) declared it to be in the top ten of most disabling diseases. So, insights in the etiology of this condition are necessary to be able to find ways to prevent it, cure it or improve the lives of people suffering from schizophrenia.

Starting with Emil Kraepelin in 1893, people have been studying a very broad range of aspects surrounding this disease. Up to now, the etiology and pathology of schizophrenia is not clear. In a meta-analysis of 12 twin studies, the heritability of schizophrenia is found to be 81% (Sullivan, Kendler & Neale, 2003). Thousands of polymorphisms in thousands of studies have been associated with schizophrenia, but no specific gene or genetic variant has been found to encode for the disease (Allen et al., 2008). Recently a study by Allen et al. (2008) was done that cleverly accumulated all this data and found substantial evidence for the association of the genes DRD1, DTNBP1, MTHFR and TPH1. These genes encode for proteins that are part of completely different functional systems in the brain (Allen et al., 2008). A variety of environmental risk factors are known to influence the chance of

developing schizophrenia. Among these are living in an urban environment, cannabis usage, early trauma and prenatal vitamin D deficiency (Van Os, Krabbendam, Myin-Germeys & Delespaul, 2005). A role for stress is also widely accepted (Gispen-de Wied, 2000). In an extensive review of schizophrenia etiology Tandon, Keshavan and Nasrallah (2008) conclude that the lack of understanding can be explained by the fact that complex gene and gene-environmental interactions are at play and that the etiology can be different for different schizophrenia patients.

So, the people who share the schizophrenia diagnosis are heterogeneous in that they may vary in the present symptoms and in that the etiology of the disease may be different. Therefore it is not surprising that the brains of schizophrenia patients are aberrant from the normal population on a variety of ways (Keshavan, Tandon, Boutros & Nasrallah, 2008). A

broad range of deviations has been found in the brain structure, physiology and chemistry, but none of the brain measures is valid enough to serve as a diagnostic criteria (Keshavan,

Tandon, Boutros & Nasrallah, 2008). So the heterogeneous characteristic of this disorder exits on multiple levels: etiology, neurobiology and symptoms. Heinrichs (2001) beautifully

formulated this as follows: ‘people who share a diagnosis of schizophrenia often share little else’.

Of all the neurobiological measures, the link between dopamine and schizophrenia is most widely accepted (Kapur, Mizrahi & Li, 2005). After an extensive review looking at post-mortem, neuroimaging, plasma levels and animal model studies, Davis, Kahn, Ko and

(5)

Davidson (1991) concluded that schizophrenia is characterized by decreased dopamine activity in the prefrontal cortex (PFC) and increased dopamine activity in the mesolimbic neurons. The finding that frontal lobe lesions result in similar effects as the negative

symptoms of schizophrenia led to the hypothesis that the dopamine hypoactivity in the PFC causes the negative symptoms. The authors found that dopamine metabolite levels in the striatum are positively associated with positive symptoms and response to antipsychotics. This led to the hypothesis that the dopamine hyperactivity in the mesolimbic neurons causes positive symptoms. So, this ‘dopamine hypothesis’ states that schizophrenia is characterized by decreased dopamine activity in the prefrontal cortex (PFC) and increased dopamine activity in the mesolimbic neurons that account for the negative and positive symptoms respectively.

To describe it bluntly, this hypothesis consists of a description of the activity of dopamine in two brain areas and a description of the two symptom groups that seem to be related to this activity. What is missing is an understanding of the intermediate mechanisms to fill this ‘explanatory gap’ (Montague, Dolan, Friston & Dayan, 2012). Insights into these intermediate levels of description could provide a means to make the information gathered by schizophrenia neuroscientists useful for schizophrenia patients in the clinic (Montague, Dolan, Friston & Dayan, 2012). The aim of this thesis is to make a start at filling this

‘explanatory gap’ by looking at the cognitive deficits in schizophrenia and attempting to link these to the dopamine and the symptom level of description.

To summarized the section above, the current state of knowledge is firstly that a consensus exists that dopamine functioning is related to schizophrenia symptoms and secondly that the majority of schizophrenia patients experience a decline in cognition. The next step to be able to come to useful information is integration. With doing this, it is important to keep in mind that the heterogeneity and dynamicity that characterize schizophrenia form methodological challenges. On this basis, here an attempt is made to answer the question: ‘How are cognitive deficits related to the association of dopamine and schizophrenia symptoms?’ Thereby keeping in mind that in this thesis, the terms ‘cognition’ and ‘cognitive deficits’ are defined as cognitive functioning as measured by

neuropsychological tests.

Specifically, the hypothesis that cognition mediates the association of dopamine with schizophrenia is investigated. This thesis addresses to this matter in three parts. Part one reviews the current knowledge about the links between dopamine and schizophrenia symptoms, between dopamine and cognition and between cognition and schizophrenia

(6)

symptoms. If the link of dopamine with schizophrenia is mediated by cognition, there should also be links of dopamine with cognition and of cognition with schizophrenia symptoms. The aim of this part is to elucidate these links. The association of dopamine with working memory will be dealt with specifically. The gating hypothesis will be described, as this is a possible mechanistic explanation. In the second part of this thesis an adjustment to the original ‘network perspective on psychopathologies’ is proposed as a method for schizophrenia research. This method is described and the benefits are discussed. In part three, this method is further elaborated into a concrete research proposal to test the hypothesis that cognition mediates between dopamine and schizophrenia symptoms. In addition, it will be investigated if the gating hypothesis is supported in this study.

PART I

In this part, empirical studies to the association of schizophrenia with dopamine, symptoms with cognition and cognition with dopamine are reviewed. In addition to the widely accepted ‘dopamine hypothesis’, evidence for the relationship between dopamine and schizophrenia is provided by empirical studies. If this relationship is mediated by cognition, associations of dopamine with cognition and of cognition with schizophrenia symptoms should also be found in studies that address these links. The purpose of this part of the thesis is to gain preliminary support for this hypothesis.

Schizophrenia and Dopamine

The first evidence for the association of dopamine with schizophrenia comes from the efficacy of antipsychotic medication in treating the positive symptoms of schizophrenia. This treatment works well in 75% of the schizophrenia patients (Franken, Muris & Denys, 2013). Up to now, all effective antipsychotics block to a greater of lesser extent receptor D2 (Frankle & Laruelle, 2002).

Other evidence comes from empirical studies to the genetic and environmental etiology. In the meta-analysis of Allen et al. (2008) of genetic associations to schizophrenia one of the four ‘strongly’ established genes has clear links with dopamine. This gene, DRD1,

(7)

encodes for the most common dopamine receptor found in the brain: dopamine receptor 1 (D1; Allen et al., 2008). The same study concluded that there was ‘moderate’ evidence for the association with gene DRD2, which encodes for dopamine receptor 2 (D2). Some of the known environmental risk factors for schizophrenia also appear to be mediated by dopamine. In an experimental study the psychoactive substance of cannabis was found to indirectly increase the dopamine release in the striatum of rats (Cheer, Wassum, Heien, Phillips & Wightman, 2004). Also, in an experimental study to the effects of prenatal stress, Rhesus monkeys born from a mother that experienced significant stress during their pregnancy resulted in an increased dopamine synthesis in the striatum compared to the monkeys from mothers that did not experience this stress (Roberts et al., 2004).

Studies correlating brain measures of dopamine with schizophrenia provide a third branch of evidence. In the striatum of schizophrenia patients, increased presynaptic dopamine availability (Howes et al., 2009; McGowan, Lawrence, Sales, Quested & Grasby, 2004) and dopamine release (Laruelle et al., 1996) has been found compared to healthy subjects in studies using PET scanning and single photon emission computerized tomography (SPECT) respectively. Furthermore, in identical twins discordant for schizophrenia, higher D1 receptor availability was found in the ‘patient’ twin using PET scanning (Hirvonen, 2006).

Symptoms and Cognition

In this section, firstly studies that investigate the structure of the cognitive domains that are affected by schizophrenia are reviewed. Secondly a review is given of the

relationships between cognitive measures and symptoms.

There have been inconsistent findings in factor analytical studies that investigated the structure of the cognitive deficits in schizophrenia (Gladsjo et al., 2004; Nuechterlein et al., 2004; Keefe et al. 2006). The aim of the MATRICS research group was to develop a valid consensus cognitive battery for cognitive functioning in schizophrenia, the MATRICS consensus cognitive battery (MCCB; Nuechterlein et al., 2008). As a part of this project Nuechterlein et al., (2004) evaluated 13 factor analytic studies with the aim to identify separate cognitive factor in schizophrenia. They found seven factors: speed of processing, attention/vigilance, working memory, verbal learning and memory, visual learning and memory, reasoning and problem solving, and verbal comprehension. The list of domains they recommended for the cognitive battery did not contain verbal comprehension because this

(8)

was thought to be insufficiently sensitive because of its resistance to change. Furthermore, the domain social cognition was added because of the promising results of the recent studies (Nuechterlein et al., 2008). In the same year another study was published that used a

confirmatory factor analysis to identify the structure of cognitive functioning in a test battery consisting of 21 neuropsychological tests in patients with schizophrenia or related psychotic disorders (Gladsjo et al., 2004). The model was made out of six factors (verbal crystallized, attention/working memory, verbal episodic memory, speed of information processing, visual episodic memory, and reasoning/problem solving) and was based on earlier exploratory factor analyses and factor analytic studies of similar test batteries (Gladsjo et al., 2004). This model fitted the data significantly better than models with fewer factors. In contrary to these multiple factor models, Keefe et al. (2006) found in their factor analysis that a model with only one factor, consisting of five cognitive domains, was the best fit for data of 1493 patients on 11 neurocognitive tests. So they concluded that schizophrenia is characterized by an overall decline of cognitive functioning.

Recent studies investigated the relationship between different cognitive domains and symptom clusters such as ‘positive’, ‘negative’ and ‘disorganized’ in schizophrenia patients. In these studies, multiple significant negative relations between a range of cognitive measures and negative symptoms were found (Gladsjo et al., 2004; Hegde et al., 2013; Heydebrand et al., 2004; Keefe et al., 2006; Müller, Sartory & Bender, 2004; Rocca, Castagna & Marchiaro, 2006; Rund et al., 2004), and no or only few relations with positive symptoms (Cohen & Docherty, 2005; Gladsjo et al., 2004; Hegde et al., 2013; Heydebrand et al., 2004; Keefe et al., 2006; Müller, Sartory & Bender, 2004; Rocca, Castagna & Marchiaro, 2006; Rund et al., 2004). In Table 1 typical results are depicted. Among the cognitive domains that were most consistently found to be related to negative symptoms are (working) memory, speed of

processing, executive functioning and attention. But even these associations are not uniformly found in all studies. The cluster ‘disorganized’ consists of disorganized thought and language or conceptual disorganization. A significant negative relation was found to between this cluster and IQ (Müller, Sartory & Bender, 2004), verbal memory (Müller, Sartory & Bender, 2004) verbal acquisition (Lucas et al., 2004), verbal recall (Lucas et al., 2004) and cognitive flexibility (Lucas et al., 2004).

(9)

Table 1. Typical results for studies investigating the relationship between cognitive measures

and positive and negative symptoms in schizophrenia. Reprinted from Rocca, P., Castagna, F., Marchiaro, L., Rasetti, R., Rivoira, E., & Bogetto, F. (2006). Neuropsychological correlates of reality distortion in schizophrenic patients. Psychiatry research, 145, 49-60.

So, the studies described above did not find a clear association between the cluster positive symptoms and cognitive measures. Cohen and Docherty (2005) compared the relationship with hallucinations and delusions separately and found that they were associated with relatively distinct patterns of neuropsychological performance. In this study it was

(10)

suggested that positive symptoms could better be investigated using individual symptom measures instead of syndrome measures. In a study investigating the relationship between auditory verbal hallucinations (AVH’s) specifically and verbal working memory in 52 first episode psychosis, a significant association has been found using a PANSS subscale

measuring AHV’s and two working memory tasks (Gisselgård et al., 2014). This suggests that the relationships between cognitive measures and symptoms can better be studied per

symptom.

In a recent series of studies, a negative relationship of meta-cognitive processes and schizophrenia symptoms is found (Chan et al., 2012; MacBeth et al., 2014; McLeod, Gumley, MacBeth, Schwannauer & Lysaker, 2014; Vohs et al., 2014). Meta-cognition can be shortly described as consciousness about cognition. To give an idea of this concept, the

Metacognitive Assessment Scale Abbreviated (MAS-A) consists of four subscales; the ability to reflect on your own cognitive functions (‘understanding of one's own mind’) the ability to think about the cognitions of others (‘understanding of others' minds’), the ability view the mind of others non-egocentrically (‘decentration’) and the ability to solve real world problems using metacognitive information (‘mastery’; McLeod et al., 2014; Vohs et al., 2014). Insight into one’s own illness is also a form of meta-cognition (Chan et al., 2012). ‘Mastery’ was found to significantly correlate with negative symptoms in two out of three studies that studied this link (McLeod et al., 2014; Vohs et al., 2014) and with disorganized symptoms in one (Vohs et al., 2014). ‘Understanding of others' minds’ was found to significantly correlate with negative symptoms in two out of three studies (MacBeth et al., 2014; Vohs et al., 2014) and with disorganized symptoms in one (Vohs et al., 2014). ‘Understanding of one's own mind’ was found to significantly correlate with negative symptoms and disorganized symptoms in one out of three studies (Vohs et al., 2014). ‘Decentration’ was found to significantly correlate with negative symptoms in one out of three studies (McLeod et al., 2014) and with disorganized symptoms in one (Vohs et al., 2014). Insight into one’s own illness was found to correlate significantly with negative, positive and disorganized symptoms (Chan et al., 2012).

To recapitulate, when the relationship between symptoms and cognition is investigated using clusters of positive, negative en disorganized symptoms three things can be concluded. One, an association between cognition and negative symptoms is consistently found, but the specific cognitive functions that are impaired with schizophrenia are hard to pin down. Two, there is some evidence that disorganized symptoms relate to cognitive deficits, but from this review the specific cognitive domains are not clear. Three, no clear association between the

(11)

clustered positive symptoms and cognitive deficits is found. Positive symptoms significantly correlated to some cognitive tests in some studies, but this was marginal.

However, when the relationship between specific hallucinations and cognitive deficits was studied, a significant association was found in one study. This suggests that in order to gain insights in the relation between symptoms and cognition, it may be beneficial to

investigate symptoms separately. Furthermore, the studies reviewed here show an association between the negative and disorganized clusters and metacognition. The positive symptom cluster is only found to be associated to insight into one’s own condition.

Cognition and Dopamine

In the section above, it is seen that various cognitive measures are linked to

schizophrenia symptoms. In this section firstly studies are reviewed that provide a descriptive association of dopamine with cognitive measures. Secondly the gating hypothesis that

provides mechanistic explanation of the links between dopamine and working memory is reviewed.

The role for dopamine in cognition has been investigated empirically by manipulating dopamine functioning with a D1 receptor agonist comparing cognitive measures with a placebo group in schizophrenia patients (Kane et al., 2010; Pietrzak, Snyder & Maruff, 2010) or with baseline functioning in schizophrenia patients and healthy subjects (Barch & Carter, 2005). These studies show that the D1 receptor agonist d-amphetamine significantly enhanced reasoning and problem solving, attention, speed of processing (Pietrzak, Snyder & Maruff, 2010) and working memory (Barch & Carter, 2005). In contrary to these results, Kane et al. (2010) found no differences in cognitive performance between schizophrenia patients that were treated with placebo and the dopamine agonist armodafinil.

The relationship between D2 receptor blockade by antipsychotic medication, estimated by plasma levels, and cognitive functions was investigated in 410 schizophrenia patients (Sakurai et al., 2013). A significant positive relationship was found between D2 occupancy levels and overall cognitive impairment and vigilance. The patients with occupancy levels above 80% showed specially reflected neurocognitive impairment, suggestion that this relationship is not a linear one (Sakurai et al., 2013).

Dopamine neurons in the midbrain have projections to the PFC, striatum and

(12)

aspects of cognition. Using PET scans, the relationship between cognitive functioning and dopamine D1 and D2 receptors specifically in PFC and hippocampus was studied in 23 healthy patients (Takahashi et al., 2008). The cognitive functions were measured with a test battery consisting of tests for learning and memory (L&M), executive functioning (EF) and working memory (WM). Significant positive relationships were found between hippocampus D2 availability and all cognitive domains measured. A significant U-shaped relationship was found between prefrontal D1 receptor availability and the domains EF and WM. No

significant relationships with cognition were found for PFC D2 receptor availability or hippocampus D1 receptor availability. However, these results were not replicated in a similar study only focusing on D1 availability, were no relationship of this availability and cognition was found in in dorsolateral PFC and a positive relationship was found between D1

availability in the hippocampus and EF, speed of processing and general knowledge. The striatum itself may also contain different functional compartments (Cervenka, Bäckman, Cselényi, Halldin & Farde, 2008). Studies that investigated dopamine availability in healthy subject used PET scanning in which the striatum was divided into sensorimotor (SMST), associative (AST) and limbic areas (LST) found different cognitive functions to be positively associated with dopamine availability in these areas (Cervenka et al., 2008; Karlsson et al., 2011). Specifically, when D1 availability was investigated, SMST was associated to speed of processing and AST and LST were associated to general knowledge (Karlsson et al., 2011). When D2 availability was tested, LST was association with episodic memory and SMST and AST with fluency and general knowledge (Cervenka et al., 2008). To summarize, studies show that there is a role for dopamine in cognition. The specific cognitive functions that are associated with dopamine availability depend on the dopamine receptor type and location under study. And even the results for these specific types of measures vary. These studies provide descriptive evidence, but the mechanisms by which dopamine influences cognition in schizophrenia is still unclear. Recently computational models have attempted to elucidate these mechanisms. The gating hypothesis tries to explain how dopamine is involved in working memory. This will be further elaborated in the next section.

Working Memory

The capacity for cognitive control depends on the balance between stability and flexibility of goal representations in the prefrontal cortex (PFC). Stability is preserving the current representation without letting other information interfere, which avoids distraction

(13)

from achieving a goal (Montague, Hyman & Cohen, 2004). Flexibility is updating contextual information, which avoids perseverative behavior (Montague, Hyman & Cohen, 2004). But when and how does the brain decide when to be flexible when to be stable? Put differently, when and how will contextual information be allowed ‘into’ the PFC to update the current goal representations? This role for opening the gate has been assigned to dopamine (Braver & Cohen, 2000). The ‘gating hypothesis’ proposes that when contextual cues signal the need to update the goal representation, the ventral tegmental area (VTA) responds with phasic

dopamine release. This release ‘opens the gate’ and hereby allows the contextual information to influence the representation. When this phasic dopamine release is not present, the PFC is resistant to the influence of incoming signals (Braver & Cohen, 2000).

Now the question remains when and how a contextual cue will signal this need to update the goal representation. The reinforcement learning theory of dopamine function may be able to answer this (Montague, Hyman & Cohen, 2004). In this theory, phasic dopamine release occurs when a positive reward-prediction error is present (Montague, Hyman & Cohen, 2004). This occurs when a contextual cue indicates that a more valuable goal can be achieved by directing behavior towards that goal (Montague, Hyman & Cohen, 2004). It makes sense that the ‘old’ goal representation needs to be updated in this situation.

Furthermore it is stated that the phasic dopamine release strengthens the association of the cue with the goal representation in the PFC, so that it becomes a learning signal as well

(Montague, Hyman & Cohen, 2004). Learning occurs because a positive outcome after gating reinforces future gating when this cue is present and a negative outcome attenuates the chance of a gating signal after this cue in the future (Montague, Hyman & Cohen, 2004). So, it is assumed that phasic dopamine release has two functions: gating and reinforcement learning. The combination of these functions provide a possible mechanism of how the appropriate cues bring about the updating of representations in the PFC, the main component of working memory.

A recent study by D’Ardenne et al. (2012) empirically tested this hypothesis. In this study, human subjects made a task in which some trials required the maintenance of a contextual cue (working memory trials) and other trials did not require this (non-working memory trials). After three experiments, D’Ardenne et al. (2012) came to the following results. Firstly, when comparing the fMRI images of these trials, the working memory trials resulted in greater bilateral dorsolateral prefrontal cortex (DLPFC) activity. Secondly, the working memory trials were disrupted when a TMS pulse was applied to the right DLPFC 150 ms after onset of the cue. Performance on the working memory trials was not disrupted

(14)

when a TMS pulse was applied at another time or at the left DLPFC. TMS pulses to the DLPFC did not influence the performance on the non-working memory trials. This shows that the right DLPFC is involved in context updating. Thirdly, using a high-resolution fMRI technique, the working memory trials did not only result in higher right DLPFC activation, but also in substantia nigra/ventral tegmental area activation. This provides support that dopamine activity is involved in right DLPFC activation, which is necessary for updating and thus working memory.

So, firstly it is seen that studies provide evidence for the role of dopamine in cognition in schizophrenia patients on a descriptive way. Secondly it is pointed out that with the gating hypothesis a mechanistic explanation is possible for the role of dopamine in working memory.

Conclusion and Discussion Part One

Evidence that dopamine plays a role in schizophrenia comes from findings from different lines of research. The first one encompasses the finding that all the effective antipsychotic medication is acting on dopamine receptors in the brain. Furthermore, it is found that genes that encode dopamine receptors are associated with this disorder, that the effect of environmental risk factors may be mediated by dopamine and that schizophrenia patients have an increased dopamine receptor availability and dopamine release.

It is states that if this role is mediated by cognition, associations of dopamine with cognition and of cognition with schizophrenia symptoms should also be found in studies that address these links. In most studies reviewed here, a link between cognitive deficits and negative symptoms has been found, but the specific cognitive domains tended to vary. For disorganized symptoms the findings are even more variable. The least clear is the association between the clustered positive symptoms and cognitive deficits. Positive symptoms

significantly correlated to some cognitive tests in some studies, but this was marginal. However, when the relationship between specific hallucinations and cognitive deficits was studied, a significant association was found in one study. Investigating per symptom, instead of per cluster of symptoms might be a more valid way to approach this matter. Furthermore, deficits in metacognition also seem be associated with schizophrenia symptoms.

The role for dopamine in cognition has been established in studies showing a significant association between dopamine receptor availability and cognitive measures. Because different cognitive functions have been associated to dopamine D1 or D2 receptor

(15)

availability in PFC, hippocampus and different striatal areas, in future research, it is advisable to compartmentalize dopamine receptor availability into these areas and to take into account that results may differ for D1 and D2 receptors. To address the hypothesis that cognition mediates the association between dopamine schizophrenia symptoms, previous studies found links of dopamine with cognition and of cognition with schizophrenia symptoms. So it is concluded that preliminary support is found, but empirical research is needed to further test this hypothesis. In the next part a method is discussed that could be used to investigate how cognition is related to the association of dopamine with schizophrenia, based on the

conclusions drawn from the reviews and taking the hurdles in schizophrenia research into account.

PART II

In this part of this thesis the ‘network perspective on psychopathologies’ is described and the potential of this approach to serve as a basis for schizophrenia research is discussed.

Network perspective

Until now, this thesis has described the neuropsychological and neurobiological factors that are involved in schizophrenia. Here, we will focus on the symptom level of description. Schizophrenia is seen as a disorder, an illness. In theory, it is treated as an underlying causal factor that causes all the specific symptoms found to be related to the disorder (see Figure 1). One can imagine that this approach works well in the medical sciences. If we take for example a brain tumor, this also causes psychological symptoms and these are all caused by one underlying entity. The most efficient way to cure this illness is to intervene at the underlying cause instead of controlling the symptoms.

Recently a new perspective on psychopathology has been developed that advocates a model in which symptoms are not caused by one underlying factor, but form dynamical networks in which they influence each other (Borsboom & Cramer, 2013; Borsboom, Cramer, Schmittmann, Epskamp & Waldorp, 2011; Cramer, Waldorp, van der Maas & Borsboom, 2010; Schmittmann, Cramer, Waldorp, Epskamp, Kievit & Borsboom, 2013). In Figure 2 this

(16)

idea is visualized. From this perspective it is assumed that symptoms themselves have causal power. They can influence other symptoms, or themselves by positive or negative feedback loops. An example of how one symptom can influence another in schizophrenia is patient A, who hears a voice in her head that tells her to save the world (hallucination). This

hallucination may strengthen her idea that she is the new messiah (delusion). An example of a symptom influencing itself via a positive feedback loop is patient B, who hears a lot of people talking to him (hallucination) and has trouble sleeping because of this. Sleep deprivation is known to cause hallucinations, even in healthy people. So this may strengthen his

hallucinations. When in schizophrenia patient, for example, symptoms are represented by nodes and the relationship between the symptoms are represented by edges, the

psychopathology can be visualized and meaningful analyses can be done (Borsboom & Cramer, 2013).

Figure 1. Model of one underlying entity (g) causing a range of symptoms (x1-x5) for

example in a brain tumor. Adapted from Van Der Maas, H. L., Dolan, C. V., Grasman, R. P., Wicherts, J. M., Huizenga, H. M., & Raijmakers, M. E. (2006). A dynamical model of general intelligence: the positive manifold of intelligence by mutualism. Psychological review, 113, 842.

(17)

Figure 2. Model of a network of symptoms (x1-x5) that cause each other, without an underlying factor. Adapted from Van Der Maas, H. L., Dolan, C. V., Grasman, R. P.,

Wicherts, J. M., Huizenga, H. M., & Raijmakers, M. E. (2006). A dynamical model of general intelligence: the positive manifold of intelligence by mutualism. Psychological review, 113, 842.

When this model is used to study schizophrenia the first thing to do is constructing a ‘schizophrenia’ network. To do this, the variables that will be added in the network as nodes need to be selected (Borsboom & Cramer, 2013). In the original network perspective of psychopathologies, the nodes in the network represent symptoms only. The first part of this thesis has shown that dopamine and cognitive measures play a major role in the

psychopathology of the disorder. That is why it is proposed here that it will be useful to add biological and cognitive measures as nodes as well. Figure 3a depicts such a possible

schematic interdisciplinary schizophrenia network. Here, node D represents data from a fMRI dopamine measure, node WM from a working memory task, node H from the hallucination item of the Positive and Negative Syndrome Scale (PANSS; Kay, Flszbein & Opfer, 1987)

(18)

Figure 3. Possible part of a schizophrenia network. Node D represents data from a fMRI dopamine measure, node WM from a working memory task, node H from the hallucination item of the PANSS and node A from the ‘blunted affect’ item of the PANSS. a) A network is depicted with all the possible links between the nodes. b) Illustrated here is what the network may look like after graphical lasso is applied. In this example this shows that dopamine is linked to the symptoms via working memory.

Furthermore, the kind of relationship that is represented by the edges need to be determined (Borsboom & Cramer, 2013). An edge can represent the average correlation between nodes in a population (Borsboom & Cramer, 2013). The thickness of the edge can represent the strength of the correlation and the color (green or red) the direction of the

correlation. Similarly, edges between two nodes can represent partial correlations in which the influence of all other nodes in the network on this correlation is controlled for. This way, the possibility is excluded that a different node in the network acts as a third variable on this correlation (Costantini et al., 2014). The correlational or partial correlation networks have the tendency to be highly interconnected because nodes are almost always somewhat correlated with each other (Costantini et al., 2014). A more parsimonious network can be obtained with the graphical lasso method, which shrinks small partial correlations to zero so that no edge will appear between those nodes (Friedman, Hastie, & Tibshirani, 2008). This way more meaningful networks can be obtained. Figure 3 shows an example of how the visualization of

(19)

a network may change after the graphical lasso is applied. Without the lasso, every node is more or less related to all other nodes. After lasso is applies, in this example it becomes clear that the dopamine node is only linked to the symptom nodes via the working memory node. The edges as described above are all derived from population averages. It is also possible to construct networks for individual persons using time series (Borsboom & Cramer, 2013). In this method, the nodes are repeatedly measured and the edges stand for the correlation between one node at a certain moment in time and other nodes at a later moment.

Once it is determined what variables will be inserted in the network as nodes and what kind of correlation the edges will represent, the measures should be done and the data can be inserted into a software program such as qgraph (Epskamp, Cramer, Waldorp, Schmittmann & Borsboom, 2012). With this R package, the network can be visualized and subsequent analyses can be done. One such analysis is the centrality of a node. This measures the

influence that a particular node has on the complete network (Borsboom et al., 2011). This is useful in psychopathology such as in schizophrenia, because the node that is very central might be fruitful to intervene at. The benefit of adding biological and cognitive nodes to the network becomes clear as well: if these nodes are central, the importance of biological or cognitive interventions can be elucidated. Another analysis is path analysis. This can be done to find out how one node is related to another (Borsboom et al., 2011). For example, a

dopamine node is related to social withdrawal via hallucination and sleep deprivation. A measure for shortest path length between two nodes can be obtained to quantify this idea

(Borsboom et al., 2011). Also the structure of the complete network can be analyzed. One can look at whether the correlations between the nodes are random or not (Borsboom et al., 2011).

And if it is not random one can look at what kind of structure it has, for example a small world structure (Borsboom et al., 2011).

Conclusion and Discussion Part Two

So, in part two, the original network perspective of psychopathologies is discussed and an adjusted version is proposed for schizophrenia networks, in which cognitive and biological factors are added on top of the symptoms. It is described how networks are constructed and what kind of analyses can be done subsequently.

Using this approach has some clear benefits for addressing schizophrenia. Firstly, by being able to describe the structure of the associations between molecular, cognitive and

(20)

symptom measures, this method makes interdisciplinary integration possible. For this thesis this means that cognition can be investigating in the context of schizophrenia symptoms and neurobiological measures. Secondly, in this model symptoms have the same causal status as biological and cognitive factors. The influence of the symptoms on each other and on other factors can be studied using this method. Thirdly, in the introduction we have seen that the heterogeneity in etiology, neurobiology and symptoms and the dynamical representations of schizophrenia symptoms form major hurdles for research progress in this disorder.

Schizophrenia as a construct is therefor not specific enough, because one schizophrenia patient can differ in etiology, neurobiology and symptoms from another schizophrenia patient. In this method, there is no need to work with the concept of schizophrenia, because the

relations that are studied are between smaller constructs like symptoms. In the first part of this thesis it is seen that even when the link between cognition and symptoms are studied

clustering symptoms into negative and positive syndromes, results are ambiguous. It was suggested that studying this link per symptom might lead to more clear results. The fourth benefit of this method is that it can be used to give direction to future research. For example, if a certain node is found to be very central in the schizophrenia network, this suggests that it would be efficient to intervene at this point. Research can built on this finding by

investigating the precise role of this node and subsequently developing interventions.

When interpreting the results of network visualization or analysis, is important to take into account that the edges represent correlations. This means the networks discussed here do not say anything about causality. Time series networks give somewhat more information about causality, but in the scope of this thesis this will not further be discussed. Also, an attentive reader can have noticed that this method is of an exploratory nature. A network with the most important actors is ‘run’ and subsequently it is analyzed. There are no a priori assumptions. However, the network approach can also be used when a priori statements are present. This is elaborated in the next part of this thesis. The hypothesis that cognitive measures mediate the association of dopamine with schizophrenia symptoms is tested and furthermore we will look at whether the network provides support for the specific association of dopamine with working memory, as predicted by the gating hypothesis.

The network approach could also be used for other research that could help our understanding or treatment of schizophrenia. For example, anti-psychotics do not seem to work with all patients and often it is a problem to find the right dose (Franken, Muris & Denys, 2013). As stated in the first part of this thesis, all the effective antipsychotic

(21)

the network predict the sensitivity to medication in patients, for example by analyzing the centrality of dopamine measures? In the research proposed here the main focus is on dopamine, cognitive measures and symptoms, but conform the complexity of the disorder genetic and environmental factors could be added to the network as well. In general, the proposed approach offers a mean to study the influence of a variety of relevant

interdisciplinary factors in context of the complex network of schizophrenia.

PART III

Research Proposal

Dopamine in Schizophrenia: Putting Cognitive Deficits in Context

Abstract

A research is proposed that uses an adjusted version of the network approach as a basis to study how cognition is related to the association of dopamine with schizophrenia symptoms. Specifically, the proposed study investigates whether cognitive measures mediate between dopamine activity and symptoms. This is an attempt to fill up the ‘explanatory gap’ between the small level of description (dopamine) and large level of description (symptoms). By elucidating the intermediate levels of descriptions, for example cognition, directions for the development of new treatments may be provided.

Introduction

This proposal is preceded by a review and a description of a methodological approach. The findings of these earlier parts that are most relevant to the proposed study will be repeated here. From the introduction of this thesis is has become clear that firstly that a consensus exists that dopamine functioning is related to schizophrenia symptoms (Kapur, Mizrahi & Li,

(22)

2005) and secondly that the majority of schizophrenia patients experience a decline in cognition (Holthausen et al., 2002). Cognition in this thesis is defined as the cognitive functioning that can be measured by neuropsychological tasks. It is stated that in order to come to useful information, integration is needed. How is cognition involved in this link between dopamine and schizophrenia? Also it has become clear with investigating this, it has to be taken into account that the heterogeneity and dynamicity that characterize schizophrenia form methodological challenges (Heinrichs, 2001; Keshavan, Tandon, Boutros & Nasrallah, 2008).

In the second part of this thesis the links between dopamine, cognition and symptoms are reviewed. Studies have supported the involvement of dopamine with symptoms with the following findings. Firstly, that all the effective antipsychotic medication is acting on dopamine receptors in the brain (Frankle & Laruelle, 2002). Secondly, it is found that genes that encode dopamine receptors are associated with this disorder (Allen et al., 2008). Thirdly it is found that the effect of known environmental risk factors may be mediated by dopamine (Cheer et al., 2004; Roberts et al., 2004). Fourthly it is found that schizophrenia patients have increased presynaptic dopamine availability (Howes et al., 2009; McGowan et al., 2004), dopamine release (Laruelle et al., 1996) and dopamine receptor availability (Hirvonen, 2006).

The association between schizophrenia and cognition was reviewed subsequently. This association was supported by studies investigating this relationship using clusters of positive, negative en disorganized symptoms. The results from these studies were threefold. One, an association between cognition and negative symptoms is consistently found (Gladsjo et al., 2004; Hegde et al., 2013; Heydebrand et al., 2004; Keefe et al., 2006; Müller, Sartory & Bender, 2004; Rocca, Castagna & Marchiaro, 2006; Rund et al., 2004), but the specific cognitive functions that are impaired with schizophrenia are hard to pin down. Two, there is some evidence that disorganized symptoms relate to cognitive deficits (Müller, Sartory & Bender, 2004; Lucas et al., 2004), but from this review the specific cognitive domains are not clear. Three, no clear association between the clustered positive symptoms and cognitive deficits is found (Cohen & Docherty, 2005; Gladsjo et al., 2004; Hegde et al., 2013;

Heydebrand et al., 2004; Keefe et al., 2006; Müller, Sartory & Bender, 2004; Rocca, Castagna & Marchiaro, 2006; Rund et al., 2004). Positive symptoms significantly correlated to some cognitive tests in some studies, but this was marginal. However, two studies investigated the relationship between specific symptoms and cognitive deficits and did find a significant result (Cohen and Docherty, 2005; Gisselgård et al., 2014). This suggests that in order to gain insights in the relation between symptoms and cognition, it may be beneficial to investigate

(23)

symptoms separately. Furthermore, the meta-cognitive measure insight into one’s own illness was found to correlate significantly with negative, positive and disorganized symptoms in one study (Chan et al., 2012).

The third review dealt with the association between dopamine and cognition. Studies are found that investigated this by manipulating dopamine activity with dopaminergic drugs

(Barch & Carter, 2005; Kane et al., 2010; Pietrzak, Snyder & Maruff, 2010) or by measuring dopamine receptor availability and relating this to cognitive measures (Sakurai et al., 2013). The studies supported the association of dopamine with cognition. A mechanistic explanation of how dopamine may be involved in working memory is given by describing the gating hypothesis (Braver & Cohen, 2000; Montague, Hyman & Cohen, 2004) and discussing a empirical study that found support for this hypothesis (D’Ardenne et al., 2012).

To summarize, dopamine is found to be involved in schizophrenia. Cognition is involved as well, but how is not yet clear. It is hypothesized that cognition mediates the relationship of dopamine with schizophrenia symptoms. The reviews on the links between dopamine and symptoms, cognition and symptoms, and dopamine and cognition provide preliminary support for this hypothesis.

In part two of this thesis it is discussed that the network perspective of

psychopathologies provides a useful basis for studying schizophrenia, because with this method the heterogeneity and dynamicity of schizophrenia psychopathology can be taken into account. It is discussed that this is originally an exploratory method that only looks at the symptoms, but this thesis proposes that it may also be used for interdisciplinary research and that it can be used when a priori hypotheses are present. To put these ideas into practice, a research is proposed here using this approach to investigate the role of cognition in the association of dopamine and schizophrenia symptoms. Furthermore, the gating hypothesis described above is tested.

Hypotheses

It is hypothesized that cognitive measures mediate the association of dopamine with schizophrenia symptoms. This will be referred to as hypothesis 1 and this hypothesis is depicted in Figure 4. This is an a priori assumption, but is still partly exploratory because no statements are done about which cognitive measures mediate the association of dopamine with which symptoms. Based on the gating hypothesis it is specifically hypothesized that

(24)

working memory is associated to dopamine. This is referred to as hypothesis two. This hypothesis will be supported if the network visualization with graphical lasso shows a link between the dopamine measure and the working memory measure.

Figure 4. Depiction of a part of the network and two possible outcomes after graphical lasso. The nodes in the yellow, blue and purple area represent symptoms, cognitive measures and dopamine measures respectively. Node D represents data from a fMRI dopamine measure, node WM from the working memory task of the MCCB, node H, D, A and E from the PANSS items ‘hallucinations’, ‘delusions’, ‘blunted affect’ and ‘emotional withdrawal’ respectively. If hypothesis one is not supported, the visualization of the network may be similar to a): there are direct associations between dopamine and symptoms. If hypothesis one is supported, the visualization of the network may be similar to b): there are no direct

associations between dopamine and symptoms, but they are linked through cognitive measures.

(25)

Method

How can the network perspective investigate how cognitive deficits are related to the association of dopamine and schizophrenia symptoms? The method proposed here is

constructing a network in which the nodes represent symptoms, cognitive measures and a dopamine measure. Fourteen nodes are assigned for seven positive and seven negative symptoms and these will be measured using the separate items of the Positive and Negative Syndrome Scale (PANSS; Kay, Flszbein & Opfer, 1987). The seven positive symptom items are  ‘Delusions’, ‘Conceptual disorganization’, ‘Hallucinations’, ‘Hyperactivity’,

‘Grandiosity’, ‘Suspiciousness/persecution’ and ‘Hostility’. The seven negative symptom items are ‘Blunted affect’, ‘Emotional withdrawal’, ‘Poor rapport’, ‘Passive/apathetic social withdrawal’, ‘Difficulty in abstract thinking’, ‘Lack of spontaneity and flow of conversation’ and ‘Stereotyped thinking’. The subjects are rated on these items from one to three based on a 45-minute clinical interview.

There are eight cognitive nodes. Seven of them are the cognitive domains of the MATRICS consensus cognitive battery (MCCB): speed of processing, attention/vigilance, working memory, verbal learning, visual learning, reasoning and problem solving and social cognition (Nuechterlein et al., 2008). The raw scores of the domains will be transformed to a T-score. To get this data, ten tests are being administered to the subjects that will take one to one-and-a-half hour in total. The eighth cognitive node represents the data from a

metacognitive measure of insight. Just like Chan et al., (2012) the first three items of the abridged version of Scale for Assessment of Unawareness of Mental Disorder (SUMD) will be used (Amador et al., 1994). These items are ‘Global awareness of mental disorder’, ‘Awareness of the effect of medications’ and ‘Awareness of the social consequences of having the illness’. The subjects are rated on these items from one to three based on a semi-structured interview, with higher scores indicating poorer insight.

The dopamine node will be represented by high-resolution fMRI measures of the substantia nigra (SN) and the ventral tegmental area (VTA), like in the study of D’Ardenne et al., (2012). Per subject, the average of a 5 second measure of blood oxygen level-dependent (BOLD) signal activity of the SN and VTA will be represented by a t-score.

One hundred and fifty schizophrenia patients are included in the study. The subjects are only included in the study if they have an official schizophrenia diagnosis. Subjects with other comorbid psychiatric disorders are excluded from the study. From these subjects, cross

(26)

sectional data is obtained and inserted into R with the package qgraph (Epskamp et al., 2012). A partial correlation network of the data will be visualized, so the edges will represent partial correlations. Graphical lasso will be applied to make the network sparser (Friedman, Hastie, & Tibshirani, 2008). It claimed that with the lasso method, meaningful data can be obtained with a number of measures that is the same as the number of variables (Friedman, Hastie, & Tibshirani, 2008). There are 22 nodes in this network so at least 22 subject need to be included. One hundred and fifty measures are proposed here because the findings get more meaningful with more measures (S. Epskamp, personal communication, the 4th

of July 2014).

Analyses and Interpretation

Based on hypothesis 1, that cognition mediates the relationship op dopamine with schizophrenia symptoms is true, it is expected that the qgraph partial variation network with graphical lasso shows a network in which there are no edges between dopamine and symptom measures directly, but that they are connected to each other via cognitive measures. Put differently, the shortest path between the dopamine node and the symptom nodes should be going via cognitive nodes. Another analysis can be done when qgraph is given the command to leave the direct dopamine to symptom edge out (S. Epskamp, personal communication, the 4th

of July 2014). The variance that is explained in this model can then be compared to the explained variance when the command is not active. If this does not differ significantly it can be concluded that the cognition node mediates the correlation of the dopamine node with the symptom nodes (S. Epskamp, personal communication, the 4th

of July 2014). It is expected that the explained variance of these models no not differ.

If the visualization of the network shows that like Figure 4b, the dopamine node is connected to the symptoms only via working memory, the gating hypothesis and our more general hypothesis 1 will be supported. If this similar construct is visualized, but then with other cognitive nodes, hypothesis 1 will be supported as well. This finding should stimulate researchers and theorists to investigate mechanical explanations of the link between dopamine and this particular cognitive measure. If qgraph shows that the shortest path between the dopamine node and the symptom nodes do not go via cognitive nodes this can be explained in various ways. The first possibility is that cognition, as defined in this thesis, does not mediate the association of dopamine with schizophrenia symptoms. Alternatively, it is possible that

(27)

cognitive measures that are not included in this network would mediate between the dopamine and symptom nodes.

It will be interesting to look at the symptoms that will be directly or indirectly

connected to the dopamine node. Which symptoms are most involved? And are they linked to specific cognitive nodes or more generally to multiple cognitive nodes? Furthermore, how do the symptoms that are linked to dopamine influence other symptoms? The current study will allow a first glance at these effects, but future research should elaborate this further. By being able to encompass levels of description from small to big: the molecular, cognitive and symptoms, this integrative approach provides a useful tool in closing the explanatory gap described by Montague, Dolan, Friston and Dayan (2012) and bridging fundamental research with clinical use.

(28)

References

Allen, N. C., Bagade, S., McQueen, M. B., Ioannidis, J. P., Kavvoura, F. K., Khoury, M. J., ... & Bertram, L. (2008). Systematic meta-analyses and field synopsis of genetic association studies in schizophrenia: the SzGene database. Nature genetics, 40, 827-834.

Amador, X. F., Flaum, M., Andreasen, N. C., Strauss, D. H., Yale, S. A., Clark, S. C., & Gorman, J. M. (1994). Awareness of illness in schizophrenia and schizoaffective and mood disorders. Archives of General Psychiatry, 51, 826-836.

American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental

Disorders (5th ed.). Washington, DC: Author.

Barch, D. M., & Carter, C. S. (2005). Amphetamine improves cognitive function in medicated individuals with schizophrenia and in healthy volunteers. Schizophrenia research, 77, 43-58. Borsboom, D., & Cramer, A. O. (2013). Network analysis: An integrative approach to the structure of psychopathology. Annual review of clinical psychology, 9, 91-121.

Borsboom, D., Cramer, A. O., Schmittmann, V. D., Epskamp, S., & Waldorp, L. J. (2011). The small world of psychopathology. PloS one, 6, e27407.

Braver, T. S., & Cohen, J. D. (2000). On the Control of Control: The Role of Dopamine in Regulating Prefrontal Function and Working Memory. In S. Monsell & J. Driver (Eds.),

Attention and Performance XVIII; Control of Cognitive Processes (pp. 713-737). London,

England: MIT.

Cervenka, S., Bäckman, L., Cselényi, Z., Halldin, C., & Farde, L. (2008). Associations between dopamine D2-receptor binding and cognitive performance indicate functional compartmentalization of the human striatum. Neuroimage, 40, 1287-1295.

Chan, S. K., Chan, K. K., Lam, M. M., Chiu, C. P., Hui, C. L., Wong, G. H., ... & Chen, E. Y. (2012). Clinical and cognitive correlates of insight in first-episode schizophrenia.

(29)

Cheer, J. F., Wassum, K. M., Heien, M. L., Phillips, P. E., & Wightman, R. M. (2004). Cannabinoids enhance subsecond dopamine release in the nucleus accumbens of awake rats.

The journal of neuroscience, 24, 4393-4400.

Comer R. J., (2011). Abnormal Psychology. New York, NY: Worth.

Cohen, A. S., & Docherty, N. M. (2005). Symptom-oriented versus syndrome approaches to resolving heterogeneity of neuropsychological functioning in schizophrenia. The Journal of

neuropsychiatry and clinical neurosciences, 17, 384-390.

Cools, R. (2011). Dopaminergic control of the striatum for high-level cognition. Current

opinion in neurobiology, 21, 402-407.

Costantini, G., Epskamp, S., Borsboom, D., Perugini, M., Mõttus, R., Waldorp, L., & Cramer, A. (2014). State of the aRt personality research: A tutorial on network analysis of personality data in R. Manuscript submitted for publication.

Cramer, A. O. J., Waldorp, L. J., Maas, H. L. J. van der, & Borsboom, D. (2010). Comorbidity: A network perspective. Behavioral and Brain Sciences, 33, 137–150.

D’Ardenne, K., Eshel, N., Luka, J., Lenartowicz, A., Nystrom, L. E., & Cohen, J. D. (2012). Role of prefrontal cortex and the midbrain dopamine system in working memory updating.

Proceedings of the National Academy of Sciences, 109, 19900-19909.

Davis, K. L., Kahn, R. S., Ko, G., & Davidson, M. (1991). Dopamine in schizophrenia: a review and reconceptualization. The American journal of psychiatry.

Epskamp, S., Cramer, A. O., Waldorp, L. J., Schmittmann, V. D., & Borsboom, D. (2012). Qgraph: Network visualizations of relationships in psychometric data. Journal of Statistical

Software, 48, 1-18.

Franken, I., Muris, P., Denys, D., (2013). Basisboek Psychopathologie. Utrecht, the Netherlands: De Tijdstroom.

(30)

Frankle, W. G., & Laruelle, M. (2002). Neuroreceptor imaging in psychiatric disorders.

Annals of nuclear medicine, 16, 437-446.

Friedman, J., Hastie, T., & Tibshirani, R. (2008). Sparse inverse covariance estimation with the graphical lasso. Biostatistics, 9(3), 432-441.

Gispen-de Wied, C. C. (2000). Stress in schizophrenia: an integrative view. European Journal

of Pharmacology, 405, 375-384.

Gisselgård, J., Anda, L. G., Brønnick, K., Langeveld, J., ten Velden Hegelstad, W., Joa, I., ... & Larsen, T. K. (2014). Verbal working memory deficits predict levels of auditory

hallucination in first-episode psychosis. Schizophrenia research.

Gladsjo, J. A., McAdams, L. A., Palmer, B. W., Moore, D. J., Jeste, D. V., & Heaton, R. K. (2004). A six-factor model of cognition in schizophrenia and related psychotic disorders: relationships with clinical symptoms and functional capacity. Schizophrenia Bulletin, 30, 739. Heinrichs, R. W. (2001). In search of madness: Schizophrenia and neuroscience. Oxford University Press.

Hegde, S., Thirthalli, J., Rao, S. L., Raguram, A., Philip, M., & Gangadhar, B. N. (2013). Cognitive deficits and its relation with psychopathology and global functioning in first episode schizophrenia. Asian journal of psychiatry, 6, 537-543.

Heydebrand, G., Weiser, M., Rabinowitz, J., Hoff, A. L., DeLisi, L. E., & Csernansky, J. G. (2004). Correlates of cognitive deficits in first episode schizophrenia. Schizophrenia research,

68, 1-9.

Hirvonen, J., van Erp, T., Huttunen, J., Aalto, S., Någren, K., Huttunen, M., ... & Hietala, J. (2006). Brain dopamine d1 receptors in twins discordant for schizophrenia. American Journal

of Psychiatry, 163, 1747-1753.

Holthausen, E. A., Wiersma, D., Sitskoorn, M. M., Hijman, R., Dingemans, P. M., Schene, A. H., & van den Bosch, R. J. (2002). Schizophrenic patients without neuropsychological

(31)

deficits: subgroup, disease severity or cognitive compensation? Psychiatry research, 112, 1-11.

Howes, O. D., Montgomery, A. J., Asselin, M. C., Murray, R. M., Valli, I., Tabraham, P., ... & Grasby, P. M. (2009). Elevated striatal dopamine function linked to prodromal signs of schizophrenia. Archives of general psychiatry, 66, 13-20.

Kane, J. M., D'Souza, D. C., Patkar, A. A., Youakim, J. M., Tiller, J. M., Yang, R., & Keefe, R. S. (2010). Armodafinil as adjunctive therapy in adults with cognitive deficits associated with schizophrenia: a 4-week, double-blind, placebo-controlled study. The Journal of clinical

psychiatry, 71, 1475.

Kapur, S., Mizrahi, R., & Li, M. (2005). From dopamine to salience to psychosis—linking biology, pharmacology and phenomenology of psychosis. Schizophrenia research, 79, 59-68. Karlsson, S., Rieckmann, A., Karlsson, P., Farde, L., Nyberg, L., & Bäckman, L. (2011). Relationship of dopamine D1 receptor binding in striatal and extrastriatal regions to cognitive functioning in healthy humans. Neuroimage, 57, 346-351.

Kay, S. R., Flszbein, A., & Opfer, L. A. (1987). The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia bulletin, 13(2), 261.

Keefe, R. S., Bilder, R. M., Harvey, P. D., Davis, S. M., Palmer, B. W., Gold, J. M., ... & Lieberman, J. A. (2006). Baseline neurocognitive deficits in the CATIE schizophrenia trial.

Neuropsychopharmacology, 31, 2033-2046.

Keshavan, M. S., Tandon, R., Boutros, N. N., & Nasrallah, H. A. (2008). Schizophrenia,“just the facts”: What we know in 2008: Part 3: Neurobiology. Schizophrenia research, 106, 89-107.

Kessels, R., Eling, P., Ponds, R., Spikman, J., & Zandvoort, M. van, (2012). Klinische Neuropsychologie. Amsterdam, the Netherlands: Boom.

(32)

Laruelle, M., Abi-Dargham, A., Van Dyck, C. H., Gil, R., D'Souza, C. D., Erdos, J., ... & Innis, R. B. (1996). Single photon emission computerized tomography imaging of

amphetamine-induced dopamine release in drug-free schizophrenic subjects. Proceedings of

the National Academy of Sciences, 93, 9235-9240.

Lucas, S., Fitzgerald, D., Redoblado-Hodge, M. A., Anderson, J., Sanbrook, M., Harris, A., & Brennan, J. (2004). Neuropsychological correlates of symptom profiles in first episode

schizophrenia. Schizophrenia research, 71, 323-330.

Maas, H. L. van der, Dolan, C. V., Grasman, R. P., Wicherts, J. M., Huizenga, H. M., & Raijmakers, M. E. (2006). A dynamical model of general intelligence: the positive manifold of intelligence by mutualism. Psychological review, 113, 842.

MacBeth, A., Gumley, A., Schwannauer, M., Carcione, A., Fisher, R., McLeod, H. J., & Dimaggio, G. (2014). Metacognition, symptoms and premorbid functioning in a First Episode Psychosis sample. Comprehensive psychiatry, 55, 268-273.

McGowan, S., Lawrence, A. D., Sales, T., Quested, D., & Grasby, P. (2004). Presynaptic dopaminergic dysfunction in schizophrenia: a positron emission tomographic [18F] fluorodopa study. Archives of General Psychiatry, 61, 134-142.

McLeod, H. J., Gumley, A. I., MacBeth, A., Schwannauer, M., & Lysaker, P. H. (2014). Metacognitive functioning predicts positive and negative symptoms over 12 months in first episode psychosis. Journal of psychiatric research, 54, 109-115.

Montague, P. R., Dolan, R. J., Friston, K. J., & Dayan, P. (2012). Computational psychiatry. Trends in cognitive sciences, 16, 72-80.

Montague, P. R., Hyman, S. E., & Cohen, J. D. (2004). Computational roles for dopamine in behavioural control. Nature, 431, 760-767.

Müller, B. W., Sartory, G., & Bender, S. (2004). Neuropsychological deficits and concomitant clinical symptoms in schizophrenia. European Psychologist, 9, 96-106.

(33)

Nuechterlein, K. H., Barch, D. M., Gold, J. M., Goldberg, T. E., Green, M. F., & Heaton, R. K. (2004). Identification of separable cognitive factors in schizophrenia. Schizophrenia research, 72, 29-39.

Nuechterlein, K., Green, M., Kern, R., Baade, L., Barch, D., Cohen, J., … & Marder, S. (2008). The MATRICS Consensus Cognitive Battery, part 1: test selection, reliability, and validity. American Journal of Psychiatry, 165, 203-213.

Os, J. van, Krabbendam, L., Myin-Germeys, I., & Delespaul, P. (2005). The schizophrenia envirome. Current opinion in psychiatry, 18, 141-145.

Pietrzak, R. H., Snyder, P. J., & Maruff, P. (2010). Use of an acute challenge with d‐ amphetamine to model cognitive improvement in chronic schizophrenia. Human Psychopharmacology: Clinical and Experimental, 25, 353-358.

Roberts, A. D., Moore, C. F., DeJesus, O. T., Barnhart, T. E., Larson, J. A., Mukherjee, J., … & Schneider, M. L. (2004). Prenatal stress, moderate fetal alcohol, and dopamine system function in rhesus monkeys. Neurotoxicology and teratology, 26, 169-178.

Rocca, P., Castagna, F., Marchiaro, L., Rasetti, R., Rivoira, E., & Bogetto, F. (2006). Neuropsychological correlates of reality distortion in schizophrenic patients. Psychiatry research, 145, 49-60.

Rund, B. R., Melle, I., Friis, S., Larsen, T. K., Midbøe, L. J., Opjordsmoen, S., ... &

McGlashan, T. (2004). Neurocognitive dysfunction in first-episode psychosis: correlates with symptoms, premorbid adjustment, and duration of untreated psychosis. American Journal of Psychiatry, 161, 466-472.

Sakurai, H., Bies, R. R., Stroup, S. T., Keefe, R. S., Rajji, T. K., Suzuki, T., ... & Uchida, H. (2013). Dopamine D2 receptor occupancy and cognition in schizophrenia: analysis of the CATIE data. Schizophrenia bulletin, 39, 564-574.

(34)

Semerari, A., Carcione, A., Dimaggio, G., Falcone, M., Nicolo, G., Procacci, M., & Alleva, G. (2003). How to evaluate metacognitive functioning in psychotherapy? The Metacognition Assessment Scale and its applications. Clinical Psychology & Psychotherapy, 10, 238-261.

Schmittmann, V. D., Cramer, A. O., Waldorp, L. J., Epskamp, S., Kievit, R. A., & Borsboom, D. (2013). Deconstructing the construct: A network perspective on psychological phenomena. New Ideas in Psychology, 31, 43-53.

Sullivan, P. F., Kendler, K. S., & Neale, M. C. (2003). Schizophrenia as a complex trait: evidence from a meta-analysis of twin studies. Archives of general psychiatry, 60, 1187-1192.

Takahashi, H., Kato, M., Takano, H., Arakawa, R., Okumura, M., Otsuka, T., ... & Suhara, T. (2008). Differential contributions of prefrontal and hippocampal dopamine D1 and D2

receptors in human cognitive functions. The Journal of Neuroscience, 28, 12032-12038.

Tandon, R., Keshavan, M. S., & Nasrallah, H. A. (2008). Schizophrenia,“just the facts” what we know in 2008. 2. Epidemiology and etiology. Schizophrenia research, 102, 1-18.

Vohs, J. L., Lysaker, P. H., Francis, M. M., Hamm, J., Buck, K. D., Olesek, K., ... & Breier, A. (2014). Metacognition, social cognition, and symptoms in patients with first episode and prolonged psychoses. Schizophrenia research, 153, 54-59.

World Health Organization. (2001). The World health report: 2001: Mental health: new understanding, new hope.

Referenties

GERELATEERDE DOCUMENTEN

The newborn rat’s stress system readily habituates to repeated and prolonged maternal separation, while continuing to respond to stressors in context dependent fashion. Chapter 3

It appeared that the experience of being kept in isolation in a novel environment during repeated maternal separation, rather than the maternal absence per se, caused priming of

(c) MS + Chronic stress: Choy and van den Buuse studied how early and later life stress affects the schizophrenia phenotype of adult Wistar rats.. 24-hMD was used as early-life

We sacrificed rat pups in two different testing conditions: basal levels (basal) and 8h of separation (separated). 2C): to determine the effects of repeated separation in home context

In the present study we demonstrated that the stressful experience of peer deprivation in a novel cage during repeated MS (NOVEL SEP) rather than the maternal absence experience

2.8.1 Experiment I - Genetic susceptibility (Hit 1): In order to explore if APO-SUS rats are phenotypicaly different from the WH, we, first, described their differences

In order to investigate the effects of the combination of High, Med and Low maternal LG history with post-weaning social isolation on psychosis susceptibility under basal conditions,

Rats, without genetic-susceptibility to psychosis (Wistar), displayed increased psychosis susceptibility, when they encountered a radically different, in terms of stress, later