Cognitive communication disorders
in German patients with Parkinson´s
disease
Bianca Spelter
Rijksuniversiteit Groningen
Begeleider:
Roel Jonkers
Studieprogramma:
Master Taalwetenschappen
Afstudeerrichting:
Neurolinguistiek
Acknowledgment
First of all, I want to thank Roel Jonkers for the supervision of the thesis. He supported me a lot during the thesis and his quick responses on my emails always helped me.
Furthermore, I would like to express my gratitude to Dr. Ingeborg Maser and Dr. Werner Hofmann for the inspiring internship and the recruitment of patients for the current thesis. I also want to thank Ruth who made this internship possible for the organisation of the intern-ship and my accommodation.
Abstract
Background Literature on communication disorders in patients with PD is scant. However,
some studies suggest language problems on a higher level of communication, the macro level. Aim of the study was to provide information whether patients with PD show communi-cation disorders on different levels of communicommuni-cation and if so, which factors influence their communication skills.
Method 24 German patients with PD were assessed with the Boston Naming Test (BNT),
the Parkinson Neuropsychometric Dementia Assessment (PANDA) and the MAKRO screen-ing for communication skills.
Results 21 participants scored below the cut-off score on the MAKRO. The factors age, age
at onset, level of education and the performance on the whole PANDA and on all PANDA subtests correlated significantly with the performances on the MAKRO while disease dura-tion, Hoehn & Yahr staging (Hoehn & Yahr, 1967) and the performance on the BNT did not.
Discussion and conclusion Communication disorders seem to be a common secondary
Table of content
Introduction ... 7
Parkinson´s disease ... 7
Mild cognitive impairment (MCI) and dementia in Parkinson´s disease (PD-D). ... 7
Cognitive Problems in Parkinson´s disease. ... 8
Language Problems in Parkinson´s disease. ...10
Communication disorders in Parkinson´s disease ...13
Research objectives ...19
Method ...20
Participants ...20
Materials ...21
MAKRO-Screening. ...21
Parkinson Neuropsychometric Dementia Assessment (PANDA). ...24
Boston Naming Test (BNT). ...26
Procedure ...27
Statistical analysis ...27
Results ...28
MAKRO: Comparison of participants with PD and controls ...29
MAKRO: Correlating factors ...30
Total score ...30
Discussion ...32
MAKRO – total score ...33
MAKRO – subtests ...37
Conclusion ...42
References ...43
Appendices ...50
Appendix A: Hoehn and Yahr scale (Hoehn & Yahr, 1967) ...50
Appendix B: Description of all participants ...51
List of abbreviations
PD
Parkinson’s disease
H&Y staging Hoehn & Yahr staging
PD-D
Dementia in Parkinson’s disease
MCI
Mild cognitive impairment
Introduction
Parkinson´s disease
Parkinson´s disease (PD) is the second most common neurodegenerative disease in Germany. The prevalence in Germany is currently estimated on 1680 cases per 100,000 persons with higher rates in men than in women (Riedel, Bitters, Amann, Garbe, & Langener, 2016). The clinical diagnosis of PD is based on the existence of at least two of the three core motor symptoms, i.e. 1) resting tremor, 2) rigidity and 3) akinesia/ bradykinesia (Levy & Cum-ings, 1999). Furthermore, in PD non-motor symptoms like cognitive dysfunctions, dementia, depression and psychotic symptoms occur frequently (Meireles & Massano, 2012). Around 75% of the patients with PD suffer from cognitive impairments in the course of their disease (Aarsland & Kurz, 2010). According to Aarsland et al. (2010) most studies report a preva-lence of dementia between 75% and 80%. Depressive symptoms occur in approximately half of PD patients (McDonald, Richard & De Long, 2003).
The present paper aims to evaluate cognitive communication disorders in patients with PD. Therefore, the following chapters focus on language, cognition and communication in these patients.
Mild cognitive impairment (MCI) and dementia in Parkinson´s
disease (PD-D).
In the current version (version five) of the Diagnostic and Statistical Manual of Mental
Disorders (DSM V; American Psychiatric Association, 2013) the terms mild cognitive
Mild cognitive impairment (MCI) represents a degree of cognitive impairment that is not normal for the age but with essentially normal functional activities (Galtier, Nieto, Lorenzo, & Barroso, 2016). The literature regarding PD patients with MCI is scant. It could be found that PD patients with MCI were older, and had less education, longer disease duration, and high-er Hoehn and Yahr stage scores than patients without MCI (Pedhigh-ersen, Larsen, Tysnes, & Alves, 2013). Additional risk-factors to develop a MCI are a rigid-akinetic phenotype, an im-pairment of semantic fluency, an imim-pairment on the pentagon test, genetic factors, as well as postural instability (Vascocellos & Pereira, 2015). Furthermore, MCI is considered as a high-risk factor to develop a dementia (Galtier et al., 2016).
The current literature contains several definitions and diagnostical criteria for a dementia in Parkinson´s disease (PD-D). Emre et al. (2007) developed diagnostical guidelines and define PD-D as a syndrome of insidious onset and progressive decline of cognition from a premorbid level, that is not attributable to motor or autonomic symptoms. Impairment in at least two of the four typically involved cognitive domains (impaired attention, executive func-tions, visuospatial abilities, and impaired free recall that improves with cueing) must be rec-ognized without prominent language dysfunction. Several risk factors for PD-D have been proposed, including certain predominant motor features such as rigidity and gait instability, MCI, and the presence of visual hallucinations (Goetz, Emre, & Dubois, 2008). Older age is broadly accepted as a risk factor for PD-D. Some authors have found parameters such as disease duration, age of onset, and motor symptom severity to be significant risk factors, but published data is at times contradictory (Aarsland & Kurz, 2010).
Cognitive Problems in Parkinson´s disease.
Mostly named in the context of cognitive deficits in patients with PD are the executive functions (EF). Executive dysfunction is often present in patients with PD from the time of diagnosis (Garcia-Ptacek & Kramberger, 2016). The impairment worsens with disease pro-gression, and in some cases, has been found to be predictive of conversion to PD-D, though this remains controversial (Gratwicke et al., 2015). Some studies suggest that a proportion of broader cognitive deficits reported in the literature may be the manifestation of underlying executive impairment (Stuss & Alexander, 2000). Kudlicka, Clara and Hindle (2011) conclud-ed in their systematic review on EF in patients with PD that not the whole spectrum of execu-tive abilities is researched yet. Still, some studies used the same tests on different subtypes of EF and a meta-analysis showed an impairment in the assessed EF.
However, not only deficits in executive functions are detectable in patients with PD from an early stage, but also attention deficits are common, particularly on tests sensitive to defi-cits in executive control. Even non-demented patients with PD demonstrate impaired orient-ing of visual and auditory attention (Gratwicke et al., 2015). With progression to PD-D these deficits worsen, and impaired vigilance also develops in fluctuating levels of alertness (Bal-lard et al., 2002). Furthermore, there is evidence for an attentional deficit in patients with PD, which in turn interferes with EF like set-switching and working memory (Colman, 2011).
Patients with Parkinson’s disease exhibit both subtle visuospatial deficits and visuoper-ceptive deficits, sometimes from early stages in the disease on. These deficits become more noticeable and more common with disease progression and show high sensitivity in detecting the transition to PD-D (Gratwicke et al., 2015).
Language Problems in Parkinson´s disease.
Patients with Parkinson´s disease (PD) often develop dysarthria including voice- and ar-ticulation problems (Miller, Andrew, Noble & Walshe, 2011). However, aphasic disturbances are generally not considered to be part of the changes associated with PD (Pahwa, 1998). In the following, only the language deficits that might be related to communication on the macro level are described.
The language production of patients with PD is often compared to that of patients with Broca´s aphasia (Bastiaanse & Leenders, 2009). One reason for this impression is a reduced syntax seen in patients with PD (Colman, 2011). Murray and Lenz (2001) stated that the re-duced syntax mirrors cognitive and motor speech impairments rather than being a pure lan-guage deficit. A second phenomenon that is comparable to the problems with lanlan-guage pro-duction seen in patients with Broca´s aphasia is that action naming is more impaired than object naming (Bertella et al., 2002; Boulenger et al., 2008; Cotelli et al., 2007; Péran et al., 2009). Carthery-Goulart et al. (2015) found the same pattern in a semantic association task. In a recently published study, Salmazo-Silva (2017) found that action fluency performance was significantly related to the severity of PD, while object semantic association deficits and noun fluency scores were associated to lower scores in measures of global cognitive func-tioning. For this, it can be concluded that semantic deficits are related to problems in specific types of cognitive processing (Salmazo-Silva, 2017).
difficulties in patients with PD differ qualitatively from patients with Broca´s aphasia. The per-formance of patients with PD in their study was not influenced by linguistic factors, i.e. lin-guistic complexity or regularity of past tense, but rather by non-linlin-guistic variables such as sentence length. Their scores on the language tests were significantly correlated to non-linguistic cognitive functions, i.e. verbal working memory and set switching/inhibition abilities. Therefore, the authors conclude that impaired EF in patients with PD lead to an impaired verb production.
Except for the discrepancy between nouns and verbs, only little is known about lexical processing in patients with PD. In a study conducted by Copland et al. (2003), patients with PD could process lexical ambiguities presented in isolation via automatic and attention-based procedures. However, attentional processing was typically limited to the facilitation of both dominant and subordinate meanings, implying a deficit in selectively engaging the se-mantic network based on meaning frequency. Thus, the patients’ sese-mantic priming patterns are different from normal (Copland et al., 2003). Ellis, Holt, and West (2015) assessed lexical diversity during a three-minute discourse production. They found no problems in the lexical diversity in patients with PD compared to controls. Thus, lexical diversity seems to be an as-pect of language production that is not impaired in the early stages of PD. Furthermore, it could be found that in the absence of dopamine, the lexico-semantic system does not func-tion properly (Herrera & Cuetos, 2013). In their study, Herrera and Cuetos (2013) assessed differences in the strength of association of the words generated by a group of non-demented PD patients on and off dopamine medication using a word-association task and found that PD patients off medication said words less associated with the target compared with when they were on medication.
explain it through an impairment of executive functions (Colman, Koerts, van Beilen, Leenders, & Bastiaanse, 2006; Geyer & Grossman, 1994; Hochstadt, Nakano, Lieberman, & Friedman, 2006; Kemmerer, 1999). Hochstadt et al. (2006) assessed the relation between distinct EF and sentence comprehension and concluded that limits on sequencing and/ or verbal working memory may be responsible for the syntactic comprehension deficits in pa-tents with PD. Grossman et al. (2002) explain sentence comprehension deficits in patients with PD through impairments in inhibition and planning.
As already mentioned earlier in this chapter, in the current literature authors do not agree on the causes of language impairment in PD. While some authors see the cause of the lan-guage problems in a reduced capacity of verbal working memory (Caplan and Waters, 1999; Hochstadt et al., 2006), others state that executive resource limitation (Grossman et al., 2003), sequencing problems (Lieberman, Friedman, & Feldman, 1990) or reduced semantic priming (Kischka et al., 1996) are responsible. Bastiaanse and Leenders (2009) compared different studies and concluded that the problems that PD patients face when presented with language tasks are not linguistic in nature, but caused by various impaired cognitive func-tions, such as verbal working memory and set switching. However, there are also authors reporting a pure language deficit in patients with PD. Bocanegra et al. (2015) compared pa-tients with PD with and without MCI on measures of linguistic, semantic and executive func-tions. Different linguistic and semantic domains are significantly impaired in patients with PD, even in the absence of MCI. The findings indicate that problems in action naming and action-semantic deficits in this population constitute a primary disturbance, whereas impairments of syntax and object semantics are secondary to executive functions. Moreover, other studies report deficits in lexical retrieval, especially for verbs (Salmazo-Silva et al., 2017; Péran et al., 2009), independent of the executive functions.
al., 2017). However, this task is highly dependent of cognitive functions (Piatt, Fields, Paolo & Tröster, 1999). Henry and Crawford (2004) did a meta-analysis to find out if the word flu-ency deficit in patients with PD reflects executive dysfunction or problems with the semantic memory. They found that tests of alternating fluency, where participants have to switch be-tween two categories, were associated with larger deficits than standard fluency tasks which supports evidence for a specific deficit in the executive functions as a reason.
A study conducted by Lewis, Lapointe, Murdoch and Chenery (1998) tried to relate cogni-tive functions in MCI patients to language performances on structured tests. The resulting data showed that with increasing cognitive impairment, the performances on tasks with greater demands of cognitive processing, such as processing ambiguity, explaining meta-phors and constructing sentences became worse.
In patients with PD-D core language functions are largely spared obtained in patients. However, impaired verbal fluency, impaired word finding (Brønnick, 2010) and confrontation naming difficulties (Caballol, Martí & Tolosa, 2007) can be found regularly. Especially the impaired verbal fluency is more severe than in patients with for example Alzheimer´s demen-tia (Emre, 2003).
Communication disorders in Parkinson´s disease
re-garding non-aphasic communication disorders and the performance of patients with PD on the macro level will be discussed.
In general, communication skills are highly associated with executive functions. How-ever, there is some controversy on which components of EF are involved. Most researchers state that working memory, cognitive flexibility and planning skills influence communication most. This relates to the conceptual contiguity of planning skills, the capacity of the working memory and the comprehension and production of linguistic macro-structure (Büttner, 2014). Dressler et al. (2004) state that the ability to differentiate between relevant and irrelevant information is the basis to understand and produce coherent texts. For this working memory and long-term memory are needed as well as monitoring skills for a targeted and flexible lan-guage planning and selective attention. Marini, Carlomagno, Caltagirone and Nocenti (2005) see the cause for the communication problems in the inability to create mental models out of visual information and empirical knowledge.
Independent of the component of EF, the current literature agrees on the high influ-ence of EF on communication skills on the macro level. Most studies concentrated on non-aphasic communication disorders following prefrontal lesions. However, Boelen, Allain, Spikman and Fasotti (2011) showed that not only prefrontal lesions lead to communication disorders. A more important factor seemed to be the impairment of EF. Executive dysfunc-tion is often present in patients with PD from the time of diagnosis (Garcia-Ptacek & Kramberger, 2016). Therefore, problems in communication can be expected.
during social interactions. In contrast, the impairment in aspects of conversational appropri-ateness (pauses, feedback, conciseness, etc.) point to a deficit in fundamental pragmatic competence and may be due to the brain dysfunction associated with PD. Since the perfor-mances of the patients with PD did not differ from controls on general measures of cognitive skills, the pragmatic deficit cannot be attributed to general cognitive impairment. In a second study, McNamara and Durso (2003) asked the participants with PD and their partners to rate their pragmatic communication skills. This showed that patients with PD are less aware of their communication problems than their partners. The authors state that self-awareness or self-monitoring may be fundamental for appropriate social communication.
McKinlay, Dalrymple-Alfrod, Grace and Roger (2009) found that patients with PD showed impairments on aspects of language, working memory, and processing speed. The cognitive skills were significantly correlated with pragmatic language functioning. The deficits in pragmatic language functioning were mediated by verbal working memory and processing speed, but not attentional set-shifting. Processing speed showed to be a stronger determiner of pragmatic language performance than verbal working memory. Thus, the results of McKin-lay et al. (2009) suggest that pragmatic language deficits may be explained in terms of defi-cits in processing speed associated with PD.
Text comprehension
Reading comprehension is the product of a complex integration of knowledge and skills. These are decoding (Lyon, 1995; Torgesen, 2000), vocabulary (Verhoeven & van Leeuwe, 2008), as well as syntactic (Cutting & Scarborough, 2006; Oakhill & Cain, 2011) and seman-tic processing (Nation et al., 1999; Torgesen, 2000). Additionally, reading comprehension depends on EF (Cain, 2006; Christopher et al., 2012), among which working memory is the most well-established predictor in adults (Daneman & Merikle, 1996). Various studies sup-port the view that processing capacities tapped by working memory tasks in both the phono-logical and the semantic domain are important in explaining variance in reading comprehen-sion (Daneman & Merikle, 1996).
It could be shown that patients with PD show pragmatic comprehension deficits (Holt-graves & McNamara, 2010). Some authors suggested that patients with PD display reduced sensitivity to context, which might complicate their comprehension (McNamara, Obler, Au, Durso, & Albert, 1992).
An important skill for text comprehension is the ability to interpret nonliteral elements of language such as metaphors, proverbs and idioms. Patients with PD have shown to be im-paired in the comprehension of metaphors (Berg, Bjornram, Hartelius, Laakso & Johnels, 2003; Colman, 2011). Additionally, the comprehension of ambiguous sentences or implied information or with complex grammar is impaired in patients with PD (Natsopoulos, et al., 1991; Hanes, Andrewes and Pantelis, 1995; Lewis, LaPointe, Murdoch and Chenery, 1998; Grossman, 1999; Kemmerer, 1999; Murray and Stout, 1999). Monetta and Pell (2007) used a metaphor comprehension task (by Gernsbacher, Keysar, Robertson, & Werner, 2001) to investigate how patients with PD process metaphors in comparison to healthy controls. Pa-tients with PD showed an impairment of comprehension of metaphors if they had a reduced capacity of working memory. In contrast, participants without deficits in working memory per-formed as good as the healthy controls. This stands in line with the results of a study con-ducted by Lewis et al. (1998) in which it was tried to relate cognitive functions in PD patients with MCI to language performances on structured tests. The resulting data showed that with increasing cognitive impairment, the performances on tasks with greater demands of cogni-tive processing, such as processing ambiguity, explaining metaphors and constructing sen-tences, became worse.
Text production
supra-structure (e.g. narrative or descriptive text) and the recognition of a pointe of the story (Büttner, 2014).
Following the ageing process, a neurological disease or injury, disruptions of global co-herence may occur. However, little is known about the specific impact that ageing, disease or injury has on global coherence during discourse production (Ellis et al., 2016). Ellis et al. (2016) compared 20 studies in their systematic review to assess the impact of these on global coherence during expressive language tasks. They found indications that global co-herence is negatively influenced by ageing and neurological disease or injury. However, the heterogeneity of included participants, measurement tools and study designs were limiting factors in determining the exact nature by which these factors impact the skill of global co-herence.
An aspect that complicates the communication of patients with PD is their weak self-monitoring capacity in discourse. This impairment is related to attentional dysfunction in PD (McNamara, Obler Au, Durso, & Albert, 1992).
Inferences
Inferences are cognitive constructs that can be seen as ideas, suggested by facts or de-tails but not explicitly said. Since inferences can only be successively made by logical rea-soning, a certain degree of cognitive skills is required (Büttner, 2014).
Berg, Bjornram, Hartelius Laakso and Johnels (2003) reported that patients with PD without cognitive abnormalities in the Mini Mental State Examination exhibit impairments in making inferences compared to healthy controls. Their results suggest that processing of implied information might be a specific problem in this group and that the task of making in-ferences could be a particularly sensitive test of high-level language dysfunction. Further-more, it was noted that the performance on this high-level language test battery was strongly correlated with the performance on the test of cognitive function.
Research objectives
dis-age, disease duration or H&Y staging. Moreover, the relation between communicative skills and cogni-tive functions measure with the Parkinson Neuropsychometric Dementia Assessment (PANDA; Kalbe et al., 2008) as well as language skills, in this case word retrieval measured with the Boston Naming
Test (BNT, Kaplan, Goodglass & Weintraub, 1983), is examined.
Method
Participants
The participants were recruited from a private practice for Speech- and Language Ther-apy and the neighbouring practice for neurology in Aschaffenburg, Bavaria. All participants were informed about the study in written and spoken word and signed a declaration of con-sent. Included were patients with the diagnosis of idiopathic PD, independent from their se-verity or duration of the disease or the development of an MCI or dementia. All participants were native speakers of German without severe visual or auditory impairment, severe dysar-throphonia as well as additional neurological impairments or a history of alcohol abuse. The stage of the severity of the motor symptoms of PD were rated by a neurologist using the Hoehn and Yahr (H&Y) scale (see Appendix A) which consists out of five stages (Hoehn & Yahr, 1967).
The patients’ medication was assessed since it is known that the dopaminergic treat-ment has an influence not only on motor functions but also on semantic word association (Herrera & Cuetos, 2013). All patients were on dopamine medication during the testing. A detailed description of all participants and a list of their medication can be found in Appendix B. Table 1 Description of participants. Median Range Age (years) 69.50 54 / 78 Years of education a 12.00 8 /19
Age at onset (years) 64.75 47 / 76.5
Disease duration b 5.50 0.5 / 13
Hoehn & Yahr staging 3.00 1 / 4
a years of education including all forms of schools, apprenticeships and universities b years since diagnosis
For the comparison of the patients with PD with healthy controls, the data of the partici-pants of the study of Büttner (2014) was used. She included 20 participartici-pants (ten males and ten females) with an age range from 32 to 65 years (Mdn. = 49 years). The participants had different levels of education.
Materials
MAKRO-Screening.
macrostruc-tural skills in different types of texts (narrative, procedural). The MAKRO contains four sub-tests, which are explained in detail in the following.
Text comprehension. In this subtest, participants have to read a narrative text. No time
limit is given. Subsequently, they have to answer ten questions about the text without being allowed to have a look at the text again. Five of these questions ask for explicit and five for implicit information given in the text. Three answers per question were presented and only one is correct. For each correct answer, three points are given.
Text production. In this subtest, the participants have to produce two stories based on
picture stories. The first story contains four, the second eight pictures. The pictures remained visible during the narration to have only a small influence of memory skills. The narrations were audiotaped and later transcribed. The number of produced obligatory propositions was analyzed, i.e. utterances that further the development of the story. These obligatory proposi-tions were developed for every story after the analysis of narratives of 20 healthy controls. If a proposition was named in more than 80% of the controls, it was defined as obligatory. One point was scored for each named obligatory proposition. In total, 30 points can be achieved. Furthermore, the amount of meta-commentaries, wrong propositions and repetitions were counted, but not included in the scoring. No analysis of grammar, vocabulary etc. was in-cluded.
Inferences. The subtest Inferences requires the generation of a causal inference through
Table 2
Example of the subtest Inferences.
Die Kinder gehen zum Spielen nach draußen. (engl.: The children go outside to play.) Am Himmel sind dunkle Wolken zu sehen. (engl.: There are dark clouds at the sky.)
Tropfnass kommen sie daheim wieder an. (engl.: They come home completely wet.)
Procedural sequences. The participants are asked to sort sequences of actions of daily
living (e.g. doing the groceries or washing the dishes). They also have to find headlines for actions or notice sequences that are not fitting to the action. Sometimes more than one se-quence is correct. One point is given for each correct headline, noticed wrong sese-quence and correctly sorted sequence.
Thus, in every subtest the participants can receive a maximum of 30 points. For each subtest, cut-off norms for a light and a severe impairment are given resulting out of the per-formance of 20 healthy controls. Table 3 provides an overview of the cut-off norms of the four subtests. The cut-off norm for the total performance is 109 points.
Table 3
Overview of the cut-off norms used in the MAKRO Screening.
Subtest Interpretation
Text comprehension < 27 points light impairment
Text production < 28 points light impairment
< 26 points severe impairment
Inferences < 27 points light impairment
< 24 points severe impairment
Procedural sequences < 27 points light impairment
< 24 points severe impairment
Parkinson Neuropsychometric Dementia Assessment
(PAN-DA).
The Parkinson Neuropsychometric Dementia Assessment (PANDA; Kalbe et al., 2008) was designed to detect cognitive decline in patients with PD. It has a high construct validity, objectivity, retest and inter-rater reliability and it is independent of age. In contrast to the MMSE and other conventional screenings, the PANDA in constructed to detect the symp-toms of subcortical dementia. Furthermore, it includes a short depression questionnaire. The mood questionnaire corresponds well with the results of the Beck´s Depression Index (Beck, Steer, & Brown, 1996) as the golden standard and can therefore be regarded as a valid tool for the clinician to objectify possible depressive symptoms. This is highly relevant considering the high prevalence of depression in patients with PD. The PANDA consists of five cognitive tasks and has a specifity of 91% and a sensitivity of 90% (Kalbe et al., 2008).
Subtests one and five are word pair associate learning tasks with immediate (subtest 1) and delayed (subtest 5) recall. Contradictory to memory tasks including single words, word pair associate learning tasks increase the demands on EF and attention as two words have to be actively linked to each other by mental strategies. The second subtest is an alternating
two different categories have to be named alternating within one minute, adds an extra de-mand on attention, working memory, and switching ability (i.e. executive functions) in con-trast to traditional verbal fluency tasks. Research showed that patients with PD especially show difficulties in semantic word fluency in contrast to phonemic verbal fluency. Subtest three is a visuospatial task which contains three items, i.e. half-masked squares with dotted patterns where the subject is expected to find the pattern emerging on removing the mask (see Fig. 1). Items have to be evaluated in the given alignment i.e. they should not be rotated mentally. This task was included as deficits in visual perception may even be among the ear-liest symptoms of cognitive decline in PD (Kalbe et al., 2008).
Figure 1. Visuospatial task of the PANDA.
Subtest four is a working memory and attention task in which rows of numbers are pre-sented in a random order (e.g. “3–1–9–4”), and the subject is expected to repeat these num-bers in reversed order (“4–9–1–3”) (Kalbe et al., 2008).
two points for four or more mistakes. The result is the total score of the PANDA. A final score of more than 18 of the maximal 30 points can be interpreted as normal cognitive functioning. Between 15 and 17 points, a light cognitive dysfunction is possible. Below 14 points, a de-velopment of dementia is presumable and further diagnosis is needed (Kalbe et al., 2008).
Boston Naming Test (BNT).
At the moment, several different versions of the manual of the 60-item Boston Naming
Test (Kaplan, Goodglass & Weintraub, 1983) are available including different procedures and
norms (Heesbeen, 2001). According to the original manual of Kaplan et al. (1983), adults without aphasia start at item 30. The items are ordered from high frequent to low frequent nouns and it can be assumed that the first 30 Items are well known if the participants do not name an item wrong until item 38. If they do, after item 60 the test continues with item 29 and goes on backwards. After eight correct items in a row, it can be assumed that the participant knows the rest. During the tests, all divergent answers are written down and analyzed later.
Procedure
All participants were tested with the three, i.e. the MAKRO, the PANDA and the BNT, tests within one session by the same speech and language therapist. Dependent of the par-ticipant, the testing lasted between 50 minutes and two hours. Depending on the preferences of the participants their session took place in the morning or the afternoon. Thus, the day times of the tests differed between the participants. The three tests were administered in randomized order. The versions A and B of the MAKRO were used alternatingly.
During the testing, the participants could have a break between the three tests or be-tween the subtests of the MAKRO if needed. Disease symptoms such as tremor, bradykine-sia, bradyphrenia, pain, fatigue, sleepiness, and mood disorders, as well as medication ef-fects, can interfere with cognitive performance and testing (Meireless & Massano, 2012). To minimalize these effects, the participants could use helping tools if needed. They could use a white sheet of paper to reduce distraction while reading or a clip board for writing. In some cases, they read out their written sentences (subtest inferences) and they were written down afterwards by the therapist to avoid wrong scoring due to handwriting.
Statistical analysis
Data was analysed with the SPSS software (IBM, 2015). Firstly, the performances on the MAKRO and on the MAKRO subtests text comprehension, inferences and procedural se-quences were compared to the ones of a control group out of the study of Büttner (2014). Dependent on the normal distribution of the data, the parametric unpaired t-test or the non-parametric Mann-Whiney U test was used. Secondly, the influence of the factors age, age at onset, disease duration and H&Y staging as well as the performances on the PANDA, on the
PANDA subtests, and on the BNT on the performance in the MAKRO was assessed using a
subtests on the four subtests of the MAKRO were assessed using correlation analyses. De-pendent on the normal distribution of the data, the parametric coefficient Person´s r or the non-parametric coefficient Spearmen’s rho (rs) was used to assess the correlations.
All tests were done two-tailed as no hypothesises could be put forward beforehand due to a lack of literature. Results were considered statistically significant at p < 0.05. Analyses were done for the whole group. Cohen’s D standard were used to evaluate the correlation coefficient to determine the strength of the relationship, or the effect size, where coefficients between .10 and .29 represent a small association; coefficients between .30 and .49 repre-sent a medium association; and coefficients above .50 reprerepre-sent a large associate or rela-tionship.
Since the data of the control group is only available for the total MAKRO score and the three subtests text comprehension, inferences and procedural sequences, the analyses could only be conducted for these scores and not for the subtest text production.
Results
Out of the 24 participants with PD 21 scored below the cut-off score of 109 on the
MAK-RO. Six participants showed impaired text comprehension (one slightly and five severe) and
21 impaired text production (two slightly and 19 severe). On the subtest inferences, 19 partic-ipants scored below the cut-off score (ten slightly impaired and nine severely) and on the subtest procedural sequences 17(eight slightly impaired and nine severely). In the BNT, sev-en participants scored below the cut-off score including the three participants that scored above the cut-off score on the MAKRO. Six participants scored below the cut-off score in the
PANDA. The individual scores of all participants can be found in Appendix C. Table 4
partici-Table 4
Median, range, minimum and maximum scores of the participants.
Median Range Minimum Maximum
BNT 159,50 52 122 174 PANDA 22,00 24 4 28 MAKRO 101,00 72 43 115 Text comprehension 27,00 15 15 30 Text production 23,00 18 11 29 Inferences 24,00 22 8 30 Procedural sequences 25,00 22 8 30
The assumption of normality was tested via the Shapiro-Wilk test for normality. A review of the results of the tests suggested that the scores of the participants with PD and the partic-ipants of the study of Büttner (2014) are not normally distributed for the total MAKRO score (SW = .806, df = 44, p = .001) as well as scores of the three subtests, i.e. text
comprehen-sion (SW = .669, df = 44, p = .001), inferences (SW = .752, df = 44, p = .001) and procedural
sequences (SW = .802, df = 44, p = .001). Therefore, the non-parametric Mann-Whitney U
test was used for the group comparison. The scores of the participants with PD of the
MAK-RO (SW = .867, df = 24, p = .005) as well as scores of all subtests, i.e. text comprehension
(SW = .803, df = 24, p = .001), text production (SW = .917, df = 24, p = .049), inferences (SW = .890, df = 24, p = .013) and procedural sequences (SW = .842, df = 24, p = .002), are not normally distributed. Therefore, the non-parametric coefficient Spearmen’s rho (rs) was used
to assess the correlations.
MAKRO: Comparison of participants with PD and controls
text comprehension task was higher for participants of the control group (Mdn = 30) than for
the participants with PD (Mdn = 27), U = -5.019, p = .001, the performance on inferences was higher for participants of the control group (Mdn = 30) than for the participants with PD (Mdn = 24), U = -5.599, p = .001 and the performance on procedural sequences was higher for participants of the control group (Mdn = 25) than for the participants with PD (Mdn = 30),
U = -5.156, p = .001.
MAKRO: Correlating factors
Total score
A significant correlation could be found between the performance on the MAKRO of the participants and their age, rs = -.525, p = 0.008, and their age at onset, rs = -.540, p = .006, their level of education, rs = -.671, p = .001, and their performance on the PANDA, rs = .743,
p = .001 and all PANDA subtests word pair association (immediate retrieval), rs = .634, p = .001, on the subtest word pair association (delayed retrieval), rs = .540, p = .006, the
alternat-ing verbal fluency task, rs = .506, p = .012, the visuospatial task, rs = .492, p = .015, or the
working memory task, rs = .709, p = .001. No significant correlation could be found between the performance on the MAKRO of the participants and their H&Y staging, rs = -.096, p = .657, their disease duration, rs = .099, p = .645 or their performance on the BNT, rs = .172, p
= .422.
MAKRO subtests
p = .056, the word pair association (delayed retrieval), rs = .375, p = .071, the verbal fluency task, rs = .264, p = .212, or the visuospatial task, rs = .348, p = .096.
A significant positive correlation could also be found between the performances of the participants on the MAKRO subtest text production and the performance on the PANDA, rs = .515, p = .010 and on the PANDA subtests word pair association (immediate retrieval), rs = .476, p = .019, the subtest word pair association (delayed retrieval), rs = .517, p = .010, the
visuospatial task, rs = .438, p = .032, and the working memory task, rs = .464, p = .022. No significant correlation could be found between the performance in verbal fluency of the partic-ipants and their performances on the subtest text production, rs = .385, p = .063.
There was also a significant positive correlation between the performances of the partic-ipants on the MAKRO subtest inferences and their total score on the PANDA, rs = .779, p = .001 as well as between the performances on the subtest inferences and the ones on the
PANDA subtests word pair association (immediate retrieval), rs = .710, p = .001, word pair
association (delayed retrieval), rs = .509, p = .011, the alternating verbal fluency task, rs = .617, p = .001, and the working memory task, rs = .712, p = .001. No significant correlation could be found between the performance on the visuo-spatial task of the PANDA and the performances on the subtest inferences, rs = .369, p = .076.
A significant positive correlation could also be found between the performances of the participants on the MAKRO subtest procedural sequences and their performance on the
PANDA, rs = .734, p = .001 and on all PANDA subtests, i.e. word pair association (immediate
retrieval), rs = .592, p = .002, word pair association (delayed retrieval), rs = .489, p = .015, the
Discussion
The aim of the present study was to provide first information whether patients with PD show communication disorders on the macro level of communication. Out of the 24 partici-pants 21 scored below the cut-off score of 109 on the MAKRO. The performance of the par-ticipants with PD on the MAKRO and the subtests text comprehension, inferences and pro-cedural sequences was significantly lower than the one of the control group from the study of Büttner (2014). Thus, it can be assumed that communication problems are a common sec-ondary non-motor symptom in PD. This could already be expected since Boelen et al. (2011) showed that EF seem to be a very important factor influencing the communication and exec-utive dysfunction is often present in patients with PD from the time of diagnosis (Garcia-Ptacek & Kramberger, 2016). Furthermore, Büttner (2014) showed that the performances on the MAKRO have high correlations with performances on a test for EF. Apart of the expecta-tion based on EF, the results of the present study fit to the current literature on communica-tion disorders in patients with PD. McNamara and Durso (2003) found that patients with PD were significantly impaired on measures of pragmatic communication skills. Still, the present study adds unique information on communication skills in patients with PD since it is the first study assessing communication skills on different levels, i.e. text comprehension and produc-tion, the ability to make inferences and procedural sequences.
Another aim of the study was to identify influencing factors, i.e. age, age at onset, disease duration and the H&Y staging. Moreover, the relation between communicative skills and cognitive functions and language skills, in this case word retrieval, was examined. First, the results for the total score on the MAKRO are discussed and later the results for the
MAKRO subtests.
MAKRO
– total score
In the correlation analysis, it could be found that the factors, age and age at onset corre-late negatively with the performance on the MAKRO, while years of educations and the per-formance on the PANDA and all PANDA subtests significantly correlated positively. In con-trast, no significant correlation could be found between the performance on the MAKRO and the H&Y staging as well as the disease duration and the performance on the BNT.
However, the BNT contains only nouns which might be important since it could be shown that patients have more problems in producing verbs than nouns (Bertella et al., 2002; Bou-lenger et al., 2008; Cotelli et al., 2007; Péran et al., 2009). Since the BNT does not include verb retrieval or more complex language functions, a hypothesis over the influence of lan-guage functions on the communication skills in general cannot be done. It can only be con-cluded that object retrieval abilities do not have influence on communication skills.
cor-related larger with the performance on the MAKRO than verbal fluency as attentional set-shifting is especially important in alternating fluency tasks.
In comparison to the other subtests of the PANDA only a medium correlation could be found between the performances on the visuospatial task and the total MAKRO score. This mirrors the fact that not between every subtest of the MAKRO and the visuospatial task a significant correlation could be found. How the different cognitive functions influenced the performance on the MAKRO subtests will be discussed in the following chapter.
Interestingly, more participants scored under the cut-off score in the MAKRO screening than in the PANDA. This can have different reasons. Firstly, the cut-off scores of the recently published MAKRO might be not representable since they are based on only 20 healthy con-trols. The patients with PD were significantly older than the controls and therefore age was an uncontrolled factor in this study that may account for the pragmatic performance differ-ences in the two groups. However, even two of the youngest patients with PD scored below the cut-off score and all participants scored lower than the control person with the lowest total score (116 points). In addition, pragmatic communication deficits may not be reliably associated with normal aging (Baltes, 1993; Nocentini, Goulet, Drolet, & Joanette, 1999). Secondly, the cut-off scores of the MAKRO might have a higher sensitivity to detect impair-ments. The tasks in the MAKRO are more complex than the ones in the PANDA. However, it could also be possible that communication deficits become evident before the cognitive de-clines. This could also be explained through the higher demands on cognitive functions.
Vascocellos & Pereira, 2015) and PD-D (Aarsland and Kurz, 2010). Normal cognitive decline in aging, however, is characterised by cognitive functions, not to a qualitatively difference (Beyreuther et al., 2002). In addition, pragmatic communication deficits may not be reliably associated with normal aging (McNamara & Durso, 2003). It can be concluded that age is rather an influencing factor since the risk for patients with PD is higher to develop a cognitive impairment and not because it mirrors the cognitive decline in aging. The same might be suggested for the factor age at onset.
As well as age, a lower level of education is a risk factor for patients with PD to develop a MCI (Pedersen et al., 2013; Vascocellos & Pereira, 2015). Furthermore, it is known that cognitive functions and especially EF are influenced by the level of education. Thus, a posi-tive influence of education on the communication skills is explainable through its influence on cognitive functions especially the EF.
The H&Y staging is also considered to be a risk factor for MCI (Pedersen et al., 2013). Even though this is not reported consistently in the current literature. In this study, the H&Y staging is not significantly correlated to the communication skills. This might be because the scale is describing the severity of motor symptoms only. Thus, the cognitive decline is not related to this scale (Goetz et al., 2004).
As age and the level of education showed to correlate significantly with the total score of the MAKRO, it could be suggested to base the cut-off norms on a larger group of healthy controls and evaluate if age and education specific norms are needed in this communication screening.
MAKRO
– subtests
Six out of the 24 participants showed impaired text comprehension, five severe and one slightly. The score on this MAKRO subtest was significantly lower in patients with PD com-pared to healthy controls. The correlation between the performance on the PANDA and the performance on the subtest text comprehension of the MAKRO can be considered as medi-um. Regarding the subtests of the PANDA, only between the performance on the working
memory task and the performance on the text comprehension task a significant correlation
was chosen to limit the influence of memory span and concentration to get a realistic impres-sion of the text comprehenimpres-sion rather than these two cognitive functions. Since her partici-pants, as well as the participants in the current study, scored below average, it can be as-sumed that the subtest is a reliable instrument to measure text comprehension.
In the current literature, an impairment of sentence comprehension is frequently report-ed. Thus, it could have been expected that more patients struggle with the text
comprehen-sion task since this is a more complex task than sentence comprehencomprehen-sion. Still, the reasons
for impairments in sentence comprehension might apply for text comprehension as well. Some authors explain these problems through an impairment of executive functions (Col-man, Koerts, van Beilen, Leenders, & Bastiaanse, 2006; Geyer & Gross(Col-man, 1994; Hochstadt, Nakano, Lieberman, & Friedman, 2006; Kemmerer, 1999). Hochstadt et al. (2006) assessed the relation between distinct EF and sentence comprehension and con-cluded that limits on sequencing and/ or verbal working memory may be responsible for the syntactic comprehension deficits in patients with PD. Grossman et al. (2002) explain sen-tence comprehension deficits in patients with PD through impairments in inhibition and plan-ning. This shows that EF and in particular working memory are considered to be an important factor for sentence comprehension and thus might also be one for text comprehension. This stands in line with the results of the current study.
Col-man, 2011). In the used text, one popular saying was used, which was understood by all par-ticipants, but no metaphors.
On the subtest text production of the MAKRO, 21 participants scored under the cut-off score, with two participants being slightly and 19 being severely impaired. Thus, it is the sub-test in which most participants scored below the cut-off. In contrast to text comprehension, text production seems to be more relying on cognitive functions. Additionally to selection and sequencing skills, verbal memory processes as well as a flexible retrieval of semantic-lexical knowledge and grammatical skills on text level are needed (Büttner, 2014). Lewis et al. (1998) showed that with increasing cognitive impairment, the performances on tasks with greater demands of cognitive processing, such as processing ambiguity, explaining meta-phors and constructing sentences became worse. As a consequence, more participants showed performances under the cut-off score. The lack of coherence and the decreased information content found in patients with PD in this study stands in line with the current liter-ature (Cummings, Darkins, Mendez, Hill and Benson, 1988; Bayles, 1990; Small, Lyons and Kemper, 1997; Pahwa et al., 1998; Emre, 2010; Caballol et al., 2007).
The performance of the participants on the PANDA as well as on four of the five subtests separately were significantly correlating with the performance on the subtest text production of the MAKRO. While the correlation between the total score and the performances on the subtest word pair association (delayed retrieval) can be considered as large, the correlation with the subtests word pair association (immediate retrieval), the visuospatial task and the
working memory task are medium. No significant correlation could be found between the
in the fact that the correlation between the delayed retrieval task with the text production was the largest of all PANDA subtests.
Apart from the influence of the cognitive functions on text production, a further explana-tion might be the motor speech impairment of patients with PD as Murray and Lenz (2001) stated that reduced syntax mirror not only cognitive but also motor speech impairments ra-ther than being a pure language deficit. A motor speech impairment could also lead to the tendency to tell a shorter story. It would be interesting to study the roll of the degree of dysar-thria in correlation with the length of text production and the information content.
In contrast to the subtests text comprehension and inferences, a significant correlation between the performances on the visuospatial task and the text production task could be found. Considering the fact that the participants had to tell a story based on presented pic-tures, the influence of visuospatial skills on this task is logic. However, it can be critically dis-cussed whether this task should be the only task measuring text production in a communica-tion screening. A person with a severe impairment of visuospatial percepcommunica-tion might still be able to tell a coherent story and communicate adequately.
On the subtest inferences, 19 participants scored below the cut-off score, with ten partic-ipants being slightly and nine being severely impaired. The particpartic-ipants with PD scored sig-nificantly lower than the control group of Büttner (2014). That stands in line with the results of Berg et al. (2003) who reported that patients with PD without cognitive abnormalities in the
Mini Mental State Examination exhibit impairments in making inferences compared to healthy
controls. The performances on the PANDA and on the subtests word pair association
(im-mediate retrieval), word pair association (delayed retrieval), the alternating verbal fluency
task, and the working memory task correlated significantly and largely with the performance
performance on alternating verbal fluency tasks with this subtest. In the current study, how-ever, a larger correlation could be found between inferences and the working memory task than between inferences and the verbal fluency task. Even if the sentences are presented during the whole task, the participants need working memory skills for example to keep the first sentences in mind while reading the last one or to remember their inference while writing which is a complex motor skill for patients with PD (Jankovic, 2008).
The ability to make inferences correlates highly with cognitive functions, which fits out-comes of earlier studies on this topic. Since inferences can only be successively made by logical reasoning, a certain degree of cognitive skills like memory, working memory and EF is always required (Büttner, 2014). Berg et al. (2003) suggested that the performance on tasks testing the ability to make inferences strongly correlated with the performance on the test of cognitive function.
The answers of the participants in the subtest inferences are analysed regarding the content and the morpho-syntaxis. Normally, an analysis of morpho-syntaxis is not part of the macro level of language but rather of the micro level. However, the panning of morph-syntactical structures is commonly impaired in patients with deficits in EF. Therefore, this analysis was included in the subtest inferences. A qualitative analysis of the answers of the patients with PD showed that morpho-syntactical mistakes occurred frequently. This might be explained through the already high demand on cognitive functions to make inferences. The additional task to produce syntactically correct sentences might have been too challenging for some patients.
Thus, this ability is often impaired in patients with dysexecutive symptoms. The ability to or-der procedural sequences is a predictor for the representation of structural complexes of events. The performances of the participants on the subtest procedural sequences of the
MAKRO correlated highly with the total PANDA score and the scores on all PANDA subtests.
The influence of the performances on the total PANDA and the subtests word pair
associa-tion (immediate retrieval) and the working memory task can be considered as larg. The high
correlation of the working memory task can be explained through the construction of the task itself. Many propositions need to be in the focus of the attention at the same time to bring them into a logical order.
Conclusion
References
Aarsland, D. & Kurz, M.W. (2010) The epidemiology of dementia associated with Parkinson´s disease. Brain Pathology, 20 (3), 633 - 639.
Aarsland, D., Zaccai, J., & Brayne, C. (2005). A systematic review of prevalence studies of dementia in Parkinson’s disease. Movement Disorders, 20 (10), 1255-1263.
Almor, A., Kempler, D., Andersen, E.S., MacDonald, M.C., Hayes, U.L., & Hintiryan, H. (2002). The production of regularly and irregularly inflected nouns and verbs in Alz-heimer and Parkinson patients. Brain and Language, 83, 149 – 151.
American Psychological Association (2013). Diagnostic and Statistical Manual of Mental
Dis-orders (5 Ed.). Washington: American Psychological Association.
Ballard, C.G., Aarsland, D., McKeith, I., O’Brien, J., Gray, A., Cormack, F., Burn, D., ... Tovee, M. (2002). Fluctuations in attention: PD dementia vs DLB with parkinsonism.
Neurology, 59, 1714 –1720.
Baltes, P.B., (1993). The aging mind: Potential and limits. Gerontologist, 33 (5), 580 – 594. Bastiaanse, R., & Leenders, K. L. (2009). Language and Parkinson’s disease. Cortex, 45,
912-914.
Bayles, K.A., (1990). Language and Parkinson disease. Alzheimer Disease and Associated
Disorders, 4, 171 – 180.
Beck, A.T., Steer, R.A. & Brown, G.K. (1996). Manual for the Beck Depression Inventory-II. Psychological Corporation: San Antonio, Taxas.
Berg, E., Björnram, C., Hartelius, L., Laakso, K., Johnels, B. (2003). High-level language dif-ficulties in Parkinson's disease. Clinical Linguistics & Phonetics, 17 (1), 63-80.
Bertella, L., Albani, G., Greco, E., Prino, L., Mauro, A., Marchi, S., …, Sementa, C. (2002). Noun verb dissociation in Parkinson´s disease. Brain and Cognition, 48, 277 – 280. Beyreuther, K., Einhäupl, K.M., Förstl, H. & Kurz, A. (2002). Demenzen Grundlagen und
Klinik. Stuttgart: Georg Thieme Verlag.
Bocanegra, Y., García, A.M., Pineda, D., Buriticá, O., Villegas, A., Lopera, F., Gómez, D., Gómez-Arias, C., Cardona, J.F., Trujillo, N., & Ibáñez, A. (2015). Syntax, action verbs, action semantic and object semantics in Parkinson’s disease: Dissociability, progression, and executive influences. Cortex, 69, 237-254.
Boelen, D.H., Allain, P., Spikman, J.M., & Fasotti, L. (2011). Script generation and the Dysexecutive Syndrome in Patients with Brain Injury. Brain Injury, 25 (11), 1091 – 1100. Brønnick, K. (2010). Cognitive profile in Parkinson’s disease dementia. In M. Emre (Ed.),
cognitive impairment and dementia in Parkinson’s disease (p. 27-46). Oxford: Oxford
University Press.
Boulenger, V., Mechtouff, L., Thobis, S., Broussolle, E., Jeannerod, M., & Nazir, T.A. (2008). Word processing in Parkinson´s disease is impaired for action verbs but not for concrete nouns. Neuropsychologia, 46, 743 – 756.
Büttner, J. (2014). Sprache und Kognition – Diskurspragmatik und Textverarbeitung bei
Caballol, N., Martí, M.J., & Tolosa, E. (2007). Cognitive dysfunction and dementia in Parkin-son disease. Movement Disorders, 22 (17), 358-366.
Cain, K. (2006). Individual differences in children’s memory and reading comprehension: An investigation of semantic and inhibitory deficits. Memory, 14 (5), 553–569.
Caplan, D. & Waters, G.S. (1999). Verbal working memory and sentence comprehension. T
he Behavioral and Brain Sciences, 22, 77 - 94.
Carlsson, E., Hartelius, L., & Saldert, C. (2014). Communication strategies used by spouses of individuals with communication disorders related to stroke-induced aphasia and Pa-krinson´s disease. International Journal of Language and Communication disorders, 49
(6), 722 – 735.
Carthery-Goulart, M.T., da Silva, H.S., Rocha, M.S.G., Baradel, R.R., Godinho, F., Cravo, A. & Sato, J.R. (2015). Semantic memory manifestations in normal agning and parkinson´s disease and their implications to the embodied cognition theory. Alzheimer´s &
Demen-tia, 11 (7), 818.
Christopher, M.E., Miyake, A., Keenan, J. M., Pennington, B., DeFries, J.C., Wadsworth, S.J., …, Olson, R.K. (2012). Predicting word reading and comprehension with executive function and speed measures across development: A latent variable analysis. Journal of
Experimental Psychology-General, 141 (3), 470–488.
Cohen, H., Bouchard, S., Scherzer, P., & Whitaker, H. (1994). Language and verbal reason-ing in Parkinson´s disease. Neuropsychiatry, Neuropsychology & Behavioral Neurology,
7 (3), 166 – 175.
Colman, K., Koerts, J., van Beilen, M., Leenders, K., & Bastiaanse, R. (2006). The role of cognitive mechanisms in sentence comprehension in dutch speaking Parkinson’s dis-ease patients: Preliminary data. Brain and Language, 99 (1-2), 120 – 121.
Colman, K., Koerts, J., van Beilen, M., Leenders, K., Post, W., & Bastiaanse, R. (2009). The impact of executive functions on verb production in patients with Parkinson’s disease.
Cortex, 45, 930 – 942.
Colman, K.S.F. (2011). Behavioral and neuroimaging studies on language processing in Dutch speakers with Parkinson's disease. Groningen dissertations in linguistics, 90, Uni-versity of Groningen.
Copland, D.A. (2003). The basal ganglia and semantic engagement: potential insights from semantic priming in individuals with subcortical vascular lesions, Parkinson’s disease and cortical lesions. Journal of the International Neuropsychological Society, 9, 1041– 1052.
Copland, D.A., Chenery, H.J., & Murdoch, B.E. (2001). Discourse priming of homophones in individuals with dominant nonthalamic subcortical lesions, cortical lesions and Parkin-son's disease. Journal of clinical and experimental neuropsychology, 23(4), 538 – 556. Costa, A., Monaco, M., Zabberoni, S., Peppe, A., Perri, R., Fadda, L., …,Carlesimo, G.A.
(2014). Free and cued recall memory in Parkinson’s disease associated with amnestic mild cognitive impairment. PLoS One, 9, 86233.
Cotelli, M., Borroni, B., Manenti, R., Zanetti, M., Arévalo. A., Cappa, S.F., & Paovani, A. (2007). Action and object naming in Parkinson’s disease without dementia. European
Cummings, J.L., Darkins, A., Mendez, M., Hill, M.A., & Benson, D.F. (1988). Alzheimer´s disease and Parkinson´s disease: Comparison of speech and language alternations.
Neurology, 38, 680 – 684.
Cutting, L.E., & Scarborough, H.S. (2006). Prediction of reading comprehension: Relative contributions of word recognition, language proficiency, and other cognitive skills can depend on how comprehension is measured. Scientific Studies of Reading, 10 (3), 277– 300.
Daneman, M., & Merikle, P.M. (1996). Working memory and language comprehension: A meta-analysis. Psychonomic Bulletin Review, 3 (4), 422–433.
De Bleser, R., Willmes, K., Graetz, P. & Hagoort, P. (1991). De Akense Afasie Test (2): Psy-chometrische kenmerken van de Nederlandstalige versie. Logopedie en Foniatrie, 63, 207 – 217.
Dressler, W.U., Stark, H.K., Vassilakou, M., Rauchensteiner, D., Tosic, J., Weitzenauer, S.M., …, Brunner, G. (2004). Textpragmatic Impariments of Figure-Ground Distinction in Right-Brain-Damaged Stroke Patients compared with Aphasics and Healthy Controls.
Journal of Pragmatics, 36 (2), 207 – 215.
Dujardin, K., Moonen, A.J.H., Behal, H., Defebvre, L., Duhamel, A. Duits, A.A., ...., Leentjes, A.F.G. (2015). Cognitive disorders in Parkinson's disease: Confirmation of a spectrum of severity. Parkinsonism and Related Disorders, 21, 1299 – 1305.
Ellis, C., Holt, Y.F., & West, T. (2015). Lexical diversity in Parkinson’s disease, Journal of
Clinical Movement Disorders, 2 (5), 1-6.
Ellis, C., Crosson, B., Gonzalez Rothi, L.J., Okun, M.S., & Rosenbek, J.C. (2015). Narrative Discourse Cohesion in Early Stage Parkinson's Disease. Journal of Parkinson's disease,
5(2), 403-11.
Ellis, C., Henderson, A., Wright, H.H., & Rogalski, Y. (2016). Global coherence during dis-course production in adults: a review of the literature. International Journal of Language
& Communication Disorders, 51 (4), 359 – 367.
Emre, M. (2003). Dementia associated with Parkinson’s disease. Lancet Neurology, 2, 229-37.
Emre, M., Aarsland, D., Brown, R., Burn, D.J., Duyckaerts, C., Mizuno, Y., …, Dubois, B. (2007). Clinical diagnostic criteria for dementia associated with Parkinson´s disease.
Movement Disorders, 22 (12), 1689 -1707.
Emre, M. (2010). General features, mode of onset, and course of dementia in Parkinson’s disease. In M. Emre (Ed.), cognitive impairment and dementia in Parkinson’s disease (p. 17-26). Oxford: Oxford University Press.
Eriksson, K., Forsgren, E., Hartelius, L., & Saldert, C. (2016). Communication partner training of enrolled nurses working in nursing homes with people with communication disorders caused by stroke or Parkinson’s disease. Disability and Rehabilitation, 38 (12), 1187 – 1203.
Galtier, I., Nieto, A., Lorenzo, J.N. & Barroso, J. (2016). Mild cognitive impairment in Parkin-son’s disease: Diagnosis and progression to dementia. Journal of Clinical and
Experi-mental Neuropsychology, 38(1), 40-50.
Garcia-Ptacek, S. & Kramberger, M. (2016). Parkinson Disease and Dementia. Journal of
