Communication impairment in Parkinson’s disease: Impact of motor and cognitive symptoms on speech and language

Brain and Language 185 (2018) 38–46

Contents lists available at ScienceDirect

Brain and Language journal homepage: www.elsevier.com/locate/b&l

Review

Communication impairment in Parkinson’s disease: Impact of motor and cognitive symptoms on speech and language

T

⁎

Kara M. Smitha, , David N. Caplanb a b

University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655, USA Massachusetts General Hospital, 175 Cambridge Street, Boston, MA 02114, USA

A R T I C LE I N FO

A B S T R A C T

Keywords: Parkinson’s disease Communication Acoustics Linguistics Syntax Grammar Pausing

Communication impairment is common in Parkinson’s disease (PD) and may have both motor speech control and cognitive-linguistic underpinnings. The neurobiology of communication impairment in PD is poorly understood, and work is needed to disentangle the relative contributions of motor and cognitive dysfunction. In clinical practice, cognitive-linguistic impairments are often overlooked despite the large body of research on this topic in neurocognitive and linguistics literature. In this review, we will discuss the roles of motor speech changes, cognitive and linguistic impairment, and other related functions in the communication disabilities of individuals with PD. We will describe the various types of communication difficulties in PD and tools for measuring these symptoms. We will discuss specific deficits that may further understanding of the neurobiology of communication impairment in PD, including voice and speech acoustic changes, linguistic processing and production difficulties, and pausing. We will emphasize the importance of an interdisciplinary approach and the patient perspective on daily communication in guiding future research.

1. Introduction Communication is one of the most important functions that defines humans as individuals, and loss of ability to communicate can have a devastating effect. Even mild communication difficulties may lead to impaired occupational function, social isolation, and depressive symptoms. Communication impairment is common in Parkinson’s disease (PD), present in up to 90% of patients (Miller et al., 2007). Communication difficulties may be present early in the course of PD or develop in later stages (Miller, 2017). Individual with PD often report difficulties with daily communication, but subjective symptoms may differ from the objective deficits detected through standard clinical evaluation. The degree of communication impairment varies widely, but can progress to inaudible and unintelligible speech, preventing communication with loved ones, caregivers and healthcare providers. Functional communication, or the real-life daily communication experiences of the patient, is an important concept as it differs from communication deficits detected in controlled research settings. Functional communication may involve conversational, extemporaneous, or prepared speaking with various speakers (e.g. familiar and unfamiliar), in various physical settings (e.g. phone and in –person, social and occupational), and is therefore critical to activities of daily living. Communication

⁎

deficits in PD are particularly complex because PD affects both motor and cognitive function. It is not entirely understood which communication deficits are based on effects on motor speech control and the musculature of the vocal apparatus, and which have a cognitive or linguistic basis. It is important to better understand the nature and the underlying pathophysiology of communication deficits in PD, in order to diagnose and monitor changes, predict prognosis, and develop new therapeutic interventions to combat these symptoms. Although PD is primarily defined by its motor features of tremor, bradykinesia, and rigidity, it is widely recognized that cognitive impairment is prevalent. Cognitive impairment may emerge at any point in the disease course, and both the severity and evolution of deficits are heterogeneous. Subtle cognitive changes may even occur before the development of clinically apparent motor features (Postuma et al., 2012). At the time of initial PD diagnosis, mild cognitive impairment (MCI) may be present in up to 30–40% of patients (Yarnall, , 2014). Those with MCI are at high risk of progression to dementia within 5 years (Pigott et al., 2015). During a typical 15–20 year disease course, the majority of all individuals with PD will develop dementia (Hely, Reid, Adena, Halliday, & Morris, 2008). Despite the prevalence of cognitive impairment in PD, the relationship of cognitive deficits and communication disability in PD has

Corresponding author. E-mail addresses: [email protected] (K.M. Smith), [email protected] (D.N. Caplan).

https://doi.org/10.1016/j.bandl.2018.08.002 Received 17 December 2017; Received in revised form 21 July 2018; Accepted 2 August 2018 0093-934X/ © 2018 Elsevier Inc. All rights reserved.

Brain and Language 185 (2018) 38–46

K.M. Smith, D.N. Caplan

predominantly akinetic-rigid clinical phenotype had impaired tip-ofthe-tongue response accuracy compared to tremor-predominant PD subjects (Yu, Wu, Tai, Lin, & Hua, 2010). There may well be a discrepancy between the anecdotally common patient-reported symptom of word finding difficulty and “tip-of-the-tongue” instances in an experimental setting. Future studies should address word finding difficulty in PD in the context of daily communication. Studies designed to assess the epidemiology and functional impact of this symptom beyond that of normal aging-related changes will be helpful to better understand and address this symptom in the clinical setting. It has been hypothesized that word finding difficulty in PD is explained at least in part by decreased speed of processing, akin to the motor slowing characteristic of the condition. Another possible explanation is impaired access to, or content of, the lexicon. Although the evidence is mixed, most studies suggest that the lexicon, as measured by confrontation naming, is relatively preserved in PD without dementia (Yarnall et al., 2014). Lexical retrieval may therefore be impaired in PD because of dysfunction of the connections between the frontal cortex and basal ganglia needed to achieve efficient retrieval. This impairment would be most obvious when there were additional frontal executive demands, such as during a conversation with rapidly shifting topics. This hypothesis of frontostriatal dysfunction in word retrieval is supported by studies of verbal fluency in PD. Some but not all studies show that verbal fluency tasks requiring frontal executive function are most consistently impaired in PD (see Section 3.3.2.4 for further discussion of verbal fluency). Also in support of this hypothesis are the results of Yu et al. mentioned above. The PD subjects who had akinetic-rigid predominant motor symptoms and showed impaired tip-of-the-tongue response accuracy also had worse executive function. While evidence for a frontostriatal etiology of word finding difficulty is still preliminary, it suggests that testing fluency, executive function, and word finding in more challenging paradigms that reflect functional communication may be necessary to capture these subtle deficits.

been relatively overlooked in comparison with studies of motor function. Clinicians treating patients with PD often focus more on motor symptoms, especially in earlier disease when cognitive deficits are subtle, and may not screen for the entire range of possible communication deficits. Speech therapy programs designed for PD, such as the Lee Silverman Voice Training Program, also focus on motor aspects of speech (Sapir, Ramig, & Fox, 2011). New therapeutic approaches incorporating cognitive as well as motor aspects of speech and language may improve outcomes on functional communication. Before such therapies can be developed, research is greatly needed to explore the complex interplay between motor and cognitive function in all aspects of communication in PD. It will be critical for this work to involve interdisciplinary teams of researchers to bridge the gaps between clinicians (neurologists and speech-language pathologists) and neuroscience and linguistics researchers. This review will attempt to elucidate, to the extent of our current knowledge, the relative roles of motor speech changes, cognitive and linguistic impairment, and other related functions in the communication deficits of individuals with PD. We will first explore patient-reported communication symptoms in PD, their functional impact, and approaches for measurement. We will then discuss specific areas of impairment that may impact communication, including voice and speech acoustic changes, linguistic processing and production deficits, and pausing. Finally, we will discuss gaps and future directions in research with a focus on the patient perspective and quality of life. 2. Material and methods We searched the PubMed database using search terms “Parkinson’s” and “language,” “speech”, “acoustics,” “voice,” “pause,” and “communication.” Inclusion criteria included English language, inclusion of human subjects with PD, and primary focus on communication or speech and language. Exclusion criteria included papers that only described technical aspects of measuring voice or speech, or computational algorithm development for speech analysis. Papers were handreviewed by K.S. and selected based on their relevance to the objectives of the review. Additional papers were searched and selected based on our areas of focus, including papers describing similar topics in neurological disorders other than PD for comparison.

3.2. Clinical assessment of communication in PD Given that communication symptoms in PD vary considerably in type and severity, clinical assessment tools are needed to gather both individual and population-level data. It is difficult to directly observe and measure a patient’s range of communication difficulties in an office-based visit. Symptoms are likely to vary with the pace and complexity of the communication, the number and identity of speakers (familiar or unfamiliar), and the surrounding environment. Possible approaches to assessing communication symptoms include patient-reported and knowledgeable informant/caregiver-reported scales and questionnaires, and objective quantitative measures designed for use in the patient’s daily life. Overall there are few assessment measures designed to capture daily communication difficulties in PD. The most common quality of life scale in PD, PDQ-39, has one question on speech and one question on communication with other people (Jenkinson, Fitzpatrick, Peto, Greenhall, & Hyman, 1997). The Penn Parkinson’s Daily Activities Questionnaire (PDAQ) has a 50 item and a 15 item version. It was developed using item response theory to capture cognitive impact on instrumental activities of daily living in PD and includes several questions pertaining to communication (Brennan et al., 2016a, 2016b). However, it was validated for a knowledgeable informant to complete rather than the patient themselves. It is unknown how patient and knowledgeable informant perspectives on communication symptoms may differ, as these particular symptoms can be quite subjective in how they are experienced by the patient. Miller et al. administered a self-developed questionnaire about communication abilities to PD patients and their caregivers. The patients’ perceptions of communication abilities did not correlate with intelligibility, or with severity of motor or cognitive symptoms. Some PD patients perceived worse deficits than their caregivers reported, suggesting that subtle subjective difficulties may still impact the patient and that some

3. Results and discussion 3.1. Common self-reported speech and communication symptoms in PD Both quantitative and qualitative research methodologies have been used to understand the scope and nature of communication difficulties in PD. Some clinical scales require a speech and language pathologist to perform, and others are questionnaires based on patient response. One cross-sectional survey of PD patients in Sweden revealed that worsened speech was the second most common symptom endorsed overall in PD, equally prevalent in early and later disease (Schalling, Johansson, & Hartelius, 2017). Qualitative research by Miller et al. found that individuals with PD commonly report challenges in carrying on a conversation (Miller, Noble, Jones, Allcock, & Burn, 2008). These communication difficulties impacted perception of self-worth and led to avoidance of social interactions. Word finding difficulty and getting off topic during conversation are common patient-reported symptoms impacting communication in PD (Schalling et al., 2017). Word finding difficulty in PD was initially described as “tip-of-the-tongue” phenomenon by Fahn et al. and measured by confrontational and category naming (Matison, Mayeux, Rosen, & Fahn, 1982). However, when tested using general knowledge questions, subjects with PD had a similar percentage of tip-of-the-tongue instances compared with elderly controls and both groups were similarly accurate in predicting their ability to recognize the correct word within a list (Oh-Lee, Szymkowicz, Smith, & Otani, 2012). One study reported that PD subjects with a 39

Brain and Language 185 (2018) 38–46

K.M. Smith, D.N. Caplan

patients with objectively “mild” speech impairment have a strong subjective impression of difficulty communicating (Miller et al., 2008). Due to the limitations of standard speech assessments to evaluate everyday communication, the Communicative Participation Item Bank (CPIB) (Baylor et al., 2013) has been proposed for use in PD. In PD, CPIB score is associated with self-reported cognitive difficulties, objective measures of global cognitive function (Montreal Cognitive Assessment, MoCA) and health-related quality of life. Other clinical scales utilized by speech-language pathologists typically focus on measurement of dysarthria, speech coordination (diadochokinesis), and intelligibility. These measurement tools may miss many of the features impacting functional communication in PD. Experts in this field advocate moving beyond informal observation of conversations and patient or caregiver-based perceptual scales of communication. Objective tools that quantitatively assess functional communication would benefit researchers and clinicians by providing a more detailed and sensitive approach to diagnosing, monitoring and managing communication deficits. Griffiths et al. recommended a conversation analysis approach as has been used in aphasia research (Griffiths, Barnes, Britten, & Wilkinson, 2011), however this is time and labor intensive. In research studies, various measures of spontaneous speech and narrative discourse have been utilized to reflect typical speech, such as narration of a wordless picture book, description of a picture of a scene, re-telling of a well-known story, or monologue based on a simple prompt. The speech samples obtained through these approaches also require laborintensive processing and analysis, making them impractical for large scale screening or monitoring of speech symptoms. Surveys of speech and language therapists and patients reveal a desire for greater focus on speech participation in daily activities rather than individual impairment measures (Miller, Deane, Jones, Noble, & Gibb, 2011). Future directions in this area include remote sampling of speech during daily experiences using mobile technology. These speech samples could either be assessed manually by the patient’s speechlanguage pathologist, or processed and interpreted automatically using computerized algorithms. In the next section, we will discuss automated speech analysis approaches that could be utilized in this fashion as technology continues to advance.

disease. The speech samples used have been as simple as sustained phonation (Tsanas, Little, McSharry, & Ramig, 2011), but some suggest that complex speech such as a monologue is more informative (OrozcoArroyave et al., 2015) Orozco-Arroyave et al. showed their approach to be robust to differences across languages, obtaining reasonable accuracy in Spanish, German and Czech (Orozco-Arroyave et al., 2016). Less work has been done in using acoustic analysis to help manage patients who have already been diagnosed, for instance by assessing the benefits and side effects of therapeutic interventions. De Letter et al. found that only a small subset of speech acoustic parameters vary significantly with levodopa, while the majority of acoustic parameters had no relationship to the timing of the dose or the global motor response (De Letter et al., 2010). This was consistent with previous literature, supporting the consensus that while some acoustic variables may improve with levodopa, others may deteriorate, and that overall there is no reliable or meaningful improvement in speech with use of levodopa (Ho, Bradshaw, & Iansek, 2008; Plowman-Prine et al., 2009). A single study in patients with PD who had undergone deep brain stimulation (DBS) found that vocal acoustic changes were associated with over-stimulation. Over-stimulation and undesired spread of electrical stimulation to non-target tissue commonly causes speech side effects in patients with DBS. It can be challenging to balance these stimulation-induced side effects with symptomatic benefit, and the use of acoustic markers may help guide clinicians toward more optimized DBS programming (Valalik, Smehak, Bognar, & Csokay, 2011). Advantages of automated speech signal analysis include the opportunity for frequent and remote monitoring of PD symptoms that could be scalable to large populations. However, it is not clear that the acoustic markers used in these models are specific to PD. In some of the approaches used, the acoustic markers vary similarly in other disorders, such as depression (Williamson, Quatieri, Helfer, Ciccarelli, & Mehta, 2014). In addition, acoustic research has focused almost exclusively on motor symptoms of PD, using speech acoustic markers to predict the Unified Parkinson’s Disease Rating Scale (UPDRS), the gold standard measure of motor severity. It is not clear how these models will detect and control for possible effects of concurrent cognitive and depressive symptoms on speech. Studies have also relied on relatively small speech databases (n = 30–80) and therefore validity and generalizability remains to be proven. In summary, automated speech analysis warrants further research and development, focusing on expanding sample size and looking at outcomes beyond basic motor symptom scales.

3.3. Speech and language impairments in PD Voice, speech, and language changes have been described in PD. Communication may be impacted by a deficit in any single area, and deficits may combine to produce more complex and significant impairment. We will next review the speech and language deficits in PD, and how study of these deficits may further our understanding of the neurobiology of communication changes in PD and the overall disease pathophysiology.

3.3.2. Language production and comprehension 3.3.2.1. Comprehension of syntax and grammar. Language deficits are often under-recognized in the clinical management of PD patients, but have been well established in linguistic research for decades (AuclairOuellet, Lieberman, & Monchi, 2017). Despite some mixed results in the literature, individuals with PD appear to have deficits in the processing and comprehension of complex grammar and syntax (Colman, Koerts, Stowe, Leenders, & Bastiaanse, 2011; Grossman et al., 2012; Hochstadt, 2009; Terzi, Papapetropoulos, & Kouvelas, 2005). For instance, individuals with PD tend to have difficulty understanding sentences of the objective-relative structure compared with subject-relative structure. In one study, PD subjects made syntax comprehension errors with no particular pattern, suggesting that they could not determine the appropriate grammar rules to apply based on the sentence structure, so instead guessed randomly (Kemmerer, 1999). The mechanism of these grammar and syntax deficits in PD is not entirely clear, but there are several lines of evidence pointing to the frontostriatal loop and dysfunctional underlying basal ganglia input. Studies have demonstrated links between language processing and setshifting (Simard et al., 2011; Simard, Monetta, Nagano-Saito, & Monchi, 2013), attention and executive function (Grossman et al., 1993; Grossman, Crino, Reivich, Stern, & Hurtig, 1992; Grossman, Lee, Morris, Stern, & Hurtig, 2002; Hochstadt, 2009; Lee, Grossman, Morris, Stern, & Hurtig, 2003) and working memory (Grossman et al., 2003;

3.3.1. Speech acoustics Production of speech is a complex task integrating motor and cognitive processing in real-time. The vocal apparatus is under motor control, and is therefore impacted by the basal ganglia dysfunction that occurs in PD. Slowness (bradykinesia) and rigidity of the vocal apparatus may lead to motor speech changes including decreased loudness, decreased pitch variation, and a decline in articulation typically summarized as hypokinetic dysarthria (Duffy, 2012). These motor speech changes can lead to speech that is slow, hypophonic, and monotone, which combine to reduce intelligibility (Canter, 1965). Speech changes may evolve as PD progresses (Ho, Iansek, Marigliani, Bradshaw, & Gates, 1998), potentially allowing use as a proxy of disease status. Speech acoustic changes have been studied as a novel way to diagnose PD and monitor symptoms. Several groups have applied signal processing and advanced analytic techniques such as machine learning to develop algorithms to predict disease status and severity (Ozkan, 2016). This technology has gained substantial enthusiasm in the innovation space, given the possibility of using mobile devices to monitor 40

Brain and Language 185 (2018) 38–46

K.M. Smith, D.N. Caplan

skill in PD, both in clinical neuropsychological testing and research. In this task, subjects are asked to generate lists of words based on a category cue. Letter-guided verbal fluency (name words beginning with “F”) is thought to depend more on frontal lobe function, while semantic category based verbal fluency (e.g. fruits and vegetables) is thought to rely more on temporal lobe function (Bak, O'Donovan, Xuereb, Boniface, & Hodges, 2001; Damasio & Tranel, 1993; Shapiro, Moo, & Caramazza, 2006). Alternating verbal fluency tasks (for instance, naming a word beginning with “F” and then a fruit or vegetable and alternating back and forth), involve set-shifting skills dependent on frontal executive function. Studies of verbal fluency in PD have shown mixed results, with some studies showing deficits in letter-guided fluency, semantic fluency, or both, while other have shown normal performance (Henry & Crawford, 2004). Some studies have shown more robust impairments in alternating verbal fluency, suggesting that the added challenge to executive function brings out this deficit (Henry & Crawford, 2004). Alternating verbal fluency impairment may therefore be related to syntax deficits in PD, since both support a link between language deficits and executive function. However, the verbal fluency literature has more inconsistencies than that for syntax, and it is possible that verbal fluency is a less sensitive marker of language impairment in PD, or that other confounders need to be examined.

Hochstadt, Nakano, Lieberman, & Friedman, 2006). In an fMRI study, subjects with PD had decreased activation of the basal ganglia and increased activation of structures involved in verbal and working memory compared with controls during comprehension of sentences of increasing syntactical difficulty. This activation pattern correlated with improved performance on the comprehension task, and was proposed to be compensatory in nature (Grossman et al., 2003). Overall, the etiology of language processing deficits is likely more related to cognitive than motor dysfunction, since language processing further deteriorates with increasing cognitive demand (Grossman et al., 2000). Furthermore, if the etiology was purely motor, than one would expect that language production, but not comprehension, would be impacted. This hypothesis is supported by Gross et al. in a study including PD subjects with and without dementia. Grammatical and working memory components of sentence processing were impaired only in the PD dementia group, and this correlated with atrophy of the prefrontal, premotor, dorsolateral prefrontal and right superior temporal cortex (Gross et al., 2012). 3.3.2.2. Language production deficits. Production of grammar and syntax in PD has been less well studied than processing and comprehension. Early studies did not find significant deficits in grammar production in PD compared to controls, but found significant deficits in other neurodegenerative diseases such as Alzheimer’s dementia and Huntington’s disease (Cummings, Darkins, Mendez, Hill, & Benson, 1988; Murray & Lenz, 2001). Ash et al. identified deficits in grammatical formation and syntactically complexity during narration of a wordless picture book in subjects with Dementia with Lewy bodies (DLB) and PD dementia (PDD), grouped together as Lewy body spectrum disorders (LBSD) (Ash et al., 2012). On evaluation of each group separately, it was noted that DLB, but not PDD subjects, demonstrated this deficit in grammaticality measures compared with non-demented PD subjects and controls. Based on this limited literature, it seems that production of grammar and syntax may be less consistently or less severely impaired than comprehension in PD. One possible explanation is that motor speech impairment leads to decreased speech output in general, thus making it more difficult to detect a difference from other disease populations. However, Ash et al. argues against this explanation since word count did not differ between LBSD and controls. More work is needed to assess different speech tasks with increased cognitive complexity. Additionally, there are no well-established rating scales or scores for grammatical complexity or “correctness,” and future development of quantitative, objective measures may help identify such deficits in PD. Other speech and language production deficits have been identified in PD. These deficits may have both motor and cognitive influences. For instance, speech rate is decreased in PD (Ash et al., 2012) potentially due to motor and cognitive input into the planning, initiation, and maintenance of output. A recent small study evaluated motor and cognitive influences on speech and language in subjects with various degrees of cognitive impairment in PD and DLB. Subjects were monitored over a mean of 3 years of follow up. Subjects with either PD with dementia or DLB had a significant decline in speech rate, utterance length, and informativeness of speech content over time, with significant impairment compared with non-demented PD subjects by the second assessment (Ash et al., 2017). However, there were no significant associations of these speech and language changes with either cognitive or motor features, perhaps due to limitations in the included assessments. Larger studies with comprehensive assessments of cognitive, motor speech control, and PD motor features, as well as neuroimaging data are needed to characterize how speech and language evolve over time and in relationship to other pathophysiological changes in PD and related diseases.

3.3.2.4. Action words. There has been a substantial amount of research in the use of words representing actions in PD, and studies have consistently found action word deficits in PD. A relationship between action words and physical movements may explain why individuals with PD have more difficulty producing and processing action words compared with non-action words. The proposed explanation for this action word deficit is that the same brain regions that are necessary for learning and executing actions are also necessary for storing the semantic meaning of the word for the action. More broadly, the theory of semantic, or embodied, cognition states that categories of content are represented in different brain regions depending on the sensory and motor processes involved in acquisition of the content (Meteyard, Cuadrado, Bahrami, & Vigliocco, 2012). Motor content would thus rely on motor planning and execution areas, such as primary motor and pre-motor cortex. Other content, for instance a noun for an object, would rely on perceptual and sensory processing areas such as infero-posterior temporal lobe. In this model, the action word deficit in PD would be related to impaired input from the striatum to the motor and associated cortex. Several studies suggest that action verb tasks are impaired in PD compared to controls (Fernandino et al., 2013a; Kemmerer, Miller, Macpherson, Huber, & Tranel, 2013; Salmazo-Silva et al., 2017). However, the results vary by experimental design, such as differences in methods used to assess action verbs, control tasks, as well as demographics of the PD subject groups. Studies evaluating naming of actions based on pictures have shown an impairment in PD compared with controls. This deficit worsened with the degree of motion-related semantic content associated with a verb (Herrera, Rodriguez-Ferreiro, & Cuetos, 2012) and improved with levodopa (Herrera & Cuetos, 2012). Peran et al. showed that action verb naming was associated with prefrontal and premotor cortex activation on fMRI, and argued that this suggests localization to the motor basal ganglia loop through the putamen (Peran et al., 2009). Cognitive function has also been studied as a potential contributor to action verb naming deficits in PD. In a study including PD subjects with and without MCI, the PD-MCI group had deficits in naming low and high motion content action verbs, as well as in naming objects with both low and high manipulability. The normal cognition group had an isolated deficit in high motion content verbs (Bocanegra et al., 2017). There was no significant correlation between naming performance and motor severity (UPDRS) in either group. Bocanegra and colleagues had previously shown that action verb naming and action semantic tasks did not depend on executive function integrity (Bocanegra et al. 2015) The authors propose that a semantic

3.3.2.3. Verbal fluency. Verbal fluency is a commonly tested language 41

Brain and Language 185 (2018) 38–46

K.M. Smith, D.N. Caplan

possible disease marker. Depending on context, pausing may represent difficulty with lexical retrieval or formulation of syntax (e.g. cognitivelinguistic), weakness or slowness of the vocal apparatus musculature (e.g. motor), changes in breath control and reserve (respiratory), or a combination thereof (Green, Beukelman, & Ball, 2004). Due to this multi-faceted nature, pausing may be particularly informative in a combined motor and cognitive disorder such as PD. Pauses also have an impact on speech intelligibility, since they convey punctuation and emphasis. Understanding pausing in PD may allow for therapeutic interventions to improve communication effectiveness. Individuals with PD pause more frequently during a variety of speech tasks. During a story narration task, non-demented PD subjects paused for a significantly longer proportion of the overall speech duration compared to controls. Subjects with PD dementia and DLB paused for longer between utterances than non-demented PD subjects (Ash et al., 2012). Pausing at the beginning of an utterance is thought to represent cognitive demand of utterance planning. Therefore, pausing between utterances appears to increase as cognition deteriorates in PD. However, the influence of motor speech control on pausing was not specifically controlled for in this work. We have found that pausing within utterances was increased in subjects with PD without dementia compared to controls (Smith, Ash, Xie, & Grossman, 2018). There was a correlation between pausing within utterances and global cognitive function, but this was no longer significant after controlling for motor severity. A small study found that the number of filled hesitations per minute separated PD subjects in the mild from moderate motor stage (Illes, Metter, Hanson, & Iritani, 1988), also suggesting that speech dysfluencies have a motor component. The effect of subthalamic deep brain stimulation (DBS) on pausing was examined in a single study. Subjects were tested on and off stimulation, and paused more frequently on stimulation compared to off stimulation, though the mean duration of pauses was shorter. Pausing occurred more often at nonlinguistic boundary locations in the on stimulation state, suggesting an executive planning or lexical retrieval problem. The impact of DBS on motor speech control vs. cognitive control was not clear from this work and requires further study. Another important area of needed research is the impact of dopaminergic medications, the mainstay of PD treatment, on pausing. Understanding the role of medical and surgical PD treatments in pausing may help clinicians manage patients with problematic communication symptoms to optimize fluency and intelligibility. This research would also elucidate the role of the dopamine system and target structures in the neurobiology of pausing. Pause location has been studied in only a handful of papers, but it may be useful in distinguishing different pausing etiologies. For instance, a pause before a noun (especially a low frequency noun) may represent lexical retrieval difficulty while a pause before an inflected verb or at a syntactic boundary may suggest cognitive-linguistic planning. Huber et al. evaluated breath pauses during a reading passage in young and older adult controls and in individuals with PD. In older compared with young adults, there was an increase in breath pauses at minor syntactic boundaries and with minor punctuation (i.e. commas). Since breath pauses remained closely related to punctuation and syntax, the authors attributed these findings to respiratory physiological changes of normal aging rather than cognitive changes. In the PD group, there were significantly more breath pauses at locations unrelated to syntax, suggesting that both cognitive and respiratory function contribute (Huber, Darling, Francis, & Zhang, 2012). A relationship between respiratory-related pausing and cognitive-linguistic demand has also been demonstrated in both PD and controls in other studies. Young adults paused more and spoke slower overall when discussing a topic without a prepared outline compared to with preparation (Mitchell, Hoit, & Watson, 1996). Both subjects with PD and controls spoke slower in extemporaneous speech than in reading (Huber et al., 2012). Respiratory measures were impacted by the type of speech task in both PD and controls, but the relationship patterns differed by disease state. Both PD and controls demonstrated higher lung volume

deficit specific to action verbs is present in PD independent of cognitive function, likely related to damage to basal ganglia and frontostriatal circuits. However, cognitive impairment causes additional deterioration in action word semantics that is less specific for degree of motion content and word type. Another possible explanation may be that despite the authors’ attempts to match the verb and noun categories for comparable complexity, the semantic complexity of high motion verbs may increase cognitive demand even in subjects without MCI. A major limitation of these studies is that action verb naming and verb fluency (e.g. list generation) are influenced by lexical retrieval and other executive functions, and do not independently reflect semantic memory (Piatt, Fields, Paolo, & Troster, 1999). These verb tasks may be impaired in PD simply because verbs are more semantically complex than nouns. Further research is needed to support a specific semantic deficit for action verbs and to identify an underlying neuroanatomical correlate. Some researchers have proposed similarity judgement (Fernandino et al., 2013a; Kemmerer et al., 2013) and association tasks (Bocanegra et al., 2015; Salmazo-Silva et al., 2017) as more appropriate assessments of action verb semantics. Fernandino et al. assessed action verbs compared with abstract verbs, in an attempt to control for the complexity and cognitive demand of verbs as a category. PD subjects performed worse on both primed lexical decision and similarity judgement tasks for action verbs compared with abstract verbs, while performance was similar for both verb types in controls (Fernandino et al., 2013a). Salmazo-Silva et al. determined that non-demented PD subjects were mildly impaired on the Kissing and Dancing Test, an assessment of action verb semantics. However, in this study action verbs were compared to nouns rather than non-action verbs. Boulenger et al. used a lexical decision task with masked priming, demonstrating reduced priming for action words compared with nouns in PD that improved with levodopa, suggesting dependency on motor circuits (Boulenger et al., 2008). A single study evaluated sentence processing by measuring response times for sentences including literal action, non-idiomatic metaphoric action, idiomatic action, and abstract verbs. PD subjects had longer reaction times for sentences involving action verbs both in literal and idiomatic contexts (Fernandino et al., 2013b). While this work suggests a specific deficit in PD for action verbs, more research is needed to evaluate action verb semantics in comparison with action nouns and non-action verbs. Additionally, larger studies including a wider range of PD subject demographics and clinical characteristics are needed. Demographics may contribute to the inconsistency in this literature, since years of education was found to be associated with action verb fluency (Salmazo-Silva et al., 2017). PD motor symptom severity also appears to play a role in various verbal fluency measures (Obeso, Casabona, Bringas, Alvarez, & Jahanshahi, 2012) and needs to be evaluated more thoroughly in relation to action word tasks. While the UPDRS is the standard measure of motor severity, it evaluates mostly limb movement speed on non-goal directed repetitive tasks, and thus should be supplemented with more detailed analysis of motor performance as well as motor speech metrics in order to more conclusively distinguish the contribution of motor system dysfunction to action word deficits. In summary, action verb processing and production may be sensitive markers of PD that could be utilized for detection and monitoring of disease. Further study of action verb deficits could lead to better understanding of the impact of PD on prefrontal and premotor circuits, and improved therapies to normalize these pathways. Preliminary evidence suggests that action verb deficits are relatively unaffected by cognitive status, and respond to levodopa. Action verb deficits could thus be contrasted with grammar and syntax deficits (i.e. markers of cognitive function), to develop symptom-specific linguistic markers for improved disease monitoring. 3.3.3. Pausing Pausing has been studied in PD and other neurological disorders as a 42

Brain and Language 185 (2018) 38–46

K.M. Smith, D.N. Caplan

logopenic, and semantic. This work has relevance to PD because nfvPPA is related to motor speech control impairment (considered an apraxia of speech), whereas logopenic and semantic PPA are thought to be cognitive disorders. Mack et al. found patients with logopenic PPA paused more before nouns compared to verbs, which distinguishes this group from other variants (Mack et al., 2015). Ballard et al. found that a global measure of silent periods did not distinguish between PPA types, but vowel duration differentiated nfvPPA with good accuracy and correlated with grey matter density on MRI in regions important for motor speech control (Ballard et al., 2014). Similar findings were recently reported by Corella et al. who found that pause measures do not distinguish PPA types, but that articulation rate has better diagnostic accuracy for nfvPPA than a clinical rating scale, suggesting articulation rate is a specific marker of motor speech impairment (Cordella, Dickerson, Quimby, Yunusova, & Green, 2017). Finally, Yunusova et al. studied pause measures in patients with amyotrophic lateral sclerosis (ALS), behavioral variant fronto-temporal dementia or nfvPPA. Total pause time and number of pauses was higher in all groups compared to controls, and pause measures did not distinguish between motor and cognitive disorders (Yunusova et al., 2016). Although this body of work seems to suggest that pausing is not successful in disentangling motor speech control from cognitive-linguistic impairment, it should be noted that pausing location and context was not evaluated in the majority of works. In the PPA literature in particular, pausing was measured using automatic pause detectors that were unable to assess location of pauses. Additional research is therefore warranted to examine influences on pauses in greater detail, and evaluate for more sensitive and distinguishing pause characteristics for various neurodegenerative disease.

initiations and terminations in extemporaneous speech compared with reading, and there was a correlation with utterance length (syllables/ breath). The PD group also displayed increased inspiratory duration in extemporaneous speech. The authors hypothesized that individuals with PD have difficulty planning and coordinating language and respiratory support simultaneously, particularly in more cognitively demanding types of speech (Huber et al., 2012). Pausing therefore appears to have cognitive, motor, and possibly respiratory influences in PD. The balance of these contributing factors may vary with patient characteristics (e.g. age, mild vs. severe motor symptoms), context and complexity of the speech task, and medical or surgical interventions. The neurobiological mechanism underlying pausing in PD may involve frontal and prefrontal cortex. In subjects with PD dementia and DLB, between-utterance pause time was associated with decreased grey matter volume in several regions on volumetric brain MRI. Atrophy of the right medial frontal (frontopolar) region (BA 10) and left ventrolateral prefrontal cortex (BA 47) were significantly associated with between-utterance pause time, speech rate, executive function, and proportion of grammatically well-formed sentences (Ash et al., 2012). These areas are thought to be important for working memory and for making associations between events in narrative discourse (Gilbert et al., 2006; Troiani et al., 2008). In the unstructured task of creating an autobiographical narrative, the frontopolar area may regulate switching between inner life and the outside world while telling a story (Burgess, Dumontheil, & Gilbert, 2007). Inferior frontal lobe dysfunction may therefore also underlie difficulties with cohesiveness of narrative discourse observed in PD (Ellis, Crosson, Gonzalez Rothi, Okun, & Rosenbek, 2015). However, this narrative discourse impairment was seen in cognitively normal PD subjects who would not be expected to have significant frontal lobe atrophy (Ellis et al., 2015). Furthermore there is no clear causal relationship between frontostriatal dysfunction and atrophy of the frontal lobe grey matter. The exact mechanism for the frontal dysfunction linked with pausing therefore remains unclear. One potential hypothesis would invoke frontostriatal dysfunction caused by dopaminergic deficiency that precedes frontal lobe structural changes. However, further research is needed to explain why certain frontal regions and their associated functions are preferentially affected among other potential frontal lobe functions that may relate to language. In addition, atrophy of BA 10 has also been identified in association with between-utterance pauses in Alzheimer’s disease (Pistono et al., 2016), suggesting that pathology other than hypodopaminergic frontostriatal dysfunction can cause this same manifestation. More work is needed to better understand the neurobiology of pausing across neurodegenerative disease, and to identify PD-specific pausing measures that can be used as markers of specific pathophysiology. We will now briefly review research on pausing in other neurodegenerative diseases. As mentioned above, subjects with DLB demonstrated similar pausing patterns to those with PD dementia (Ash et al., 2012). This finding fits with the clinical and pathologic overlap between these two conditions. In Alzheimer’s disease (AD), increased pausing has been reported in various speech tasks from picture description to autobiographical discourse. Lexical-semantic deficits likely underlie pausing in AD, but involvement of memory (Ullman, 2004) and executive function (Cannizzaro & Coelho, 2013) has also been explored. During autobiographical discourse, AD-MCI patients paused for longer between utterances, but not within utterances (neither grammatical nor non-grammatical locations). Between-utterance pauses correlated with memory performance on neuropsychological tests. As mentioned above, neuroimaging results were consistent with those in Lewy spectrum disorders, showing a correlation between BA 10 (frontopolar area) grey matter density and between-utterance pauses (Pistono et al., 2016). Primary Progressive Aphasia (PPA) is also associated with increased pause rates (Mack et al., 2015). Researchers have attempted to use pause-related and other linguistic characteristics to help distinguish between the three PPA variants: non-fluent agrammatic (nfvPPA),

3.3.4. Non-verbal communication impairment In addition to the language impairments we have described, individuals with PD may also have subtle deficits in processing and comprehension of non-verbal and emotional aspects of communication. Hearing loss and auditory processing deficits have been observed in PD. Hearing loss may impair conversational participation, or cause individuals to miss registering information that they are later expected to remember. One study found PD patients have more severe aging-related hearing changes compared with age-matched controls (Vitale et al., 2012). Impairment in sensorimotor processing has also been described, impacting the use of auditory feedback to adjust one’s own speech and language. As a result, individuals with PD may have difficulty adjusting their speech output for appropriate prosody (Albuquerque et al., 2016), pitch (Mollaei, Shiller, Baum, & Gracco, 2016), and rhythm (Spath et al., 2016). Since such real-time adjustments are integral to effective communication, these deficits could make the speech of individuals with PD seem less empathetic or cause others to diminish its importance. Mood is another area often impacted by PD that may play a role in language and communication. Individuals with PD may be incorrectly perceived as depressed due to reduced facial expression and a soft, monotone voice. Vice versa, depression is common in PD, but may go undetected if psychomotor manifestations of mood are misinterpreted as motor symptoms. It is unknown how depressive symptoms combine with motor features to impact speech and language in PD. It has been suggested that depression in the general population has a specific vocal acoustic signature (Trevino, Quatieri, & Malyska, 2011), but depression has not been considered in the work on vocal acoustics or speech and language in PD. Individuals with PD have difficulty differentiating between emotional or affective tones in speech, for instance sarcastic vs. serious. They are also impaired in recognizing emotional faces and gestures in the visual form (Gray & Tickle-Degnen, 2010). These difficulties can have a major impact on communication with loved ones and caregivers. More work must be done to evaluate the impact of emotion and mood-related deficits on speech and language, and to discern how the impact may be similar or different from other symptoms impacting speech in PD (e.g. cognitive impairment), as well as from depression in 43

Brain and Language 185 (2018) 38–46

K.M. Smith, D.N. Caplan

Statement of significance to the neurobiology of language

the general population.

This review addresses the neural mechanisms underlying speech and language changes in Parkinson’s disease (PD). We review studies using neuroimaging and theoretical knowledge of neuroanatomy to understand these changes. We explain how speech and language dysfunction in PD may tell us about the mechanism of such changes in other disorders.

4. Conclusions Communication disorders in PD are complex and multifaceted, with dramatic heterogeneity between patients and evolution over the course of the disease. Future research must take into account both the patient perspective, including symptoms such as word finding difficulty and conversational limitations, as well as quantitative and objective data. Clinical scales and questionnaires are being used with increasing frequency in clinical settings and research, but need to be further validated by objective measures. However, it remains unclear exactly which objective measures accurately represent communication impairment. The deficits observed in speech and language in PD include speech acoustic changes, production and processing of grammar and syntax, verbal fluency, action word use, and pauses. Much work remains to be done to elucidate the mechanisms involved in these various deficits, in order to better define them as quantitative markers. With improved characterization of speech and language deficits, it may be possible to develop markers specific for motor and non-motor symptoms of PD. Speech acoustic deficits appear to correlate with motor impairment, including motor speech systems, but further work should explore potential cognitive influences as well. Linguistic deficits may relate more to cognitive than motor impairment. For instance, syntactical processing depends on verbal and working memory, and may serve as a marker of cognitive function. Since research has not consistently identified syntax production or comprehension deficits in PD subjects without dementia, more research is needed to determine if these deficits occur early enough to make them useful markers. Subjects with milder forms of cognitive impairment should be further assessed on tasks with increasing cognitive demand and tasks more relevant to daily communication. Another possible linguistic biomarker specific to PD is action verb use. There is a relatively large body of consistent literature demonstrating a specific impairment in action verb use in PD. This appears to relate to motor more than cognitive dysfunction, caused by striatal dopaminergic deficit. Finally, pausing is increased in the speech of individuals with PD, and is likely impacted by both motor and cognitive systems. Pausing is also increased in other motor and cognitive neurodegenerative diseases, and more work is needed to identify disease-specific pausing patterns. More detailed study of different patterns of pausing location may lead to development of PD-specific pause measures and also help distinguish the motor and cognitive impact on pausing in PD. In order for speech and language research in PD to be more clinically relevant, these potential speech and language markers must be further studied in the context of functional communication. It will also be important to consider the involvement of motor, cognitive, respiratory and affective systems. Assessment of speech and language has the potential to provide a wealth of information about a patient’s clinical status in each of these domains. However, controlled research and clinic-based measures are likely insensitive to early changes and to the variability of these speech and language deficits in daily communication. Mobile technology may be a powerful tool to advance this field of study by providing frequent, remote, and objective assessments of speech and language. As this research advances, the communication difficulties experienced by individuals with PD could be utilized to better understand the disease, and to develop novel diagnostic and therapeutic tools for future patients experiencing these disabling symptoms.

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.bandl.2018.08.002. References Albuquerque, L., Martins, M., Coelho, M., Guedes, L., Ferreira, J. J., Rosa, M., & Martins, I. P. (2016). Advanced Parkinson disease patients have impairment in prosody processing. Journal of Clinical and Experimental Neuropsychology, 38(2), 208–216. Ash, S., Jester, C., York, C., Kofman, O. L., Langey, R., Halpin, A., ... Grossman, M. (2017). Longitudinal decline in speech production in Parkinson's disease spectrum disorders. Brain and Language, 171, 42–51. Ash, S., McMillan, C., Gross, R. G., Cook, P., Gunawardena, D., Morgan, B., ... Grossman, M. (2012). Impairments of speech fluency in Lewy body spectrum disorder. Brain and Language, 120(3), 290–302. Auclair-Ouellet, N., Lieberman, P., & Monchi, O. (2017). Contribution of language studies to the understanding of cognitive impairment and its progression over time in Parkinson's disease. Neuroscience & Biobehavior Reviews, 80, 657–672. Bak, T. H., O'Donovan, D. G., Xuereb, J. H., Boniface, S., & Hodges, J. R. (2001). Selective impairment of verb processing associated with pathological changes in Brodmann areas 44 and 45 in the motor neurone disease-dementia-aphasia syndrome. Brain, 124(Pt 1), 103–120. Ballard, K. J., Savage, S., Leyton, C. E., Vogel, A. P., Hornberger, M., & Hodges, J. R. (2014). Logopenic and nonfluent variants of primary progressive aphasia are differentiated by acoustic measures of speech production. PLoS One, 9(2), e89864. Baylor, C., Yorkston, K., Eadie, T., Kim, J., Chung, H., & Amtmann, D. (2013). The Communicative Participation Item Bank (CPIB): Item bank calibration and development of a disorder-generic short form. Journal of Speech Language and Hearing Research, 56(4), 1190–1208. Bocanegra, Y., Garcia, A. M., Lopera, F., Pineda, D., Baena, A., Ospina, P., ... Cuetos, F. (2017). Unspeakable motion: Selective action-verb impairments in Parkinson's disease patients without mild cognitive impairment. Brain and Language, 168, 37–46. Bocanegra, Y., Garcia, A. M., Pineda, D., Buritica, O., Villegas, A., Lopera, F., ... Ibanez, A. (2015). Syntax, action verbs, action semantics, and object semantics in Parkinson's disease: Dissociability, progression, and executive influences. Cortex, 69, 237–254. Boulenger, V., Mechtouff, L., Thobois, 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(2), 743–756. Brennan, L., Siderowf, A., Rubright, J. D., Rick, J., Dahodwala, N., Duda, J. E., ... Weintraub, D. (2016b). The Penn Parkinson's Daily Activities Questionnaire-15: Psychometric properties of a brief assessment of cognitive instrumental activities of daily living in Parkinson's disease. Parkinsonism and Related Disorders, 25, 21–26. Brennan, L., Siderowf, A., Rubright, J. D., Rick, J., Dahodwala, N., Duda, J. E., ... Weintraub, D. (2016a). Development and initial testing of the Penn Parkinson's Daily Activities Questionnaire. Movement Disorders, 31(1), 126–134. Burgess, P. W., Dumontheil, I., & Gilbert, S. J. (2007). The gateway hypothesis of rostral prefrontal cortex (area 10) function. Trends in Cognitive Science, 11(7), 290–298. Cannizzaro, M. S., & Coelho, C. A. (2013). Analysis of narrative discourse structure as an ecologically relevant measure of executive function in adults. Journal of Psycholinguistics Research, 42(6), 527–549. Canter, G. J. (1965). Speech characteristics of patients with Parkinson's disease. Ii. Physiological support for speech. Journal of Speech and Hearing Disorders, 30, 44–49. Colman, K. S., Koerts, J., Stowe, L. A., Leenders, K. L., & Bastiaanse, R. (2011). Sentence comprehension and its association with executive functions in patients with Parkinson's disease. Parkinsons Disease, 2011, 213983. Cordella, C., Dickerson, B. C., Quimby, M., Yunusova, Y., & Green, J. R. (2017). Slowed articulation rate is a sensitive diagnostic marker for identifying non-fluent primary progressive aphasia. Aphasiology, 31(2), 241–260. 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 alterations. Neurology, 38(5), 680–684. Damasio, A. R., & Tranel, D. (1993). Nouns and verbs are retrieved with differently distributed neural systems. Proceedings of the National Academy of Sciences USA, 90(11), 4957–4960. De Letter, M., Van Borsel, J., Boon, P., De Bodt, M., Dhooge, I., & Santens, P. (2010). Sequential changes in motor speech across a levodopa cycle in advanced Parkinson's disease. International Journal of Speech and Language Pathology, 12(5), 405–413. Duffy, J. (2012). Motor speech disorders: Substrates, differential diagnosis, and management (3rd ed.). St. Louis, MO: Elsevier Mosby. 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 Parkinsons

Funding sources This research did not receive any specific grant form funding agencies in the public, commercial, or not-for-profit sectors. 44

Brain and Language 185 (2018) 38–46

K.M. Smith, D.N. Caplan

17(4), 266–274. Miller, N., Allcock, L., Jones, D., Noble, E., Hildreth, A. J., & Burn, D. J. (2007). Prevalence and pattern of perceived intelligibility changes in Parkinson's disease. Journal of Neurology Neurosurgery and Psychiatry, 78(11), 1188–1190. Miller, N., Deane, K. H., Jones, D., Noble, E., & Gibb, C. (2011). National survey of speech and language therapy provision for people with Parkinson's disease in the United Kingdom: Therapists' practices. International Journal of Language and Communication Disorders, 46(2), 189–201. Miller, N., Noble, E., Jones, D., Allcock, L., & Burn, D. J. (2008). How do I sound to me? Perceived changes in communication in Parkinson's disease. Clinical Rehabilitation, 22(1), 14–22. Mitchell, H. L., Hoit, J. D., & Watson, P. J. (1996). Cognitive-linguistic demands and speech breathing. Journal of Speech and Hearing Research, 39(1), 93–104. Mollaei, F., Shiller, D. M., Baum, S. R., & Gracco, V. L. (2016). Sensorimotor control of vocal pitch and formant frequencies in Parkinson's disease. Brain Research, 1646, 269–277. Murray, L. L., & Lenz, L. P. (2001). Productive syntax abilities in Huntington's and Parkinson's diseases. Brain and Cognition, 46(1–2), 213–219. Obeso, I., Casabona, E., Bringas, M. L., Alvarez, L., & Jahanshahi, M. (2012). Semantic and phonemic verbal fluency in Parkinson's disease: Influence of clinical and demographic variables. Behavioral Neurology, 25(2), 111–118. Oh-Lee, J. D., Szymkowicz, S. M., Smith, S. L., & Otani, H. (2012). Metacognitive performance, the tip-of-tongue experience, is not disrupted in parkinsonian patients. Parkinsons Disease, 2012, 174079. Orozco-Arroyave, J. R., Belalcazar-Bolanos, E. A., Arias-Londono, J. D., Vargas-Bonilla, J. F., Skodda, S., Rusz, J., ... Noth, E. (2015). Characterization methods for the detection of multiple voice disorders: Neurological, functional, and laryngeal diseases. IEEE Journal of Biomedical and Health Informatics, 19(6), 1820–1828. Orozco-Arroyave, J. R., Honig, F., Arias-Londono, J. D., Vargas-Bonilla, J. F., Daqrouq, K., Skodda, S., ... Noth, E. (2016). Automatic detection of Parkinson's disease in running speech spoken in three different languages. Journal of the Acoustics Society of America, 139(1), 481–500. Ozkan, H. (2016). A comparison of classification methods for telediagnosis of Parkinson's disease. Entropy, 18(Apr), 14. Peran, P., Cardebat, D., Cherubini, A., Piras, F., Luccichenti, G., Peppe, A., ... Sabatini, U. (2009). Object naming and action-verb generation in Parkinson's disease: A fMRI study. Cortex, 45(8), 960–971. Piatt, A. L., Fields, J. A., Paolo, A. M., & Troster, A. I. (1999). Action (verb naming) fluency as an executive function measure: Convergent and divergent evidence of validity. Neuropsychologia, 37(13), 1499–1503. Pigott, K., Rick, J., Xie, S. X., Hurtig, H., Chen-Plotkin, A., Duda, J. E., ... Weintraub, D. (2015). Longitudinal study of normal cognition in Parkinson disease. Neurology, 85(15), 1276–1282. Pistono, A., Jucla, M., Barbeau, E. J., Saint-Aubert, L., Lemesle, B., Calvet, B., ... Pariente, J. (2016). Pauses during autobiographical discourse reflect episodic memory processes in early Alzheimer's disease. Journal of Alzheimers Disease, 50(3), 687–698. Plowman-Prine, E. K., Okun, M. S., Sapienza, C. M., Shrivastav, R., Fernandez, H. H., Foote, K. D., ... Rosenbek, J. C. (2009). Perceptual characteristics of Parkinsonian speech: A comparison of the pharmacological effects of levodopa across speech and non-speech motor systems. NeuroRehabilitation, 24(2), 131–144. Postuma, R. B., Aarsland, D., Barone, P., Burn, D. J., Hawkes, C. H., Oertel, W., & Ziemssen, T. (2012). Identifying prodromal Parkinson's disease: Pre-motor disorders in Parkinson's disease. Movement Disorders, 27(5), 617–626. Salmazo-Silva, H., Parente, M. A., Rocha, M. S., Baradel, R. R., Cravo, A. M., Sato, J. R., ... Carthery-Goulart, M. T. (2017). Lexical-retrieval and semantic memory in Parkinson's disease: The question of noun and verb dissociation. Brain and Langage, 165, 10–20. Sapir, S., Ramig, L. O., & Fox, C. M. (2011). Intensive voice treatment in Parkinson's disease: Lee Silverman voice treatment. Expert Review of Neurotherapeutics, 11(6), 815–830. Schalling, E., Johansson, K., & Hartelius, L. (2017). Speech and communication changes reported by people with Parkinson's disease. Folia Phoniatrica et Logopaedica, 69(3), 131–141. Shapiro, K. A., Moo, L. R., & Caramazza, A. (2006). Cortical signatures of noun and verb production. Proceedings of the National Academy of Sciences USA, 103(5), 1644–1649. Simard, F., Joanette, Y., Petrides, M., Jubault, T., Madjar, C., & Monchi, O. (2011). Fronto-striatal contribution to lexical set-shifting. Cerebral Cortex, 21(5), 1084–1093. Simard, F., Monetta, L., Nagano-Saito, A., & Monchi, O. (2013). A new lexical card-sorting task for studying fronto-striatal contribution to processing language rules. Brain and Language, 125(3), 295–306. Smith, K. M., Ash, S., Xie, S. X., & Grossman, M. (2018). Evaluation of Linguistic Markers of Word-Finding Difficulty and Cognition in Parkinson's Disease. Journal of Speech, Hearing and Language Research, 61(7), 1691–1699. Spath, M., Aichert, I., Ceballos-Baumann, A. O., Wagner-Sonntag, E., Miller, N., & Ziegler, W. (2016). Entraining with another person's speech rhythm: Evidence from healthy speakers and individuals with Parkinson's disease. Clinical Linguistics and Phonetics, 30(1), 68–85. Terzi, A., Papapetropoulos, S., & Kouvelas, E. D. (2005). Past tense formation and comprehension of passive sentences in Parkinson's disease: Evidence from Greek. Brain and Language, 94(3), 297–303. Trevino, A. C., Quatieri, T. F., & Malyska, N. (2011). Phonologically-based biomarkers for major depressive disorder. EURASIP Journal on Advances in Signal Processing, 1, 1–18. Troiani, V., Fernandez-Seara, M. A., Wang, Z., Detre, J. A., Ash, S., & Grossman, M. (2008). Narrative speech production: An fMRI study using continuous arterial spin labeling. Neuroimage, 40(2), 932–939. Tsanas, A., Little, M. A., McSharry, P. E., & Ramig, L. O. (2011). Nonlinear speech analysis algorithms mapped to a standard metric achieve clinically useful quantification of

Disease, 5(2), 403–411. Fernandino, L., Conant, L. L., Binder, J. R., Blindauer, K., Hiner, B., Spangler, K., & Desai, R. H. (2013b). Where is the action? Action sentence processing in Parkinson's disease. Neuropsychologia, 51(8), 1510–1517. Fernandino, L., Conant, L. L., Binder, J. R., Blindauer, K., Hiner, B., Spangler, K., & Desai, R. H. (2013a). Parkinson's disease disrupts both automatic and controlled processing of action verbs. Brain and Language, 127(1), 65–74. Gilbert, S. J., Spengler, S., Simons, J. S., Steele, J. D., Lawrie, S. M., Frith, C. D., & Burgess, P. W. (2006). Functional specialization within rostral prefrontal cortex (area 10): A meta-analysis. Journal of Cognitive Neuroscience, 18(6), 932–948. Gray, H. M., & Tickle-Degnen, L. (2010). A meta-analysis of performance on emotion recognition tasks in Parkinson's disease. Neuropsychology, 24(2), 176–191. Green, J. R., Beukelman, D. R., & Ball, L. J. (2004). Algorithmic estimation of pauses in extended speech samples of dysarthric and typical speech. Journal of Medical SpeechLanguage Pathology, 12(4), 149–154. Griffiths, S., Barnes, R., Britten, N., & Wilkinson, R. (2011). Investigating interactional competencies in Parkinson's disease: The potential benefits of a conversation analytic approach. International Journal of Language and Communication Disorders, 46(5), 497–509. Gross, R. G., McMillan, C. T., Chandrasekaran, K., Dreyfuss, M., Ash, S., Avants, B., ... Grossman, M. (2012). Sentence processing in Lewy body spectrum disorder: The role of working memory. Brain and Cognition, 78(2), 85–93. Grossman, M., Carvell, S., Gollomp, S., Stern, M. B., Reivich, M., Morrison, D., ... Hurtig, H. I. (1993). Cognitive and physiological substrates of impaired sentence processing in Parkinson's disease. Journal of Cognitive Neuroscience, 5(4), 480–498. Grossman, M., Cooke, A., DeVita, C., Lee, C., Alsop, D., Detre, J., ... Hurtig, H. I. (2003). Grammatical and resource components of sentence processing in Parkinson's disease: An fMRI study. Neurology, 60(5), 775–781. Grossman, M., Crino, P., Reivich, M., Stern, M. B., & Hurtig, H. I. (1992). Attention and sentence processing deficits in Parkinson's disease: The role of anterior cingulate cortex. Cerebral Cortex, 2(6), 513–525. Grossman, M., Gross, R. G., Moore, P., Dreyfuss, M., McMillan, C. T., Cook, P. A., ... Siderowf, A. (2012). Difficulty processing temporary syntactic ambiguities in Lewy body spectrum disorder. Brain and Language, 120(1), 52–60. Grossman, M., Kalmanson, J., Bernhardt, N., Morris, J., Stern, M. B., & Hurtig, H. I. (2000). Cognitive resource limitations during sentence comprehension in Parkinson's disease. Brain and Language, 73(1), 1–16. Grossman, M., Lee, C., Morris, J., Stern, M. B., & Hurtig, H. I. (2002). Assessing resource demands during sentence processing in Parkinson's disease. Brain and Language, 80(3), 603–616. Hely, M. A., Reid, W. G., Adena, M. A., Halliday, G. M., & Morris, J. G. (2008). The Sydney multicenter study of Parkinson's disease: The inevitability of dementia at 20 years. Movement Disorders, 23(6), 837–844. Henry, J. D., & Crawford, J. R. (2004). Verbal fluency deficits in Parkinson's disease: A meta-analysis. Journal of the International Neuropsychology Society, 10(4), 608–622. Herrera, E., & Cuetos, F. (2012). Action naming in Parkinson's disease patients on/off dopamine. Neuroscience Letters, 513(2), 219–222. Herrera, E., Rodriguez-Ferreiro, J., & Cuetos, F. (2012). The effect of motion content in action naming by Parkinson's disease patients. Cortex, 48(7), 900–904. Ho, A. K., Bradshaw, J. L., & Iansek, R. (2008). For better or worse: The effect of levodopa on speech in Parkinson's disease. Movement Disorders, 23(4), 574–580. Ho, A. K., Iansek, R., Marigliani, C., Bradshaw, J. L., & Gates, S. (1998). Speech impairment in a large sample of patients with Parkinson's disease. Behavioral Neurology, 11(3), 131–137. Hochstadt, J. (2009). Set-shifting and the on-line processing of relative clauses in Parkinson's disease: Results from a novel eye-tracking method. Cortex, 45(8), 991–1011. Hochstadt, J., Nakano, H., Lieberman, P., & Friedman, J. (2006). The roles of sequencing and verbal working memory in sentence comprehension deficits in Parkinson's disease. Brain and Language, 97(3), 243–257. Huber, J. E., Darling, M., Francis, E. J., & Zhang, D. (2012). Impact of typical aging and Parkinson's disease on the relationship among breath pausing, syntax, and punctuation. American Journal of Speech-Language Pathology, 21(4), 368–379. Illes, J., Metter, E. J., Hanson, W. R., & Iritani, S. (1988). Language production in Parkinson's disease: Acoustic and linguistic considerations. Brain and Language, 33(1), 146–160. Jenkinson, C., Fitzpatrick, R., Peto, V., Greenhall, R., & Hyman, N. (1997). The Parkinson's Disease Questionnaire (PDQ-39): Development and validation of a Parkinson's disease summary index score. Age and Ageing, 26(5), 353–357. Kemmerer, D. (1999). Impaired comprehension of raising-to-subject constructions in Parkinson's disease. Brain and Language, 66(3), 311–328. Kemmerer, D., Miller, L., Macpherson, M. K., Huber, J., & Tranel, D. (2013). An investigation of semantic similarity judgments about action and non-action verbs in Parkinson's disease: Implications for the Embodied Cognition Framework. Frontiers in Human Neuroscience, 7, 146. Lee, C., Grossman, M., Morris, J., Stern, M. B., & Hurtig, H. I. (2003). Attentional resource and processing speed limitations during sentence processing in Parkinson's disease. Brain and Language, 85(3), 347–356. Mack, J. E., Chandler, S. D., Meltzer-Asscher, A., Rogalski, E., Weintraub, S., Mesulam, M. M., & Thompson, C. K. (2015). What do pauses in narrative production reveal about the nature of word retrieval deficits in PPA? Neuropsychologia, 77, 211–222. Matison, R., Mayeux, R., Rosen, J., & Fahn, S. (1982). “Tip-of-the-tongue” phenomenon in Parkinson disease. Neurology, 32(5), 567–570. Meteyard, L., Cuadrado, S. R., Bahrami, B., & Vigliocco, G. (2012). Coming of age: A review of embodiment and the neuroscience of semantics. Cortex, 48(7), 788–804. Miller, N. (2017). Communication changes in Parkinson's disease. Practical Neurology,

45

Brain and Language 185 (2018) 38–46

K.M. Smith, D.N. Caplan

Workshop: Depression Challenge, Orlando, FL. Yarnall, A. J., Breen, D. P., Duncan, G. W., Khoo, T. K., Coleman, S. Y., Firbank, M. J., Nombela, C., Winder-Rhodes, S., Evans, J. R., Rowe, J. B., Mollenhauer, B., Kruse, N., Hudson, G., Chinnery, P. F., O'Brien, J. T., Robbins, T. W., Wesnes, K., Brooks, D. J., Barker, R. A., Burn, D. J., & Group, I. P. S. (2014). Characterizing mild cognitive impairment in incident Parkinson disease: The ICICLE-PD study. Neurology, 82(4), 308–316. Yu, R. L., Wu, R. M., Tai, C. H., Lin, C. H., & Hua, M. S. (2010). Feeling-of-knowing in episodic memory in patients with Parkinson's disease with various motor symptoms. Movement Disorders, 25(8), 1034–1039. Yunusova, Y., Graham, N. L., Shellikeri, S., Phuong, K., Kulkarni, M., Rochon, E., ... Green, J. R. (2016). Profiling speech and pausing in amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). PLoS One, 11(1), e0147573.

average Parkinson's disease symptom severity. Journal of the Royal Society Interface, 8(59), 842–855. Ullman, M. T. (2004). Contributions of memory circuits to language: The declarative/ procedural model. Cognition, 92(1–2), 231–270. Valalik, I., Smehak, G., Bognar, L., & Csokay, A. (2011). Voice acoustic changes during bilateral subthalamic stimulation in patients with Parkinson's disease. Clinical Neurology and Neurosurgery, 113(3), 188–195. Vitale, C., Marcelli, V., Allocca, R., Santangelo, G., Riccardi, P., Erro, R., ... Barone, P. (2012). Hearing impairment in Parkinson's disease: Expanding the nonmotor phenotype. Movement Disorders, 27(12), 1530–1535. Williamson, J. R., Quatieri, T. F., Helfer, B. S., Ciccarelli, G., & Mehta, D. D. (2014). Vocal and Facial Biomarkers of Depression Based on Motor Incoordination and Timing. In: Paper presented at the 4th International Audio/Visual Emotion Challenge and

46