The gradual deterioration of syntax and semantics in a patient with progressive aphasia

How to constrain and maintain a lexicon for the treatment of progressive semantic naming deficits: Principles of item selection for formal semantic therapy

The progressive degradation of semantic memory is a common feature of many forms of dementia, including Alzheimer's disease and the semantic variant of primary progressive aphasia (svPPA). One of the most functionally debilitating effects of this semantic impairment is the inability to name common people and objects (i.e., anomia). Clinical management of a progressive, semantically based anomia presents extraordinary challenges for neurorehabilitation. Techniques such as errorless learning and spaced-retrieval training show promise for retraining forgotten words. However, we lack complementary detail about what to train (i.e., item selection) and how to flexibly adapt the training to a declining cognitive system. This position paper weighs the relative merits of several treatment rationales (e.g., restore vs. compensate) and advocates for maintenance of known words over reacquisition of forgotten knowledge in the context of semantic treatment paradigms. I propose a system for generating an item pool and outline a set of core principles for training and sustaining a micro-lexicon consisting of approximately 100 words. These principles are informed by lessons learned over the course of a Phase I treatment study targeting language maintenance over a 5-year span in Alzheimer's disease and SvPPA. Finally, I propose a semantic training approach that capitalises on lexical frequency and repeated training on conceptual structure to offset the loss of key vocabulary as disease severity worsens.

Assessment of explicit and implicit linguistic impairments in patients with aphasia after resection of tumour of the left cerebral hemisphere. Preliminary results

BACKGROUND AND PURPOSE

Classical definitions of aphasia des-cribe deficits of different language levels (syntactic, semantic, phonologic) hindering the ability to communicate. Recent studies indicate, however, that impairment of particular aspects of linguistic competencies in aphasia differs in severity. Contemporary approach to the aphasic symptoms presents them as disturbed access of linguistic representations to the awareness system. Accordingly, such an approach requires different types of tasks: direct, involving explicit language processes, and indirect, based on implicit language representations. The aim of our study was to examine explicit and implicit language processes in patients with aphasia after resection of the tumour of left cerebral hemisphere along with characterization of relationships between explicit and implicit language processes.

MATERIAL AND METHODS

Our cohort included 28 right-handed patients who were divided into four equal groups: two clinical (brain tumours) and two control (lumbar disc disease). Four tasks that assess and compare language processes: lexical decisions (at explicit and implicit levels), sorting of picture captions and word monitoring were implemented.

RESULTS

In direct tasks, patients with aphasia provided less correct lexical decisions at word level, but did not show deficits in sentence comprehension. In both groups, no priming effect was observed in tasks requiring implicit lexical decisions. The longest time was found in non-primed words, the shortest in pseudowords. The differences between groups regarding word monitoring were also observed. Patients with aphasia obtained longer reaction times in all types of sentences (of different grade of language correctness), with respect to low- and high- frequency words.

CONCLUSIONS

Patients with aphasia after brain tumour resection show more pronounced impairments of explicit than implicit linguistic behavior; the same effect was found in studies on forgetting in amnestic syndrome.

The neural basis of syntactic deficits in primary progressive aphasia

Patients with primary progressive aphasia (PPA) vary considerably in terms of which brain regions are impacted, as well as in the extent to which syntactic processing is impaired. Here we review the literature on the neural basis of syntactic deficits in PPA. Structural and functional imaging studies have most consistently associated syntactic deficits with damage to left inferior frontal cortex. Posterior perisylvian regions have been implicated in some studies. Damage to the superior longitudinal fasciculus, including its arcuate component, has been linked with syntactic deficits, even after gray matter atrophy is taken into account. These findings suggest that syntactic processing depends on left frontal and posterior perisylvian regions, as well as intact connectivity between them. In contrast, anterior temporal regions, and the ventral tracts that link frontal and temporal language regions, appear to be less important for syntax, since they are damaged in many PPA patients with spared syntactic processing.

The relation between content and structure in language production: an analysis of speech errors in semantic dementia

In order to explore the impact of a degraded semantic system on the structure of language production, we analysed transcripts from autobiographical memory interviews to identify naturally-occurring speech errors by eight patients with semantic dementia (SD) and eight age-matched normal speakers. Relative to controls, patients were significantly more likely to (a) substitute and omit open class words, (b) substitute (but not omit) closed class words, (c) substitute incorrect complex morphological forms and (d) produce semantically and/or syntactically anomalous sentences. Phonological errors were scarce in both groups. The study confirms previous evidence of SD patients' problems with open class content words which are replaced by higher frequency, less specific terms. It presents the first evidence that SD patients have problems with closed class items and make syntactic as well as semantic speech errors, although these grammatical abnormalities are mostly subtle rather than gross. The results can be explained by the semantic deficit which disrupts the representation of a pre-verbal message, lexical retrieval and the early stages of grammatical encoding.

Lexical retrieval and semantic knowledge in patients with left inferior temporal lobe lesions

BACKGROUND: It has been proposed that anomia following left inferior temporal lobe lesions may have two different underlying mechanisms with distinct neural substrates. Specifically, naming impairment following damage to more posterior regions (BA 37) has been considered to result from a disconnection between preserved semantic knowledge and phonological word forms (pure anomia), whereas anomia following damage to anterior temporal regions (BAs 38, 20/21) has been attributed to the degradation of semantic representations (semantic anomia). However, the integrity of semantic knowledge in patients with pure anomia has not been demonstrated convincingly, nor were lesions in these cases necessarily confined to BA 37. Furthermore, evidence of semantic anomia often comes from individuals with bilateral temporal lobe damage, so it is unclear whether unilateral temporal lobe lesions are sufficient to produce significant semantic impairment. AIMS: The main goals of this study were to determine whether anomia following unilateral left inferior temporal lobe damage reflected a loss of semantic knowledge or a post-semantic deficit in lexical retrieval and to identify the neuroanatomical correlates of the naming impairment. METHODS #ENTITYSTARTX00026; PROCEDURES: Eight individuals who underwent left anterior temporal lobectomy (L ATL) and eight individuals who sustained left posterior cerebral artery strokes (L PCA) completed a battery of language measures that assessed lexical retrieval and semantic processing, and 16 age- and education-matched controls also completed this battery. High-resolution structural brain scans were collected to conduct lesion analyses. OUTCOMES #ENTITYSTARTX00026; RESULTS: Performance of L ATL and L PCA patients was strikingly similar, with both groups demonstrating naming performance ranging from moderately impaired to unimpaired. Anomia in both groups occurred in the context of mild deficits to semantic knowledge, which manifested primarily as greater difficulty in naming living things than nonliving things and greater difficulty in processing visual/perceptual as opposed to functional/associative semantic attributes. Lesion analyses indicated that both patient groups sustained damage to anterior inferior temporal lobe regions implicated in semantic processing. CONCLUSIONS: These results contribute to a better understanding of the cognitive mechanism of naming impairment in patients with temporal lobe damage and support the notion that pure anomia and semantic anomia represent two endpoints along a continuum of semantic impairment. Unilateral left temporal lobe lesions in our patients resulted in relatively mild semantic deficits that were apparent primarily in lexical production tasks, whereas severe semantic impairment likely requires bilateral temporal lobe damage.

Syntactic and thematic components of sentence processing in progressive nonfluent aphasia and nonaphasic frontotemporal dementia

We used an online word-monitoring paradigm to examine sentence processing in healthy seniors and frontotemporal dementia patients with progressive nonfluent aphasia (PNFA) or a nonaphasic disorder of social and executive functioning (SOC/EXEC). Healthy seniors were sensitive to morphosyntactic, major grammatical subcategory, and selection restriction violations in a sentence. PNFA patients were insensitive to grammatical errors, but showed reasonable sensitivity to thematic matrix violations, consistent with a differential grammatical processing impairment. By contrast, SOC/EXEC patients showed partial sensitivity to grammatical errors but were insensitive to thematic violations. These findings support a dissociation between grammatical and thematic components of sentence processing. Specifically, they are consistent with a grammatical processing deficit in PNFA patients, and impairment in the formation of a coherent thematic matrix in SOC/EXEC patients.

Progressive anomia without semantic or phonological impairment

We describe a 59-year-old woman, M.T., with a progressive language impairment and neuroimaging findings of decreased perfusion (SPECT) and focal atrophy (MRI) in the left temporal region. The most prominent feature of her cognitive profile was a profound and progressive impairment in naming. In spite of this, she performed normally on tests of semantic processing and phonological output. Her spontaneous speech was fluent with preserved syntax and articulation but with notable word-finding problems. All other cognitive abilities were relatively stable and intact. These features are not typical of either fluent or non-fluent forms of neurodegenerative language disturbance. The cognitive mechanisms that may underlie this case are discussed.

'Non-semantic' aspects of language in semantic dementia: as normal as they're said to be?

This article describes a 52-month longitudinal study of a patient, ILJ, whose semantic profile fits the criteria for a classical case of semantic dementia (SD). As in all such cases, it was the semantic aspects of receptive and expressive language processing--essentially vocabulary--that were most dramatically affected. The novel observation from this study is ILJ's performance on a comprehensive language examination. Results from this assessment, even early in the course of his disease, are compatible with the hypothesis that phonological, morphological and/or syntactic aspects of language processing may be disrupted by their interaction with degraded information from the semantic system.

Sentence processing in frontotemporal dementia

Patients with frontotemporal dementia (FTD) have sentence comprehension difficulty. We examined the hypothesis that both grammatical and resource factors contribute to their impaired sentence comprehension with a traditional, off-line sentence comprehension task, and an on-line sentence processing procedure that minimizes task-related resources. This was investigated in subgroups of FTD patients with Progressive Non-fluent Aphasia (PNFA; n = 5) who have effortful speech; non-aphasic patients with an executive deficit (EXEC; n = 8); and Semantic Dementia (SD; n = 3) patients with poor single word comprehension. The results were correlated with measures of executive resources. We found that PNFA patients are significantly impaired in their off-line sentence comprehension, and that their performance correlated with auditory-verbal short-term memory. PNFA patients also demonstrated a pattern of slowed processing for the on-line sentence measure. This is consistent with the hypothesis that information relevant for constructing sentence representations during comprehension degrades in working memory as it is activated over an abnormally slowed time course. EXEC patients had modest off-line sentence comprehension difficulty, and this correlated with performance on measures of working memory, planning, and inhibitory control. On-line processing in EXEC patients demonstrated their insensitivity to sentence-based information. This raises the possibility that a limitation in material-neutral executive resources not dedicated to grammatical processing may play a role in their sentence comprehension deficit. SD patients' pattern of on-line sentence comprehension paralleled control subjects' performance. We conclude that grammatical and executive components both contribute to sentence comprehension, and that the profile of sentence comprehension difficulty varies across FTD subgroups depending on the sentence processing component that is impaired.

A longitudinal study of sentence comprehension difficulty in primary progressive aphasia

CONTEXT

Patients with primary progressive aphasia have sentence comprehension difficulty, but the longitudinal course of this deficit has not been investigated.

OBJECTIVE

To determine how grammatical, single word meaning, and working memory factors contribute to longitudinal decline of sentence comprehension in primary progressive aphasia. We hypothesised partially distinct patterns of sentence comprehension difficulty in subgroups of patients with progressive non-fluent aphasia (PNFA) and semantic dementia (SD).

DESIGN

Cohort.

SETTING

Institutional out patient referral centre.

PATIENTS

PNFA (n = 14), SD (n = 10).

MAIN OUTCOME MEASURE

Sentence comprehension accuracy.

RESULTS

PNFA patients were significantly impaired at understanding grammatically complex sentences when first seen, and this was more evident than impairment of their comprehension of grammatically simple sentences (p<0.05). Comprehension of grammatically complex sentences correlated with their working memory deficit at presentation (p<0.05). PNFA patients showed modest decline over time in grammatical comprehension. In SD, comprehension of grammatically complex sentences was not more impaired than comprehension of grammatically simple sentences when first seen, but these patients demonstrated a significant longitudinal decline in understanding grammatically complex sentences (p<0.05). Cox regression analyses showed that a deficit in single word meaning contributes to the progressive impairment for grammatically complex sentences in SD (p<0.05), but working memory does not contribute to longitudinal decline in PNFA.

CONCLUSION

Patients with PNFA and SD have sentence comprehension difficulty, but distinct factors contribute to this impairment during the course of their disease.

Paradoxical features of word finding difficulty in primary progressive aphasia

Impaired word retrieval is a main symptom of primary progressive aphasia (PPA). The cognitive features of this impairment in PPA are poorly understood. We studied 12 patients with PPA (6 English-speaking and 6 Dutch-speaking), 7 patients with early-stage clinically probable Alzheimer's disease (PRAD), 5 patients with mild cognitive impairment (MCI), and 15 age-matched, cognitively intact, control subjects. Subjects had to name a picture (the probe), which was preceded by a written word (the prime) that could be the correct name of the picture, a noun belonging to the same semantic subcategory (related prime), a semantically unrelated noun (unrelated prime), or a pseudoword (neutral control). Naming latencies were longer in PPA and PRAD patients than in control subjects. Critically, the interaction between group and prime type was highly significant. PPA patients named the probe more slowly after a related compared with an unrelated prime. In contrast, PRAD patients, mild cognitive impairment patients, and healthy control subjects tended to name the probe faster when it was preceded by a related prime. The semantic interference effect in PPA generalized across languages and PPA subtypes. Selection among competing word forms sharing a same semantic field is abnormal in PPA. The semantic interference effect constitutes a positive distinguishing feature between PPA and PRAD.

Deficits for Semantics and the Irregular Past Tense: A Causal Relationship?

The regular and irregular past tense has become a focus for recent debates about the structure of the language processing system, asking whether language functions are subserved by different neural and functional mechanisms or whether all processes can be accommodated within a single unified system. A critical claim of leading single mechanism accounts is that the relationship between an irregular stem and its past tense form is primarily semantic and not morphological in nature. This predicts an obligatory relationship between semantic performance and access to the irregular past tense, such that a semantic deficit necessarily leads to impairments on the irregulars. We tested this claim in a series of studies probing the comprehension and production of regular and irregular past tense forms in four semantic dementia patients, all of whom had profound semantic deficits. In two elicitation tasks and one auditory priming study, we found that three out of the four patients did not have a deficit for the irregular past tense, in spite of their semantic deficits. This argues against the view that the relationship between irregular past tense forms and their stems is primarily semantic, and more generally against the single system claim that morphological structure can be captured solely based on phonological and semantic relationships.

Deficits in thematic integration processes in Broca's and Wernicke's aphasia

This study investigated how normal subjects and Broca's and Wernicke's aphasics integrate thematic information incrementally using syntax, lexical-semantics, and pragmatics in a simple active declarative sentence. Three priming experiments were conducted using an auditory lexical decision task in which subjects made a lexical decision on a 'target' (the last word of each sentence) preceded by a 'prime' (a subject noun phrase and verb). The presence and magnitude of priming was compared to a baseline condition in which non-words systematically replaced real word primes. Normal subjects showed evidence for combinatorial thematics by exhibiting significant and larger amounts of thematic priming in the condition where two real words were present in the prime than in the conditions in which only one real word was present in the prime. Additionally, normal subjects showed sensitivity to both syntactic structures and pragmatics. In contrast, Broca's patients did not show significant priming for any condition nor did they show a difference in the magnitude of priming among the conditions. Nonetheless, they showed sensitivity to pragmatics. Wernicke's patients showed significant priming for all conditions, but did not show differences in the magnitude of priming among the conditions. However, they showed sensitivity to sentence grammaticality and pragmatics. The distinct patterns of performance of Broca's and Wernicke's aphasics are discussed in terms of the nature of their impairments in the processes of combinatorial thematics.

No right to speak? The relationship between object naming and semantic impairment: neuropsychological evidence and a computational model

The processes required for object naming were addressed in a study of patients with semantic dementia (a selective decline of semantic memory resulting from progressive temporal lobe atrophy) and in a computational model of single-word production. Although all patients with semantic dementia are impaired in both single-word production and comprehension, previous reports had indicated two different patterns: (a) a parallel decline in accuracy of naming and comprehension, with frequent semantic naming errors, suggesting a purely semantic basis for the anomia and (b) a dramatic progressive anomia without commensurate decline in comprehension, which might suggest a mainly postsemantic source of the anomia. Longitudinal data for 16 patients with semantic dementia reflected these two profiles, but with the following additional important specifications: (1) despite a few relatively extreme versions of one or other profile, the full set of cases formed a continuum in the extent of anomia for a given degree of degraded comprehension; (2) the degree of disparity between these two abilities was associated with relative asymmetry in laterality of atrophy: a parallel decline in the two measures characterized patients with greater right- than left-temporal atrophy, while disproportionate anomia occurred with a predominance of atrophy in the left-temporal lobe. In an implemented computational model of naming, semantic representations were distributed across simulated left- and right-temporal regions, but the semantic units on the left were more strongly connected to left-lateralized phonological representations. Asymmetric damage to semantic units reproduced the longitudinal patient profiles of naming relative to comprehension, plus additional characteristics of the patients' naming performance. On the basis of both the neuropsychological and computational evidence, we propose that semantic impairment alone can account for the full range of word production deficits described here.

The rise and fall of frequency and imageability: noun and verb production in semantic dementia

This study examines the impact of progressive degeneration of conceptual knowledge on the content words used in connected speech elicited using the Cookie Theft picture description (Goodglass & Kaplan. 1983). We began with an analysis of control subjects' descriptions with regard to word types and their frequency and imageability. Because the impairment of conceptual knowledge in semantic dementia is graded by concept familiarity, we created a model of a standardized normal Cookie Theft description that was then progressively degraded by the systematic removal of lower bands of word frequency. We drew two main predictions from this model: reduced availability of the lower bands of word frequency should result in (a) an apparent deficit for noun retrieval in relation to verb retrieval and (b) an apparent reverse imageability effect. Results from a longitudinal study. in which three patients with semantic dementia each described the Cookie Theft picture on three occasions during the progression of their disease, confirmed these predictions. An additional cross-sectional analysis, adding narratives from a larger number of cases, demonstrated that the decline in ability to produce suitable words for the picture description is closely related to the extent of semantic impairment as measured in tests of word comprehension and production. Both verbs and nouns are affected by the degradation of semantic memory; the fact that the impairment to noun production is manifested earlier and more catastrophically may be attributed to the relatively lower frequency of these terms.

A progressive category-specific semantic deficit for non-living things

We report a longitudinal study of a patient, ES, with a progressive degenerative disorder resulting from generalised cerebral atrophy. Across a range of tasks, ES showed a greater difficulty in recognising and naming artifacts than living things. This deficit for artifacts emerged over time, as she became more severely impaired. In one task, picture naming, there was a crossover from an initial deficit for living things to the later artifact deficit. All materials were carefully controlled to rule out potential confounding factors such as concept familiarity or age of acquisition. There was no evidence that ES's deficit for artifacts was associated with a greater loss of functional than visual information. The pattern of results are consistent with a recently proposed distributed connectionist model, in which a deficit for artifact concepts can emerge as the result of severe, general damage to semantic memory.

Dual-process model in semantic priming: A functional imaging perspective

In this positron emission tomography study, we investigated the neural correlates of semantic priming, where response to a word is facilitated when preceded by a semantically related word. Nine normal subjects were scanned while performing a lexical decision task. Within this condition, the proportion of related prime-target word pairs was varied across scans from 0 to 100%. The control task involved letter decision on consonant letter strings, controlling orthographic processing and response selection. First, lexical decision (relative to letter decision) activated regions previously observed in lexicosemantic tasks, i.e., the left anterior and inferior temporal lobe and left inferior frontal gyrus. Behavioral analysis confirmed significant facilitation of lexical decision to related targets (mean priming effect 68 ms). It also suggested the contribution of both automatic and strategic processes, consistent with theoretical accounts of priming. Automatic priming was indicated by consistent RTs to related targets irrespective of the proportion of related word pairs per scan. Strategic processing was indicated by decreases in RTs to nonwords as the proportion of related targets increased. Nonlinear correlational analysis of cerebral activity during lexical decision revealed a neurophysiological correlate of these behavioral effects in (i) the left anterior temporal lobe (BA 38), a region involved in lexicosemantic processing; (ii) the anterior cingulate cortex, right premotor region (BA 6), and right superior parietal lobe (BA 7), regions associated with attentional processes. We conclude that in this experimental context, semantic priming involves both automatic and strategic processing.

Episodic memory: new insights from the study of semantic dementia

The study of patients with semantic dementia, the temporal variant of frontotemporal dementia, has provided new insights into the interaction between episodic and semantic memory, and the different roles played by various structures in the temporal lobe. Recent findings indicate that the syndrome of semantic dementia can inform us about the organisation of long-term memory and the relationship between semantic memory and other cognitive systems.

Non-Alzheimer degenerative dementias

Recent progress in diagnostic criteria of non-Alzheimer degenerative dementias is reviewed. These dementias comprise frontotemporal dementias (including hereditary dementias), primary progressive aphasia and anarthria, corticobasal degeneration, progressive supranuclear palsy and dementia with Lewy bodies. The approach of studying these diseases has changed considerably with genetic and biochemical analyses. A molecular classification is suggested and the clinical significance of this classification is discussed.

A reversal of the temporal gradient for famous person knowledge in semantic dementia: implications for the neural organisation of long-term memory

On tests of autobiographical memory, patients with semantic dementia demonstrate significantly better retrieval of episodic events from the recent past compared with the distant past. This reversal of the Ribot effect has been attributed to the relative sparing of the hippocampal complex in the disorder. Current computational models of long-term memory predict a similar time-dependent pattern of impairment on tests of remote semantic memory. Five patients with semantic dementia were tested on recognition (familiarity) and identification (knowledge) of famous names selected from four different time-periods: 1950's, 1980's, 1990-1993 (early 1990's) and 19941996 (current). As expected, it was found that one patient DM (who had focal left temporal lobe atrophy) showed no significant impairment on recognition of famous names, but was significantly better at producing information about people who were currently famous compared to people famous in the other three time-periods. The other four patients (who had bilateral temporal lobe damage) showed better recognition of famous names from the current time-period (and to a lesser extent the 1950's), yet were profoundly impaired on the identification component, producing very little information across all four time-periods. The results are discussed with respect to current views of the neural organisation of person-specific and general semantic memory.

Linguistic dissociations in Williams syndrome: evaluating receptive syntax in on-line and off-line tasks

Williams syndrome (WS) is a neurodevelopmental disorder of genetic origin which results in relatively spared language in the face of serious non-verbal deficits. There is controversy, however, about how intact WS language abilities are. The discussion has focused on impairments of lexico-semantics and of morphological feature analysis, with the presumption that WS syntax is intact. We challenged this view and assessed WS receptive syntax by using two tasks testing various syntactic structures: an on-line word monitoring task and an off-line picture-pointing task. WS performance on the off-line task was generally poor. By contrast, their performance on the on-line task was far better and allowed us to ascertain precisely which aspects of WS receptive syntax are preserved and which are impaired. WS participants were sensitive to the violation of auxiliary markers and phrase structure rules but, unlike both the normal young and elderly controls, they did not show sensitivity to violations of subcategory constraints. The present study suggests that there exist dissociations within WS language which are not restricted to lexico-semantics or to morphological feature analysis, but which also invade their processing of certain syntactic structures. We conclude by arguing that WS syntax is not intact and that their language might turn out to be more like second language learning than normal acquisition.