Measures of lexical diversity in aphasia
Abstract
Background: Important to the assessment of aphasia are analyses of discourse production and, in particular, lexical diversity analyses of verbal production of adults with aphasia. Previous researchers have used type-token ratio (TTR) to measure conversational vocabulary in adults with aphasia; however, this measure is known to be sensitive to sample size, requiring that only samples of equivalent length be compared. The number of different words (NDW) is another measure of lexical diversity, but it also requires input samples of equivalent length. An alternative to these measures, D, has been developed (Malvern & Richards, 1997) to address this problem. D allows for comparisons across samples of varying lengths. Aims: The first objective of the current study was to examine the relationships among three measures of productive vocabulary in discourse for adults with aphasia: TTR, NDW, and D. The second objective was to use these measures to determine in what ways, and to what degree, they each can differentiate fluent and nonfluent aphasia. Methods & Procedures: Eighteen adults with aphasia participated in this study (nine with nonfluent aphasia; nine with fluent aphasia). Participants completed the Western Aphasia Battery (WAB) and produced language samples consisting of conversation and picture description. Samples were then subjected to the three lexical diversity analyses. Outcomes & Results: Results indicated that, although the measures generally correlated with each other, adults with fluent aphasia evidenced significantly higher D and NDW values than those with nonfluent aphasia when whole samples were subjected to analyses. Once samples were truncated to 100- and 200-word samples, groups differed significantly for all three measures. Conclusions: These findings add further support to the notion that because TTR and, although to a lesser extent, NDW are sensitive to sample size, length differences across samples tend to confound results. As an alternative to these measures, the use of D for the measurement of conversational vocabulary of adults with aphasia enables the analysis of entire language samples, so that discarding language sample data is not necessary. In the present study, D values differed for fluent and nonfluent aphasia samples.
... We additionally contrasted the group of PWA with a neurotypical control group in order to ascertain whether the relationships between descriptive discourse and cognitive skills might differ between groups. In order to capture the multifaceted pattern of discourse we focused on different measures (Alyahya et al., 2020): (a) lexical quantity and diversity, such as type, token and type-token ratio of lexical items (Wright et al., 2003), (b) utterance length and complexity (c) quality and informativeness of the discourse, which included percentage of different kinds of paraphasias in PWA, as well as percentage of correct information units (Nicholas & Brookshire, 1993;Pritchard et al., 2017) and Main Concept Analysis (Nicholas & Brookshire, 1995), and (d) fluency, quantified as words per minute. ...
... a) Token: the total number of words used by each participant was measured as a quantitative index of word production. b) Type: the number of different words used c) Type-token ratio (TTR) (Wright et al., 2003): It is defined as the relationship between the total number of different words and the total number of words in the sample and was considered a measure of lexical diversity ...
Aims:
To study the relationship between cognitive and linguistic skills (as measured through standardized tasks) over spontaneous speech elicited during a picture description task.
Methods & procedures:
21 controls and 19 people with fluent aphasia matched by age and sex were evaluated using transcripts made from a picture description task coded using the CHAT format and analyzed using Computerized Language Analysis (CLAN). Indices obtained from the speech samples contained measures of lexical quantity and diversity, morphosyntactic complexity, informativeness, and speech fluency, along with different kinds of speech errors. We studied their correlations with attentional measures from Conners' Continuous Performance Test and with standardized measures of naming, pseudoword repetition and semantic non-verbal association. We further used stepwise linear regression to analyze the predictive value of standardized linguistic and cognitive skills over discursive indices.
Outcomes & results:
Contrary to our initial hypothesis, there were no significant correlations between attentional scores and discourse variables in aphasic participants. Moreover, semantic association, along with naming, was the measure more related with discourse performance in people with fluent aphasia, but cognitive and linguistic standardized measures had overall little predictive power on most discourse indices. In the control group, there was a certain association of naming skills and attentional reaction time with discourse variables, but their predictive power was also low.
Conclusions & implications:
The current results do not support a strong relationship between basic attentional skills and performance in descriptive discourse in fluent aphasia. Although some of the standardized tasks seem to bear some relationship with spontaneous speech, there is a high amount of interindividual variability in discourse that is not captured by classical cognitive tasks routinely used in assessment. Further work on the determinants of discourse performance in aphasia and on the clinical application of discourse analysis is warranted.
... To determine the language ability of the person with dementia, we included measures for lexical diversity, propositionality of an utterance and the complexity of an utterance. For determining the lexical diversity, the Type Token Ratio (TTR) was used, provided that a minimum of at least 300 word was uttered (Wright et al., 2003). We note that in two cases, this minimum requirement was not achieved, see Results below. ...
... Additionally, interpreting and scoring a part of the conversation as a breakdown without participating in that conversation is subjective and complex, for example not resolving a misunderstanding can be even pragmatically seen a beneficial for the flow of the Lexical diversity as measured by the type token ratio is a good index of the language ability of the person with dementia. We would recommend to keep this measure, with the note that there should be enough words included in the calculation (Wright et al., 2003). The differentiation between propositional and non-propositional utterances is also recommended to keep as a measure for language ability of the person with dementia, but we note that utterance boundaries should have a better and clearer definition, as this turned out to be problematic in the comparisons between the two raters. ...
Background and Objective: This study explored the feasibility and usefulness of a set of observer rated outcome measures for the joint verbal functional communication of people with dementia and their communication partners, in combination with a set of quantitative measures for the language ability of the person with dementia. We hypothesized that the joint verbal functional communication would at least remain stable, despite an expected progressive deterioration of language ability of the person with dementia.Research Design and Methods: This was an exploratory study with audio and video recordings of 13 dyadic conversations before, directly after, and three and six months after a communication intervention. Four dyads of a home-dwelling person with dementia and their primary informal caregiver were included. Proportion of speaking time as well as occurrence and repair of communicative breakdowns were included as measures for joint verbal functional communication. Lexical diversity, propositionality and complexity of utterances were included as measures for language ability. Results: We found evidence that lexical diversity of the persons with dementia decreased over time. By contrast, there was no evidence that the proportion of speaking time by the persons with dementia changed over time. Discussion and Implications: Our combination of measures has primarily been proven feasible and useful for assessing joint verbal functional communication in persons with dementia and their communication partner, and seems to have potential for measuring the impact of a communication intervention. We recommend expanding our measures and investigating them in a larger sample.
... MATTR first computes the TTR for a section of selected length (e.g., 100 words) and then re-computes it in successive windows of N + 1 (e.g., Word 2 to Word 101) and then averages the results. This measure has been used to evaluate the very short samples typically generated by adults with aphasia (Fergadiotis et al., 2013;Harris Wright et al., 2003) and one large-scale study of children ages 4-9 years (Charest et al., 2020) that used a standard elicitation procedure (Edmonton Narrative Norms Instrument) as well as a study of bilingual children with language impairment (Kapantzoglou et al., 2017). In that study, MATTR (using a window of 50 words) was the best distinguishing measure of lexical diversity, accounting for 98% of the variance, whereas VocD accounted for 73% and TTR much less at 32% of the variance. ...
Purpose
Prior work has identified weaknesses in commonly used indices of lexical diversity in spoken language samples, such as type–token ratio (TTR) due to sample size and elicitation variation, we explored whether TTR and other diversity measures, such as number of different words/100 (NDW), vocabulary diversity (VocD), and the moving average TTR would be more sensitive to child age and clinical status (typically developing [TD] or developmental language disorder [DLD]) if samples were obtained from standardized prompts.
Method
We utilized archival data from the norming samples of the Test of Narrative Language and the Edmonton Narrative Norms Instrument. We examined lexical diversity and other linguistic properties of the samples, from a total of 1,048 children, ages 4–11 years; 798 of these were considered TD, whereas 250 were categorized as having a language learning disorder.
Results
TTR was the least sensitive to child age or diagnostic group, with good potential to misidentify children with DLD as TD and TD children as having DLD. Growth slopes of NDW were shallow and not very sensitive to diagnostic grouping. The strongest performing measure was VocD. Mean length of utterance, TNW, and verbs/utterance did show both good growth trajectories and ability to distinguish between clinical and typical samples.
Conclusions
This study, the largest and best controlled to date, re-affirms that TTR should not be used in clinical decision making with children. A second popular measure, NDW, is not measurably stronger in terms of its psychometric properties. Because the most sensitive measure of lexical diversity, VocD, is unlikely to gain popularity because of reliance on computer-assisted analysis, we suggest alternatives for the appraisal of children's expressive vocabulary skill.
... Like VocD, MATTR must be computed using a computer algorithm. This measure has been used to evaluate the very short samples typically generated by adults with aphasia (Harris Wright et al., 2003;Fergadiotis et al., 2013) and one large-scale study of children ages 4-9 years (Charest et al., 2020) that used a standard elicitation procedure (Edmonton Narrative Norms Instrument) as well as a study of bilingual children with language impairment (Kapantzoglou et al., 2017). In that study, MATTR (using a window of 50 words) was the best distinguishing measure of lexical diversity, accounting for 98% of the variance, while VocD accounted for 73% and TTR much less at 32% of the variance. ...
Background
Type-Token Ratio (TTR), given its relatively simple hand computation, is one of the few LSA measures calculated by clinicians in everyday practice. However, it has significant well-documented shortcomings; these include instability as a function of sample size, and absence of clear developmental profiles over early childhood. A variety of alternative measures of lexical diversity have been proposed; some, such as Number of Different Words/100 (NDW) can also be computed by hand. However, others, such as Vocabulary Diversity (VocD) and the Moving Average Type Token Ratio (MATTR) rely on complex resampling algorithms that cannot be conducted by hand. To date, no large-scale study of all four measures has evaluated how well any capture typical developmental trends over early childhood, or whether any reliably distinguish typical from atypical profiles of expressive child language ability.
Materials and Methods
We conducted linear and non-linear regression analyses for TTR, NDW, VocD, and MATTR scores for samples taken from 946 corpora from typically developing preschool children (ages 2–6 years), engaged in adult-child toy play, from the Child Language Data Exchange System (CHILDES). These were contrasted with 504 samples from children known to have delayed expressive language skills (total n = 1,454 samples). We also conducted a separate sub-analysis which examined possible contextual effects of sampling environment on lexical diversity.
Results
Only VocD showed significantly different mean scores between the typically -developing children and delayed developing children group. Using TTR would actually misdiagnose typical children and miss children with known language impairment. However, computation of VocD as a function of toy interactions was significant and emerges as a further caution in use of lexical diversity as a valid proxy index of children’s expressive vocabulary skill.
Discussion
This large scale statistical comparison of computer-implemented algorithms for expressive lexical profiles in young children with traditional, hand-calculated measures showed that only VocD met criteria for evidence-based use in LSA. However, VocD was impacted by sample elicitation context, suggesting that non-linguistic factors, such as engagement with elicitation props, contaminate estimates of spoken lexical skill in young children. Implications and suggested directions are discussed.
... Many measurements have been introduced and employed to capture learners' lexical complexity. Overall, previous scholars suggested MTLD (a measure of textual lexical diversity) and the vocd-D scale of learners' lexical diversity (e.g., Harris Wright et al., 2003;Suzuki & Kormos, 2020). Both measurements are derived from scaling lexical complexity. ...
It is crucial to design learning strategies for English as Foreign Language (EFL) learners in the context of proliferating and cultivating argumentative speaking. Argumentation mapping aims to provide a visual depiction of collaborative arguments in order to unravel students' engagement in constructing argumentative discussion. The employment of argumentation mapping is fundamental to maintaining students' virtual social practice. However, in the conventional argumentation mapping approach, the acquisition of argumentation skills does not yield successful argumentative dialogue-to-speaking practice. The collective reflection aims to promote a group of learners’ reflective thinking and higher order thinking by way of collaborative tasks and group monitoring. This strategy allows students to reflect on their group works when dealing with argumentative knowledge. Hence, this study referred to the educational theory of collective reflection practice and proposed its practice in the Collective Reflection-based Argumentation Mapping (CR-AM) learning strategy. This strategy was applied to an EFL speaking course to enhance students' argumentative speaking. Twenty-four students were recruited as the CR-AM strategy group, with 22 in the conventional Argumentation Mapping (AM) strategy group. The findings indicated that the CR-AM learning strategy could significantly improve students' argumentative speaking performance and their lexical complexity. Although students' critical thinking was at an equivalent level in both groups, students' communication and collaboration tendencies were much higher in the CR-AM strategy group. Based on the argumentation mapping discourse, it was verified that instructing students to extend their activities of classifying collective arguments and monitoring group reflection could effectively support their engagement in argumentation mapping discourse and enhance their argumentation skill performance.
Aphasia has had a profound influence on our understanding of how language is instantiated within the human brain. Historically, aphasia has yielded an in vivo model for elucidating the effects of impaired lexical-semantic access on language comprehension and production. Aphasiology has focused intensively on single word dissociations. Yet, less is known about the integrity of combinatorial semantic processes required to construct well-formed narratives. Here we addressed the question of how controlled lexical-semantic retrieval deficits (a hallmark of aphasia) might compound over the course of longer narratives. We specifically examined word-by-word flow of taxonomic vs. thematic semantic distance in the storytelling narratives of individuals with chronic post-stroke aphasia (n = 259) relative to age-matched controls (n = 203). We first parsed raw transcribed narratives into content words and computed inter-word semantic distances for every running pair of words in each narrative (N = 232,490 word transitions). The narratives of people with aphasia showed significant reductions in taxonomic and thematic semantic distance relative to controls. Both distance metrics were strongly predictive of offline measures of semantic impairment and aphasia severity. Since individuals with aphasia often exhibit perseverative language output (i.e., repetitions), we performed additional analyses with repetitions excluded. When repetitions were excluded, group differences in semantic distances persisted and thematic distance was still predictive of semantic impairment, although some findings changed. These results demonstrate the cumulative impact of deficits in controlled word retrieval over the course of narrative production. We discuss the nature of semantic flow between words as a novel metric of characterizing discourse and elucidating the nature of lexical-semantic access impairment in aphasia at multiword levels.
Statistical properties of language provide important cues for language learning and may be processed by domain-general cognitive systems. We investigated the relationship between implicit statistical learning (the unconscious detection of statistical regularities in input) and language production. Twenty typically developing (TD) children and nine children with acquired language disorders (ALD) (aged 6–18 years) took part in a Boston Cookie Theft picture description task. Using a computerized analysis, we investigated statistical properties, such as usage frequency of words and collocation strength of word combinations. Participants also completed a non-linguistic serial reaction time (SRT) task, which tested non-verbal, implicit statistical learning in the visual-motor modality. We determined age effects, and compared language production and SRT performance between both groups. Older TD children produced more connected language, more words, less frequent function words, more rare or novel combinations, and showed better statistical learning. Children with ALD produced less connected language, more weakly collocated combinations, displayed less lexical diversity and showed poorer statistical learning. Post-hoc analyses found correlations between statistical learning and statistical properties of spoken language. Given the rarity and heterogeneity of children with ALD, group size was small and the study should be considered exploratory. However, we note that results are compatible with the view that language production draws on statistical learning and that impairment of statistical learning can be related to language disorders.
The abstract for this document is available on CSA Illumina.To view the Abstract, click the Abstract button above the document title.
This study reports the reliability and concurrent validity of the information units (IU) metric as an efficient method for quantifying the amount of information comprehended and reproduced on the Story Retelling Procedure (SRP) (Doyle et al., 2000). Subjects were 31 normal adults and 15 adults with aphasia. Significant and moderately high correlation coefficients were obtained for subjects with aphasia between %IUs and most linguistic measures including the correct information unit (Nicholas & Brookshire 1993, 1995) while low and non-significant correlations were found for many measures of language productivity, efficiency, and disruption. The %IUs among the four SRP forms within group was non-significant (p > .05) and correlations were significant and high. Normal speakers produced significantly greater %IUs than aphasic speakers. Standard error of measurement was low across forms for both groups (3-4%) and the range of individual subjects' performance overlapped between 20 and 27%for the group with aphasia and between 36 and 55% for the normal group. These results support the conclusion that %IU is a reliable and valid measure and differentiates aphasic from normal individuals better than normal individuals from persons with aphasia.
Language research thrives on data collected from spontaneous interactions in naturally occurring situations. However, the process of collecting, transcribing, and analyzing naturalistic data can be extremely time-consuming and often unreliable. This book describes three basic tools for language analysis of transcript data by computer that have been developed in the context of the "Child Language Data Exchange System (CHILDES)" project. These are: the "CHAT" transcription and coding format, the "CLAN" package of analysis programs, and the "CHILDES" database. These tools have brought about significant changes in the way research is conducted in the child language field. They are being used with great success by researchers working with second language learning, adult conversational interactions, sociological content analyses, and language recovery in aphasia, as well as by students of child language development. The tools are widely applicable, although this book concentrates on their use in the child language field, believing that researchers from other areas can make the necessary analogies to their own topics. This thoroughly revised 2nd edition includes documentation on a dozen new computer programs that have been added to the basic system for transcript analysis. The most important of these new programs is the "CHILDES" Text Editor (CED) which can be used for a wide variety of purposes, including editing non-Roman orthographies, systematically adding codes to transcripts, checking the files for correct use of "CHAT," and linking the files to digitized audio and videotape. In addition to information on the new computer programs, the manual documents changed the shape of the "CHILDES/BIB" system--given a major update in 1994--which now uses a new computer database system. The documentation for the "CHILDES" transcript database has been updated to include new information on old corpora and information on more than a dozen new corpora from many different languages. Finally, the system of "CHAT" notations for file transcript have been clarified to emphasize the ways in which the codes are used by particular "CLAN" programs. The new edition concludes with a discussion of new directions in transcript analysis and links between the "CHILDES" database and other developments in multimedia computing and global networking. It also includes complete references organized by research topic area for the more than 300 published articles that have made use of the "CHILDES" database and/or the "CLAN" programs. LEA also distributes the "CLAN" programs and the complete "CHILDES" Database--including corpora from several languages and discourse situations--described in "The CHILDES Project." Be sure to choose the correct platform (IBM or Macintosh) for the "CLAN" programs; the "CHILDES" Database CD-ROM runs on both platforms.
The free speech of each of twelve adult aphasic patients was examined with reference particularly to (1) the distribution of words according to grammatical function, (2) sequential dependencies in form-class usage, and (3) stereotypy in vocabulary. The majority of the aphasic records departed considerably from normal usage (as defined by analysis of twelve control records), with similarity among some patients in the pattern of divergence. The measures used appear to be of particular value in revealing (i) semantic difficulties in word selection and (ii) difficulties in the sequencing of speech that occur along with syntactic losses.
A retrospective longitudinal study of 9 preschool-aged children, all meeting the criteria for specific language impairment (SLI), was conducted. Language growth was documented while the children were between the ages of 3 and 5 years and enrolled in a language intervention program. Three language measures were obtained across this period, including mean length of utterance (MLU) in morphemes, lexical diversity (i.e., number of different words used per 50 utterances), and a finite verb morphology composite (i.e., percent correct production of regular past -ed, present third person singular -s, and both the copula and auxiliary forms of is, are, and am). These longitudinal data were compared with cross-sectional data from 99 normally developing children who resided in the same community. Age-based comparisons revealed that, for most children with SLI, lexical diversity approached normal levels by the second year of data collection, whereas the production of finite verb morphology continued to be significantly delayed. A second set of comparisons, in which the children with SLI were compared with younger controls matched for MLU, further accentuated the persistence of difficulties with finite verb morphology. Applications of these language growth measures to assessment and intervention are discussed.
The present investigation is concerned with the relation of certain language variables to the length of sample from which they are derived and certain psychologically pertinent factors. The object of this study may be oriented around the analysis of the number of different words (D) as a function of the total number of words (N). As part of a remedial education survey, sponsored by the Iowa Child Welfare Research Station and financed by the Federal Work Projects Administration, approximately 1,000 public school children with schizophrenia wrote manuscripts of 3,000 words each under conditions to be specified below. It should be realized that the recording, tabulating, and counting operations in this study were unusually extensive. Three-thousand-word written language samples were obtained from 108 Iowa school children who had been selected to fill the cells of a factorial design which consisted of three levels each of I.Q., C.A., and locality (city, town, rural) as well as two equal groups of boys and girls. The subjects were asked to write about whatever they wanted to write about and in a free-writing situation for a short time each day until they had reached their quota of 3,000 words. Each language sample was edited and the words tabulated according to a set of predetermined rules. From these tabulations, 20 language measures were obtained for each sample, and individual cumulative type-frequency curves were computed for a selected group of subjects. (PsycINFO Database Record (c) 2012 APA, all rights reserved)