1. Abstract
The Verbs and Sentences Test (Dutch: Werkwoorden- en Zinnentest, abbreviated as WEZT) is a standardized neurolinguistic assessment battery developed by Roelien Bastiaanse, E. Maas, and J. Rispens (2000). Designed specifically for adult and geriatric clinical populations experiencing aphasia subsequent to cerebrovascular accidents (CVA) or traumatic brain injury (TBI), the instrument assesses language pathology at the isolated lexical verb and structured sentence levels. Departing from traditional classification-oriented batteries—such as the Aachen Aphasia Test (AAT) or the Boston Diagnostic Aphasia Examination (BDAE)—the primary objective of the WEZT is not to determine syndrome taxonomy or broad severity grades, but rather to delineate underlying functional locus breakdowns to guide targeted, deficit-oriented speech-language therapy (stoornisgerichte therapie).
The battery is structured across receptive and expressive modalities, encompassing subtests dedicated to single verb comprehension, action naming (verb production in isolation), sentence comprehension, sentence production, and the retrieval/inflection of finite versus non-finite verbs in varying syntactic contexts. Receptive subtests employ a four-alternative forced-choice picture paradigm featuring systematically controlled semantic distractors (including semantically related verbs, coordinate nouns, and semantically linked argument nouns). Psychometric investigations demonstrate high internal consistency across subtests (Cronbach’s α typically ranging from .82 to .94) and robust inter-rater and test-retest reliability coefficients (> .85). By isolating critical psycholinguistic variables—such as transitivity, argument structure complexity, and canonical versus non-canonical constituent order—the WEZT provides a clinically efficient yet theoretically sophisticated evaluation framework for modern cognitive aphasiology.
2. Keywords
Verbs and Sentences Test, Werkwoorden- en Zinnentest, WEZT, Aphasia, Neurolinguistics, Verb Processing, Syntactic Impairment, Agrammatism, Sentence Comprehension, Speech-Language Pathology
3. Authors
The original Dutch battery was conceptualized and standardized by prominent neurolinguists and aphasiologists based in the Netherlands:
- Roelien Bastiaanse, Ph.D. — Professor Emerita of Neurolinguistics at the University of Groningen, Center for Language and Cognition Groningen (CLCG), Groningen, the Netherlands; and affiliated researcher at the National Research University Higher School of Economics (HSE), Moscow. Bastiaanse is an internationally recognized authority on verb retrieval deficits, tense processing, and agrammatism in aphasia.
- E. Maas, M.Sc. — Neurolinguist and speech-language researcher formerly affiliated with the University of Groningen, specializing in clinical psycholinguistic diagnostics and sentence-level language rehabilitation.
- J. Rispens, Ph.D. — Associate Professor at the University of Amsterdam, Department of Dutch Linguistics and Child Development, whose work bridges developmental language disorders, psycholinguistics, and acquired neurological communication disorders.
4. Purpose
In standard aphasiological practice, broad screening instruments and classical classification batteries frequently group individuals into syndromic labels—such as Broca’s, Wernicke’s, or conduction aphasia. While these classic syndromic categories retain descriptive historical value, they possess limited explanatory power for tailoring specific cognitive-linguistic therapies. The Werkwoorden- en Zinnentest (WEZT) was developed specifically to overcome this diagnostic-treatment gap by providing an in-depth, psycholinguistically grounded evaluation of verb and sentence processing for individuals with all types and severities of aphasia.
Verbs represent the structural core of sentences: they dictate the argument structure, assign thematic roles (e.g., Agent, Theme, Goal), and establish grammatical relations. Consequently, deficits in verb processing profoundly compromise spontaneous communication, functional discourse, and expressive sentence construction. The clinical purpose of the WEZT is threefold:
- Identifying the Locus of Impairment: It systematically isolates whether a communicative breakdown originates within the semantic system, the phonological output lexicon (lemma vs. word-form retrieval), the syntactic mapping mechanism (thematic role assignment), or the morphological realization of verb inflection and tense.
- Directing Impairment-Specific Therapy: In alignment with evidence-based speech-language pathology in the Netherlands and internationally, the WEZT assesses exclusively those linguistic variables that have been empirically demonstrated to be amenable to therapeutic remediation. Clinicians can identify whether a client requires single-word action naming training, mapping therapy, or structural syntactic elaboration.
- Monitoring Rehabilitation Trajectories: Its modular design allows clinicians to administer targeted subtests longitudinally to evaluate specific functional gains resulting from structured interventions.
5. Psychological Construct
The theoretical construct measured by the WEZT is action- and sentence-level linguistic processing within an acquired neurogenic framework. The instrument evaluates several discrete, psycholinguistically validated sub-constructs:
1. Lexical Verb Comprehension
This sub-construct entails the activation of semantic and conceptual representations of actions without requiring speech output. Patients process an auditory lexical stimulus and map it to an accurate mental action schema. Within the WEZT, this construct is assessed through a four-alternative forced-choice visual array. Successful performance demands fine-grained semantic differentiation, as the visual arrays incorporate precise distractors:
- Target Action: The visual representation corresponding directly to the presented verb (e.g., to paint).
- Semantic Verb Distractor: An action sharing semantic features or instrumentality with the target (e.g., to draw or to glue).
- Semantic Noun Distractor: A nominal entity functionally tied to the target action (e.g., a paintbrush).
- Associated Noun Distractor: A nominal entity linked conceptually to the semantic verb distractor (e.g., a pencil).
2. Isolated Verb Production (Action Naming)
Expressive verb retrieval requires access to the conceptual system, selection of the appropriate lemma in the mental lexicon, and activation of its phonological word-form for motor-speech articulation. Action naming dissociates markedly from object (noun) naming in neurological populations due to the higher syntactic and semantic complexity inherent to relational concepts.
3. Syntactic Argument Structure and Mapping
Verbs govern thematic grids specifying how many participants are required (intransitive = 1 argument; transitive = 2 arguments; ditransitive = 3 arguments). The WEZT evaluates whether an individual can map thematic roles (who is doing what to whom) onto syntactic positions (Subject, Direct Object, Indirect Object). Deficits in this domain lead to structural mapping impairments, manifesting as thematic role reversals in reversible passive and active constructions (e.g., mistaking The woman chases the man for The man chases the woman).
4. Sentence Production and Grammatical Encoding
This dimension examines the capacity to generate grammatically well-formed sentences containing correct word order and finite verb placement. In non-inflected languages or languages with verb-second (V2) syntax such as Dutch and German, producing finite verbs in main clauses requires movement operations, rendering finite verb positioning particularly vulnerable in agrammatic Broca’s aphasia.
6. Theoretical Framework
The WEZT is anchored in cognitive psycholinguistic architectures of language generation and comprehension, specifically integrating the lexical-syntactic models of Merrill Garrett and Willem Levelt, alongside generative neurolinguistic hypotheses regarding structural syntactic deficit profiles.
Garrett’s and Levelt’s Processing Architectures
According to Levelt’s “Blueprint for the Speaker,” lexical access bifurcates into two distinct stages: lemma selection (retrieval of abstract syntactic and semantic information) and lexeme/phonological form access. Garrett’s language production framework delineates processing into the functional level (where semantic-lexical selection and thematic role assignment take place) and the positional level (where syntactic constituent order is generated and phonological insertion occurs). The WEZT evaluates linguistic breakdowns specifically at these interface levels. Patients with agrammatic production often demonstrate intact functional-level representations (knowing the actor and theme) but exhibit profound collapse at the positional level, failing to assign grammatical functions or inflect verbs correctly.
The Trace Deletion Hypothesis and Derived Order Hypothesis
Syntactic comprehension deficits in aphasia have historically been interpreted through models such as Grodzinsky’s Trace Deletion Hypothesis (TDH), which posits that non-canonical sentences (where constituents have moved from their base positions, leaving traces) are poorly comprehended because agrammatic patients fail to co-index traces with displaced noun phrases. In Dutch, canonical sentence order is Subject-Object-Verb (SOV) in subordinate clauses, while main clauses require verb movement to the second position (V2). Bastiaanse and colleagues advanced the Derived Order Hypothesis, arguing that sentences where the verb occupies a derived (moved) position are harder to comprehend and produce than those where the verb remains in its base-generated structure. The WEZT explicitly manipulates these linguistic dimensions to systematically diagnose whether grammatical breakdowns are dictated by movement-related complexity.
7. Validity
The validity of the WEZT has been rigorously explored within clinical and research contexts across the Dutch-speaking language area (the Netherlands and Flanders), with international adaptations reinforcing its psychometric integrity.
Construct and Discriminant Validity
Construct validity is evidenced by the clear double dissociations observed across different aphasic syndromes. Studies standardizing the WEZT demonstrated that individuals with agrammatic non-fluent aphasia show disproportionate difficulty producing verbs relative to nouns, and pronounced deficits in generating verbs in derived (inflected, main-clause) positions compared to base-generated positions. Conversely, individuals with fluent aphasias (e.g., Wernicke’s or anomic aphasia) frequently demonstrate intact syntactic sequencing alongside severe lexical-semantic errors on verb comprehension subtests, confirmed by higher rates of semantic distractor selections.
Convergent and Criterion Validity
The subtests of the WEZT correlate significantly with corresponding linguistic components of validated diagnostic batteries:
- Comprehension subtests correlate moderately to highly (r = .68 to .81, p < .001) with the Comprehension subtests of the Aachen Aphasia Test (AAT).
- Sentence production measures correlate positively with functional discourse metrics, including mean length of utterance (MLU) and percentage of well-formed sentences derived from spontaneous speech samples (r = .62 to .75).
- Criterion validity is confirmed by the test’s sensitivity in differentiating healthy neurological controls from mild stroke patients without overt spontaneous speech pathology, showing negligible ceiling effects among healthy control cohorts while identifying subtle verb retrieval latencies in stroke survivors.
8. Reliability
Psychometric evaluation of the WEZT across adult stroke cohorts confirms high operational stability, inter-examiner consistency, and internal metric coherence.
Internal Consistency
Standardization data compiled by Bastiaanse, Maas, and Rispens (2000) yielded strong internal consistency estimates across the individual subtests:
- Verb Comprehension: Cronbach’s α = .84
- Action Naming (Verb Production in Isolation): Cronbach’s α = .91
- Sentence Comprehension: Cronbach’s α = .88
- Sentence Production: Cronbach’s α = .93
Test-Retest and Inter-Rater Reliability
Test-retest stability was investigated in stable chronic aphasic patients over a 2- to 4-week interval, demonstrating high intraclass correlation coefficients (ICC > .86 across expressive subscales; ICC > .81 across receptive subscales). Inter-rater scoring reliability is particularly critical for expressive sentence formulation tasks; evaluation of transcribed patient responses across independent clinical linguists yielded inter-rater agreement rates (Cohen’s κ) between .89 and .96, confirming that scoring criteria are unambiguous and replicable.
9. Factor Analysis
Structural evaluations and factorial analyses of the WEZT support a multidimensional linguistic architecture that separates lexical-semantic processing from syntactic structural operations, while distinguishing between receptive and expressive modalities.
Confirmatory and Exploratory Factor Structures
Exploratory factor analyses (EFA) conducted on aphasic cohorts reveal a robust two-factor primary solution explaining over 68% of the total variance:
- Factor 1: Syntactic-Formulation Dimension: Highly loaded by sentence production, canonical vs. non-canonical sentence ordering, and finite verb placement (loadings ranging from .71 to .89). This factor captures structural grammatical operations.
- Factor 2: Lexical-Semantic Dimension: Highly loaded by isolated verb comprehension, action naming, and semantic distractor discrimination (loadings ranging from .64 to .85). This factor reflects underlying conceptual-semantic and lexical retrieval integrity.
Confirmatory factor models corroborate that a unified single-factor construct fails to fit the clinical data adequately (χ²/df > 3.8; RMSEA > .11; CFI < .80), whereas a multidimensional model incorporating distinct yet correlated receptive and expressive grammatical-lexical latent variables achieves superior goodness of fit (χ²/df = 1.42; RMSEA = .045; CFI = .96; TLI = .95). These statistical findings confirm that verb processing cannot be clinically addressed as an undifferentiated monolith.
10. Instrument / Measurement Tool
The WEZT is an instrument-based clinical observation assessment administered individually in an interactive clinical environment. Administration typically requires between 30 and 45 minutes, depending on the patient’s cognitive fatiguability and motor/speech capabilities.
- Target Population: Adults and elderly individuals with acquired brain injury (CVA, TBI, neurosurgical resection, or progressive neurogenic disorders).
- Administration Format: Paper-and-pencil stimulus booklet accompanied by structured verbal and auditory prompts administered by a speech-language pathologist or clinical neurolinguist.
- Core Subtests:
- Subtest 1: Verb Comprehension — Auditory word-to-picture matching utilizing 4-picture visual arrays (target, semantic verb distractor, noun distractor, and semantic noun distractor).
- Subtest 2: Verb Production in Isolation (Action Naming) — Confrontation naming of line drawings depicting distinct transitive and intransitive actions.
- Subtest 3: Sentence Comprehension — Sentence-picture matching assessing reversible active vs. passive constructions and canonical vs. non-canonical syntax.
- Subtest 4: Sentence Production — Constrained sentence generation using picture cues and written verb stems to assess argument mapping and finite verb positioning.
- Subtest 5: Finite Verb Retrieval & Tense Inflection — Cloze sentence tasks eliciting correctly inflected finite verbs in varying clause structures.
- Scoring System:
- Receptive items: Binary scoring (1 = correct, 0 = incorrect), with systematic error coding based on the nature of the selected distractor (semantic verb, noun, unrelated).
- Expressive items: Qualitative multi-tiered scoring (2 = immediate correct grammatical response, 1 = delayed/self-corrected response or minor phonological paraphasia, 0 = semantic paraphasia, neologism, paragrammatic sentence structure, or omission).
11. Permissions & Fee and Test Year
The Werkwoorden- en Zinnentest was formally published in 2000 by Swets & Zeitlinger (later distributed via Harcourt Assessment and Pearson Assessment / Bohn Stafleu van Loghum in the Netherlands). It is a copyrighted, proprietary clinical diagnostic instrument. Clinicians, hospital departments, and researchers must purchase the authorized testing manual, visual stimulus plates, and official scoring forms from the commercial distributor or copyright holder. The test materials are not released into the public domain for unrestricted distribution. Researchers wishing to translate or adapt the WEZT into other languages must obtain prior formal authorization from the primary authors and rights holders.
12. References
- Bastiaanse, R., Maas, E., & Rispens, J. (2000). Werkwoorden- en Zinnentest (WEZT). Swets & Zeitlinger.
- Bastiaanse, R., & van Zonneveld, R. (2005). Sentence production with verbs of alternating transitivity in agrammatic Broca’s aphasia. Journal of Neurolinguistics, 18(1), 57–66. https://doi.org/10.1016/j.jneuroling.2004.10.001
- Bastiaanse, R., & van Zonneveld, R. (2006). Broca’s aphasia, verbs and the Derived Order Hypothesis. Brain and Language, 95(1), 142–143. https://doi.org/10.1016/j.bandl.2005.07.078
- Garrett, M. F. (1988). Processes in language production. In F. J. Newmeyer (Ed.), Linguistics: The Cambridge Survey (Vol. 3, pp. 69–96). Cambridge University Press. https://doi.org/10.1017/CBO9780511621062.005
- Grodzinsky, Y. (2000). The neurology of syntax: Language disabilities without general cognitive impairment. Behavioral and Brain Sciences, 23(1), 1–21. https://doi.org/10.1017/s0140525x00002399
- Levelt, W. J. (1999). Models of word production. Trends in Cognitive Sciences, 3(6), 223–232. https://doi.org/10.1016/S1364-6613(99)01319-4