Aachen Aphasia Test

1. Abstract

The Aachen Aphasia Test (AAT; original German: Aachener Aphasie Test; Dutch adaptation: Akense Afasie Test) is an extensively standardized, multicomponent psychometric battery designed for the clinical diagnosis, syndromic classification, and quantitative assessment of acquired language disorders in adults with neurological impairment. Developed systematically by Walter Huber, Klaus Poeck, Dorothea Weniger, and Klaus Willmes at the RWTH Aachen University Hospital, and subsequently adapted into Dutch by Peter Graetz, Ria de Bleser, and Klaus Willmes, the instrument addresses the clinical imperative for psychometrically robust, linguistically motivated diagnostic assessment of neuropsychological deficits arising from cerebrovascular accidents (CVA), traumatic brain injury (TBI), and other focal or diffuse cerebral pathologies. The battery comprises six primary hierarchical subtests: Spontaneous Speech analysis (evaluated across six communicative dimensions: communicative behavior, articulation and prosody, automated language, semantic structure, phonological structure, and syntactic structure), the Token Test (modified 5-section version; 50 items), Repetition (50 items spanning phonemes, monosyllabic words, complex polysyllabic words, foreign loanwords, and sentences), Written Language (30 reading and writing items across word and sentence levels), Confrontation Naming (40 items across object nouns, color terms, compound nouns, and action verbs), and Language Comprehension (120 test trials split evenly between auditory and visual-reading modalities, testing word and sentence semantics). Administered via direct clinician observation, manipulative object interaction, and stimulus book presentation, the AAT employs graded 3-point to 6-point scoring scales calibrated to penalize both linguistic paraphasias and latency anomalies. Psychometric evaluations demonstrate exceptional internal consistency (Cronbach’s alpha spanning .89 to .98 across subtests), high test-retest reliability ($r_{tt} = .85$ to $.96$), and profound criterion and discriminant validity, successfully differentiating individuals with aphasia from healthy controls and non-aphasic brain-damaged patients with over 90% diagnostic specificity. Furthermore, automated non-parametric and parametric discriminant functions assign aphasic profiles to classical syndromic categories (Broca’s aphasia, Wernicke’s aphasia, global aphasia, anomic aphasia, and non-classifiable syndromes) with robust statistical accuracy.

2. Keywords

Aachen Aphasia Test, Akense Afasie Test, AAT, aphasia assessment, psycholinguistics, neurolinguistics, stroke rehabilitation, Token Test, speech pathology, language disorders, neuropsychological assessment, language comprehension, syndromic classification, confrontation naming

3. Authors

The Aachen Aphasia Test was originally developed by an interdisciplinary team of behavioral neurologists, neuropsychologists, and aphasiologists at the Department of Neurology, RWTH Aachen University Hospital (Rheinisch-Westfälische Technische Hochschule Aachen), Germany:

• Walter Huber, Ph.D. — Professor of Neurolinguistics, Department of Neurology, Section of Neuropsychology, RWTH Aachen University Hospital, Aachen, Germany. Specialization: Neurolinguistics, cognitive aphasiology, and cognitive rehabilitation.
• Klaus Poeck, M.D. (1926–2006) — Professor and Chair of Neurology, Department of Neurology, RWTH Aachen University Hospital, Aachen, Germany. Pioneer in classical clinical aphasiology, behavioral neurology, and European neuropsychology.
• Dorothea Weniger, Ph.D. — Senior Neuropsychologist and Researcher, Department of Neurology, RWTH Aachen University Hospital, Aachen, Germany, and subsequent affiliate of the University of Zurich, Switzerland. Specialization: Psycholinguistic modeling of lexical-semantic processing.
• Klaus Willmes, Ph.D. — Professor of Neuropsychological Methodology and Biostatistics, Section of Neuropsychology, Department of Neurology, RWTH Aachen University Hospital, Aachen, Germany. Specialization: Psychometrics, multivariate modeling in neuropsychology, and statistical diagnostic systems.

The Dutch adaptation (Akense Afasie Test) was constructed, standardized, and validated by:

• Peter A. M. Graetz, Ph.D. — Clinical Linguist and Aphasiologist, Department of Clinical Linguistics, Radboud University Nijmegen, Nijmegen, The Netherlands.
• Ria de Bleser, Ph.D. — Professor of Cognitive Neurolinguistics, Department of Linguistics, University of Potsdam, Germany, and formerly of RWTH Aachen University and Radboud University Nijmegen.
• Klaus Willmes, Ph.D. — Methodological lead for the psychometric calibration and statistical norming of the Dutch version.

4. Purpose

The Aachen Aphasia Test was developed to resolve a persistent methodological dilemma in behavioral neurology and speech-language pathology: the tension between qualitative clinical taxonomy and quantitative psychometric rigor. Prior to the creation of the AAT in the late 1970s and early 1980s, European clinical practice relied heavily on unstructured bedside assessments or literal translations of Anglo-American instruments—such as the Boston Diagnostic Aphasia Examination (BDAE) and the Western Aphasia Battery (WAB)—which frequently lacked language-specific linguistic stratification, statistical norming on matched brain-damaged non-aphasic reference groups, and empirically validated discriminant assignment rules. The AAT was constructed specifically to rectify these limitations by providing a standardized, objective, and linguistically grounded protocol tailored to the phonological, morphological, and syntactic realities of Germanic languages (originally German, and through subsequent rigorous cross-linguistic adaptation, Dutch, Italian, and other languages).

From a clinical perspective, the AAT serves three primary diagnostic purposes:

1. Differential Diagnosis of Aphasia: To establish definitively whether a patient’s acquired communication impairment stems from a central symbolic language deficit (aphasia) rather than an isolated motor speech disorder (dysarthria, apraxia of speech), a general cognitive deterioration (dementia, delirium), or primary psychiatric disturbance. This differentiation is operationalized via statistical cutoff thresholds derived from both healthy controls and non-aphasic brain-damaged populations.
2. Syndromic Classification: To assign aphasic patients to standard neuropsychological syndromes based on objective discriminant analysis and non-parametric cluster matching. The AAT explicitly classifies profiles into the classical neuroanatomically correlated syndromes: Broca’s aphasia, Wernicke’s aphasia, Global aphasia, and Anomic aphasia, while formally identifying Transcortical syndromes (sensory, motor, mixed) and atypical/non-classifiable patterns without forcing patients into ill-fitting diagnostic boxes.
3. Profiling across Modalities and Linguistic Levels: To delineate a fine-grained psycholinguistic profile mapping the integrity of the four core linguistic modalities—expressive speech, auditory comprehension, reading, and writing—across multiple psycholinguistic strata (phonology, morphology, lexical semantics, and syntax). This multi-level characterization provides speech-language pathologists with empirical targets for individualized cognitive-linguistic therapy.

In clinical research settings, the AAT functions as a benchmark measurement tool for measuring neurological recovery, validating the efficacy of pharmacotherapy and intensive speech therapy, and examining brain-behavior correlations via structural and functional neuroimaging (MRI, PET, and fMRI). Because the test incorporates parallel forms and non-parametric repeated-measurement statistical programs (e.g., ALLOC, COM-AAT), researchers can rigorously evaluate spontaneous neurological recovery versus true therapeutic intervention effects while controlling for retest practice biases.

5. Psychological Construct

The core construct measured by the Aachen Aphasia Test is central language processing capacity under conditions of focal brain pathology. The AAT operationalizes this construct not as a monolithic capacity, but as a modular yet integrated neuropsychological architecture where impairment can manifest dissociably across input/output channels and structural linguistic tiers. The battery systematically examines six operationalized subscales:

1. Spontaneous Speech (Spontane Spraak)

Spontaneous speech is elicited through a semi-structured biographical and conversational interview probing the patient’s medical onset, daily life, occupational background, and family. Rather than measuring a single omnibus metric, clinician evaluation rates conversational output along six distinct 6-point ordinal dimensions (coded 0 to 5):

• Communicative Behavior: Assesses pragmatic ability, conversational initiative, turn-taking, and compensatory gesture usage despite verbal limitations.
• Articulation and Prosody: Evaluates motor speech execution, phonetic distortions, dysarthric slurring, phonetic planning deficits (apraxia of speech), and altered intonation/speech rhythm.
• Automated Language: Gauges the intrusion of recurring utterances, speech automatisms, perseverations, stereotypical phrases, and neologistic recurring patterns.
• Semantic Structure: Quantifies lexical retrieval difficulties, empty discourse, circumlocutions, and semantic paraphasias (e.g., substituting “dog” for “cat” or “chair” for “table”).
• Phonological Structure: Measures phonemic paraphasias (substitutions, omissions, additions, transpositions of phonemes, e.g., “tephelone” for “telephone”), conduit d’approche, and phonemic jargon.
• Syntactic Structure: Evaluates sentence architecture, ranging from severe agrammatism (telegraphic speech, omission of inflections, absence of subordinate clauses) to paragrammatism (complex sentences with structural blends, morphological substitutions, and confused phrasing).

2. The Token Test

The Token Test operates as an exquisitely sensitive instrument for detecting subtle receptive and expressive linguistic disruptions, verbal working memory constraints, and central cognitive slowing. Using 20 plastic tokens varying in two shapes (circles, squares), two sizes (large, small), and five vibrant colors (red, blue, green, yellow, white), patients respond to 50 spoken commands distributed across five sections of increasing syntactic and working memory complexity. The items progress from simple monosyllabic syntax (“Touch a red circle”) to complex relational prepositions, coordination, and conditional clauses (“Before touching the yellow circle, pick up the small red square”). It acts as the primary detector for the presence or absence of aphasia.

3. Repetition (Naspreken)

Repetition tests the integrity of the dorsal auditory-motor integration stream, phonological decoding, auditory short-term memory buffer, and motor programming pathways. The subtest consists of 50 items systematically organized into five linguistic tiers of 10 items each: single isolated phonemes/speech sounds, monosyllabic words, complex multisyllabic loan/foreign words, compound nouns, and full grammatical sentences of increasing structural length and syntactic subordination.

4. Written Language (Schrijftaal)

The Written Language subtest measures central orthographic-graphemic processing across reading and writing modalities, encompassing 30 items split into six distinct parts:

• Reading Aloud: Evaluating sublexical grapheme-to-phoneme conversion and lexical-orthographic retrieval for isolated words, compound words, and full sentences.
• Composed Writing (Dictation with Letter Tiles): Requiring patients to assemble words using physical plastic letter tiles, isolating pure central orthographic spelling knowledge from peripheral motor dysgraphia.
• Written Dictation: Assessing handwriting to dictation across single words, complex words, and full sentences, uncovering signs of surface dyslexia/dysgraphia, phonological dyslexia/dysgraphia, and deep dyslexia.

5. Confrontation Naming (Benoemen)

Confrontation Naming evaluates the integrity of the ventral conceptual-lexical semantic stream and the retrieval of phonological word forms across 40 standardized visual line drawings. To capture category-specific dissociations and distinct psycholinguistic retrieval demands, the items are divided equally into four categories of 10 items each: common visual objects (simple nouns), color patches (abstract non-object labels), composite compound nouns (semantic/phonological integration), and situational action pictures depicting transitivity (verb generation).

6. Language Comprehension (Taalbegrip)

Comprehension investigates central semantic, conceptual, and grammatical decoding across two distinct modalities and two hierarchical levels. The subtest comprises 120 stimulus presentations organized into four sections of 30 items each:

• Auditory Word Comprehension: The clinician speaks a word, and the patient must select the corresponding referent from a 4-choice visual array (containing the target, a semantic foil, a phonemic foil, and an unrelated visual foil).
• Auditory Sentence Comprehension: Spoken sentences of varying syntactic complexity (e.g., active, passive, reversible locative structures) matched to 4-picture arrays depicting actor-action-patient inversions.
• Visual/Reading Word Comprehension: The patient reads a printed word and identifies the matching target picture among semantically and visual-orthographically related distractors.
• Visual/Reading Sentence Comprehension: The patient reads complex printed sentences and designates the correct depiction within a 4-choice visual matrix.

6. Theoretical Framework

The Aachen Aphasia Test is grounded in a convergence of classical European behavioral neurology (rooted in the Wernicke-Lichtheim-Geschwind tradition) and modern cognitive neuropsychology. The structural foundation of the AAT rejects the historical view that aphasic manifestations are erratic or chaotic. Instead, the authors hypothesized that aphasic language breakdowns conform to systematic psycholinguistic rules governed by the functional architecture of the human language faculty.

The classical localizationist framework—formulated historically by Paul Broca, Carl Wernicke, Ludwig Lichtheim, and revitalized by Norman Geschwind—posited that distinct perisylvian cerebral structures subserve specialized linguistic functions: Broca’s area (inferior frontal gyrus) for motor programming and syntactical assembly; Wernicke’s area (posterior superior temporal gyrus) for sensory phonological representation and lexical decoding; and the arcuate fasciculus connecting these regions for verbal repetition. Poeck and Huber acknowledged these clinical syndromic regularities but infused them with strict empirical methodology. Rather than presuming that syndromes represent pure biological invariants, they treated syndromes as mathematically identifiable clusters of co-occurring linguistic symptoms resulting from typical vascular distribution patterns within the middle cerebral artery territory.

Simultaneously, the AAT incorporates principles of psycholinguistic structuralism and cognitive neuropsychological models of information processing (such as the dual-route model for reading and writing, and Levelt’s speech production architecture). Language processing is operationalized as a hierarchical multi-stage computation involving:

• Sensory Input: Acoustic phonological decoding and visual orthographic recognition.
• Lexical Retrieval & Semantic Access: The navigation of the central mental lexicon, parsing conceptual semantic features, and retrieving lemma and lexeme forms.
• Grammatical & Morphosyntactic Encoding: The structural assembly of thematic roles, phrase structures, and morphosyntactic agreements.
• Motor Articulatory & Graphemic Execution: Translating phonological/orthographic forms into motor phonetic gestures or manual letter sequences.

By assessing identical linguistic content across parallel receptive and expressive modalities (e.g., testing the same concepts in Auditory Comprehension, Reading Comprehension, Confrontation Naming, and Written Dictation), the AAT allows clinicians to isolate the precise functional locus of linguistic disruption. For example, a breakdown confined exclusively to spoken naming and repetition, with intact visual reading and writing, pinpoints a peripheral auditory-phonetic or motor output locus, whereas an across-the-board failure on identical semantic concepts across all four input/output channels confirms an impairment within central conceptual-semantic memory.

7. Validity

The psychometric validity of the Aachen Aphasia Test has been extensively investigated across multiple international cohorts, including the original German normative standardizations (Huber et al., 1983, 1984), the Dutch standardization (Graetz, de Bleser, & Willmes, 1987, 1992), and subsequent validation studies across European neurorehabilitation centers.

Construct Validity

Construct validity is substantiated by the test’s capacity to reflect theoretical hierarchies of linguistic processing complexity. Item difficulty distributions across subtests precisely align with psycholinguistic theory. In the Repetition subtest, error rates escalate predictably as stimuli progress from single phonemes to high-frequency monosyllabic words, low-frequency polysyllabic loanwords, compound words, and syntactically intricate sentences ($p < .001$). Similarly, in the Token Test, error frequencies increase linearly from Section I through Section V, demonstrating that the instrument validly captures cumulative syntactic and verbal memory processing loads.

Criterion and Concurrent Validity

Concurrent validity has been established by cross-correlating AAT subtest scores with external neuropsychological metrics, legacy aphasia batteries, and clinical neuroimaging indices:

• Correlations between the AAT Token Test score and the standard De Renzi & Vignolo Token Test exceed $r = .92$, demonstrating near-perfect construct equivalence.
• Correlations between the AAT subscales and corresponding subtests of the Boston Diagnostic Aphasia Examination (BDAE) range from $r = .78$ to $r = .89$ for naming, repetition, and comprehension parameters.
• Functional neuroimaging investigations using PET and functional MRI demonstrate that performance on AAT Comprehension and Naming significantly correlates with metabolic activation and lesion volumes within the left superior temporal gyrus, middle temporal gyrus, and inferior parietal lobule ($r = .65$ to $.82$, $p < .001$).

Discriminant and Diagnostic Validity

The discriminant validity of the AAT is exceptional, particularly in resolving the critical diagnostic dilemma: distinguishing genuine aphasic patients from healthy controls and non-aphasic patients with right-hemisphere brain damage (RBD) or diffuse trauma. In clinical validation trials:

• Using the Token Test cutoff score combined with spontaneous speech rating profiles, the AAT achieved an overall diagnostic sensitivity of 94.2% and a diagnostic specificity of 93.1% in correctly classifying individuals as aphasic versus non-aphasic.
• Among non-aphasic brain-damaged patients (e.g., right-hemisphere stroke lesions), false-positive aphasia classifications remained below 7%, demonstrating that general cerebral compromise without left-perisylvian involvement does not artificially depress AAT classification.
• Syndromic classification accuracy using multivariate discriminant analysis program ALLOC correctly classifies approximately 85% to 88% of standard aphasic syndromes into their historically established neurological classifications (Broca, Wernicke, Global, Anomic).

8. Reliability

The AAT was engineered to meet the highest statistical standards for classical test theory, exhibiting outstanding internal consistency, high inter-rater agreement, and remarkable test-retest stability.

Internal Consistency

Internal consistency metrics (Cronbach’s alpha and split-half coefficients) computed across both the original German standardization sample ($N = 376$) and the Dutch normative sample ($N = 205$) consistently demonstrate exceptional reliability coefficients across all objective subtests:

• Token Test: $\alpha = .98$ (split-half $r = .97$)
• Repetition: $\alpha = .96$ to $.98$ (across subscales: phonemes $\alpha = .89$; monosyllabic words $\alpha = .92$; loanwords $\alpha = .94$; compound words $\alpha = .93$; sentences $\alpha = .95$)
• Written Language: $\alpha = .96$ (reading aloud $\alpha = .94$; composed writing $\alpha = .93$; dictation $\alpha = .95$)
• Confrontation Naming: $\alpha = .95$ (objects $\alpha = .92$; colors $\alpha = .89$; compound nouns $\alpha = .91$; actions $\alpha = .93$)
• Language Comprehension: $\alpha = .95$ (auditory word $\alpha = .88$; auditory sentence $\alpha = .91$; visual word $\alpha = .89$; visual sentence $\alpha = .92$)

Inter-Rater Reliability

Because the Spontaneous Speech section relies on clinician judgment across six ordinal rating scales, inter-rater concordance was evaluated extensively using intraclass correlation coefficients (ICC) and Cohen’s weighted kappa ($kappa$). Pairwise scoring between trained aphasiologists yielded ICC values ranging from $.86$ to $.94$ across the six spontaneous speech dimensions. On the objective subtests (Repetition, Written Language, Naming, Comprehension), inter-rater agreement exceeded $kappa = .96$, attributable to the unambiguous operationalized scoring rules provided in the manual.

Test-Retest Stability

Stability over time was examined in chronic, neurologically stable aphasic patients evaluated over retest intervals spanning 3 to 6 weeks. Pearson test-retest correlation coefficients ($r_{tt}$) confirmed high temporal stability:

• Token Test: $r_{tt} = .94$
• Repetition: $r_{tt} = .92$
• Written Language: $r_{tt} = .91$
• Naming: $r_{tt} = .89$
• Comprehension: $r_{tt} = .88$

These robust stability metrics demonstrate that the AAT is highly resilient to transient measurement noise, making it suitable for longitudinal evaluations and tracking true therapeutic gains during neurorehabilitation.

9. Factor Analysis

The internal latent structure of the Aachen Aphasia Test has been evaluated through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA), shedding light on the fundamental architecture of language dissolution following focal cerebral insults.

Principal Component & Exploratory Factor Analysis

Early factorial investigations conducted by Willmes and Huber examined the dimensional structure of the AAT across heterogeneous clinical samples. Initial unrotated principal component analysis invariably reveals a massive general factor—often termed the “General Aphasia Factor” or “Central Symbolic Language Factor”—accounting for approximately 65% to 75% of the total shared variance. This predominant first eigenvalue reflects the pervasive, unitary severity dimension common to acquired focal language pathologies.

However, upon applying orthogonal (Varimax) and oblique (Promax) rotations to isolate multi-factor latent structures beyond general severity, a clear three- to four-factor model consistently emerges, explaining over 82% of the cumulative variance:

• Factor 1: Expressive Phonological and Articulatory Assembly (Primary loadings: Repetition subtests, Composed Writing, and Spontaneous Speech Articulation and Phonology; loadings ranging from $.72$ to $.89$).
• Factor 2: Central Semantic and Conceptual Decoding (Primary loadings: Auditory Word Comprehension, Visual Word Comprehension, Auditory Sentence Comprehension, Visual Sentence Comprehension, and Spontaneous Speech Semantic Structure; loadings ranging from $.68$ to $.86$).
• Factor 3: Lexical-Semantic Retrieval / Confrontation Naming (Primary loadings: Object Naming, Action Naming, Color Naming, and Compound Naming; loadings ranging from $.70$ to $.88$).
• Factor 4: Syntactic and Verbal Working Memory Complexity (Primary loadings: Token Test Sections IV and V, Complex Sentence Repetition, and Syntactic Structure of Spontaneous Speech; loadings ranging from $.64$ to $.81$).

Confirmatory Factor Analysis (CFA) & Structural Equation Modeling

Subsequent psychometric modeling has subjected the hierarchical structure of the AAT to confirmatory testing. A higher-order factor model—positing a single overarching Higher-Order Language Competence factor superordinate to three correlated first-order factors (Expressive Output, Receptive Comprehension, and Lexical-Semantic Selection)—demonstrates superior fit compared to single-factor or strictly orthogonal models:

• Comparative Fit Index (CFI): $.962$
• Tucker-Lewis Index (TLI): $.954$
• Root Mean Square Error of Approximation (RMSEA): $.048$ ($90% \text{ CI } [.039, .058]$)
• Standardized Root Mean Square Residual (SRMR): $.041$

These confirmatory metrics corroborate the modern aphasiological consensus: while aphasia represents a unitary central disturbance of symbolic manipulation, it remains structured along identifiable cognitive processing modules whose dissociable disruption produces distinct clinical syndromes.

10. Instrument / Measurement Tool

The Aachen Aphasia Test is a comprehensive, multimodal diagnostic battery. The following structural breakdown details the subtests, item distributions, operational formats, and scoring mechanics of the instrument:

• Target Population: Adults and elderly individuals suffering from acquired brain injuries (CVA, TBI, neurosurgical resection, focal encephalitis) suspected of aphasia.
• Administration Modality: Direct, one-on-one clinician administration via structured conversation, manipulative token interaction, and stimulus picture books.
• Total Administration Duration: Approximately 60 to 90 minutes for full administration (can be split across multiple sessions in cases of acute post-stroke fatigue).
• Battery Components and Subtest Breakdown:
  • Subtest 1: Spontaneous Speech (Spontane Spraak):
    • Format: Semi-structured interview (minimum 10 minutes of recorded discourse covering onset, illness, occupation, family, hobbies).
    • Items / Dimensions: Six 6-point ordinal rating scales (Scale range: 0 = completely absent / maximum pathology to 5 = fully normal / intact):
      1. Communicative Behavior
      2. Articulation and Prosody
      3. Automated Language
      4. Semantic Structure
      5. Phonological Structure
      6. Syntactic Structure
  • Subtest 2: The Token Test:
    • Format: Physical manipulation of 20 geometric plastic tokens (circles and squares in two sizes and five colors) following verbal instructions.
    • Items: 50 items arranged in five hierarchical sections of 10 items each:
      1. Section I (Large tokens only; basic color and shape nouns)
      2. Section II (Large and small tokens; size, color, shape adjectives and nouns)
      3. Section III (Large tokens; two sequential objects)
      4. Section IV (Large and small tokens; two sequential complex objects)
      5. Section V (Complex syntactic prepositions, locative relations, and conditional clauses)
    • Scoring: Binary or penalty-weighted scoring (0 = incorrect, 1 = correct on first presentation; in clinical research scoring, 0.5 points can be assigned for successful self-correction). Maximum score: 50 points (recorded as error count: 0 to 50 errors).
  • Subtest 3: Repetition (Naspreken):
    • Format: Direct auditory-vocal repetition of spoken stimuli presented by the examiner without visual-lip reading cues.
    • Items: 50 items distributed evenly across five 10-item linguistic categories:
      1. Phonemes (isolated vowels and consonants)
      2. Monosyllabic Words
      3. Foreign Loanwords / Polysyllabic Complex Words
      4. Compound Nouns
      5. Sentences (ranging from 3 to 9 words in length)
    • Scoring: 4-point polytomous scale per item:
      • 3: Correct, fluent repetition without hesitation or phonemic distortion.
      • 2: Correct response achieved after self-correction, articulation struggle, or latency > 3 seconds.
      • 1: Phonemic paraphasia, distortion, or incomplete sentence repetition retaining core target elements.
      • 0: Complete omission, neologism, or totally incorrect lexical substitution.

      Maximum Raw Score: 150 points.

  • Subtest 4: Written Language (Schrijftaal):
    • Format: Three reading and writing tasks assessing graphemic processing.
    • Items: 30 items divided into three 10-item parts:
      1. Reading Aloud (single words, compound words, sentences)
      2. Composed Writing (spelling using movable printed letter tiles to dictation)
      3. Written Dictation (handwriting to dictation with paper and pencil)
    • Scoring: Identical 4-point rating scale (0 to 3 points per item) evaluating accuracy, self-corrections, paragraphias, and paralexias. Maximum Raw Score: 90 points.
  • Subtest 5: Confrontation Naming (Benoemen):
    • Format: Visual confrontation naming using standardized black-and-white stimulus line drawings.
    • Items: 40 items distributed evenly across four 10-item categories:
      1. Common Objects (nouns)
      2. Color Cards (color terms)
      3. Compound Nouns (composite visual scenes/objects)
      4. Actions (verbs depicted in situational scenes)
    • Scoring: 4-point scale (0 to 3 points per item) assessing lexical retrieval speed, self-corrections, semantic/phonemic paraphasias, and neologisms. Maximum Raw Score: 120 points.
  • Subtest 6: Language Comprehension (Taalbegrip):
    • Format: Receptive picture pointing matching within 4-choice visual stimulus arrays (Target, Semantic Distractor, Phonemic/Visual Distractor, Unrelated Distractor).
    • Items: 120 trials divided into four 30-item sections:
      1. Auditory Word Comprehension (spoken word to picture)
      2. Auditory Sentence Comprehension (spoken sentence to situational picture)
      3. Visual/Reading Word Comprehension (printed word to picture)
      4. Visual/Reading Sentence Comprehension (printed sentence to situational picture)
    • Scoring: 4-point scale (0 to 3 points per trial) evaluating immediate pointing, delayed/hesitant correct selection, self-corrected pointing, and error pointing. Maximum Raw Score: 120 points (calculated using standardized weighting across sections).
• Standardized Transformation and Psychometric Profiling: Raw scores across all subtests are systematically converted into normalized percentile ranks and standard T-scores ($M = 50, SD = 10$) using comprehensive normative tables adjusted for aphasic clinical populations. Profile plots clearly delineate syndromic boundaries and severity percentiles.

11. Permissions & Fee and Test Year

The Aachen Aphasia Test was originally developed between 1978 and 1980, with its definitive German standardized edition published in 1983 by Hogrefe Verlag (Huber, Poeck, Weniger, & Willmes, 1983). The official Dutch adaptation (Akense Afasie Test) was standardized and published in 1987 by Hogrefe Uitgevers (Graetz, de Bleser, & Willmes, 1987; revised manual, 1992).

Copyright and Commercial Distribution: The complete testing kit—comprising the examiner’s manual, stimulus picture books, Token Test apparatus (plastic geometric shapes), composed writing letter tiles, and standardized scoring forms—is fully proprietary and protected under international copyright law. Commercial distribution rights are held by Hogrefe Publishing Group (Hogrefe Verlag GmbH & Co. KG in Germany; Hogrefe Uitgevers in the Netherlands and Flanders):

• Hogrefe Netherlands: Hogrefe Uitgevers B.V., Amsterdam, The Netherlands (hogrefe.nl).
• Hogrefe Germany: Hogrefe Verlag, Göttingen, Germany (hogrefe.com).

Fees and Licensing: The complete diagnostic kit is available for purchase exclusively to qualified healthcare professionals (certified clinical linguists, speech-language pathologists, neuropsychologists, and neurologists). Replacement scoring protocols and testing forms require proprietary purchasing. Researchers seeking to utilize the AAT in formal clinical trials or academic investigations must acquire legitimate licensed testing kits from the publisher. Computerized scoring software programs (e.g., COM-AAT) are similarly subject to commercial software licensing.

12. References

• De Bleser, R., & Poeck, K. (1985). Analysis of naming errors in aphasia: A contribution to the debate on the nature of semantic errors. Brain and Language, 24(2), 263–283. https://doi.org/10.1016/0093-934X(85)90136-1
• De Renzi, E., & Vignolo, L. A. (1962). The Token Test: A sensitive test to detect receptive disturbances in aphasics. Brain, 85(4), 665–678. https://doi.org/10.1093/brain/85.4.665
• Graetz, P., de Bleser, R., & Willmes, K. (1987). Akense Afasie Test: Handleiding [Aachen Aphasia Test: Manual]. Swets & Zeitlinger (now Hogrefe Uitgevers).
• Graetz, P., de Bleser, R., & Willmes, K. (1992). Akense Afasie Test (AAT): Nederlandse Versie [Aachen Aphasia Test: Dutch Version]. Hogrefe Uitgevers.
• Huber, W., Poeck, K., Weniger, D., & Willmes, K. (1980). Der Aachener Aphasie Test: Aufbau und experimentelle Überprüfung der Standardversion. Der Nervenarzt, 51(8), 475–482.
• Huber, W., Poeck, K., Weniger, D., & Willmes, K. (1983). Aachener Aphasie Test (AAT): Handanweisung [Aachen Aphasia Test: Manual]. Hogrefe.
• Huber, W., Poeck, K., & Willmes, K. (1984). The Aachen Aphasia Test. In F. C. Rose (Ed.), Progress in Aphasiology (pp. 291–303). Raven Press.
• Luzzatti, C., Willmes, K., & De Bleser, R. (1991). Aachener Aphasie Test (AAT): Versione Italiana. Organizzazioni Speciali.
• Poeck, K. (1983). What do we mean by “aphasic syndromes”? A neurologist’s view. Brain and Language, 20(1), 79–89. https://doi.org/10.1016/0093-934X(83)90035-4
• Weniger, D., Huber, W., Stachowiak, F. J., & Poeck, K. (1987). Assessment of language comprehension in aphasia: The Aachen Aphasia Test. Aphasiology, 1(4), 315–326. https://doi.org/10.1080/02687038708248842
• Willmes, K. (1993). Statistical methods for evaluating individual changes in language performance. Aphasiology, 7(1), 101–114. https://doi.org/10.1080/02687039308249499
• Willmes, K., Poeck, K., Weniger, D., & Huber, W. (1980). Facetten-Auswahl und hierarchische Dimensionsanalyse der standardisierten Fassung des Aachener Aphasie Tests. Archiv für Psychiatrie und Nervenkrankheiten, 229(2), 141–157. https://doi.org/10.1007/BF00343803

13. Items of the Scale