Aphasia Test 2
Anatomy of an *Assessment*
Case History/Intake/Consult Collaborators/Referral Sources Patient Reported Outcomes Oral Motor Exam Diagnostic Impressions Hearing Assessment Formal Assessment Informal Assessment
Anatomy of an Assessment
Case History/Intake/Consult Collaborators/Referral Sources Patient Reported Outcomes Oral Motor Exam Diagnostic Impressions Hearing Assessment Formal Assessment Informal Assessment all of this leads to diagnostic impressions
WAB overview
Criterion-referenced •Based on a fixed set of criteria, not on sample data •e.g. what you should know as a 6th grader •WAB: 0 to 100 scale based on arbitrary item weightings Subtest scores used to classify patients according to the Wernicke-Lichtheim-Geschwind model Limited usefulness for treatment planning •e.g, tx for Wernicke's Aphasia
Verbal paraphasias
Degree of semantic relatedness Response "cap" "lid" "top" "brush"
LSA Transcription
Materials and skills needed: 1) Patience! 2) Good recording equipment 3) Headphones Listen, type, repeat!
Impairment types
Impaired lexical-semantic *representations* Impaired *access* to lexical-semantic representations
Formal Assessment chart
Impairment vs. Activity vs. Participation-Oriented Assessment- refer to slide 31 powerpoint 6a
The Classic Model
Influential hypotheses: brain mechanisms involved in word access/storage Norman Geschwind (1926 - 1984)American Behavioral Neurologist •1970s - revived classic work by Broca, Wernicke, others •Emphasized language localization •"neoconnectionism" based on neuroanatomical pathways è "Wernicke-Geschwind" model •Led to the repackaging of Wernicke's classifications •i.e. Boston or BDAE classifications (Fun fact: the Geschwind legacy...)
Semantic Memory:
"...which of these corner pictures goes best with this middle one?..."
Line Bisection:
"I want you to cut these lines in half..."
Word Fluency:
"I'm going to give you a minute. Say out loud all the words that go in a category..." •Practice: "clothes" •"animals" •"Words that begin with s________"
Comprehension- Spoken Words example
"find me mouse" with photos
measures review •Intelligibility •
% of Intelligible utterances % of Intelligible words
Motor plan
- movement goals of the articulators. Example: to make the /s/ sound, move your tongue up to the alveolar ridge. It's a tongue-tip movement where contact is made between the front of the tongue and the alveolar ridge. like a recipe, or furniture assembly manual
Motor planning
Broca's Wernicke's Prefrontal Pre-Motor Supplementary Motor Area (SMA) Influence of contextual factors Core motor plan recall Phoneme 1....... Phoneme 2.......Phoneme 3 Adaptation of core motor plan of phoneme within the context of the planned unit. (Movement goal)
Informal Assessments
Provides both quantitative and qualitative data •May not developed with objective population data •Percent correct of probe set •Observed strengths (self cues; motivation) and weaknesses •Important part of obtaining the whole picture •Examples: Discourse analyses, observations, patient reported outcome measures (PROMs)
WAB Charts
Refer to slide 2, 3, 4 on 6b powerpoint
Refer to slide 24
Refer to slide 24 powerpoint 6a
Test-Retest Reliability
Slide 13,14
"impairment-based" or "restorative" approaches
Target: specific language impairment (e.g., phonological) Goal: generalization beyond trained items/tasks, and to the communicative environment of the person; reducing impairment > improving success of communication
CAT to measure change
The CAT manual provides data supporting its sensitivity to change, as did Winans-Mitrik et al. (2014): Minimal detectable change for overall scores individual and subtest scores •p < 0.05, 1-tailed: 3.2. • TAKE AWAY: CAT is more sensitive to change
Execution
SMA Cerebellum Basal ganglia Motor cortex Thalamus Brain stem Motor neurons Muscle Movements Manifested speech
Impaired lexical-semantic representations
Semantic dementia Difficulty naming objects, places, people Early stage: can describe but not name Late stage: do not recognize
What's the difference (perception and recognition) ?according to Hickok & Poeppel *recognition*
ventral stream Access the lexicon/stored words Recognizing words "rhubarb" I recognize that word!
Naming objects
name objects based on pictures
Raymer et al. (1993) Phonological Cueing Hierarchy
Therapy uses a phonological hierarchy of least to most cues. Twenty pictures were taught in two sets of 10. 1.Subjects spontaneously named pictures. 2.SLP provides a word that rhymed with target word, and the subject attempts to name. 3.SLP provides a phonemic cue, and the subject attempts to name. 4.SLP says the target word, and the subject repeats it. 5.At whatever level correct naming is achieved, the subject repeats the word 5 times. •Four subjects greater 1+ year unilateral left CVA, Broca's aphasia. • *Outcome Measures* •3 probe tasks: oral picture naming, oral word reading, and written picture naming. •Improvement in naming of target set once treatment began, but no subject reached 80% criterion. •Generalization >30% to untrained items was noted in 2 of 4 subjects. •Oral naming results were maintained in 2 of 3 subjects.
Generalization
To see treatment effects extend to •untrained stimuli and •untrained contexts in other words... The transfer of treatment gains to untreated items and to everyday, non-clinical environment(s) ULTIMATE TREATMENT GOAL
Phonemic paraphasias
Types: additions, substitutions, anticipatory, perseverative Response "hato" "har" "tat" (prev item) "arm"
omission
no response
Goals of Assessment
Understanding a client's communicative function and activity limitations, participation restrictions, determining rehab objectives, potential treatment approaches, determining a diagnosis, potential for improvement/realistic prognosis, factors that may impact recovery
Wernicke-Geschwind Repetition of spoken words
auditory cortex -> Wernicke's area-> arcuate fasciculus -> Broca's area -> motor cortex
measures review syntax/ morphology
o # MLU (mean length of utterance) in Words o # MLU in Morphemes (type/s) o SI (subordination index) composite score oMeasures syntactic complexity oRatio of total clauses to number of utterances o(He got on the rock [S1-1])
Writing to dictation
can you write the word pen, man, yacht, idea, etc
unrelated error
cat to desk
Neologism
cat to peesh
Measure review semantics
o # of total words (outside of mazes) o # of different words o Type Token Ratio (TTR)- Number of different words to the total number of words o Moving Average TTR- Estimates TTR for small segments of sample, then calculated as a whole
Goals for LSA
o 5 minutes of connected speech OR to the end of the prescribed LS task o o 4-5 stimuli o o 300-400 words in length o o 50 utterances o o Ultimately, you want a GOOD representation of the client's natural language skills
LSA Skills and Equipment
o Audio/Video recorder o Consent to audio/video record o Watch/clock/timer o Elicitation protocol script o Story book (if narrative sample) o Quiet area o Rapport with your client
Skills for Effective LSA Solicitation
o Intelligibility issues oGlossing: Exact, immediate repetition of client's speech by examiner o o Encouraging participation (Grandma's Rules) o oFriendly and enthusiastic oGenuine interest oFollow the PWA's lead oPatience oPerseverance oAdapt your language
Discourse barriers
o Limited recognition as a valid assessment measure o o Limited resources o o Limited time o o Limited training and expertise
Measure review discourse
o% Responses to Questions o Mean Turn Length (words) o Utterances with overlapping speech o Interrupted other speaker o Main concept analysis (Kong, 2009)
Measures Review verbal facility
oWords/Minute o # Maze words as a % of total words o Pauses within utterances o Pauses between utterances o Abandoned utterances
Writing picture names
point to the picture of a hat move on to the picture of the boy
Some errors are mixed or unrelated
double errors mixed error omission unrelated
additions
fake to flake
Perseverative
fed the dog to fed the fog
deletion
flake to fake
process
how language is activated/accessed
Representation:
how language phoneme, morpheme, grapheme/letter, lemma/word form, lexical item/word, concept is represented
Communicative participation item bank- general short film
in slides (6a overview of assessment slide 20)
Unrelated behaviors example
looks at door when someone walks by; fidgets
Comprehension of Spoken Paragraphs
read paragraphs and ask questions
CAT chart
refer to slide 10 on 6b
Comprehension- spoken words scoring
refer to slide 29-31
Bidirectional; "network neighborhood"
slide- Models of Word RetrievalDell's Interactive Activation Model Activation of phonological representations
Motor program
specifies which muscles will be used in moving the relevant articulators specified in the motor plan Example: specifying which particular muscles are going to make the tongue move to that location. like the ingredients, or furniture pieces and tools
Reading (aloud)
Words Complex words Nonwords Function words
•Word frequency •Word imageability
Tap more into lexical and lexical-semantic properties
•Word length •Lexicality
Tap more into lexical and phonological properties
Semantic Feature Analysis
-
Verb Network Strengthening Treatment
-
Neologistic paraphasias
(Neologisms) Response "chith"
Mixed paraphasias
(aka, "formal errors") Both semantic and phonologic Response "hood" (/h/ initial + covers head)
Impaired access to lexical-semantic representations
(most often) Stroke-related aphasia Difficulty naming objects, places, people Can describe or provide synonym Recognize, but show variability in naming ability Related to processing of language
*Comprehensive Aphasia Test*
*Comprehensive Aphasia Test*
Basic connectionist model principles
*Knowledge (language) can be described by interconnected networks of units*. •"Neurally plausible" •units = neurons, connections = synapses *These networks or patterns of units are activated depending on the strength of the network/patterns* *or "greater connection weight" [in a few slides] *Spreading activation* •When a neuron/unit is activated, it spreads to other neurons/units connected to it •interaction is multi-directional *Learning* (e.g. speech therapy) •Learning is modifying the connections between units.
Benefits discourse analysis
.More naturalistic form of assessment. Supplements standardized assessment data. Can be repeated frequently. Can be used with any speaker in many contexts. .Certain measures corroborate aphasia diagnosis and severity
phonomotor Two phases of treatment
1. . Phonemes in Isolation Improve awareness of specific linguistic (phoneme) representations across modalities 2. Phoneme Sequences •Train mono- and multi-syllabic phonological sequences improve access to phonological representations •Help participants learn how phonemes behave in words
Wambaugh Semantic Cueing
1. picture of target item 2. picture + verbal description of target 3. picture + semantically non-specific sentence completion phrase 4. picture + semantically loaded completion phrase •For cueing levels 1-4 a naming response was requested • Verbal feedback was provided for correct or incorrect responses 5. picture + verbal model of target word. •repetition of target word was requested. •Picture (duck) •"swimming bird, wide beak" •"He carved a wooden..." •"At the lake I heard the quack of a..." •Model
Wambaugh Phonological Cuing
1. picture of target item 2. picture + verbal production of non-real word that rhymes 3. picture + verbal first sound cue 4. picture + sentence completion that included rhyme and sound cue •For cueing levels 1-4, naming response requested. •Verbal feedback provided for correct or incorrect responses. 5. picture + verbal model of target word •Repetition of target word requested. •Picture (cake) •Rhymes with "zake" •Starts with /k/ •"name rhymes with zake, it's a /k/..." •Model •Three individuals with chronic aphasia, subsequent to a single left-hemisphere stroke (2 subjects) or two left-hemisphere strokes (1 subject). •Subject 1 (predominantly semantic deficit) was classified as having anomic aphasia •Subject 2 (mixed semantic/phonologic deficit) was classified as having Wernicke's aphasia •Subject 3 (predominantly phonologic deficit) was classified as having conduction aphasia •All participants responded positively to both treatments, though degree of response varied. •The subject with the semantic-level deficit responded similarly to both treatments; •the subject with the mixed deficit reached criterion on both phases of PCT but only on the first phase of SCT; •the subject with the phonologic-level deficit responded better to SCT in that he achieved higher levels of accuracy with fewer sessions with SCT as compared with PCT.
Rules for Scoring and Counting words
1.COUNT all intelligible words in context. 2.Delete statements made before or after task. 3.Cross out all words which are unintelligible, even with context. 4.Cross out all nonword fillers (e.g. um, er, uh) 5.Words included CAN contain sound substitutions, omissions, distortions, or additions if the word is intelligible in context. 6.Count commentary on the task. 7.Count filler words and phrases. 8.Count interjections and informal terms (e.g. uh-huh) 9.Count common contractions or simplifications of words. 10.Standard contractions count as TWO words. 11.Count each word in hyphenated words and each word in numbers. 12.Compound words are one word (e.g. pancake) 13.Each word counts in proper names. 14.Count acronyms as one word (e.g. VA)
How to complete LSA
1.Choose your type of language sample •Contexts •Descriptive •Narrative •Procedural •Conversational
Rules for Scoring and Counting CIUs
1.Count all words that are intelligible in context, accurate in relation to the picture/topic, and relevant to/informative about the content of the pictures. 2.Words do not have to be grammatically correct to count UNLESS the misuse obscures the meaning of the word/sentence. 3.Count all types of words (nouns, adjectives, pronouns, verbs, adverbs, articles, prepositions, and conjunctions). 4.Production of a non-target, intelligible word counts (e.g. school for stool). 5.Last attempt of the conduit d' approach counts. 6.Informal terms that convey meaning (e.g. Uh-huh). 7.Embellishments on the pictures (e.g. This is one accident-prone family!) 8.Verbs and auxiliary verbs are 2 separate CIUs (one for "is" and one for "done"). 9.Contractions are 2 CIUs (e.g. won't = will not) 10.Each word in hyphenated words (e.g. good-bye) •DO NOT COUNT: •Words that are not on topic to the picture/conversation. •Words that are not regionally-appropriate (e.g. calling a mid day meal "dinner") •Grammatically incorrect words that obscure the meaning of the word. •Attempts to correct sound errors in words except for the final attempt. •Dead ends, false starts, revisions that are not informational. •Repetitions of words/ideas that do not add new information and are not used purposefully to intensify meaning. •The first use of a pronoun for which an unambiguous referent has not been provided. •Vague, nonspecific word use which does not add meaning to the utterance •Conjunctive terms (e.g. so and then) •Qualifiers and modifiers if they are used as fillers (e.g. Apparently, this is a kitchen). •Filler words and phrases, interjections, and tag questions that don't add meaning •THE CONJUNCTION "AND" •Commentary on the task and lead in phrases •Commentary on the subject's performance/personal experiences.
The Western Aphasia Battery (WAB)
1.Information content rating 2.Fluency rating 3.Yes-no questions 4.Auditory Word Recognition 5.Sequential Commands 6.Repetition 7.Object Naming 8.Semantic Fluency 9.Sentence Completion 10.Responsive Naming
I. COGNITIVE BATTERY
1.Line bisection 2.Semantic memory 3.Word fluency 4.Recognition memory 5.Gesture object use 6.Arithmetic
Best
1.Participants provided with item to name. 2.If unable to name the picture after 5 sec, cue with a single phoneme plus schwa and/or single grapheme. 3.Next, first syllable of the word or C.V. if the target was monosyllabic. Cues provided approximately 5 sec apart. 4.Next, client given the word to repeat in the presence of the picture.
Diagnostic Methods of the Process Approach Clinician
1.Record and clearly describe every correct answer and error. 2. Note any related or unrelated behaviors that may influence performance 3. Note behavior surrounding the task to assess awareness of difficulties, orientation to task. (ugh- i know this.. this is hard!) Note all modalities of communication throughout test Carefully monitor client's behaviors between test items; note fatigue, decreased motivation, frustration, etc. Observe immediate environment and impact of client's performance, such as distractions.
Phonomotor Treatment
A phonologically based treatment to improve word retrieval issues in persons with aphasia •Modified from Lindamood Phoneme Sequencing Program (LiPS) (Lindamood & Lindamood, 1998) •Treatment mirrored on how we learn language early on •Based on connectionist models (Dell, Nadeau) •Good example of strengthening multiple linguistic networks across modalities sounds (flower -> F + L + OW + R) Distributed phoneme representation- "Press tongue against alveolar ridge and turn on voice"
MODERN MODELS- MODERN MODELS OF NORMAL LANGUAGE PROCESSES
A select few (explain different stages of language) 1.Functional anatomical model (Hickok and Poeppel) 2.Cognitive neuropsychological models •Interactive Activation (Dell, Schwartz, et al) •Parallel Distributed Processing (Plaut, McClelland; Nadeau, others)
Why know models of language processing?
Allows us to consider why and how those impairments occur. •To select more appropriate treatments. •To develop better treatments (researchers) •Researchers are still learning the "whys" and "hows" and (working with clinicians) how they can be applied to assessment and treatment. •So, we all continue to learn à improve treatment outcomes. • Evidence-based models to think critically about the client's impairment AND how it may be maximally remediated.
Lawton IADL Scale
Assessment of functional status "For each category, circle the item description that most closely resembles the client's highest functional level (either 0 or 1)."
Be aware of cultural/dialectal/generational differences
Be aware of cultural/dialectal/generational differences
Gary Dell (1997) Interactive activation model Lexical access in aphasic and nonaphasic speakers
Bidirectional; "network neighborhood"
phonomotor bidirectional
By strengthening phonological level representations, bidirectional spread of activation: •Bottom up to conceptual semantics •Top-down: when a word is activated, previously weak phonological representations will now be stronger and more easily accessible, Results in improved word retrieval
The Comprehensive Aphasia Test (CAT)
COGNITIVE SCREEN DISABILITY QUESTIONNAIRE LANGUAGE BATTERY •Comprehension •Aud, Written •Naming •Repetition •Oral Reading •Writing •Spoken Picture Desc •Written Picture Desc
Models of language processing
Classic model of language processing Modern models of language processing
LANGUAGE BATTERY CAT examples
Comprehension: Spoken words Comprehension: Spoken sentences Comprehension: Spoken paragraphs Comprehension: Written words Comprehension: Written sentences Naming objects Naming actions Semantic and phonemic word fluency Repetition: Digit strings Repetition: Words Repetition: Complex words Repetition: Nonwords Reading (aloud): Words Reading: Complex words Reading: Nonwords Reading: Function words Writing: Copying Writing: Picture names Writing to dictation Spoken picture description Written picture description
Phonomotor Treatment supported by 2 models
Connectionist model of phonological processing Interactive Activation model of language
What's the difference (perception and recognition) ?according to Hickok & Poeppel *perception*
Dorasl stream Activate and maintain sublexical items, like phonemes related articulation Perceiving speech sounds "rhughet" I perceive speech sounds (and I can say them)
Formal (Standardized) Assessment
Formal (Standardized) Assessment Developed with data that informs conclusions •Based on diagnostic or other criteria •(criterion-referenced assessment) •Based on population sample data •(norm-referenced assessment) •How does your client's performance compare with others in the test sample? •Results - statistically derived •percentiles, standard scores, etc. •Examples: Aphasia Batteries, Tests of Specific Linguistic Skills, Tests of Specific Cognitive Skills, Screens
I.Cognitive Battery II.Language Battery III.Disability Questionnaire
I. COGNITIVE BATTERY: all subtests II. LANGUAGE BATTERY 1.Auditory comprehension 2.Visual Comprehension 3.Repetion 4.Naming 5.Reading 6.Writing 7.Spoken picture description 8.Written picture description III. DISABILITY QUESTIONAIRE 1. Disability... 2. ...and its impact you can administer one or several of these subtest sections and compare your client's results against normed scores.
Comprehensive Aphasia Test
I.Cognitive Battery II.Language Battery III.Disability Questionnaire 2-3 hours to administer all 3 Subtest sections have norm-referenced scoring
Patient (and Family Member) Reported Outcomes (PROs)
Include... ●Communication Effectiveness Index ●Functional Communication Profile ●Lawton IADL Scale ●Communication Confidence Rating Scale ●Communication Participation Item Bank ●ASHA Functional Assessment of Communication Skills (FACS) ●Stroke Severity Quality of Life ●ASHA Communicative Quality of Life Scale ●Assessment of Living with Aphasia ●CAT Disability Questionnaire
Van Der Merwe (1997) Model of speech motor processing
Intention Linguistic - Symbolic Planning Motor-Planning phonologically based specifications, e.g. place, manner of articulation (movement goals) Motor-Programming Plan fitted to specific musculature initiating action (muscle involvement) Execution
LEXICAL RETRIEVAL TREATMENT
LEXICAL RETRIEVAL TREATMENT
Treatment is Learning
Learning is modifying the brain. Building new and strengthening established multimodal linguistic networks. But...we need to know which networks are established and which need strengthening.
Lexical Retrieval Impairments and Treatment
Lexical Retrieval Impairments and Treatment
MODELS OF LANGUAGE PROCESSING
MODELS OF LANGUAGE PROCESSING
Models of Language Processing &Lexical Retrieval
Models of Language Processing &Lexical Retrieval
A few cognitive neuropsychological models
Modular vs. Distributed
Naming actions
Naming actions based on photos of people doing things
CAT overview
Norm-referenced •Based on performance of a sample; scores tell you how 1 compares to many using a bell-shaped curve •T-scores with mean = 50, sd = 10 based on a large sample of PWA Subtest scores & error types used to: •Descriptive; profiles strengths and weaknesses by modality •generate hypotheses about underlying deficits (current models of language processing) More useful for treatment planning based on current literature and evidence: •e.g., treatments for naming, single-word comprehension, sentence comprehension and production, reading, script production
Communicative Effectiveness Index (CETI)
Please Rate ________'s ability at... 1. Getting somebody's attention. 2. Getting involved in group conversations that are about him/her. 3.Giving yes and no answers appropriately. 4.Communicating his/her emotions. 5.Indicating that he/she understands what is being said to him/her. 6.Having coffee-time visits and conversations with friends and neighbors (around the bedside or at home). 7.Having a one-to-one conversation with you. 8.Saying the name of someone whose face is in front of him/her. 9.Communicating physical problems such as aches and pains. 10.Having a spontaneous conversation (i.e., starting the conversation and/or changing the subject). 11.Responding to or communicating anything (including yes or no) without words. 12.Starting a conversation with people who are not close family. 13.Understanding writing. 14.Being part of a conversation when it is fast and there are a number of people involved. - 16 situations - Can be used with patients or caregivers - 1=extremely poor to 7=excellence
Research results of Phonomotor Treatment (PMT)
RCT: 58 PWA received 56-60hrs over 6 weeks of either PMT or SFA (randomized) (Kendall et al 2019) •Results: treatment-specific generalization effects immediate post-tx and 3-months post for BOTH treatments (PMT: phonological; SFA: semantic) PMT trial à discourse improvement (Silkes et al, 2019) PMT treatment response does not depend on impairment type, severity, age, or time post onset (Hunting Pompon et al, 2017 •PMT Phase 1/early case study (2006): 77yo female •54 years post stroke (post-labor/delivery) •Non-fluent aphasia; single words and simple phrases •Treatment: 1-2 hrs day/3-4 days week/6 months (74 hours total treatment) •Phonemes in isolation, combination, and non-words •Anecdotal: participant had always wanted to read a book to her child (now 54 years old). She was able to read a simple children's book to her son at the conclusion of therapy.
Examples of semantic-lexical treatments
Semantic Feature Analysis (SFA; Boyle, 2004) Verb Network Strengthening Treatment
Motor-Programming
Supplementary Motor Area (SMA) Basal ganglia Cerebellum Fronto-limbic system Motor Cortex Selection and sequencing of motor programs for movements of the muscles of articulatory structures.
WAB to measure change
The WAB has demonstrated sensitivity to change at the group level in published studies of a wide variety of aphasia treatments, including Shewan & Kertesz (1980) Minimal detectable change: •p < 0.2, 1-tailed: 6.2 AQ points •p < 0.1, 1-tailed: 9.4 AQ points •p < 0.05, 1-tailed: 12 AQ points
2 General Treatment Considerations
Treatment 1. Objective and 2. Generalization
Wernicke-Geschwind Reading aloud of written text
Visual signals -> angular gyrus -> Wernicke's area à arcuate fasciculus -> Broca's area -> motor cortex
intention/ linguistic symbolic planning
Voluntary fronto-limbic system Temp-parietal Broca's and adjacent Intent- motivation to communicate Message compiled Syntactic/Morphological planning Phonological planning
The Process Approach Clinician
We tend to think about assessment as test administration, but effective assessment is a process that encompasses many elements. by using a process approach to the quantitative exam (numbers) , we can obtain critical qualitative information about adaptive and maladaptive behaviors, as well as problem-solving strategies used by examinees.
Hearing Screen
Why do we screen hearing, specific to speech and language assessment? "SPEECH BANANA" High frequency hearing loss impacts assessment
Repetition
repetition of words Manual: "Verbal, phonemic, neologistic, and dyspraxic errors are NOT accepted [INCORRECT -2]. Dysarthric distortion is permissible provided it is clear that each phoneme within the word has been correctly selected..."
DORSAL STREAM
speech perception and production *perceiving speech sounds* mapping sound to sublexical/articulatory representations "speech sound/artic route" activate and maintain sublexical items (phonemes, related articulation) Left-hemisphere dominant. Acoustic information received and... A. Auditory sensory/phonological + motor interface Parieto-temporal boundary ("Spt") B. Articulatory network Posterior IFG Pre-motor cortex Anterior insula •Measures of speech motor impairment
VENTRAL STREAM
speech recognition *recognizing words* mapping sound to meaning / recognizing words / "word route" Access the lexicon and stored words Bilateral. Acoustic information received and... 1. Phonological + semantic lexical Posterior MTG & ITG 2. Higher-level syntax/ combinatorial network Anterior MTG & ITG ...before moving on to articulation... •Measures of impaired speech comprehension
Raymer and Ellsworth Semantic treatment
subject was asked to name the verb, (e.g. bake) then given immediate accuracy feedback. If incorrect, subject was cued with two questions designed to help develop information about the meaning of the target word •"Is this similar to grilling?" "Does this have to do with pie?
Raymer & Ellsworth- phonologic
subject was asked to name the verb, then given immediate accuracy feedback. If incorrect, subject was cued with two questions designed to help develop information about the sound of the target word •"Does 'pay' start with /p/?" •"Does 'pay' sound like 'way'?"
Oral Motor ExamAssessing CranialNerves
table 7.1 on slides
III. DISABILITY QUESTIONNAIRE
talking day to day do you feel isolated
Spoken picture description
tell me what's happening in this picture
Comprehension:Written Sentences example
the butcher shoots the nurse (choose photo with a butcher shooting the nurse)
measure review - error
• o Percentage of utterances with errors o Number of omissions o Number of error codes
Measures review- tanscript length
• o Total Utterances o C&I verbal utterances (complete and intelligible) o All words (including mazes) o Elapsed Time
SALT Transcript Entry ConventionsTaken from the 2016 Summary of SALT Transcription Conventions
• Transcript Format. Each entry begins with one of the following symbols. If an entry is longer than one line, continue it on the next line. •$ Identifies the speakers in the transcript; always the first line of the transcript. •Example: $ Child, Examiner •C Child/Client utterance. The actual character used depends on the $ speaker line. •E Examiner utterance. The actual character used depends on the $ speaker line. •+ Typically used for identifying information such as name, age, and context. •Example of current age: + CA: 5;7 •- Time marker. •Example of two-minute marker: - 2:00 •: Pause between utterances of different speakers. •Example of five-second pause: : :05 •; Pause between utterances of same speaker. •Example of three-second pause: ; :03 = Comment line. This information is not analyzed in any way, but is used for transcriber comments 2. End of Utterance Punctuation. Every utterance must end with one of these six punctuation symbols. . Statement, comment. Do not use a period for abbreviations. ! Surprise, exclamation. ? Question. ~ Intonation prompt. Example: E And then you have to~ ^ Interrupted utterance. The speaker is interrupted and does not complete his/her thought/utterance. > Abandoned utterance. The speaker does not complete his/her thought/utterance 3. { } Comments within an utterance. Example: C Lookit {C points to box}. Nonverbal utterances of communicative intent are placed in braces. Example: C {nods}. 4. Unintelligible Segments. X is used to mark unintelligible sections of an utterance. Use X for an unintelligible word, XX for an unintelligible segment of unspecified length, and XXX for an unintelligible utterance. Example 1: C He XX today. Example 2: C XXX. Bound Morphemes. Words which contain a slash "/" indicate that the word is contracted, conjugated, inflected, or pluralized in a regular manner. The root word is entered in its conventional spelling followed by a slash "/" and then the bound morpheme. /Z Possessive inflection. Examples: dad/z, Mary/z. Do not mark possessive pronouns, e.g., his, hers, ours, yours. /S/Z Plural and Possessive. Example: baby/s/z /ED Past tense. Predicate adjectives are not slashed. Examples: love/ed, die/ed, was tired, is bored /3S 3rd Person Singular verb form. Irregular forms are not slashed. Examples: go/3s, tell/3s, does /ING Verb inflection. The gerund use of the verb form is not slashed. Examples: go/ing, run/ing, went swimming /N'T, /'T Negative contractions. Irregular 7. Mazes. Filled pauses, false starts, repetitions, and reformulations. ( ) Surrounds the words/part-words that fall into 8. Omissions. Partial words, omitted words, omitted bound morphemes, and omitted pronominal clitics are denoted by an asterisk (*). * Following one or more letters this indicates that a word was started but left unfinished. Example: C I (w* w*) want it. * Preceding a word indicates that an obligatory word was omitted. Example: C Give it *to me. /* Following a slash the * is then followed by the bound morpheme which was omitted, indicating the omission of an obligatory bound morpheme. Example: C The car go/*3s fast. +* Following a plus sign the * is then followed by the Spanish clitic which was omitted, indicating the omission of an obligatory pronominal clitic. Example: C Él está gritándo+*le a la rana. Overlapping Speech. When both speakers are speaking at the same time, the words or silences that occur at the same time are surrounded by angle brackets < >. Example 1: C I want you to do it < > for me. Example 2: C Can I have that <one>? E <Ok>. E <Uhhuh>. 10. Linked words. The underscore "_" is used to link multiple words so they are treated as a single word. Examples include titles of movies and books, compound words, proper names, and words or phrases repeated multiple times. 11. Root identification. The vertical bar "|" is used to identify the root word. English uses: The root words of irregular verb forms such as "went" or "flew" are not identified. Linked words repeated for emphasis. Example: C The boy ran very very_very|very fast. Non-words used in error. Example: C He goed|go[EO:went] by hisself|himself[EW:himself]. Shortened words. Example: C He was sad cuz|because they left. 12. Sound Effects and Idiosyncratic Forms %. The percent sign is used to identify sound effects which are essential to the meaning or structure of the utterance. Non-essential sound effects are entered as comments. Strings of the same sound are linked together. Example 1: C The dog went %woof_woof. Example 2: C The dog barked {woof woof}. The percent sign is also used to identify idiosyncratic forms used by very young children. These are immature productions which are consistent in reference to an object, person, or situation. Example 1: C See %vroom {car}. Example 2: C My %coopa {cookie}. 13. Spelling Conventions. • Filled pause words: AH, EH, ER, HM, HMM, UH, UM, MM, and any word with the code [FP] • Yes words: OK, AHA, MHM, UHHUH (English & Spanish) • No words: NO, AHAH, MHMH, UHUH (English & Spanish) NAH, NOPE (English only) • Numbers (examples): 21 or TWENTYONE, 17 or DIECISIETE 14. [ ] Codes. Codes are used to mark words or utterances. Codes are placed in brackets [ ] and cannot contain blank spaces. Codes used to mark words are inserted at the end of a word with no intervening spaces between the code and the word. a) Codes used to mark errors in the reference database samples: [EO:__] used to mark overgeneralization errors. C He falled|fall[EO:fell]. [EW:__] used to mark other word-level errors. C He were[EW:was] look/ing. [EW] used to mark extraneous words. C And then the boy is a[EW] sleep/ing. [EU] used to mark utterance-level errors. C And they came to stop/ed [EU]. [FP] used to mark non-standard filled pause words. C The dog (um like[FP]) fell down.
Phonomotor Treatment Program
•Multi-modal articulatory-kinematic approach •Starting at sounds in isolation, building to single, multi-syllable combinations •Start with consonants then vowels •Progress to nonwords (VC, CV, CVC, CVCV) •Progress to 1 and 2 syllable real words
mixed error
•(lexical + phono) •Cat (for) rat
Cueing Hierarchies
•A structured approach to supporting response selection •A common form and component of aphasia treatment •May help strengthen semantic, phonologic networks •Some evidence of generalization •Raymer (1993) •Wambaugh et al (2001) •Raymer & Ellsworth (2002) •Best et al (2013)
Tucker Hierarchy of Support
•A systematic way to support communication •5 levels of support if no or ambiguous response: 1. Repeat 2. Restate more simply, review response options Does your aphasia impact talking with friends and family? 3. Re-explain task and response options This item (point) means your aphasia does not impact talking with people you know" and this item (point) means your aphasia impacts talking with people you know a lot." 4. Restate using Yes/No options Does your aphasia impact talking to people you know? Yes or No. 5. Move on to next question CPIB Item 1: "Does your condition interfere with... talking with people you know?"
Modular language models
•Also called 'box-and-arrow' or 'local' or 'discrete' •Representation: language units are stored locally •Processing: knowledge is processed via pathways in a serial fashion
Distributed language models
•Also called connectionist •Representation: language is represented in terms of learned patterns/networks of activation between different knowledge units Processing: knowledge is processed in an interactive manner; not in a sequential, serial manner Types of connectionist models Parallel distributed processing (PDP) Interactive activation
Word retrieval impairment in aphasia
•Anomia is the most common symptom of aphasia- across all types and severities •Word retrieval breakdown is variable, and due to difficulty accessing •Meaning (semantics) •Sounds (phonology) •or both • Note: variability is less seen in global aphasia, but still occurs
Some errors attributed to phonological level deficit
•Anticipation Substitution Exchange •Neologism Addition Deletion Perseverative •Formal errors: phonological errors, real words
Highlights- models
•Aphasia is an impairment of processing. •Models help us understand specifics of impairments and deepen clinical decision-making
Aphasia is an impairment of "processing" more than "representation"
•Aphasia is not "loss of language" (representation) •Aphasia is a disorder of linguistic access •Lexical retrieval is variable in aphasia •Representations are not gone, just inaccessible •Other cognitive processes necessary for retrieval •Connectionist models can help us characterize the impairment •Emphasize processing over representation •Help identify treatment stimuli and approaches
model APPLICATION
•Based on what you know so far about aphasia and language errors, how could you use (i.e. for your own understanding) this model in assessment of aphasia?
using the CAT to diagnose aphasia
•CAT: Cutoff for aphasic performance is 62.8, based on comparison between PWA and healthy control sample •Alternative cutoff at below normal performance on any 2 subtests, as defined in the manual •Per the manual, these criteria correctly classified 97-100% of healthy controls and 92% of PWA.
Core Lexicon MeasuresDalton & Wright, 2020
•Compares most frequently used words in story retell • •Does not (necessarily) require transcription • •Identifies impaired word retrieval and structural construct (macro/microlinguistic features)
II. LANGUAGE BATTERY
•Comprehension: Spoken words •Comprehension: Spoken sentences •Comprehension: Spoken paragraphs •Comprehension: Written words •Comprehension: Written sentences •Naming objects •Naming actions •Semantic and phonemic word fluency •Repetition: Digit strings •Repetition: Words •Repetition: Complex words •Repetition: Nonwords •Reading (aloud): Words •Reading: Complex words •Reading: Nonwords •Reading: Function words •Writing: Copying •Writing: Picture names •Writing to dictation •Spoken picture description •Written picture description
Parallel Distributed Processing Plaut, McClelland, Seidenberg & Patterson (1996)
•Connectionism: interconnected networks = representation •Strength of representation: dependent on network strength •More experience/familiarity with representation = Greater network strength ("connection weight") •Language must be PROCESSED (not just retrieved) more experience/familiarity = "greater connection weight" Networks in each modality are interconnected to each other, and connected across modalities
LSA Choose your protocol
•For SALT samples: •If you want to compare to database o Standard story o Copy of story with words for examiner o Story without words for client to use (story recall/retell) o Comprehension questions oRemember, understand, apply, analyze, evaluate, and create o •Contexts •Descriptive •Narrative •Procedural •Conversational
Lexical Diversity Measurement Fergadiotis, Wright, & West (2013)
•Moving Average Type Token Ratio (MATTR) •Measure of Textual Lexical Diversity (MTLD) • •Moving Average Type Token Ratio (MATTR) • •Strong indictors of lexical diversity (LD) • •Resistant to length effects • •Small variance, high validity • •High MATTR or MTLD = High level of lexical diversity
Discourse Analyses goals
•Goals of procedure •Contexts •Descriptive •Narrative •Procedural •Conversational •Target Collection •4-5 stimuli •300-400 words in length
•What might collaborators tell us that we can use?
•Health status, stroke details •Interactions so far •Status of other impairments, assessments, treatments •Preferences, priorities, personality, supports, challenges... •...
CLASSIC & MODERN Processing...Performance
•Hildred Schuell (1907-1970) Clinician & Professor, Univ of Minnesota Book: Aphasia In Adults (1964) •Aphasia is about language performance •Language is not lost in aphasia, but instead inaccessible •Aphasia treatment: "...get the language machinery going again"
assess tips
•If on the fence, code for the shorter utterance • •Different languages have different impactful measures of discourse analysis. • •Do your homework to ensure that you are using measures that best reflect your client.
1. Objective
•Impairment-focused vs. participation focused? •In observing confrontation naming response is the impairment... Semantic < or > Phonologic? (greater than or less than
Connectionist models
•Interactive networks of sublexical units •Networks connect w/i and between modalities •Strength of networks based on experience (learning)
Influential Variables
•Item-Related • • •Task-Related • • •Patient-Related
Discourse Measurement Protocols
•Lexical Diversity • •Correct Information Units • •Core Lexicon • •Main Concept Analysis
Modular models
•Localized representations, serial processing •Useful to consider some speech/language phenomena
Test Administration Pro Tips
•Make sure that you are qualified as a professional to administer. •Give yourself time to read the manual! •Carefully read directions about repetitions and rephrasing before starting •Practice administering the test before the evaluation. •Ensure that the person is medically stable. •Develop rapport before starting the test. •If the test book says, "Say: XYZ", Say EXACTLY what is written. •Make sure you have a good test environment and all materials BEFORE you start. ●Watch your timing - Do you have enough time to finish the test in one administration? ●General feedback- OK, specific feedback- NOT OK ●"Thanks" "Got it" NOT "Right!" or "good" ●You CAN give specific feedback on practice items, as indicated. ●Watch your presentation/pacing. ●Constantly monitor for impact of fatigue. ●Use your process approach for more comprehensive, holistic data. ●Do your work in advance for bilingual speakers or polyglots. ●Prep your interpreter ahead of time.
Main Concept Analysis (MCA)Nicholas and Brookshire, 1995
•Measures effective communication of the gist of a story •Verb/Subject/Object and related clauses (if applicable) coded. •Coding: •AC- Accurate/complete •AI- Accurate/incomplete (one or more essential elements not present, those elements present are correct) •IC- Inaccurate/complete (all essential elements provided, one or more element inaccurate) •II- Inaccurate/incomplete- (essential elements not produce, one of the elements produced is inaccurate )
HICKOK AND POEPPELDUAL STREAM MODEL (2007)
•Model (typical speaker) of two auditory speech processing networks, connected to BOTH motor and conceptual systems •Dorsal stream—mapping sound onto sublexical and articulatory-based representations "speech perception and production" •Ventral stream—mapping sound onto meaning "speech recognition"
Measures review narrative
•Narrative/Expository/Persuasion Structure Score (score on each aspect) • o Introduction o Character Development o Mental States o Referencing o Conflict/Resolution o Cohesion o Conclusion
Correct Information Units (CIUs)Nicholas and Brookshire (1993)
•Percent Correct Information Units • •Words per Minute • •Correct Information Units per Minute
Formulas
•Percent Correct Information Units: •(total number of CIUs)/(total number of counted words) •Words per Minute •((total number of counted words)*60)/(total number of seconds client speaks) •Correct Information Units per Minute •((total number of CIUs)*60)/(total number of seconds client speaks)
Examples of phonologic treatments
•Phonological Component Analysis (PCA; Leonard et al., 2008) •Phonomotor Treatment (Kendall et al., 2008; 2015; 2019)
Collaborators
•Physicians •Nurses •PT, OT •Family members!
Phonomotor trains phonological awareness
•Production and comprehension tasks •Target •Phoneme discrimination •Parsing/blending •Phoneme production and comprehension - both verbal and written •Start with nonword tasks before real word tasks
What is considered a test modification?
•Rewording and providing additional test instructions other than those allowed when presenting trial items •Providing additional cues or repeating stimuli on items that do not permit these •Allowing extra time for responses on timed subtests •Skipping items that are inappropriate (e.g., items with which the client has had no experience) •Asking the client for an explanation of correct or incorrect responses (when not standard procedure) •Using alternate scoring rubrics
Some errors attributed to lexical-semantic word production deficit-
•Semantic substitutions •Cat (for) dog •Knife (for) fork •Mixed errors / formal errors •(lexical + phono) •Cat (for) rat
CATEThe Complexity Account of Treatment Efficacy
•Semantically atypical vs. typical stimuli in semantic treatment (SFA) Greater generalization occurred when training atypical items •Example: shared semantic features activate together •Typical (widely shared) features: feathers, beak, lay eggs •Other features: only some birds run, swim, skinny legs Train word 'robin' won't generalize to ostrich Train word 'ostrich' will generalize to robin Contrary to common belief, treatment need not begin with simpler items before more difficult ones - and in fact, the opposite is true.
Aphasia Classification Using the WAB
•Wertz, Deal, & Robinson (1984) found in a sample of 45 PWA that aphasia classifications based on the WAB and the BDAE agreed in only 27% of cases. •Swindell, Holland, & Fromm (1984) found that WAB classification matched clinical impression in 37 of 69 (54%) of cases. •The fluency ratings on the WAB and the BDAE have been found to be unreliable (Gordon, 1998; Trupe, 1984). •In their analysis of fluency-related variables collected in the standardization of the Comprehensive Aphasia Test, Howard, Swinburn, & Porter (2004) found no evidence for a bi-modal distribution, i.e., for a fluent vs. non-fluent category.
•Patient Related
•Skills of the communication partner •Ability of the patient to use a clinician-produced/self-generated cue •What type of cueing is most effective •Level of cueing required for success
Case History
•Starts the moment you see them - why? -How do they respond to you calling their name? -What do you notice about movement, getting up out of chair, walking -- hints about impairment -Who else is with them? -Do they engage in chat, stay quiet? -NOTE: Be careful about engaging in chit-chat while walking; some patients cannot do both safely (motor impairment = concentration required)
item related
•Target word frequency •Age of acquisition •Length •Concreteness •Semantic category •Grammatical class •Degree of certainty •Operativity
CoreLex Elicitation
•Task instructions depend on discourse context
CoreLex Scoring
•Task instructions depend on discourse context • •Synonyms don't count • •Plurals/verb conjugations/inflections for target core lexicon count • •Target word only scores 1 point, regardless of frequency •Count each CoreLex word ONE TIME, regardless of the number of times used •Percent agreement: •Number of CoreLex Words used/Overall number of CoreLex wordsx 100
•Task Related
•Type of sensory input •Stimulus novelty •Length of trial and inter-trial time •Type of naming task •Concomitant gesturing •Verbal reinforcement •Attention condition •Word-retrieval success rate
double error
•Unicorn to horse to "house"
WAB and CAT summary
•WAB is more commonly used, familiar to many clinicians •WAB and CAT have comparable test-retest reliability •WAB is criterion-referenced - scoring is based on W-L diagnostic criteria • WAB matched clinical impression in only 54% cases •CAT is norm-referenced - scoring (overall; subtests) based on aphasia population norms •CAT more descriptive of impairments; useful for treatment planning •More "ceiling effects" with WAB than CAT (people with mild aphasia misclassified as Within Normal Limits) •CAT more sensitive measure of change than WAB •CAT takes longer to administer than WAB
Using the WAB to Diagnose Aphasia
•WAB: Cutoff for aphasic performance is 93.8, based on mean of sample with diffuse/subcortical brain damage •Wertz et al. (1984) found that this criterion correctly classified 40 of 45 (89%) of PWA. •The term "Not Aphasic by WAB (NABW)" has entered the lexicon of the clinical aphasia research community.
Important to consider the characteristics of the stimuli
•Word frequency •Word imageability •Word length •Lexicality
exchange
•blue moon to mue bloon
Anticipation
•blue moon to mue moon
Substitution
•moon to loon
●Standardization
○Based on population norms ○Reflects demographic characteristics of YOUR client
Validity
○Content ○Construct ○Ecological ○Criterion-related (predictive)
Reliability
○Intra and interexaminer reliability ○Test-retest reliability
Process Approach to Assessment of Aphasia- benefits
●Develop understanding of examinee's abilities ●Assess effective compensation strategies for weaknesses ●Determine effective supports for areas of weakness ●Identify cognitive-linguistic areas of strength ●Easier to identify these areas during the assessment
Note modalities of communication examples
●Gestures, writing, drawing, motoric cues ●Facial expression, eye direction
Goals of the Process Approach Clinician
●Observe all behavior ● Score productions according to standardized tests guidelines AND ● Analyze response patterns ○What is easy? ○What is challenging? Stimulability testing/informal probes
Psychometric Properties of Formal Assessment
●Standardization ●Reliability ●Validity Psychometric "psychological measurement" Objective measurement of knowledge, skills, attitudes, traits, performance
Oral Motor Exam
●What structures are we examining? ●What are we looking for?
Related behaviors example
●air-spells to self cue
What does the process approach include?
❖Includes quantitative AND qualitative data making it a holistic diagnosis ❖Maximizes the data obtained from assessment Our patients/clients are far more than their test scores - just like you.