Phonomotor Treatment:

Principles and Methods

JoAnn P. Silkes2
Megan Oelke 1, 2
Wesley M. Allen1, 2
Diane L. Kendall1, 2, 3

1 VA Puget Sound Medical Center, Seattle, WA

2 University of Washington, Dept. of Speech and Hearing Sciences, Seattle, WA
3 University of Pretoria, Pretoria, South Africa

This work was supported by VA RR&D Merit Review Grant #C6572R from the United States
(U.S.) Department of Veterans Affairs Rehabilitation Research and Development Service. The
contents do not represent the views of the U.S. Department of Veterans Affairs or the United
States Government.
Phonomotor Treatment manual
Page 1 of 116

Contents
Introduction .................................................................................................................................................. 5
What is Phonomotor Treatment (PMT)? ...................................................................................................... 6
Quiz module – Theoretical motivations for PMT ...................................................................................... 9
How is the PMT program structured? ........................................................................................................ 10
Quiz module – The structure of the PMT program ................................................................................. 12
What qualifications do I need to deliver PMT? .......................................................................................... 13
Client candidacy for PMT ............................................................................................................................ 14
What supplies are needed to deliver PMT? ................................................................................................ 15
Socratic questioning: A fundamental element of PMT ............................................................................... 16
What if my client has poor auditory comprehension? How can I implement Socratic questioning? .... 19
Quiz Module – Socratic questioning ....................................................................................................... 20
How to implement PMT .............................................................................................................................. 21
First session: Pre-treatment conceptual explanation ............................................................................. 21
Quiz module – Conceptual introduction to PMT ................................................................................ 24
A framework for introducing individual sounds in isolation................................................................... 25
Consonant training sequence ............................................................................................................. 25
Consonant family relationships .......................................................................................................... 26
Vowel training sequence..................................................................................................................... 26
Incorporating different modalities in treatment tasks ........................................................................... 29
Training single phonemes ....................................................................................................................... 30
Procedure for introducing consonants: the first cognate pair – p/b .................................................. 30
Procedure for training consonants: subsequent consonant sets ....................................................... 32
Procedure for introducing single vowels ............................................................................................ 33
Suggested tasks for reinforcing single phonemes .................................................................................. 34
Training sounds in combination.............................................................................................................. 38
Suggested tasks for training sounds in combination .............................................................................. 39
Anatomy of a treatment session............................................................................................................. 46
Anatomy of a treatment trial – merging tasks within a single trial ........................................................ 48
Introducing orthography......................................................................................................................... 49
Progression through the task hierarchy.................................................................................................. 50
Adjusting task difficulty and facilitating client success ........................................................................... 51
Treatment stimuli ....................................................................................................................................... 54
Phonomotor Treatment manual
Page 2 of 116

Devising non-word stimulus combinations............................................................................................. 54

Introducing real words ............................................................................................................................ 55
Choice of real words to incorporate into treatment .......................................................................... 55
Introducing and using real words in treatment .................................................................................. 55
Keeping data for PMT ................................................................................................................................. 56
Delivering PMT in the real world – practical considerations ...................................................................... 57
Case studies ................................................................................................................................................ 69
Case study #1: Morgan ........................................................................................................................... 69
Case study #2: Phil .................................................................................................................................. 71
Case study #3: Suzanne........................................................................................................................... 73
Case study #4: Lynn................................................................................................................................. 75
Appendix ..................................................................................................................................................... 78
Treatment example videos ..................................................................................................................... 79
Mouth pictures ....................................................................................................................................... 80
Blank vowel circle for use in therapy sessions ........................................................................................ 82
PMT graphemes ...................................................................................................................................... 83
Non-words to use in treatment .............................................................................................................. 84
Real word stimuli to be used in treatment ............................................................................................. 86
Examples of word and syllable chains .................................................................................................... 88
Examples of phonotactically legal consonant clusters in English ........................................................... 89
Examples of real words with conventional spelling in PMT orthography .............................................. 90
Phonemes with no features shared (differ on 3 features) – Easiest to differentiate ............................. 91
Phonemes with 1 feature shared (differ on 2 features) – Moderate difficulty to differentiate ............ 92
Phonemes that share 2 features (differ on 1 feature) – Most difficult to differentiate......................... 93
PMT Supplies Checklist ........................................................................................................................... 94
Client candidacy for PMT ............................................................................................................................ 94
Sample PMT Goals ...................................................................................................................................... 96
Sample Data Sheets .................................................................................................................................... 98
Glossary of Terms...................................................................................................................................... 106
Answers and Explanations for Quiz Modules ........................................................................................... 109
Quiz Module #1 – Theoretical motivations for PMT ............................................................................. 109
Quiz Module #2 – The structure of the PMT program.......................................................................... 110
Quiz Module #3 – Socratic Questioning................................................................................................ 111
Phonomotor Treatment manual
Page 3 of 116

Quiz Module #4 – Conceptual introduction to PMT ............................................................................. 113

References ................................................................................................................................................ 115
Phonomotor Treatment manual
Page 4 of 116

Table 1 - Descriptions of each consonant ................................................................................................... 25

Table 2 - Descriptions of each vowel .......................................................................................................... 28
Table 3 – Examples of treatment tasks by input and output modalities .................................................... 35

Figure 1 - Parallel Distributed Processing (PDP) model of language (reprinted with permission from
Kendall et al., 2015). ..................................................................................................................................... 6
Figure 2 - Illustration of the components of the multi-modality aspects of each phoneme's
representation. ............................................................................................................................................. 7
Figure 3 - A partial lexical network as represented in an interactive activation model. .............................. 8
Figure 4 - Overall general progression of tasks for each consonant in PMT. ............................................. 10
Figure 5 - Overall progression of PMT program.......................................................................................... 11
Figure 6 - Display of a complete set of supplies for PMT............................................................................ 15
Figure 7 - Examples of visual supports for Socratic questioning with clients who have poor auditory
comprehension. .......................................................................................................................................... 19
Figure 8 ....................................................................................................................................................... 21
Figure 9 ....................................................................................................................................................... 21
Figure 10 ..................................................................................................................................................... 21
Figure 11 ..................................................................................................................................................... 22
Figure 12 ..................................................................................................................................................... 22
Figure 13 - Explanation of a two-step interactive activation model of language. ...................................... 23
Figure 14 - Complete vowel circle ............................................................................................................... 27
Figure 15 - Examples of category label drawings created by different clients receiving PMT. .................. 32
Figure 16 - Full consonant array before graphemes have been introduced. ............................................. 32
Figure 17 - the full array of consonant mouth pictures .............................................................................. 33
Figure 18 - Complete desktop workspace for a task using multi-syllable stimuli....................................... 46
Figure 19 - Colored blocks spaced for blended vs. parsed sequences ....................................................... 51
Figure 20 - One way to keep previous stimuli in sight during a chaining task. ........................................... 52
Figure 21 - A second example of how to keep previous stimuli in sight during a chaining task. ............... 53
Phonomotor Treatment manual
Page 5 of 116

Introduction
Phonomotor Treatment (PMT) is a treatment for aphasia that is based in strengthening
phonological skills. It is multi-faceted and complex, so this manual has been designed to help speech-
language pathologists learn how to deliver PMT. It provides detailed explanations and examples that
you can use to learn the basics of the program, along with quiz modules to help you solidify and
integrate your understanding of PMT.
This is a comprehensive manual, intended to provide detailed information about the theory and
practice of using Phonomotor Treatment (PMT) for treatment of individuals with aphasia. Information
about PMT is already available through a number of avenues. To date, these include:

1) the research literature (Kendall, Hunting Pompon, Brookshire, Minkina, & Bislick, 2013; Kendall et al.,
2006; Kendall, Oelke, Brookshire, & Nadeau, 2015; Kendall et al., 2008), with the Kendall et al. (2015)
paper providing a quick reference guide for a speech-language pathologist wanting to implement this
treatment in the clinic;

2) A fee-based video course published through MedBridge, available for continuing education credit at
[Link]
aphasia-evidence-based-practice) (Disclosure: the second and last authors of this manual receive
royalties from this course);

3) Educational presentations at national and state speech-language pathology (SLP) conferences; and

4) Resources that have been made available on the University of Washington Aphasia Research Lab
website ([Link] This site is
updated regularly as new information and resources become available.

Our hope is that these resources, collectively, will provide you with the tools you need to be
successful with PMT.
We encourage you to take your time going through each section of this manual. Read the
information and examples, watch the videos that are linked throughout, and review the case studies
provided. Use the section quizzes to help you solidify and integrate your understanding of each aspect
of PMT. Practice the tasks, keeping the written and pictorial aids that are provided here nearby for
reference, and then review the relevant manual section and check your performance against what has
been described.
Thank you for your interest in Phonomotor Treatment!
Phonomotor Treatment manual
Page 6 of 116

What is Phonomotor Treatment (PMT)?

Phonomotor Treatment is an approach for treating impairments of phonology in people with
aphasia. We define aphasia as an acquired impairment of access to language representations and
processes that crosses all language modalities (verbal expression, auditory comprehension, reading, and
writing) and involves impairments in attention and working memory, as well. This section reviews the
models that PMT is based on. These models may or may not be familiar to you; those new to this
information may benefit from reviewing this section several times and/or reading the articles referenced
for these models. Having a solid understanding of these models will help make the rest of this manual,
and the PMT program, much easier to understand and explain to your clients.
Originally based on the Lindamood Phoneme Sequencing (LiPS) program for developmental
dyslexia (Lindamood & Lindamood, 1998), and using some of the same terminology, the underlying
assumption of PMT is that phonology provides the fundamental building blocks for all language
processing. This assumption is rooted in a parallel distributed processing model of phonology (Nadeau,
2001), an expanded version of which (Kendall, Oelke, Brookshire, & Nadeau, 2015) is illustrated in Figure
1.

Figure 1 - Parallel Distributed Processing (PDP) model of language (reprinted with permission from
Kendall et al., 2015).

This model posits that there are three primary domains involved in lexical processing: concept
representations (semantics), articulatory motor representations and acoustic representations. These
domains are all connected to each other through reciprocal connections, represented by the
bidirectional arrows between domains (the “hidden units” represented in the model serve a function
similar to old-time telephone operators, facilitating appropriate network connections between
domains). In addition, each domain comprises smaller networks that are all connected within the
domain, represented by the circular arrows attached to each bubble in the figure. Representation is the
result of the massive inter-connectivity of the network, with each word and concept representation
emerging from the co-activation of a neural pattern that involves elements from each domain. Two of
the domains represented above, articulatory motor and acoustic representations, collectively underlie
phonology. For individuals who have learned to read, orthographic representations are bound onto the
network in the same fashion, with the same extensive connections.
Because of the highly inter-connected and distributed nature of representation in this system,
treatment that focuses on one aspect of the system is expected to have far-reaching effects on its
Phonomotor Treatment manual
Page 7 of 116

connected elements in other domains. Specifically, because a limited number of phonemes and
phoneme combinations are used in each language, addressing the limited domains that support
phonology (articulatory motor and acoustic representations) would be expected to lead to improved
processing throughout the language system, since all domains are so heavily interconnected. Changes in
any domain should result in improved word retrieval because they are all critical subcomponents that
work together for this purpose.
PMT, therefore, is designed to facilitate and strengthen phonological representations and the
ability to manipulate them, in both production and perception, with the ultimate goal of improving
the efficiency, effectiveness, and accuracy of language processing across language tasks. This is
achieved through using a combination of tasks that reflect the interconnected nature of processing in all
domains, represented in Figure 2. Importantly, the focus of PMT is on improving phonological awareness
and phonological sequence knowledge in general. This is done by engaging visual, acoustic,
orthographic, motor, and tactile representations and processes to “exercise” phonology.

Visual

Orthographic Acoustic
/t/

Tactile
Motor kinesthetic
Figure 2 - Illustration of the components of the multi-modality aspects of each phoneme's
representation.

The star depicted in Figure 2 underlies, motivates, and guides all therapeutic tasks in PMT. All
tasks used in PMT address the domains shown here in a variety of ways:
- Visual tasks involve looking at the articulators (the SLP’s, the client’s in a mirror, and/or the
mouth pictures) and focusing on their specific positions and movements.
- Acoustic tasks involve listening to and discriminating between stimuli based on their
acoustic properties.
- Tactile kinesthetic tasks involve attending to and describing what a sound or sound
sequence feels like; they do not necessarily involve saying the stimulus.
- Motor tasks involve production of stimuli, either in repetition or in response to some other
stimulus.
- Orthographic tasks involve focusing on visual letter representations of sounds for both
reading and writing.
Phonomotor Treatment manual
Page 8 of 116

Each task targets at least one of these domains, and often several at once. Stronger domains are
routinely paired with weaker domains to build all aspects of the network. For instance, for someone
with good ability to recognize and describe what they see but a weaker ability to do the same for what
they hear, the clinician may have them look in the mirror to observe and identify notice oral movements
while also asking questions and giving examples that highlight and contrast what the phonemes they’re
producing sound like.
As a client progresses in treatment, it is appropriate to reduce the pairing between modalities to
provide opportunities to increase reliance on, and the strength of, each modality alone.
In addition to directly targeting these different linguistic domains, PMT also engages verbal
short-term and working memory processes – which are fundamental to language function – in a number
of ways (e.g., see Martin & Reilly, 2012). Tasks are designed to rely on these processes, gradually
becoming more difficult by varying the number of linguistic elements and processing domains being
used and manipulated, the amount and type of supporting cues being provided, and the length of time
various elements need to be retained to complete a task.
A second model that motivates PMT is the Interactive Activation Model of language (Dell &
O'Seaghdha, 1992). This model suggests that words are retrieved from a network that includes three
levels of information: semantic, word form, and phonological (see Figure 3). These levels have reciprocal
connections, so that activation at one level automatically spreads activation to related items at the
other levels. Final selection of words occurs when feedback between levels causes the target word to
have a higher level of activation than the other words surrounding it. Importantly for understanding why
PMT works on phonemes, not real words, is that phonemes are fundamental to assembling word forms,
and strengthening phoneme representations can influence multiple word forms because of each
phoneme’s bidirectional connections with all related words.

Figure 3 - A partial lexical network as represented in an interactive activation model.

Note that PMT is NOT intended to directly train the production and perception of specific single
phonemes; instead, it is intended to build phonological awareness and phonological sequence
knowledge. For many clinicians, this is a very different way to think about treatment. We encourage you
to participate in the ongoing discussion on the PMT Manual website to help you adjust to and
incorporate this new perspective on aphasia treatment.
Phonomotor Treatment manual
Page 9 of 116

Quiz module – Theoretical motivations for PMT

1. Phonomotor treatment is based on the idea that lexical processing occurs…
a. in a unidirectional manner with discrete units of information.
b. in networks of individual elements that are affected only by elements in their own
domain.
c. in networks that are massively interconnected, with representation resulting from the
co-activation of elements across domains.
d. in a serial, sequential manner with little interaction between elements.

2. PMT is rooted in the understanding that lexical processing is supported by…

a. discrete representations of words that contain all relevant information.
b. the interaction of information across sensory-motor domains.
c. a feed-forward system in which auditory and visual information are wholly independent.
d. a system that relies more heavily on semantic than phonological processing.

3. In PMT, various behavioral modalities are routinely…

a. paired between stronger and weaker modalities to facilitate improvement.
b. treated independently of each other.
c. treated in a sequential manner, with auditory information being established before
other domains are engaged.
d. separated to allow the client to focus on perfecting each domain.

4. The goal of PMT is to…

a. Teach functional words.
b. Teach specific sounds that are commonly used in English.
c. Build phonological awareness and phonological sequence knowledge.
d. Improve perception and production of single phonemes.

See answers and explanations for this quiz here: Quiz Module #1 – Theoretical motivations for PMT.
Phonomotor Treatment manual
Page 10 of 116

How is the PMT program structured?

PMT involves both production and perception tasks, beginning at the level of single phonemes
and moving up as appropriate to working with combinations of phonemes into syllables of increasing
length. Individuals with aphasia typically progress to 2- and 3-syllable stimuli over the course of 60 hours
of treatment. The general sequence of treatment progression is represented in Figure 4.

Figure 4 - Overall general progression of tasks for each consonant in PMT.

Note that several different phonemes will be in play at the same time, so several phonemes would be available for any task (e.g.,
to create a CVCC, three different consonants may be used). The dotted lines reflect critical transition points: some phonemes
may be used in combination with other phonemes while some are still being introduced in isolation, but no 2+ syllable sequences
should be introduced until all phonemes have moved out of the isolation stage. See text for details of how this is implemented.

While the general progression of treatment is from single phonemes to syllables, with
orthography introduced late in the program, there is not a strictly linear progression. Once a client has
learned a few consonants and is highly accurate (consistently ~80%) on a variety of tasks using those
phonemes, you may begin manipulating them in simple CV and VC contexts while still developing
knowledge of the remaining consonants. Think of it like learning to play an instrument: once you know a
few notes or chords, you can begin stringing them together to make music even as you’re learning other
notes or chords for the first time. This idea is discussed in greater detail in the section on Progression
through the task hierarchy, and is represented in Figure 5.
Phonomotor Treatment manual
Page 11 of 116

Figure 5 - Overall progression of PMT program.

Phonemes are first introduced in isolation, and as each is established it is moved into 1-syllable combinations while other
phonemes are still being introduced. Multi-syllable combinations are not used until all phonemes have been introduced.
Graphemes are generally introduced later in treatment but may be introduced earlier depending on client abilities.

To isolate and maintain a focus on phonology, treatment uses exclusively non-words initially;
real words are introduced late in the program to facilitate connections with concept representations.
Even when real words are used, however, the focus of treatment tasks remains on phonological aspects
of the words, with no attention paid to their meanings. Details of treatment tasks, cueing
methodologies, and progression hierarchies are provided below (see How to implement PMT, Socratic
questioning: A fundamental element of PMT, and Anatomy of a treatment session).
Phonomotor Treatment manual
Page 12 of 116

Quiz module – The structure of the PMT program

1. PMT begins with introducing single phonemes and then moves to working with phoneme
combinations.
a. True
b. False

2. Treatment focuses on production of specific sounds, and a sound is considered “mastered” when
the client produces it 100% accurately.
a. True
b. False

3. PMT uses only non-words in treatment.

a. True
b. False

4. Training of syllables only happens after all training of single phonemes is complete.
a. True
b. False

See answers and explanations for this quiz here: Quiz Module #2 – The structure of the PMT program
Phonomotor Treatment manual
Page 13 of 116

What qualifications do I need to deliver PMT?

Phonomotor Treatment is intended for use by certified, licensed speech-language pathologists
who have been adequately trained in its use. This training may include thorough study of the resources
provided in this manual, along with practice prior to administering the treatment to clients. We also
strongly encourage clinicians interested in using PMT to view the MedBridge course on Phonomotor
Treatment, and to take advantage of the materials and videos provided on the University of Washington
Aphasia Research Lab website ([Link]
lab/professionals) and linked in this training manual.
Phonomotor Treatment manual
Page 14 of 116

Client candidacy for PMT

Good candidates for PMT typically meet the following criteria:
- Have a phonological processing impairment as part of aphasia, as documented by the
presence of phonologic errors in a variety of linguistic contexts and/or performance on tests
of phonological processing;
- Can repeat at least some single sound stimuli with cues;
- Are willing and able to be actively engaged in the treatment;
- Have sufficient auditory comprehension skills to participate in Socratic questioning;
- Are able to self-reflect (i.e., identify errors and change behavior to fix them);
- Have adequate speech motor programming, planning, and execution skills to participate in
verbal treatment tasks;
- Have good executive function, including abstraction and set shifting.

In our experience, people who exhibit the following characteristics do not respond well to PMT:
- They do not have good buy-in, usually because they do not understand or like the concept
of a treatment using primarily nonwords;
- They are unwilling to actively engage in the treatment tasks;
- They cannot be facilitated to repeat single sounds accurately;
- They have untreated depression.

PMT has been developed specifically for, and tested with, people with aphasia due to a left
hemisphere stroke. No data are available that support its use with other etiologies; it may or may not be
appropriate to use with other diagnoses that may yield acquired language impairments, such as
traumatic brain injury or primary progressive aphasia.
Phonomotor Treatment manual
Page 15 of 116

What supplies are needed to deliver PMT?

A clinician delivering PMT should have:
- A small mirror
- Laminated photos of mouth postures for each phoneme in the language
- Laminated icons for voiced and voiceless consonants
- Laminated blank vowel circle chart
- Grapheme tiles reflecting the orthographic system used in PMT (described below)
- A wipe-off board, markers, and eraser
- Small colored blocks – 3-4 blocks of each of 6 colors. Other colored objects, such as
laminated paper squares, crayons, or beads, may also be used if blocks are unavailable.
- Colored felt squares – approximately 4” x 4” – they must be large enough to be able to lay
out colored block or letter tile sequences on top of them
- Blank index cards

Also recommended:
- A clipboard for laying out the vowel chart, so it can be removed from the table quickly and
easily without disturbing the array.
- A second set of laminated mouth pictures, to allow the full consonant array to remain in
view while also having pictures available to manipulate.
- A second set of grapheme tiles to allow replication across stimuli presented simultaneously
- A “reset” button – this is typically just a small piece of paper with a red circle on it labeled
“reset”, which can be used as a physical and visual cue when the clinician or client feels the
need to abandon an ongoing attempt at a task and start fresh. This may not be used with all
clients, but some who are highly perseverative or persistent benefit from its presence.
- A large covered plastic bin for storing and transporting all treatment supplies.
- A tray to hold grapheme tiles (e.g., Scrabble racks).

Figure 6 - Display of a complete set of supplies for PMT

Phonomotor Treatment manual
Page 16 of 116

Socratic questioning: A fundamental element of PMT

Socratic questioning is a process of asking leading questions to help a person discover truths on
their own. A real-life example of Socratic questioning can be seen in this exchange between a parent
and a child:
(Child is walking out the door to the school bus stop wearing socks, carrying shoes. It is raining.)
Parent: I see you’re heading out to the bus.
Child: Yep.
Parent: What’s the weather like?
Child: It’s raining.
Parent: What do you have on your feet?
Child: Socks.
Parent: What do you think will happen to your socks if you go out in the rain like that?
Child: Oh – they’ll get wet.
Parent: What happens if your socks get wet?
Child: They’ll be wet all day at school.
Parent: Is that going to be comfortable?
Child: Not really.
Parent: So what can you do to avoid that?
Child: I guess I should put on my shoes.

While any parent can tell you that this exchange is idealized, you can see in this example that,
rather than just telling the child what’s wrong and what to do about it, the series of statements and
questions leads the child through the logical steps needed to come to a conclusion about what they
should do. The process began with helping the child focus on and state the obvious aspects of the
situation and then moved toward helping them synthesize the information to make appropriate
behavioral change.
Similarly, PMT uses Socratic questioning as the primary form of both stimulus presentation and
clinician feedback to client responses. Rather than the clinician providing all of the information, or telling
the client whether they got something right or wrong, Socratic questioning is used to help the client
discover and understand the information and evaluate and correct their own responses. Questions may
encourage explanation and recall of concepts, analysis of stimuli, or analysis of their own or the
clinician’s productions or responses.
Socratic questioning in PMT typically serves one of three purposes:
Type of question Examples
Questions to encourage What kinds of things can make sounds different from each other?
explanation, exploration, What do you remember from last time about this vowel chart?
and recall
Questions for analysis of Is this a quiet or a noisy sound?
stimuli What would your mouth do for this sound?
What did my lips do?
Were these the same or different?
Where should your jaw be?
How many syllables, or beats, does that word have?
You said there were three sounds in that syllable. What were they?
Questions for analysis of Did we match?
responses (both correct and Was the sound you said the same as the sound I said?
incorrect) I said “p” and you said “b”. Same or different?
Phonomotor Treatment manual
Page 17 of 116

As much as is possible, clinicians should guide clients through treatment tasks and feedback
through a series of questions that the client answers themselves, rather than having the clinician
providing most of the information and right/wrong feedback. If you are doing most of the talking,
something is wrong!
It is important to recognize that the process of Socratic questioning involves asking multiple
questions in a sequence that will guide the client to self-discovery. The various types of questions
presented here will always be used in conjunction with each other to facilitate this process. Questions
should generally be tied to physical (tactile-kinesthetic), acoustic, or visual cues that focus the client’s
attention on relevant aspects of the stimulus, task, or their own behavior. For example, these cues could
include:

- Looking in the mirror to see their own articulator placement (using visual cues to support
motor and acoustic modalities)
- Watching the clinician’s mouth (using visual to support acoustic modality)
- Feeling their own or the clinician’s neck for vocal fold vibration (using tactile-kinesthetic cues
to support acoustic modality)
- Focusing on specific articulator movements (e.g., Is your tongue moving? Is it the front of
your tongue or the back of your tongue? Is your neck vibrating?) (using tactile-kinesthetic to
support visual and/or acoustic modalities)
- Plugging their nose (to determine whether a sound is a “nose sound”) (pairing tactile-
kinesthetic and auditory cues, possibly to support visual modality)
- Listen to a model or their own production for specific features (e.g., Is it quiet or noisy? Do
these sounds match? What’s the difference between these two words?) (pairing acoustic
and motor modalities to reinforce explicit phonemic awareness)

Because Socratic questioning focuses on the client’s attention on one or more aspects of the
sound and its connections across domains, as represented in Figure 2, it is used to connect weaker with
stronger modalities. For instance, if a client has difficulty with recognizing acoustic characteristics of
phonemes but is particularly good at identifying their visual aspects, questioning may link the two, as
follows:

(Target phoneme: /p/)

SLP: “p”. What did my lips do? (FOCUS ON VISUAL ASPECTS)
Client: They came together and then opened.
SLP: Look in the mirror and you say it.
Client (looking in the mirror): “p”
SLP: What did your lips do? (FOCUS ON VISUAL ASPECTS)
Client: Came together and then opened.
SLP: Now listen to it again. “p”. Was it a long sound or a short sound? (FOCUS ON ACOUSTIC
ASPECTS)
Client: “p”. A short sound.
SLP: What did the air do? (LINK THE VISUAL AND ACOUSTIC ASPECTS)
Client: It stopped and then came out when my lips opened.
SLP: And what about your voice? Put your hand on your throat and say it again. Was this sound
quiet or noisy? (FOCUS ON ACOUSTIC AND ARTICULATORY KINEMATIC)
Client (with hand on throat): “p”. It’s quiet.
Phonomotor Treatment manual
Page 18 of 116

In the event that a client gives an incorrect response to a question, the clinician may narrow the
questions to focus even further on fewer or more concrete aspects of the task, such as:

SLP: Listen to this sound: “p”. Is it a long sound or a short sound?

Client: A long sound. (INCORRECT RESPONSE)
SLP: Listen to it again: “p”. Does the air stop, or does the air keep going when I say it?
Client: The air keeps going. (INCORRECT RESPONSE)
SLP: Put your hand in front of your mouth and you say it: “p”. (PAIR AUDITORY AND TACTILE)
Client (with hand in front of lips): “p”.
SLP: Did the air flow smoothly or did it stop and then puff out? (FOCUS ON THE SAME ASPECTS
IN MORE DETAIL)
Client: It stopped and then puffed out.
SLP: So is it a long sound, where the air keeps moving, or a short sound, where the air stops?
Client: It’s a short sound.

If the client were to still get it incorrect after adding the tactile cue, the exchange could end like
this:
SLP: Did the air flow smoothly or did it stop and then puff out?
Client: It flowed smoothly. (INCORRECT RESPONSE)
SLP: Put your hand in front of your mouth again and say “p”, “h”.
Client: “p”, “h”.
SLP: Do you feel that “p” has a puff of air and “h” has the air just flowing the whole time?
Client: Yes.
SLP: That puff of air means it’s a short sound. It’s short because you can’t keep it going. Let’s try
another one…

Remember that the goal is not 100% accurate production; it is increasing awareness and
knowledge of phonology. Incorrect responses provide an opportunity to build awareness and knowledge
that can then be applied to different targets.
The focus of Socratic questioning will shift as treatment progresses; questions will be more
exploratory at first, to establish and strengthen awareness of phonemic representations, and then will
shift to become more analytic as skills and representations improve. At any level, Socratic questioning
initially aims to link stronger modalities with weaker modalities, allowing the weaker skills to improve by
“bootstrapping” on the stronger. Later in the treatment program, Socratic questioning may focus on a
single modality to encourage its further strengthening and development. Examples of Socratic
Questioning in action can be found in all videos linked throughout this document. Examples of feedback
and cueing that are NOT Socratic Questioning, and are to be avoided, can be seen at
[Link]

For better understanding of the idea of pairing modalities, refer back to the section on What is
Phonomotor Treatment (PMT)? and ahead to the section on Suggested tasks for reinforcing single
phonemes.

At later stages of treatment, Socratic questioning may be less directive and more open-ended.
Questions for higher level clients could include:
- Listen to these two words. Are they the same or different?
o What sound is different between them?
o How are those two sounds different?
 Phonomotor Treatment manual
Page 19 of 116

o What does your mouth do for that sound in the first word? The second word?
- What sound in this word changed?
o How are the old sound and new sound different?
o Show me (using blocks or grapheme tiles) which sound changed.

For any of these, Socratic questioning may return to more a basic, feature-directed focus if the client has
difficulty with these more open-ended questions or does not capture all appropriate aspects in their
response.

What if my client has poor auditory comprehension? How can I implement Socratic
questioning?
A client with poor auditory comprehension can still participate in PMT, although the Socratic
questioning process will need to be adjusted. Some strategies to consider include:

- Supplement questions with key word writing, drawing, or gesture.

- Supplement questions with icons and symbols found in the PMT materials (e.g., the quiet/noisy
icons, mouth pictures, letter tiles, etc.).
- Make questions shorter and simpler.
- Ask yes/no questions with concrete choices, or forced choice questions with a limited set of
linguistic or non-linguistic response options, rather than requesting complex descriptions (see
examples in Figure 7).

Figure 7 - Examples of visual supports for Socratic questioning with clients who have poor auditory comprehension.
The drawing on the right represents "lips spread" versus "lips puckered".
Phonomotor Treatment manual
Page 20 of 116

Quiz Module – Socratic questioning

1. Socratic questioning involves…
a. Using carefully crafted questions to guide the client through exploration of phonemes,
stimuli, and their own responses to treatment tasks.
b. Giving explicit feedback on every production that a client makes.
c. Having the client ask questions about each sound and sound sequence to learn about it.
d. Giving multiple choice questions after each task is complete so that the client can assess
how they did.

2. Socratic questioning for a single trial usually involves just a single question.
a. True
b. False

3. The clinician should always use the same questions for the same tasks during PMT.
a. True
b. False

4. Socratic questioning evolves over the course of treatment, depending on the client’s strengths,
challenges, and current level of success.
a. True
b. False

5. Socratic questioning should facilitate linking relatively stronger with relatively weaker
modalities.
a. True
b. False

6. A client with impaired auditory comprehension is not a good candidate for PMT, because
auditory comprehension needs to be fully intact to participate in Socratic questioning.
a. True
b. False

7. The best way to support a client’s ability to engage in the Socratic questioning process is to…
a. Provide written key words.
b. Simplify questions.
c. Provide yes/no or multiple choice questions.
d. All of the above.

See answers and explanations for this quiz here: Quiz Module #3 – Socratic Questioning
Phonomotor Treatment manual
Page 21 of 116

How to implement PMT

First session: Pre-treatment conceptual explanation

PMT is unusual in that it focuses on sub-lexical elements and does not introduce real words until
late in the treatment process. Since most people with aphasia have experienced treatment focused on
functional vocabulary, and it is counterintuitive to not work on real words when the ultimate goal is
improved verbal communication, it can be difficult for clients to understand the purpose of extensive
exercises that do not incorporate real words at all. Therefore, in the first treatment session, before
treatment begins, it is important that the clinician explain the purpose of the treatment approach in a
way that is accessible and clear to the client and any caregivers or family members who may be
involved. You may also find it useful to revisit this explanation during the treatment program if your
client seems to have lost motivation or understanding of the purpose of the treatment approach.
This process begins with drawing a rough sketch of a brain (see Figures 8-12) and using this,
along with gestures toward your own head and the client’s head, to explain that:

1) Different parts of the brain are involved in different things

- Broca’s area plays an important role in speech
- Wernicke’s area plays an important role in auditory
comprehension
- the angular gyrus plays an important role in letter processing
- the occipital lobe is important for visual processing
- the sensorimotor cortex is important for feeling and movement Figure 8

(Note that you do not need to name each area as you point it out
and describe it, if you think that will be too much information for the client.)

2) These areas are usually connected to each other and

communicate with each other – they all work together to make
language happen.

Figure 9

3) The front of the brain is also involved, working as a conductor to

let you think about language and keep it all running smoothly.

Figure 10
Phonomotor Treatment manual
Page 22 of 116

3) When you have a stroke and it causes aphasia,

some of those connections are damaged or broken.

Figure 11

4) The goal of therapy is to work on making those connections

strong again.

5) That’s why there will be so many different types of tasks Figure 12

involved – they all help the whole network.

To assist comprehension and memory, create the drawing as you speak, rather than just using a
completed drawing and pointing to it. This helps maximize client understanding by breaking the ideas
into manageable pieces, adding each element as it’s discussed, and providing multi-modal
communication support.

The second concept to be discussed in this session is why treatment focuses on sounds in non-
words. The points to include in this discussion are typically:

1) All of the tasks used in this therapy are based on working with speech sounds.

2) Sounds are the building blocks of all language processing. When you learned to speak, you
started with babbling individual sounds and then learned how to string them together. Now,
when you speak, listen, read, or write, you use this knowledge of speech sounds and how to
combine them.

3) Aphasia leads to problems with using speech sounds and being able to string them together
into larger units, like words and sentences, that make up connected language.

4) The goal of this treatment is to make those speech sounds in your brain, and to make it
easier for you to manipulate them and string them together. If we worked with real words,
your brain would rely on the meanings of the words to help process them, and it wouldn’t
have to work as hard on the sounds. So we’ll be using made-up words so that your brain has
to focus on the sounds.
Phonomotor Treatment manual
Page 23 of 116

5) It takes a lot of practice for your brain to get better at doing this, so treatment is going to
involve a lot of repetition and observation.1

6) (Refer back to drawing) Every sound has lots of different things that help describe it – what
it feels like to say it, what it sounds like, what it looks like to say it, what letters represent it
in writing.

7) So our treatment activities will focus on all of these different aspects of the sounds, making
the connections in your brain stronger.

It can be useful to accompany this discussion with a simplified drawing of a lexical network from
a 2-stage interactive activation model of language (e.g., Dell, 1986), particularly for higher level clients.
This drawing should highlight that phonemes, word forms, and semantics are connected but separate, in
a network that lets information flow both ways between levels. This drawing helps to show how working
with phonemes in non-words can have an influence on processing of real words. For an example of this
type of drawing, see Figure 13. This drawing can then be used as needed throughout therapy if the
client needs reminders about why treatment is focused on non-words.

Figure 13 - Explanation of a two-step interactive activation model of

language.

In the case of lower level clients, this model of word retrieval is not typically appropriate to
present at the start of therapy. Instead, when non-words are first introduced after a few hours of
treatment, a simple “Non-words help us find real words”, accompanied by a simplified version of this
drawing, may suffice.

You can see examples of how this initial conversation before treatment begins may unfold at
[Link]

1
Pro tip: Some clients are inclined to try to analyze and understand every aspect of every task, but this can distract
from treatment. You can return briefly to basic explanation, but work to keep them focused on doing the tasks
rather than understanding them.
Phonomotor Treatment manual
Page 24 of 116

Quiz module – Conceptual introduction to PMT

1. PMT is intuitive enough that you can jump into treatment without any preliminary explanation.
a. True
b. False

2. The explanation and drawing that represent the purpose of PMT should be provided at the start
of each treatment session.
a. True
b. False

3. The explanatory drawing that represents the purpose of PMT should be created anew for each
client as you describe the purpose of the treatment, rather than having it pre-drawn.
a. True
b. False

See answers and explanations for this quiz here: Quiz Module #4 – Conceptual introduction to PMT
Phonomotor Treatment manual
Page 25 of 116

A framework for introducing individual sounds in isolation

The first step in treatment is to train each sound in
the language in isolation. Initial sound training is designed
to build a strong representation of all aspects of a
phoneme, including knowledge of its articulatory shape,
tactile-kinesthetic aspects, voicing status, sound, and
orthographic representation. In addition, a framework for
organizing sounds that are similar is also introduced early in
treatment. All consonants are introduced first, followed by
all vowels. The descriptions and cues introduced early in
treatment become the framework for cueing and Socratic
questioning that will be used throughout the treatment program, and will be used repeatedly
throughout treatment progression to reinforce knowledge or facilitate progression to more complex
stages of processing.

Consonant training sequence

Consonants are trained in cognate pairs in the following sequence, organized by place or
manner of articulation, with characteristic descriptors as noted in Table 1.

Table 1 - Descriptions of each consonant

Consonant Sound category Characteristics

p/b Lips Air stops, lips together and pop open, puff of air,
quiet (p)/noisy (b)
f/v Lips Continuous air, top teeth on bottom lip (bite the
bottom lip), quiet (f)/noisy (v)
t/d Tongue Air stops, tongue tip taps up behind front teeth,
puff of air, quiet (t)/noisy (d)
k/g Tongue Air stops, back of tongue goes up and down (taps
the back of the palate), puff of air, quiet (k)/noisy
(g)
th/th Tongue Continuous air, tongue between teeth (like you’re
biting your tongue), quiet (th)/noisy (th)
s/z Air* Continuous air, “skinny” (or “thin”) air, lips back
(smile), teeth closed, quiet (s)/noisy (z)
sh/zh Air* Continuous air, “fat” (or “big”) air, lips rounded,
teeth together, quiet (sh)/noisy (zh)
ch/j Air* “Fat air, burst” - Air stops, lips rounded, teeth
together, puff of air, quiet (ch)/noisy (j)
l/r Tongue placement Continuous air, tongue lifts up (small, front vs. big,
back)
m/n/ng Nose Continuous air, sound comes through nose, noisy,
close off the air from coming out of the mouth so it
has to come out the nose (close off at lips, front of
tongue, and back of tongue).
h/w/wh Windy Continuous air, quiet, mouth wide open. For w/wh:
two mouth movements (from puckered to open)
Phonomotor Treatment manual
Page 26 of 116

*When training air sounds, use the back of your/the client’s hand to provide tactile feedback for airflow.

Consonant family relationships

The lip, tongue, and air sounds are often referred to as “brother” sounds, because they all fit nicely into
articulatory categories. The remaining consonants are more abstractly grouped, and are often referred
to as “cousin” sounds. These terms are used throughout this manual as an easy way to differentiate the
two broad groups. You can see how this may be explained to clients at
[Link]
phonemes.

Vowel training sequence

Before beginning formal training on vowels, explain that you’re going to start now with a new
set of sounds. The sounds you’ve been doing up until now have been consonants, and this new set is
vowels. There is no need to provide detailed explanation of what makes a sound a consonant or a vowel;
simply labeling them is enough.
Vowels are trained by introducing the extremes of the vowel triangle (see Figure 14) –
contrasting the three extreme corner sounds (ee, o/au/aw, and oo2, known as anchor vowels) first, and
then building gradations between them, from ee down to o/au/aw and then up to oo. Monophthongs
are taught first, and then rhotics and diphthongs are introduced. The grapheme tiles for vowels are used
from the beginning of their training, as there are no mouth pictures for them.
The training sequence, with descriptive characteristics, is presented in Table 2. Along with the
descriptive characteristics of lip shape, the SLP should demonstrate and reinforce changes in jaw height
for each vowel by placing their hand under their own jaw, and having the client do the same, as vowels
are contrasted verbally. Brief descriptions of tongue position and movement may be included, as well.
Examples of how vowels are introduced and practiced may be seen at
[Link]
phonemes.
Note that some people have a very difficult time with the vowels, often due to apraxia of speech
and the lack of concrete articulatory details as compared with consonants. If a client is doing well with
vowels, then it is worth spending time on them. If a client is really struggling, though, due to
concomitant apraxia of speech, then it is appropriate to just establish the three anchor vowels (ee,
o/au/aw, and oo), and then incorporate the others in other tasks without the specific detail to jaw
height, lip rounding, etc. that would have been introduced in more thorough direct vowel training.

2
NOTE: Unless otherwise noted, all examples of phonemes and words presented in this manual are written using
PMT orthography using italics. Consonants are all consistent with regular English orthography (see Table 1), but
vowels differ (as described in Figure 14 and Table 2). References to specific letters of the alphabet are written in
capitals with single quotations around them. Real English words given as examples in standard English spelling are
written in lower case letters with single quotation marks.
Phonomotor Treatment manual
Page 27 of 116

Figure 14 - Complete vowel circle

Phonomotor Treatment manual
Page 28 of 116

Table 2 - Descriptions of each vowel

PMT IPA Pronunciation Characteristics

orthographic representation
representation

MONOPHTHONGS:

Anchor vowels
ee* /i/ bee, sea, me Jaw high, lips in a smile, tongue high
o / aw / au** /a/ and/or /ᴐ/ cot, saw, caught Jaw all the way down, lips loose and
open
oo /u/ food, do, stew Jaw high, lips very round

Intermediate
vowels
From ee down
to o / aw / au
(smile to open):
i /I/ pit One step down the ladder from “ee”
e /ɛ/ red One step down the ladder from “i”
ae* /eI/ ray, sleigh, obey One step down the ladder from “e”
a /æ/ cat One step down the ladder from “ae”
u /ə/ or /ʌ/ cup, some One step down the ladder from “a”
From o / aw /
au up to oo
(open to
round):
oe* /o/ home, boat, comb One step up the ladder from “ah”
uu.. /ʊ/ put, could One step up the ladder from “oo”

DIPHTHONGS
and RHOTICS:
ie* /aI/ ride, sigh Slider; start with “a” and slide to “ee”
ue* /ju/ use, you, cue Slider; start with “ee” and slide to “oo”
oy / oi /ɔI/ boy, join Slider; start with “uu” and slide to “ee”
ow / ou /aʊ/ cow, mouth Slider; start with “a” and slide to “oo”
er / ir / ur /ɚ/ her, fir, fur R vowel
or.. /or/ door R vowel
ar.. /ar/ jar R vowel
* If a client has difficulty remembering the sounds associated with these graphemes, some people
understand it and remember it better if they are told that having an ‘E’ after the vowel makes it say its
own name. For instance, a letter ‘A’ written with a letter ‘E’ after it (i.e., ae) says the name of the letter
‘A’, ‘O’ with an ‘E’ after it (i.e., oe) says the name of the letter ‘O’, etc. Laying out the vowels ‘A’, ‘E’, ‘I’,
‘O’, and ‘U’ in canonical order can help to demonstrate this point.
Phonomotor Treatment manual
Page 29 of 116

** In some dialects, the vowel in ‘saw’ and ‘caught’ is different from the vowel in ‘cot’. In these cases,
you may choose to use aw and au interchangeably to reflect the former and o to reflect the latter.
Alternatively, some clients will automatically name the letter ‘O’ when you show the grapheme o, in
which case it may be more effective to only use aw and au for all instances. If a client is confused by the
three different ways of representing this vowel, it generally helps if you put all three out at the same
time and circle them to indicate that they’re interchangeable.

Incorporating different modalities in treatment tasks

One final concept that’s important to clarify before thinking about specific procedures in PMT is
how different modalities (think of the yellow star presented in What is Phonomotor Treatment (PMT)?)
get incorporated and manipulated within a single task.
As discussed previously (see the section beginning on page 6), a primary objective of PMT is to
pair modalities to help the stronger support the weaker, with the ultimate goal of having all modalities
strong enough to be able to support themselves. As a general rule, the more modalities a single task
incorporates, the easier it should be. As a client progresses through PMT, multi-modality supports can
be reduced, with the highest level, most challenging tasks relying on just a single modality.
For any task in PMT, you should think separately about a) What is its primary goal? and b) How
you are going to achieve that goal? Your answers to these questions will reflect where the client is in the
program and what their relatively stronger and weaker modalities are; the modalities used to achieve
the tasks may differ based on these answers. For instance, consider a task that involves parsing a
syllable into its component phonemes (e.g., “bif”  b, i, f). The primary objective of this task is the
ability to recognize and isolate phonemes in a longer string. There are many ways this task can be
presented, though, to achieve the same outcome. For a client who is doing this for the first time, the
input may be auditory, visual, and orthographic; you may place colored blocks representing the three
sounds on the table in front of them, place the appropriate orthographic tiles in front of each block, and
say the word for them while scanning your finger across the blocks so that you’re pointing to each sound
as it’s pronounced (see the section on Suggested tasks for training sounds in combination for details on
tasks using blocks). This presentation combines modalities to give visual and orthographic support to the
auditory task of phoneme discrimination. With a higher level client who is further along in the PMT
program, you may present the same task with just an auditory stimulus and the blocks laid out. For
someone even more advanced, you may remove the blocks altogether and have them do the task using
only the auditory modality for input and the verbal modality for output. In all of these cases, the primary
objective remains the same but the way it is achieved differs based on the client’s needs.
In the following sections on PMT procedures, tasks are described in terms of their primary
objectives and their input and output modalities. Keep in mind that the input and output modalities may
differ as you adjust tasks within and between sessions and clients. As long as you are focusing on multi-
modality awareness of phonology and phonological sequences, you are still operating within the
framework of PMT.
Phonomotor Treatment manual
Page 30 of 116

Training single phonemes

Procedure for introducing consonants: the first cognate pair – p/b

This is the general procedure for introducing the first cognate pair (p/b), during which you also
introduce many of the basic elements of PMT, including explaining the mouth pictures, the quiet/noisy
icons, and first use Socratic questioning. This procedure should be adjusted as needed based on the
client’s response. Examples of how this procedure may be implemented can be seen at
[Link]
phonemes.

1. Demonstrate the sound while the client just watches and listens.
2. Ask the client to tell you what they saw moving when you made the sound.
3. Have them say the sound following your model.
4. Have them look in the mirror while they say the sound a few times.
5. Restate the movement that they said they saw, and ask for more detail if appropriate (e.g.,
“You lips are moving. Can you be more specific?)
6. Ask about other articulators (e.g., “Is your tongue doing anything when you make that
sound?”).
7. Introduce the mouth picture, tell the client that it represents that sound, and have them say
the sound 10 times while looking at the mouth picture.
8. Introduce the voiced cognate – Have them say the sound a few times, including while looking
in the mirror.
9. Ask them to describe what they saw happening when they made the sound (i.e., it’s the same
as the other sound).
10. Reinforce that the sounds look alike, but they sound a little different.
11. Have them place their hand on their throat and make the two sounds.
12. Ask them what they felt happening with the second sound and then use their response to
explain “noisy” sounds. Introduce “noisy” icon and place it on one the bottom right corner of
the mouth picture.3
13. Have them say the second (voiced) sound again 5-10 times while looking at the combined
mouth picture/noisy icon.
14. Tell them that the first sound had no moving, no vibration at all, so it’s a quiet sound.
Introduce “quiet” icon and place it bottom left corner of the mouth picture.
15. Have them say the quiet sound 5-10 times, while looking at the combined mouth
picture/quiet icon.
16. Have them say each sound 10 times in a row (p then b), then 5 times each, then 3 times
each, then alternating 1 time each.
17. Have them listen to you say one and they choose which you’ve said (random order).
18. Have them say each sound as you point to it, with no verbal model (random order). If they
cannot do this, have them do blocks of 5 repetitions again with a model, then move on to the
next cognate pair.

3
Pro tip: Use h vs. o / aw / au to demonstrate the concepts of quiet vs. noisy.
Phonomotor Treatment manual
Page 31 of 116
Phonomotor Treatment manual
Page 32 of 116

Procedure for training consonants: subsequent consonant sets

Introduce each consonant pair in a similar way, with appropriate descriptors (see Table 1,
above). When a client has been introduced to all of the cognate pairs in one of the three primary
articulatory groups (lips, tongue, air), give them a blank index card and ask them to draw a picture to
represent that group in a way that makes sense to them (see Figure 15). Then place that card at the top
of the column of mouth pictures representing that group. That label card should be used every time the
array of mouth pictures is presented.

Figure 15 - Examples of category label drawings created by different clients receiving PMT.
The pictures in the column on the left represent lip sounds, those in the middle column represent tongue sounds, and
those in the column on the right represent air sounds.

A complete array of consonant sounds, as it would be displayed to a client who has been
introduced to all consonants can be seen in Figure 16 (before graphemes are introduced, using the
“quiet” and “noisy” icons to differentiate voicing cognates) and Figure 17 (once graphemes have been
introduced).

Figure 16 - Full consonant array before graphemes have been introduced.

Phonomotor Treatment manual
Page 33 of 116

Figure 17 - the full array of consonant mouth pictures

laid out by sound group, once orthography has been introduced.

Procedure for introducing single vowels

Once a client has learned a few consonants and is highly accurate (consistently ~80%) on a
variety of tasks using those consonants, initial training may begin on the three anchor vowels: ee, o / aw
/ au, and oo. Trained consonants can then be presented in CV and VC contexts using these vowels, while
continuing to develop knowledge of the remaining consonants. In this case, the vowels are introduced
only through demonstration and brief teaching of the associated mouth pictures and/or graphemes;
detailed analysis of the motor movements is deferred until vowel training officially begins.
Once all consonants have been introduced, it is time to start focused training with vowels. This
training begins with the anchor vowels, but this time they are fully introduced and explored, much in the
same way that consonants were taught (though the motor descriptions are quite different). Once the
anchor vowels are established, the intermediate vowels are introduced, moving from ee down to o / aw
/ au (i, ae, e, a), and then from o / aw / au up to oo (see Table 2).
Vowel training follows the same general format, and uses the same types of tasks, as consonant
training; however, the focus is on jaw height and lip rounding, rather than articulator position and
voicing. Tongue position is generally not addressed, given that this parameter is not as prominent for
vowels as for consonants.
Vowel training typically occurs over approximately 5-6 hours of treatment in consecutive
sessions. In the first 1-2 hours, because vowels have fewer obvious distinguishing features than
consonants, the clinician provides a significant amount of guidance to explore the properties of each
vowel. This guidance evolves over time into Socratic questioning with more client-led exploration.
Once all vowels have been introduced, this knowledge should be reinforced by putting the
vowels in VC syllables with the “brother” consonants. The “cousin” consonants may also be used, but
they tend to change the nature of the vowel (e.g., nasalizing the o in om vs. op) so are less clear and may
be more difficult for some clients. To begin introducing VCs, start with one vowel and move from
consonant to consonant. Once you have gone through all of the consonants, have the client repeat the
vowel in isolation 5 times and then move to another vowel. Note that this process will create some real
words (e.g., ich is pronounced as ‘itch’ and ash looks and is pronounced like ‘ash’). This is not a problem;
Phonomotor Treatment manual
Page 34 of 116

just treat them as you would any other phoneme sequence, with no focus on the semantic aspects of
the word).

Suggested tasks for reinforcing single phonemes

Once a phoneme has been successfully introduced, it is important to reinforce that phonological
knowledge across tasks and modalities. As described earlier relative to Figure 2, this is done through a
variety of perception and production tasks, which may be used in combination. The object is to have
each task isolate one modality, or link a stronger with a weaker modality (if you haven’t already, this
would be a great time to review the section on incorporating multiple modalities in PMT before
proceeding). As such, each task can be described, and should be implemented, based on its input and
output modalities, as demonstrated in Table 3.
Phonomotor Treatment manual
Page 35 of 116

Table 3 – Examples of treatment tasks by input and output modalities

Output
Stimulus analysis and
Verbal Visual – Visual – Grapheme Visual – Blocks and/or Felt
judgment tasks
Mouth pictures tiles Writing squares
Acoustic - Repetition - Clinician says a - Clinician says a sound - Clinician says a - Clinician says a sound or - Listen to two sounds or
- Elision (“Say bis sound or sequence, or sound sequence, sound or sound sound sequence, client sound sequences and
without b”) client finds client finds sequence, client identifies corresponding determine if they are the
- Auditory corresponding mouth corresponding writes corresponding number and color of same or different, if they
discrimination picture(s) grapheme tile(s) grapheme(s) blocks / felt squares rhyme, or if they are real
tasks (e.g., words or not. The visual
parsing, modality can be included or
blending) not based on whether the
client can see the clinician’s
mouth during production.
Visual – - Clinician shows - Clinician shows a - Translation to PMT - Translation from - Clinician shows a series
Orthographic a grapheme or grapheme or orthography: Clinician PMT orthography: of grapheme tiles that
Input

grapheme grapheme sequence, writes a real word in Clinician shows a create 1 or more syllables,
sequence, client client finds the standard spelling and sequence in client identifies the
says the corresponding mouth client uses tiles to spell grapheme tiles that corresponding number
corresponding picture(s) using PMT orthography would be and color of blocks / felt
sound(s) (e.g., jeans  jeenz) pronounced like a squares
real word, client
writes the
corresponding real
word (e.g., jeenz 
jeans)
Visual –Mouth - Look at a mouth - Look at a mouth - Look at a mouth
pictures picture (or series picture (or series of picture (or series of
of them) and say them) and select the them) and write the
the corresponding corresponding
corresponding grapheme tile(s) grapheme(s)
sound(s)
NOTE: These are just sample tasks defined by basic input-output relationships. In reality, any given task may merge several input and output
modalities, depending on the client’s strengths, weaknesses, responsiveness to cues, and current level of success.
Phonomotor Treatment manual
Page 36 of 116

Examples of common single phoneme tasks include:

- TASK: Listen to and watch the SLP say a sound and select the correct sound from an array of
mouth pictures or on the vowel circle. This array will get larger as the client has more
trained phonemes.
PRIMARY PURPOSE OF TASK: To build understanding of the relationship between speech
sounds and mouth movements.

- TASK: Listen to and watch the SLP say two sounds and determine whether they are the
same or different from each other.
PRIMARY PURPOSE OF TASK: To develop the ability to discriminate between speech
sounds.

- TASK: Listen to the SLP say a sound and select the appropriate grapheme. This task will be
used only later in the treatment program, once grapheme tiles have been introduced (see
section below on Introducing orthography). Variant: have them match graphemes to mouth
pictures.
PRIMARY PURPOSE OF TASK: To strengthen connections between speech sounds and
orthography.
Phonomotor Treatment manual
Page 37 of 116

- TASK: Sound repetition

PRIMARY PURPOSE OF TASK: To strengthen connections between speech sounds and their
associated motor movements.

- TASK: Produce a sound in response to a mouth picture (with voicing icons, as appropriate)
PRIMARY PURPOSE OF TASK: To strengthen connections between motor movements and
production of speech sounds.

- TASK: Produce a sound in response to a description of a motor pattern (e.g., a noisy sound
with your lips popping apart)
PRIMARY PURPOSE OF TASK: To encourage explicit awareness of the motor movements
involved in speech sound production.

- TASK: Produce a sound in response to a written grapheme (after orthography has been
introduced)
Phonomotor Treatment manual
Page 38 of 116

PRIMARY PURPOSE OF TASK: To strengthen connections between orthography and speech

sounds.

In each of these tasks, Socratic questioning should be used to encourage analysis of stimuli (particularly
if an error was made) and assessment of client responses.

You can watch examples of introducing, practicing, and reinforcing single phonemes at
[Link]
phonemes.

Training sounds in combination

Once a phoneme has been well established in
isolation, across a variety of single-phoneme tasks, activities
should be introduced that implement and reinforce that
phoneme in progressively longer, more complex contexts,
such as combining and manipulating phonemes in sequences.
As tasks become more difficult and longer phoneme strings
are introduced, the same fundamental principles used for
single phoneme training are maintained, such as Socratic
questioning and pairing weaker with stronger modalities. A few new tools (colored blocks, felt squares)
are also introduced later in treatment, to represent the more complex sound combinations in a concrete
way (you can read more about these tools in the next section, Suggested tasks for training sounds in
combination). When a client has difficulty with a task, you may revert to the same kinds of questions or
tasks used during single-phoneme training (e.g., Are your lips moving? Is it a quiet sound or a noisy
sound?) to get them back on track with individual phonemes and then incorporate that single phoneme
knowledge more strongly at the syllable level. Training beyond single phonemes begins with single
syllables (VC, CV, CVC, CCVC, CVCC) and then progresses to multi-syllable items.
As explained earlier, PMT was designed to focus on phonology, with the goal of improving
phonological processing and phonologic-semantic connections across the entire language network, in all
domains, rather than to train specific words. As discussed earlier, the treatment primarily employs non-
words to allow clients to focus on phonology without assistance from their preserved semantic
knowledge. Real words are incorporated once syllables have been introduced, however, to strengthen
the links between phonology and semantics for the sake of maximizing treatment effects on functional
language use. Please see the section on Introducing real words for additional details and explanation.
It is important to note that you will begin working on sounds in combination for some
phonemes, particularly those introduced first, while still training at the single-phoneme level with
others. The one exception is that all individual phonemes (consonants and vowels) should be introduced
before you begin with multi-syllabic stimuli and tasks.
Phonomotor Treatment manual
Page 39 of 116

You can see examples of tasks involving sounds in combination in single syllables at
[Link]
syllables, and in multi-syllable stimuli at [Link]
sounds-in-combination.

Suggested tasks for training sounds in combination

As with single-phoneme tasks, every task that involves sounds in combination can be described,
and should be chosen, based on its input and output modalities (see Table 3). Tasks for sounds in
combination may include one syllable or multiple syllables, keeping in mind (as mentioned earlier) that
multi-syllable stimuli should not be used until all individual phonemes have been introduced. Examples
of common tasks for sounds in combination include:
- TASK: Listen to the SLP say a sound sequence and choose all of the mouth pictures that
occur in the sequence.
PRIMARY PURPOSE OF TASK: To reinforce individual phoneme knowledge in the context of
permissible sound sequences and increasing short-term memory demands.

- TASK: Listen to the SLP say a sound sequence and lay out the corresponding grapheme tiles.
PRIMARY PURPOSE OF TASK: To reinforce the association between phonemes and
graphemes in the context of permissible sound sequences and increasing short-term
memory demands.

- TASK: Sound or syllable chains - Listen to the SLP say a sound sequence and lay out colored
blocks to represent the different sounds4. In this task, it does not matter which color

4
Before beginning tasks that use blocks, the clinician should introduce the idea that each block represents a
sound. This is done by placing a colored block on the table and saying a single sound. You then remove the block
Phonomotor Treatment manual
Page 40 of 116

represents which sound; however, if the same sound occurs more than once in a sequence
(e.g., bebee), it should be represented by using blocks of the same color in that sequence
(e.g., bebee could be represented by the block sequence red-yellow-red-green). Once an
initial sequence is on the table, the clinician can ask the client to manipulate the blocks to
represent different sound combinations (e.g., ib to ish; less difficult) or present a new
sequence (e.g., ib to eef; most difficult). For instance, a less difficult task of manipulating the
sequence already present would be:
o If that says bebee, make it say febee (client would change the red block for another
color not already in use)
o If that says febee, make it say feb (client would remove the green block)
o If that says feb, make it say fub (client would swap the yellow block for another
color not already in use)
o If that says fub, make it say fuboo (client would add a different color block on the
end)
PRIMARY PURPOSE OF TASK: To provide practice manipulating phonemes in the context of
phoneme sequences with increasing working memory demands.

- TASK: When stimuli begin to include more than one syllable, an extra layer is added to tasks.
The colored felt squares are used to represent syllables (generally described in PMT as
“beats”). Introduction of syllables is the only task in the program that purposefully deviates
from the proscribed trained word lists. Typically syllables are introduced using familiar one
syllable real words, then spondees (e.g., ‘hotdog’, ‘mailman’, ‘baseball’, etc.), followed by 2-
syllable real words. The initial real words used are not necessarily from the list of proscribed
words but, after the initial introduction of the concept, practice should move to using these
words. Once the task has been established with real words, non-words from the proscribed
list should be included.
o Before introducing 2-syllable stimuli, these squares are introduced as follows:
 Explain that words have beats.
 Say a one-syllable target (e.g., ‘cat’) and tap the table one time as you say it,
and have the client do the same.
 Say a two-syllable target (e.g., ‘baseball’) and tap the table two times, once
with each beat. Have the client do the same.

and put out either the same block or a different block with a different sound. Continue doing this for 4-5 trials until
the client understands that each block represents a sound. If needed, you can explicitly state that each block
represents a sound, and that a block might represent a different sound every time it comes out.
Phonomotor Treatment manual
Page 41 of 116

 Provide a few more one- and two-syllable targets and have the client tap
out the number of beats in each one, providing assistance as needed.
 Show the felt squares and explain that they are used to show how many
beats a word has.
 Put one felt square on the table, and tap it as you say a one-syllable target.
Have the client do the same.
 Put down a second felt square to the right of the next one (from the client’s
perspective), and tap the two in sequence as you say a one-syllable target.
 Have the client practice tapping out one and two syllable targets as you say
them.
 Once this is well established, introduce placing the blocks on top of the felt
squares, with the blocks from each syllable on the appropriate square.
You can see examples of this process at
[Link]
practicing-multi-syllables.
- PRIMARY PURPOSE OF TASK: To establish understanding of syllabic structure as part of
phoneme sequence knowledge.

- TASK: Sound sequence repetition (single- or multi-syllable)

PRIMARY PURPOSE OF TASK: To strengthen connections between sound sequences and
their associated motor movements, with increasing short-term memory demands.

- TASK: Produce a sound sequence in response to mouth pictures

Phonomotor Treatment manual
Page 42 of 116

PRIMARY PURPOSE OF TASK: To strengthen connections between motor movements and

production of speech sounds in the context of sound sequences.

- TASK: Produce a verbal sound sequence in response to a written sound sequence (non-word
or real word written in PMT orthography; only after orthography has been introduced)
PRIMARY PURPOSE OF TASK: To reinforce grapheme-phoneme associations in increasingly
complex sequences.

- TASK: Listen to the SLP say a sound sequence, and break it down (parse it) into its
component phonemes, either verbally or with colored blocks, grapheme tiles, or mouth
pictures.5
PRIMARY PURPOSE OF TASK: To increase awareness of individual phonemes in the context
of complex sound sequences and increasing short-term memory demands.

5
Pro tip: This task can be made more complex by using multi-syllable stimuli. The same is true for blending and
elision tasks.
Phonomotor Treatment manual
Page 43 of 116

- TASK: Listen to the SLP say two or more separate phonemes and blend them together to
form a multi-phoneme sequence, either verbally or with colored blocks, grapheme tiles, or
mouth pictures. Input could also be written letters or grapheme tiles rather than verbal.
PRIMARY PURPOSE OF TASK: To increase awareness of, and facility with, forming sound
sequences.

- TASK: Elision - Clinician provides a target syllable and the client produces that syllable
without a given phoneme (e.g., “Say bis without saying b“). Note: This task relies heavily on
verbal working memory and is best reserved for higher level participants. It may be done
with real-word or non-word stimuli.
PRIMARY PURPOSE OF TASK: To provide practice with phoneme manipulation in the
context of sound sequences and increasing working memory demands.
Phonomotor Treatment manual
Page 44 of 116

- Complex verbal working memory tasks:

o TASK: Syllable comparisons and transitions (you can see examples of this task at
[Link]
practicing-multi-syllables) –
 Clinician provides a target word (or non-word) and asks the client how many
sounds are in the word.
 The client lays out the correct number of colored blocks (and grapheme tiles
for a lower level client)
 The client then produces the target, parses it, blends it back together for 3
repetitions.
 The same sequence is then repeated for a second target word that differs
from the first one on a single phoneme (e.g., peef and peeg), while keeping
the blocks/grapheme tiles from the first stimulus in view.6 The clinician uses
Socratic questioning as needed to draw the client’s attention to the relevant
changes and facilitate accurate productions.
 Once both sequences have been established, the clinician facilitates having
the client move between the two targets, using this format (NOTE: the
clinician should not say the two target words again; the client must rely on
their own memory throughout this task):
 The clinician says “tell me the two words” (Client says peef, peeg)
 “Now touch and say the old word” (Client physically touches each
block of the old word as they parse it (p, ee, f) and then scans their
finger across the whole set as they blend it (peef).
 “Now touch and say the new word” (Client physically touches each
block of the old word as they parse it (p, ee, g) and then scans their
finger across the whole set as they blend it (peeg).
 “So you take out ___” (Client responds with f, indicating which
sound was removed from the first word)

6
Pro tip: Typically, this task would be done without any memory supports because it is intended to push the
boundaries of working memory. If you have a low level client for whom you think it would be beneficial, though,
you may add in supports initially and then fade them over time. For example, you can do a whole sequence of this
task with support (letter tiles or pictures or both) and then repeat the entire sequence with those supports
removed.
Phonomotor Treatment manual
Page 45 of 116

 “And you put in ___” (Client responds with g, indicating which

sound is new in the new word)
 “Now tell me the new word” (Client says peeg)
 As a final step, the clinician may remove letter tiles, if present, and have the
client reconstruct the words.
PRIMARY PURPOSE OF TASK: To integrate and synthesize phoneme and phoneme
sequence knowledge, and independence in phoneme manipulation, in a highly
complex verbal working memory task.

o TASK: Syllable transposition – Clinician provides two syllables and asks the client to
reverse them. This could be designed such that the client’s production is a real word
(e.g., bee-bae  baebee [‘baby’] or a non-word (e.g., fee-tae  taefee). Felt
squares and grapheme tiles may be used to provide visual cues as needed.
PRIMARY PURPOSE OF TASK: To practice manipulation of phoneme sequences
(rather than individual phonemes) in a complex working memory task.
Phonomotor Treatment manual
Page 46 of 116

Anatomy of a treatment session

Every treatment session after the first one (see above for details of that session) should begin
with a 5 minute warmup to give some initial success, re-establish the task mindset, and reinforce skills
that have already been learned. This warmup time may include a variety of tasks that the client has
already completed successfully, including repetition, matching of sounds to mouth pictures, or saying
sounds associated with the various mouth pictures.
Early in the treatment program, many clients benefit from actually setting up the mouth picture
and vowel tile stimuli in their ordered arrays (see Figure 16 and Figure 17 for the complete mouth
picture arrays without and with graphemes, and Figure 18 for a sample complete workspace setup); this
task helps to reinforce the properties, similarities, and differences of the individual phonemes and
phoneme groups. While some clients are able to lay out the entire array independently in just a minute
or two after they’re familiar with it, others will need more help to do it in a reasonable time. This task
can be scaled back to meet individual needs by providing partial information or structure, such as laying
out the sound group labels that they’ve drawn (see Figure 15), providing the anchor vowels on the
vowel circle (see Figure 14), or having them lay out only a portion of the items independently after you
have set up a number of them already. As the client progresses in the program and there is less of a
focus on the mouth pictures, this initial setup may be faded.

Figure 18 - Complete desktop workspace for a task using multi-syllable stimuli.

Note that the blocks and felt squares are the primary manipulative for the task being introduced, but all
tools, including pictures, vowel tiles, and mirror, are within view to be used as needed.

Within the bulk of a session, the exact structure will differ from participant to participant (the
case studies provided at the end of this manual highlight the wide variety of forms that treatment
sessions can take). Some people do well with extended periods of time working on a single task, while
others benefit most from frequent changes of activity. For instance, one participant may do well doing a
long period of one production task and then a long period of one perception task, while another may be
Phonomotor Treatment manual
Page 47 of 116

more comfortable and successful with going back and forth multiple times between production and
perception tasks. The exact content of any given session (i.e., which tasks to do and which phonemes to
target) will be based on a variety of factors, including the syllable length and complexity they are ready
for, their abilities with blocks vs. letters, and what skills they’re struggling with vs. strong with. Each
session should end with a few minutes of a task on which the client is likely to be successful. Keep in
mind that the goal of treatment is to improve phonologic awareness and phoneme sequence
knowledge, so whatever tasks are used (perception, production, orthography, etc.) will work toward
that goal.
Phonomotor Treatment manual
Page 48 of 116

Anatomy of a treatment trial – merging tasks within a single trial

Unless you are working on isolating a single phoneme or element, a single treatment trial
generally intersperses multiple tasks that reinforce the knowledge being learned and manipulated. For
instance, in a syllable chain task in which the client is asked to change a sequence of blocks from
representing babee to representing fabee (described earlier), the clinician may use Socratic questioning
to have the client explain the change that they made (e.g., “Which sound changed between words?
[“The first sound”] … “What parts of the mouth do you use for the new sound?” [“Lips and teeth”]
“Point to the mouth picture/letter that goes with the new sound.”). The clinician may then ask them to
parse the new word into syllables and/or sounds (e.g., “What’s the new word? [fabee] ... Tell me each
syllable in the new word [fa… bee] ... Tell me each sound in the new word.” [f…a…b…ee]). Then the
clinician may have them blend the sounds back together to bring them back to the full word before
presenting the next stimulus in the sequence.
This combination of tasks and levels of processing can and should be done at nearly every level
of treatment once the initial phonemes are introduced. Doing so links all domains of language
representation (i.e., visual, acoustic, orthographic, motor, and tactile kinesthetic), a fundamental goal of
PMT.
Phonomotor Treatment manual
Page 49 of 116

Introducing orthography
As described earlier, for people who have learned to read, orthographic knowledge is
fundamentally embedded in the lexical network, heavily connected with phoneme knowledge. For some
people with aphasia, orthography remains a relatively preserved modality, so they are able to use
orthography to support phoneme retrieval. For these people, orthography should be introduced
relatively late in the program, after a phoneme has been trained in isolation and is being used in
combination. This will allow the client to focus on the information carried by other modalities to build
strong phoneme representations without relying on orthographic knowledge. This may feel
counterintuitive, as most aphasia treatment encourages relying on stronger modalities to assist with
functional communication. In this case, though, it can interfere with the goal of treatment, which is to
strengthen weaker modalities. Other people with aphasia have poorly preserved orthographic
knowledge. These people may benefit from introducing orthography earlier in the program so that
orthographic knowledge can be strengthened concurrently with other modalities.
To determine whether a client has strong or weak orthographic representation, the SLP can do a
brief assessment of residual orthographic knowledge early in the treatment program. This assessment
may include asking the client to write, or point to the letters that go with a few phonemes, or say the
sound (i.e., NOT the letter name) that goes with a letter presented. Again, if someone is weak in
orthography, the SLP may choose to introduce the letters earlier in the program; if someone is strong in
orthography, letters should be introduced later so that the client can’t overly rely on that modality to
the exclusion of focusing on others.
Whenever the graphemes are introduced, there are a few options for how to approach this task,
and the SLP may use some combination of techniques such as:
- Using Socratic questioning to ask the client which letter goes with a given sound
- Giving the client a letter tile and asking them to place the letter on the corresponding mouth
picture (blocking by type of sound, such as lip, tongue, and air sounds).
- Saying a sound and asking the client to write the letter that goes with the sound.
Regardless of the method used to introduce the graphemes, the SLP should show it to them if
they do not get it correct initially. As each letter tile is introduced, it is placed on the appropriate mouth
picture in place of the “quiet” and “noisy” icons. The letter tiles are then included in all future activities
that involve the mouth pictures. The appropriate graphemes can be reinforced across sessions by having
the client lay them out at the start of each session.
You can see examples of introducing and using orthography at
[Link]
Phonomotor Treatment manual
Page 50 of 116

Progression through the task hierarchy

It may help to think of the program as linear, or sequential, for each phoneme but not globally;
you may be working on single-syllable tasks for some phonemes while still introducing and establishing
some of the individual sounds in single-phoneme tasks (see Figure 5). While PMT presents a hierarchy of
stimulus and task difficulty in theory, treatment may not progress directly through this hierarchy. Most
clients will be ready to start working with a few early-established phonemes in short syllables while
continuing to introduce other individual phonemes. This frequently happens when a client has mastered
some phonemes at the single-phoneme level but has not mastered others; the mastered phonemes may
be moved into CV tasks before the other phonemes are ready for that progression. There is one notable
limitation to this flexibility in the task hierarchy: all individual phonemes should be introduced and
explored before moving a client into tasks that involve multi-syllable stimuli with well-established
phonemes. If only a limited set of phonemes are being trained (see the section on Delivering PMT in the
real world – practical considerations), all phonemes that you intend to treat should be introduced
before moving on to multi-syllable stimuli. In addition, any given treatment session or task will likely
involve cues at many levels of complexity; for instance, a multi-syllable chaining task may involve
discussion of the individual phonemes involved as part of the Socratic questioning process.
PMT is a dynamic process; on any given task, the clinician should be responsive to the client’s
performance. There are many ways to do this; for instance, you may increase task difficulty as they do
very well and then scale back when you see that they are starting to have less success, or you may
choose a level where they are doing well consistently and then continue at that level until their accuracy
declines due to fatigue. See the section on Adjusting task difficulty and facilitating client success for
further discussion of this issue.
In general, if the client has no success in the first 4-5 trials with a new task, this indicates that
the task is too difficult and the clinician should drop down to a simpler task.7 If the client has complete
success in the first 4-5 trials, this indicates that the task is probably too simple and task difficulty should
be increased. Examples of methods of cueing and manipulations that can change task difficulty are
provided in the section on Socratic questioning: A fundamental element of PMT.

7
Pro tip: The exception to this is the client who has a hard time acclimating to tasks or shifting task set. These
problems are generally evident across tasks, including tasks that you know they can do well, making these clients
relatively easy to identify. In their case, before moving to an easier task the clinician should do more than 4-5 trials,
or take a break and then re-initiate the new task, to be certain that the client’s problems are not just due to task
shifting problems.
Phonomotor Treatment manual
Page 51 of 116

Adjusting task difficulty and facilitating client success

In addition to carefully crafting Socratic questions to facilitate client processing, the clinician also
needs to consider and adjust task complexity if a client is not having success at a particular level, or is
doing particularly well. While PMT is not specifically intended to involve errorless learning, it is
important that clients are challenged enough to make progress but not so challenged that they do not
experience frequent success. Ultimately, it is the clinician’s responsibility to assure that tasks are not too
easy or difficult. If a client is making a lot of errors, there is something wrong; the clinician should
evaluate the modalities and specific stimuli and tasks being used to find ways to challenge the client
while facilitating success.
Tasks can be made more or less challenging by:
- Incorporating and combining more or fewer modalities to support processing. This might
include:
o Adding or removing visual cues (e.g., mouth pictures, blocks, grapheme tiles, or the
ability to see the clinician’s mouth)
o Adding or removing writing
o Adding or removing tactile-kinesthetic descriptions (e.g., “This is the sound that has
the top teeth biting the bottom lip and you let air out.”)
- Physically manipulating the space between blocks or tiles, particularly in parsing and
blending tasks (or removing physical cues, such as blocks or tiles, to make the task more
difficult; see Figure 19.)

Figure 19 - Colored blocks spaced for blended vs. parsed sequences

(e.g., bif on the left and b - i - f on the right).

- Increasing or decreasing time between trials - NOTE: For some clients, increasing time
between trials provides time for activation to die down, reducing interference and
perseveration, making the task easier. For others, increasing time between trials increases
the working memory load, making the task more difficult.
- Making the stimulus shorter or longer (in terms of either phonemes or syllables). Note that
including consonant blends makes a stimulus more difficult.
- Changing the relationship between contrasting stimuli – phonemes that share phonologic
features (e.g., “b” and “g”, which share both manner of articulation and voicing) are
typically more difficult to differentiate than phonemes that do not (e.g., “b” and “sh”) (see
Appendix for tables of phonemes that share
Phonomotor Treatment manual
Page 52 of 116

- Phonemes with no features shared (differ on 3 features) – Easiest to differentiate,

Phonemes with 1 feature shared (differ on 2 features) – Moderate difficulty to differentiate,
and Phonemes that share 2 features (differ on 1 feature) – Most difficult to differentiate
phonemes). Therefore, making stimuli more similar will typically make the task more
difficult, and making them more easily differentiable will typically make the task harder.
o When starting to pair vowels with consonants, many people find voiceless
consonants easier, likely because of their greater difference from the vowels.
- Adjusting which element within a syllable is changed in a sound chaining task; clients
typically find syllable-initial position to be the easiest to manipulate, followed by syllable-
final, and then medial position. In addition, many people have more difficulty with changing
vowels, particularly if they are in the middle of the syllable.
- Changing the sequence of production, from blocked to random (e.g., once all vowels on the
left side of the vowel chart have been introduced, you might have the client work from the
top, left corner of the chart down to the bottom of the chart and then back up again; you
could then move to having them produce them in random order as you point to the
different graphemes).
- Moving between production and perception tasks – which task is easier or harder depends
on the individual client.
- Moving to phonemes with which the client has previously had either good success (easier)
or more difficulty (harder).
- In chaining tasks, keeping previous stimuli in view (blocks, grapheme tiles, or written) after
the string has been changed, so that the client can follow the progression of changes over
the course of the task (see Figure 20 and Figure 21).
-

Figure 20 - One way to keep previous stimuli in sight during a chaining task.
The clinician added each item as it was introduced during the
task, and then had the client go back through the previous items
after introducing each new stimulus, focusing on which element
had changed.
Phonomotor Treatment manual
Page 53 of 116

Figure 21 - A second example of how to keep previous stimuli in sight during a chaining task.
In this case, changing from fird to firp, the blocks and grapheme tiles that stayed the same
were moved down, while the old and new changed phoneme remained in sight.

- Providing more or less physical cueing for a task. For example, if a client is having difficulty
learning to parse and blend, this task can be made easier by providing hand-over-hand
guidance to demonstrate separation or blending of phonemes (i.e., pointing to each colored
block or letter tile individually while saying each in isolation [f-e-p], and then running the
finger across all blocks/tiles in the word while saying it as a single blended unit [fep]). As the
client becomes more adept with parsing and blending, the hand-over-hand cues may evolve
to the clinician providing visual cues by pointing, having the client point by him/herself, or
having the client do the task without visual cues.
- Encourage or discourage verbal rehearsal of target stimuli (e.g., in a complex syllable chain
task, in which a client is moving blocks or letter tiles to represent which sound is changing, it
is easier if the client verbalizes the stimulus and more difficult if you instruct them not to).
- Shifting between replacing or subtracting single phonemes in a sequence (easier), adding
phonemes to a given sequence (harder), and transposing phonemes within a sequence
(hardest).

A single treatment trial will likely include multiple forms of questions and cues as you work
through both the stimulus presentation and the response.
Phonomotor Treatment manual
Page 54 of 116

Treatment stimuli
Devising non-word stimulus combinations
During PMT, the clinician comes up with non-word phoneme combinations based on the
phonemes being addressed and the task circumstances. The clinician may do this in real time, during the
session, or may prepare lists of stimuli ahead of time so that the tasks flow more smoothly in treatment.
While any non-word phoneme combination is possible, there are a few guidelines to keep in mind:

- Non-words should be structured in ways that are permitted in English, in terms of their
structure and sound combinations. For example, English words cannot start with zh or ng, or
end on h, w, or the lax vowels i or e, so non-words used in PMT tasks should follow those
same rules. If a client is having a particularly difficult time with a phoneme sequence,
consider whether it may include sound combinations or structures that are not permissible
in English and adjust accordingly if it does.

- VC combinations are less likely to yield real words than CV combinations (e.g., ees and eeb
vs. see and bee). For this reason, when working with strings of two phonemes it is often
easier to use VCs than CVs.

- There will be times when a real word is accidentally created in the course of treatment,
particularly during tasks in which a single element of the stimulus is being switched out to
work through all of the vowels or consonants (e.g., using the same consonant and working
through all of the vowels in combination with it). This is unavoidable and generally not a
problem. When this happens, just recall that these stimuli are to be treated simply as
phoneme strings, with the focus remaining on the sounds rather than the meaning. If a
client has particular difficulty maintaining a focus on phonology when real words arise (e.g.,
they perseverate on the real word when the stimulus shifts back to non-words), then
consider switching tasks to interrupt the perseveration and returning to the problematic
task later, being sure to use only non-words when you return.

Along with all of the non-word phoneme sequences that are created in the moment, studies of
PMT effectiveness have included certain non-words used repeatedly and intentionally. These non-words
(see Appendix) were devised to have properties (low phonotactic probability and high neighborhood
density) that should facilitate learning (Storkel, Armbruster, & Hogan, 2006). They are used in the same
way that all other non-words are in this treatment, but with a specific focus on using them repeatedly
across tasks and sessions (see Kendall et al., 2015, for a detailed discussion of the rationale for using
prescribed non-word stimuli). The clinician should keep track of which non-words are used in each
session, and rotate through them all across sessions.
Phonomotor Treatment manual
Page 55 of 116

Introducing real words

Choice of real words to incorporate into treatment

PMT was designed to focus on phonology, with the goal of improving phonological processing
and phonologic-semantic connections across the entire language network, in all domains, rather than to
train specific words. As discussed earlier, the treatment primarily employs non-words to allow clients to
focus on phonology without assistance from their preserved semantic knowledge. Real words are
incorporated once syllables have been introduced, however, to strengthen the links between phonology
and semantics for the sake of maximizing treatment effects on functional language use.
This treatment has been tested using a specific set of real words. As described earlier for the
non-word stimuli, these words were chosen to have properties (low phonotactic probability and high
neighborhood density) that should facilitate learning (Storkel et al., 2006). Because the treatment is
designed to generalize across language representations and domains, and based on the model that
motivates PMT and research outcomes to date, improvements in functional communication are
expected even without specifically training personally relevant vocabulary. To keep clients focused on
phonology rather than on semantic and other aspects of stimuli, we do not recommend choosing client-
specific real words.
Introducing and using real words in treatment
Real words are typically introduced late in treatment, frequently as part of tasks that involve
parsing, blending, and elision, in pseudo-homophone tasks, and in tasks that use colored blocks to
practice swapping out phonemes to create changing strings of stimuli. In these cases, the task may start
with a non-word, evolve to a real word, and then evolve back to a non-word. For instance, a task
sequence could unfold as follows:

Clinician: Put out blocks to show the sounds in feeb.

(Client puts out a red block, a white block, and a green block, in that order)

Clinician: If that says feeb, make it say neeb.

(Client swaps out the red block for a black block)

Clinician: If that says neeb, make it say nee (‘knee’).

(Client removes the green block)

Clinician: If that says nee, make it say neek.

Regardless of the task, it is important to remember that real words should always be treated as
phoneme strings, with the focus remaining on the sounds rather than the meaning.
For real words being used intentionally (see Appendix for the full list of recommended real
words, presented in standard orthography and in PMT orthography for reading-based tasks), the
clinician should keep track of which words have been used in each session and make an effort to rotate
through all of them repeatedly throughout the treatment program.
Phonomotor Treatment manual
Page 56 of 116

Keeping data for PMT

Keeping data for PMT is critical but difficult, as could be expected for any complex treatment.
There is not a single way that it must be done, but there are some general principles that can guide your
development of a data keeping system that works for you.
In general, it is important to keep data in each session on:
- The input and output modalities of each task.
- The length of stimuli being used (single sounds or sounds in combination).
- The consistency of the position being manipulated in a string of sounds (e.g., if you are
doing constant or random practice).
- The amount of cueing needed (e.g., minimal, moderate, maximum) and examples of the
types of cues used (e.g., looking in the mirror, repeating before pointing to a stimulus, etc.).
- How accurate the client was on the task (given the parameters described by the previous
items).

It is also useful to have a place to document general observations about each session, and to
keep a record of which sounds have already been trained and mastered at which level (in isolation,
syllables, and blends). This information could be maintained on a separate reference sheet that stays in
the client’s chart, or could be incorporated into your data sheets. Having this information readily
available will make it easier for you to keep track of what level to address with each sound from session
to session, and will provide critical information for a therapist who cover a session if you are not
available.
A few blank data sheets are provided in Sample Data Sheets; you will see that they are quite
different from each other, with some being more structured and others being less structured, but each
reflects a system that has worked for a clinician experienced with PMT. Some of the demonstration
videos provided also have completed data sheets provided as examples. In addition, some clinicians find
it useful to record data during the session (typically at the end of each task, rather than during the task,
since tasks typically require high levels of clinician engagement), and others make their notes at the end
of the entire session. We encourage you to experiment with different data-keeping formats, and
establish one that works for you.
Phonomotor Treatment manual
Page 57 of 116

Delivering PMT in the real world – practical considerations

Phonomotor Treatment manual
Page 58 of 116
As with any treatment approach, the decision to use PMT with a particular client should be based on information from all three pillars of evidence-based
practice: the best available research evidence, your own clinical expertise, and your client’s preferences. This perspective requires you to incorporate a wide
variety of factors such as your own level of comfort and skill with PMT, the amount of time you have available for treatment, and your client’s abilities and
priorities. While research is continuing to explore and validate a variety of practical modifications to PMT, there are not yet data available to directly inform all
possible circumstances, such as limited treatment time or intensity; therefore, we can only speculate at this point in time about options for modifying PMT to
meet individual needs. In the meantime, we encourage you to take a translational approach, determining how best to take the currently available evidence and
apply it to your clinical setting. With all of this in mind, there are a number of questions frequently asked about the logistics and practicality of using PMT:

FAQ #1: All of the PMT data are from programs that lasted 60 or 96 hours. There’s no way I could provide that many hours of treatment to one client! How can I
implement PMT when I don’t have that many hours to devote to it?
Response to FAQ #1:
While all of the published research on PMT has involved at least 60 hours of treatment (i.e., program duration) at high intensity (i.e., frequency and
length of individual sessions), we recognize that it can be difficult or impossible to match those parameters in most clinical settings. If your client appears to be a
good candidate for PMT, the language models that motivate PMT (see What is Phonomotor Treatment (PMT)?) suggest that the tasks used in PMT could be
helpful, due to their multi-modal focus on phonology, even if all sounds are not specifically trained. Remember: the goal of PMT is to train phonological
awareness, not necessarily the entire inventory of sounds in the language. You might start with the treatment package as described and decide, as you go along,
what to prioritize based on client performance. For instance, you could specifically train just a sampling of the phonemes in the early stages, to establish the
ability to think about individual sounds in multi-modal terms, but then use all sounds in treatment tasks. Alternatively, you could choose to train all phonemes in
a cursory manner, just making sure that they understand how to explore sounds in a multi-modal manner before moving on to syllable-length tasks. These
approaches would mean that treatment tasks are implemented in less of a step-wise manner than has been tested, but would still incorporate all of the basic
elements of PMT.
With all that said, and consistent with the available literature on treatment effects in aphasia (Robey, 1998), we suggest that PMT sessions should occur
at least twice weekly. In addition, due to the number of activities involved in PMT, and the time it can take to work through the Socratic questioning process with
each stimulus, we do not suggest sessions shorter than 45 minutes. Even with 45-50 minute sessions, treatment time should be highly focused to maximize work
time, with little to no socializing. All of this is supported by review of the aphasia treatment literature, which suggests that greater treatment intensity leads to
greater improvement in language skills (Raymer et al., 2008).

FAQ #2: All of the PMT data are from programs that have provided treatment two hours per day, 5 days per week. What if I can only see my clients one or two
times each week, and my sessions are only 50 minutes long?
Response to FAQ #2:
PMT has not yet been tested at a lower intensity, such as one hour per day a few times per week. We are currently planning a research trial to address
this question. The models on which PMT are built predict that positive outcomes should still be possible with lower intensity. If you choose to implement PMT at
a lower intensity or frequency than has been tested, we suggest beginning treatment at relatively high intensity and frequency (e.g., 3 one-hour sessions in the
first week), to allow you to quickly determine if a client is a good candidate, and then reducing the intensity or frequency once this has been confirmed. In our
experience, if a client is still struggling with understanding the task format for learning single phonemes after 3-4 hours of treatment, and is not showing signs of
Phonomotor Treatment manual
Page 59 of 116
being able to take what has already been learned and apply it to new stimuli as they are introduced, they may not be a good candidate for PMT. Note that this
assessment does not rely on response accuracy on treatment tasks, but is related more to whether the client demonstrates conceptual understanding of the
tasks and cues and can actively participate (e.g., see Case study #4: Lynn). If they do not at this point, it would be appropriate to consider a different treatment
approach.

FAQ #3: This is a really complex program. What are the critical elements that I need to be sure are maintained if it doesn’t seem to be a perfect fit for my client
in all respects?
Response to FAQ #3: The only non-negotiable elements of PMT are multi-modality manipulation of phonemes in progressively longer strings and the pervasive
use of Socratic questioning. Both of these are essential to developing phonological awareness, which is the ultimate goal of the program. Despite its complexity,
the multi-faceted nature of PMT means that there is quite a bit of flexibility in how the program is implemented. The clinician can, and should, adjust every task
to meet each client’s unique combination of strengths and weaknesses. For example, the clinician may selectively manipulate the modalities being paired in a
task, the modalities focused on in Socratic questioning, or the response modalities used. The clinician may also choose to alter some of the conceptual
organization of the stimuli; for instance, after initially introducing all of the consonants in the groupings described here (i.e., lip, tongue, and air sounds, which
combine place and manner of articulation), some clients benefit from discussing or grouping the sounds in terms of just one category of phonological feature.
For instance, some people find it easier to think about stop consonants versus continuants, while others find place of articulation to be a more helpful
organization scheme. So long as treatment remains focused on multi-modality phonological awareness and uses Socratic questioning, you are working within the
boundaries and guidelines of PMT.

FAQ #4: When should my client be discharged from treatment?

Response to FAQ #4:
The question of when to discontinue treatment is a difficult one, and will depend on a combination of factors such as client progress, motivation to
continue treatment, and availability of treatment resources (e.g., insurance coverage and clinic schedule). In terms of client progress, which is the factor best
addressed here, it is useful to think about the skills that the client has gained to date.
The research literature on PMT demonstrates that many people who have completed the program as tested (i.e., 60 hours of treatment in 6 weeks)
continue to improve in language skills for at least a year after treatment ends. This is predicted by the parallel distributed processing model on which this
treatment is built, which describes the language processing system as a complex, multi-modal network; once established, the network connections are self-
reinforcing through exposure to everyday language use. To facilitate this continued growth of language skills after treatment ends, we suggest that treatment
continue at least until the client has solidified a majority of phonemes (consonants and vowels) and developed some skill with sequence manipulation (rather
than just working at the single-phoneme level).
We do believe that people who respond well to PMT could potentially benefit from continuing indefinitely. As long as they are continuing to make
progress and have not yet hit the limits of typical performance (as measured by the clinician’s ability to process the stimuli and tasks), treatment tasks can
continue to be made progressively more difficult. There may be no maximum treatment duration. At the same time, however, continuous treatment is often not
feasible, whether because of funding, client commitments, or changes in treatment priorities. If continued treatment is not feasible or desired, it is possible to
discontinue treatment and then arrange for the client to return for reassessment at some future date and determine if they could benefit at that time from
another short course of treatment, to either reinforce previously learned skills or to realize new gains.
Phonomotor Treatment manual
Page 60 of 116

FAQ #5: My client really wants homework to do on the days she doesn’t come in for therapy. What kind of homework can I have my PMT client do?
Response to FAQ #5:
Because PMT is such a multi-layered, complex treatment, and it is not advisable to have clients practice things incorrectly, we do not recommend having
clients practice specific PMT treatment tasks at home. You can, however, encourage them to do tasks that reinforce phonological awareness and processing.
These could include:
- Reading aloud (Dr. Seuss books and some forms of poetry are particularly good for highlighting and manipulating phonology);
- Listening to audio books while following along with a print version;
Record strings of stimuli in an elision task (e.g., aerplaen without plaen is aer; chootee without tee is choo; see the
Phonomotor Treatment manual
Page 61 of 116
Phonomotor Treatment manual
Page 62 of 116
Phonomotor Treatment manual
Page 63 of 116
Phonomotor Treatment manual
Page 64 of 116
Phonomotor Treatment manual
Page 65 of 116
Phonomotor Treatment manual
Page 66 of 116
Phonomotor Treatment manual
Page 67 of 116

- Glossary of Terms) and have them passively listen to them;

- Simple phoneme, real word, and non-word repetition tasks using combined audio-visual
stimuli that you record on their phone or iPad;
- Listen to lists of word pairs and judgments on whether or not they rhyme.
There are also some relevant defined treatment programs that lend themselves to independent
practice:
- Copy and Recall Treatment (Beeson, Rising, & Volk, 2003);
- The National Aphasia Association has a list of available apps for aphasia therapy:
[Link]/aphasia-resources/aphasia-apps/. Some of these apps reinforce
phonological skills, including letter to sound matching, same/different judgment on minimal
pairs, reading aloud, sound identification, sound matching, spoken rhyming, word copying,
and word repetition.
This is not an exhaustive list, and we encourage you to use your creativity and knowledge of
phonological awareness to develop tasks that are appropriate for your client. Just keep in mind that the
overarching goal is phonological awareness, and tasks should ideally allow practice without a high risk of
error.

FAQ #6: How do I decide if I should do PMT or some other type of therapy with my client?
Response to FAQ #6:
As with any intervention, the decision whether to use PMT should be made after considering
client priorities, abilities, and practical circumstances. If a client’s highest priority is verbal
communication, you have identified that they have phonological processing impairments, and they
already have a system in place for basic daily communication, then PMT may be a great way to address
this goal. If a client’s priority is functional communication, however, regardless of the amount of
difficulty they’re having with verbal communication and acquired alexia, then PMT may not be the best
treatment choice for them.
If a client is interested in balancing both restorative and compensatory treatments, consider
whether you can use treatment time for PMT to address restorative communication goals and train
family members, caregivers, or friends to support compensatory communication goals outside of
treatment. The balance of treatments may also change over time; you may start with more word-based
and compensatory treatment approaches as you establish basic communication systems early in
treatment and build rapport, and then explore doing PMT on a trial basis. In this case, you can explain
that you believe this is an appropriate treatment
Note that PMT is not appropriate for acute care settings. These settings tend to lack the
intensity needed for PMT, and there are likely other, higher priority treatment objectives at that early
stage.

FAQ #7: It seems like PMT would take a lot of time to learn to do it well. I only see people with aphasia
occasionally. Is it worth my learning PMT?
Response to FAQ #7: PMT is a complex approach to treatment, and there is definitely a steep learning
curve. Once it is learned, though, it is very flexible and can be used with people with aphasia at all levels
of severity. If you only see a few people per year with aphasia, though, you may decide it is not worth
investing the time and energy to become skilled in this treatment approach. In this case, we would
encourage you to refer clients who are appropriate candidates for PMT to another SLP who carries a
larger caseload of people with aphasia and is trained to use PMT.
Phonomotor Treatment manual
Page 68 of 116

FAQ #8: I’d love to have some support as I learn and implement PMT. Is there a way to be in touch with
other SLPs who are using PMT?
Response to FAQ #8: We have established a discussion board on the PMT training website
([Link] where clinicians can exchange ideas.
If you would like to participate in the online discussion group, please send an email request to
aphasia@[Link] with either a University of Washington email address or a Gmail address and we will
arrange for you to have access to the group.

In addition, you can hear two individuals who received intensive PMT discuss their experiences with this
treatment program at [Link]
Phonomotor Treatment manual
Page 69 of 116

Case studies
Case study #1: Morgan
Case description:
Age: 62
Time post-onset: ~2 years
Diagnosis: Aphasia subsequent to left CVA
Language profile:
- Verbal production:
o He speaks in complete sentences with accurate grammatical structure, with
primarily anomic errors characterized by frequent omissions and phonemic
paraphasias.
o He is aware of his production errors in speech, and attempts to repair them when
they occur and is almost always able to do so.
o He benefits from phonemic cues if he is unable to find a word himself.
- Comprehension:
o His auditory comprehension is good for complex, novel information in a quiet
environment, but breaks down in the presence of distraction.
o He is generally aware when he has not understood something and asks for
clarification as needed; this happens infrequently, typically when he is tired.
- Other:
o He has no dysarthria or apraxia of speech.
o He is highly motivated and engaged in the treatment process.

Socratic questioning with Morgan: Morgan has very good auditory comprehension and excellent insight
and self-awareness. He tends to notice small relevant details, at a level that is generally reserved for
trained clinicians, and is capable of doing high level analysis as a result. For instance, during exploration
of vowels, he is aware of fine differences in tongue position (front/back, high/low) and can apply this
awareness to describing and distinguishing the phonemes. Socratic questioning with this client can begin
fairly open-ended, and can involve complex questions. For example, in later stages of treatment the
clinician might say two CVC words (e.g., feep and feesh) aloud with no supporting visual stimuli, and the
client would likely be able to explain the difference between the two targets and identify the
corresponding mouth pictures and graphemes.

Single phoneme training: When the first sound pair is introduced, he quickly catches on to the types of
information being asked for, and tunes in to the various characteristics and properties of the sounds
with little cueing. After the first sound pair is introduced, he applies the training protocol to future
sound pairs so he learns them quickly, with little direct questioning (e.g., he addresses the topics of the
Socratic questions that he’d been led through with p/b as he explores f/v). He needs only a few rounds
of structured practice with each consonant cognate pair for both production and perception before
successfully moving into random practice conditions, and needs little drill practice to differentially
produce vowels after they are initially introduced. He retains knowledge and skills across sessions and
days. He is quickly successful taking on independent management of treatment materials, such as
sorting consonant mouth pictures into categories, arranging grapheme tiles on the vowel circle, and
matching grapheme tiles to the mouth pictures.

Sounds in combination: In general, this client moves through the progression of tasks very
systematically. He readily progresses from high to low levels of cueing and from highly contrastive to
Phonomotor Treatment manual
Page 70 of 116

minimally contrastive pairs (e.g., initially contrast fip-foop and then move to fip-fep). At times, the
clinician increases task difficulty or reduces cueing faster or in larger steps than she might typically use
because she senses that he is able to handle it. In tasks that combine sounds, he is quick to implement
his knowledge from single-phoneme training to analyze and correct errors that he makes (e.g., he says
fip for fap, and then describes that his jaw was not open enough). Even when the clinician is not asking
Socratic questions, he often talks about his processing in these analytical terms while he’s working to
process a stimulus or task. Tasks typically involve single modalities, as he does not need the support of
multiple modalities to achieve high levels of success. He carries over information well between tasks and
from session to session. Despite his overall high levels of performance, he often becomes overloaded as
his cognitive resources for a task he’s been doing well with seem to become depleted; clinical
observation suggests that this is due to a build-up of interference over time. On these occasions, he
becomes perseverative, shows increasing phonemic errors in production tasks, and has poor accuracy in
perception and discrimination tasks. As a result, he benefits from switching tasks frequently, when the
clinician notes that errors are becoming more frequent. In addition to this decline in performance over
time, it is also noted there are behavioral management issues that interfere with task performance; at
times, he is overenthusiastic and gets himself confused as he attempts to over-analyze stimuli or do
more than the task at hand requires. In these cases his performance also suffers, but he is generally
successful getting back on track with a cue to take a short break and “reset” and then resuming the task
at hand, rather than switching tasks.

Outcomes and next steps:

- Overall assessment: Morgan has been an ideal candidate for PMT.
- Status after several weeks:
o He is moderately to very successful (though not yet 100% accurate) manipulating
complex, multi-syllabic stimuli across many tasks and modalities, such as:
 2-syllable non-word repetition with no visual cues;
 reading aloud CCVC and CVCC strings in a random presentation sequence;
 sequencing mouth pictures and felt squares to auditory input at the level of
CVCVC strings.
o He continues to have significant difficulty with complex phonological working
memory tasks, such as identifying the phonemic change in a chaining task and
maintaining it through a series of Socratic questions about the old and new targets
without additional auditory models from the clinician.
o Repeated standardized language testing reveals that he has improved on
confrontation naming, auditory comprehension, reading, and communication
efficiency and effectiveness in discourse, and his overall disability scores have
improved.
o In conversation, he shows fewer instances of anomia (omissions), fewer phonemic
paraphasias, and more efficient repairs when word retrieval breakdown occurs.
o He reports several functional changes:
 He has been able to resume activities at home that involve significant
linguistic working memory load (e.g., working with his wife to build planter
boxes, in which she gives him complex instructions that he needs to process
and retain long enough to execute).
 He can now read newspaper articles and understands them on the first
reading, whereas he had previously needed to read articles 2-3 times to
fully understand.
Phonomotor Treatment manual
Page 71 of 116

 He is still somewhat limited in writing emails, keeping them shorter than he

would have liked, but he reports that his wife now makes fewer edits to
them, suggesting that they are more accurate and effective than previously.
- Given what has been demonstrated in the research on PMT, this may be an appropriate
point to discharge him from treatment because:
o his confrontation naming and discourse skills are now only minimally impaired, and
are meeting most of his daily needs, and
o he has received 60 hours of treatment and the literature suggests that many clients
reach a plateau after that much treatment.
- However, it may also be appropriate to continue treatment because:
o he continues to be engaged and interested in therapy,
o he has areas of language use that he would like to continue developing, and
o he has the resources available to continue treatment
- If treatment continues:
o it could continue to focus on increasing stimulus complexity and abstraction, with a
particular focus on the domains most difficult for him: writing and verbal working
memory.
o Treatment might also branch out to include more functional tasks (such as writing
emails), with an emphasis on using strategies and skills developed in PMT to support
his success.

Case study #2: Phil

Case description:
Age: 62
Time post-onset: ~4 years
Diagnosis: Aphasia subsequent to left CVA
Language profile:
- Verbal production:
o He produces grammatically complete sentences, characterized by frequent anomic
episodes and use of non-specific language.
o He often uses sound effects, intonation, and gestures to help convey his ideas.
o He often recognizes when he makes in errors in verbal production, but does not
often try to fix those errors. When he does try, he is generally unsuccessful and is
typically unable to retrieve words that he is missing, even when he takes extra time.
o When he takes extra time to look for a word, he often loses track of the message he
was trying to convey, indicating impaired working memory.
o He only occasionally benefits from phonemic cues, and does not benefit from
semantic cues.
- Comprehension:
o His auditory comprehension is good for single words and highly familiar or
contextualized information. His comprehension benefits from repetition, rewording
of the message, supplemental gestures, and keyword writing at times.
o His reading comprehension is only fair, with frequent semantic confusions noted.
o He frequently recognizes when he has not understood a message, but does not
always ask for clarification.
- Other:
o He has no dysarthria or apraxia of speech.
Phonomotor Treatment manual
Page 72 of 116

Socratic questioning with Phil: Because of Phil’s impaired auditory comprehension, he cannot manage
open ended questions as well as Morgan. Instead, he does best with short questions and a limited field
of response choices. For example, when exploring the p sound the clinician presents drawings with key
words (e.g., see Figure 7) and asks, “What’s moving?”. Because of his impaired self-monitoring and error
awareness, he often needs significant support to analyze and correct responses. For example, if the
clinician says eep in a repetition task and he responds with eesh, he may not accurately identify that the
two productions do not match. The clinician would then structure Socratic questions to take him
through analysis of each modality, with frequent reminders and control of timing of response
presentation to accommodate his WM impairments. This exchange might look like this:

Clinician: Close your eyes and just listen – eep - eesh… are they the same?
Phil: No.
Clinician: Now look at me and watch my mouth… eep - eesh. Did they look the same?
Phil: No.
Clinician: Watch my lips at the end. eep… did my lips pop at the end?
Phil: Yes.
Clinician: eesh… did my lips pop at the end?
Phil: No.
Clinician: So that’s how they were different. eep had my lips popping open and eesh didn’t. Now
try to repeat after me – eep.
Phil: eep.
Clinician: Do they match?
Phil: Yes.
Clinician: Did your lips pop at the end?
Phil: Yes.
Clinician: Yep, they did. Both of us popped our lips at the end, so they matched.

Single phoneme training: On introducing consonants, he cannot explain mouth movements due to poor
articulatory motor awareness; even when provided visual supports (e.g., written words ‘lips’, ‘teeth’,
‘tongue’, watching the clinician, and looking in the mirror) he requires maximum cueing to identify
which articulators are engaged. This continues after many hours of practice. Adding letter tiles does not
help but the clinician decides to keep the letter tiles present as an attempt at engaging and
strengthening orthographic representations in the course of the phoneme learning process. To build his
articulatory awareness, the clinician has him watch her, repeat what she says, judge whether they sound
the same, and then has him look in the mirror and judge whether they look the same. She then covers
her mouth and has him repeat the target again while watching in the mirror and deciding if they sound
the same. He has greatest success in single phoneme training with judging whether his productions
match a clinician model and, if it does, producing/repeating the sound 5 times consecutively. If his
productions don’t match, the clinician begins Socratic questioning to identify errors and attempts to
produce the correct target, but there is no consistent pairing of modalities that is notably helpful and
the clinician often ends a trial with just telling him what the answer is and having him try to repeat it
again. He is not successful in learning to differentiate vowels in any kind of explicit way; vowels are
engaged primarily through simple repetition tasks.

Sounds in combination: As task complexity increases, his articulatory motor awareness remains poor, so
the clinician continues to engage the same types of strategies as in single-phoneme tasks. In perception
tasks, he generally needs to have small fields of choices; his performance degrades significantly if there
Phonomotor Treatment manual
Page 73 of 116

are more than 3-4 options, and he is only successful with that many choices if he had already had a
significant amount of practice with the target in isolation immediately before the more complex task.
Using single letters neither helps nor hinders his performance, but full written words and non-words
generally improve his performance. This relative strength is used at times by overlaying strings of letters
on strings of mouth pictures or blocks for an initial trial, and then removing them and having him
complete the trial again. Blending tasks are particularly difficult for him, as he needs to add just a single
element at a time (e.g., rather than blending m-i-p to say mip, he produces/repeats m, then mi, then
mip), so these tasks are more like chaining tasks than blending. Initially when increasing task difficulty by
increasing phoneme string length (e.g., going from CV to CVC targets) the clinician only asks him to
identify where in the string the change is taking place, and then writes in the accurate letter or selects
the appropriate mouth picture for him. As he gets better with identifying the location of the changed
phoneme, she makes the task more difficult by having him identify where the change occurs and select
the changed element from a field of two widely contrastive choices (e.g., z vs. p). Overall, Phil does
better with many trials of just a few tasks per session.

Outcomes and next steps:

- Overall assessment: Phil has been a good candidate for PMT.
- Status after several weeks of treatment:
o repeated standardized testing indicates that Phil has improved in naming and
auditory comprehension, although his reading and disability scores have not
improved.
o He is continuing to progress through the treatment program, with gradually
increasing accuracy for progressively more complex tasks.
o He continues to be motivated and has the resources available to continue therapy.
o For all of these reasons, it would be appropriate to continue treatment with him and
re-assess his progress after another 5-10 hours of therapy to determine if he has
reached a plateau or is continuing to make gains.

Case study #3: Suzanne

Case description:
Age: 70
Time post-onset: ~2 years
Diagnosis: Aphasia and apraxia of speech subsequent to left CVA
Language profile:
- Verbal production:
o Her verbal output is severely limited, characterized by single word utterances that
are slow, with prolonged phoneme durations, lengthened inter-syllabic durations,
and phoneme distortions, indicating a concomitant apraxia of speech.
o She has fair-good awareness (60-75% of the time) when she makes errors in verbal
production.
o She is frequently frustrated by her limited verbal output.
o She is particularly frustrated with her consistent inability to say the name of the
daughter with whom she lives.
o When she recognizes incorrect verbal productions, she often tries to correct them
but is typically not successful.
Phonomotor Treatment manual
Page 74 of 116

o She does not typically benefit from phonemic cues, and can only repeat a given
word approximately 50% of the time on the first or second attempt when it is
provided to her after a word retrieval failure.
- Comprehension:
o Her auditory comprehension is good for short, familiar sentences.
o Her comprehension benefits from repetition, supplemental keyword writing, and
drawing.
o She has good recognition of when she has not understood a message, and asks for
clarification as needed.
- Other:
o She has no functional reading or writing ability.

Socratic questioning with Suzanne: Her awareness of stimulus characteristics is best in highly structured
tasks with multimodality stimulus cues and maximal difference between stimuli (e.g., few shared
features). Most of the time, she needs very specific yes/no questions (e.g., “Was your voice on for that
sound?”) along with multimodality cues (e.g., having her hand on her neck to feel for vocal fold
vibration). She is unable to respond to open-ended (e.g., “Tell me about that sound.”) or forced choice
questions (e.g., “Is that sound noisy or quiet?”).

Single phoneme training: During exploration of sounds, she cannot independently describe mouth
movements due to extremely limited verbal output. Her independent production of consonant sounds
based on visual input (e.g., pointing to a mouth picture) is approximately 30% accurate, mostly with
quiet sounds. The only vowels she can independently produce and perceive are the anchor vowels (ee,
oo, aw). Even though she doesn’t have a lot of sounds she can produce independently, her repetition for
single phonemes is approximately 70% accurate, though with apraxic distortions. In treatment, her
relatively stronger skills are engaged in a variety of ways. She can accurately point to stimuli that
represent the various aspects of phoneme production (e.g., category labels for lips, teeth, and tongue,
“quiet” and “noisy” icons, etc.), so this response modality is used often to explore how sounds are
made. In addition, combined auditory and visual input (e.g., listen to the clinician, watch the clinician,
watch herself repeat the sound in the mirror) are used to develop knowledge of the mouth pictures and
their corresponding phonemes, which increases her success to approximately 80%. Finally, because her
grapheme-phoneme correspondence knowledge is poor, so she does not use printed letters as a crutch,
the letter tiles are introduced early in her program to provide additional support to her processing
networks.

Sounds in combination: She has difficulty switching between tasks, struggling at first when a new task is
introduced; once a task is established, however, she tends to have good success over many trials.
Therefore, the clinician chooses not to vary tasks much within a session; instead, she does many trials of
just a few tasks. In more difficult repetition and parsing sequences (e.g., CCV), Suzanne has a hard time
blending the consonants in non-words, although she is generally successful if the blends are in real
words. The clinician chooses to take advantage of this by putting target phoneme combinations into real
words and then doing elision tasks to remove the extra phoneme and create a non-word (e.g., elicit
accurate production of smok (‘smock’) and then have her “say smok without the k “ to produce the non-
word smaw).

Outcomes and next steps:

- Overall assessment: Suzanne has not been an ideal candidate for PMT at this point in time,
but has benefitted to some degree.
Phonomotor Treatment manual
Page 75 of 116

- Status after several weeks of treatment:

o Suzanne is approximately 80% accurate on CVC repetition tasks, with independent
parsing of the sounds given auditory and orthographic input.
o In VC chaining tasks (e.g., “if this says eeb make it say ib “), she is most successful
with changing just the vowel sound from trial to trial, a relatively low level version of
the task.
o In tasks using auditory input paired with watching the clinician’s mouth, she
achieves ~60% accuracy identifying sound changes in a random condition using a
combination of felt squares, colored blocks, pictures, and letters for CVCV structures
(e.g., chootee  gootee  gooter), and is getting better at independently parsing
out the sounds.
- Repeated language testing shows modest improvements in picture naming, but no changes
in reading, auditory comprehension, and only minimal improvement in overall disability
scores.
- In conversation, she continues to be very anomic and apraxic, with speech output still
generally limited to single words, but she is now coming up with more accurate vocabulary
during spontaneous speech.
- For comprehension, she requires less support from her conversation partners in the way of
gestures and supplemental key word writing.
- Functionally, she is usually using the right name for her daughter (with whom she lives) as
compared with 0% accuracy at the start of treatment.
- Because her improvement with PMT is slow, and she has expressed that she would like to
focus more on establishing functional communication, it is appropriate to discontinue PMT.
The clinician might:
o discharge her from treatment, with recommendations to her family regarding how
to continue to support her functional communication needs (she has indicated that
she may be interested in returning to PMT at a later date as her skills and priorities
change), or…
o keep her enrolled in treatment but shift the focus to compensatory strategies or
direct training of specific functional verbal vocabulary.
 If the clinician chooses to work on functional verbal vocabulary, some of the
cueing strategies established in PMT may be incorporated to support this
work.

Case study #4: Lynn

Case description:
Age: 70
Time post-onset: ~6 years
Diagnosis: Aphasia subsequent to left CVA
Language profile:
- Verbal production:
o Her verbal output is typically grammatically correct but characterized by anomia.
o When she has an anomic episode, she is generally aware of it, gets frustrated, and
abandons the message with few attempts at repair.
o She has frequent stereotypic phrase production such as, “Oh honey, you’re so
sweet” or “Oh honey, you’re so pretty.”
Phonomotor Treatment manual
Page 76 of 116

- Comprehension:
o Her auditory comprehension is good for familiar or contextualized short
information.
o She benefits from having messages reworded, although she often gets anxious
about not having understood the original message.
o She has fair-good awareness when she does not understand a message, but will not
ask for clarification; instead she will respond with “I don’t know.”
- Other:
o She has a prolonged history of depression and anxiety disorder.
o She typically refuses tasks if she does not think she will be successful with them.
o She does not like working with nonsense words.

Socratic questioning with Lynn: Despite her reasonably good auditory comprehension, Lynn has a very
hard time with Socratic questioning for a few reasons. Her anxiety makes her defensive nervous about
being wrong, leaving her feeling like she is being attacked when errors are pointed out. This is
compounded by her inability to understand that Socratic questioning occurs whether a response is
correct or not. In addition, her poor evaluative skills make it difficult for her to recognize or understand
her errors when they occur, and then correct them, even with maximal cueing through Socratic
questioning.

Single phoneme training: Single phoneme tasks are very difficult for Lynn, although she is willing to
work sounds in isolation. Nearly all tasks used are based in repetition, as she is unable to independently
produce a sound in response to a description or visual cue, analyze articulator movements, or identify
external stimuli (such as mouth pictures or letters) in response to what she hears. She can repeat a
single production, but when asked to do ten repetitions of a target she counts from 1-10 instead of
saying the target; visual, non-linguistic cues are used successfully to elicit multiple repetitions. She is
unable to inhibit reading aloud of the words “quiet” and “noisy” when asked to use the relevant icons to
differentiate cognates, although she does better when icons without words written on them are used.
The clinician attempts to introduce letters early in the process to support her success, but she cannot
inhibit naming the letters, creating interference. She benefits from cues to “reset” when she gets
frustrated or perseverative. Vowel training confuses her, as she does not understand the concept of the
“ladder” between the vowels, so explicit training of all vowels is abandoned and only the anchor vowels
are used. She successfully alternates between consonants in tasks that use sound pairs in random
practice conditions (e.g., 10 productions of each, then 5, then alternate, then random), but then cannot
resume the task after a break without going through the entire sequence again. This is characteristic of
her poor maintenance of knowledge, even within a single session. After 6 hours of treatment, she is 80%
accurate identifying sounds presented auditorily with a field of two choices of mouth pictures held up to
the clinician’s mouth while she says the sound; her accuracy drops below 50% if there are more than
two choices or if the mouth pictures are not shown right next to the clinician’s mouth. After 11 hours of
practice, she has been introduced to most of the consonants. Given her struggles with single phoneme
tasks, the clinician concludes that the mouth pictures are too abstract for her, and her inability to inhibit
unrelated responses interferes with single sound practice. The clinician knows that it isn’t always clear if
someone is not a good candidate for PMT, as many people have trouble with the abstract nature of
single phoneme tasks; therefore, he decides to move to tasks that use sounds in combination in hopes
that the greater context and variety of tasks might yield greater success.

Sounds in combination: Lynn gets very frustrated working at the syllable level because the non-words
do not make sense to her. Despite multiple explanations she does not understand the rationale for using
Phonomotor Treatment manual
Page 77 of 116

them; simply put, she does not have buy-in for these activities. In addition, these tasks are very difficult
for her, increasing her frustration; the number of stimuli involved are overwhelming so the clinician can
present only one visual stimulus at a time, significantly limiting opportunities to link modalities and
compare or manipulate contrasting sounds. The colored blocks are too abstract for her, and she is
offended to be asked to use a childish toy. She is unable to do parsing tasks, and cannot participate in
Socratic questioning beyond making same/different judgments and completing lexical decision tasks.
Ultimately, she is only successful with real word repetition and copying tasks, which is not enough for
PMT. In addition to her significant language impairments, her participation is also negatively impacted
by her depression and anxiety, and possible changes noted in her general cognitive ability.

Outcomes and next steps:

- Overall assessment: Lynn has been a poor candidate for PMT
- Status after a few weeks of treatment:
o She has had poor success with both single phonemes and sounds in combination.
o She has exhibited significant frustration with treatment, and has not been able to
fully engage in treatment activities, including the critical element of Socratic
questioning.
o She shows no evidence of progressing through the treatment hierarchy, or
improving in expressive or receptive communication in functional tasks or on
standardized tests.
- This treatment approach is discontinued and the clinician chooses to focus, instead, on
client and caregiver education and support.
o Lynn responds well to participating in aphasia groups and having regular lunch dates
with another woman with aphasia, so the clinician facilitates her initial involvement
with those activities and encourage her family to continue providing these
opportunities.
Phonomotor Treatment manual
Page 78 of 116

Appendix
Printable resources for implementing PMT
Phonomotor Treatment manual
Page 79 of 116

Treatment example videos

The video examples referenced throughout the text can be viewed as follows:

Client perspectives on PMT:

[Link]

Pre-treatment conceptual explanation:

[Link]

Introducing and practicing single phonemes:

[Link]
phonemes

Introducing and practicing multi-syllables:

[Link]

Introducing and using orthography:

[Link]

How NOT to do PMT:

[Link]
Phonomotor Treatment manual
Page 80 of 116

Mouth pictures
Phonomotor Treatment manual
Page 81 of 116
Phonomotor Treatment manual
Page 82 of 116
Blank vowel circle for use in therapy sessions
Phonomotor Treatment manual
Page 83 of 116

PMT graphemes
These graphemes and icons may be cut out and laminated for manipulation in treatment.
Phonomotor Treatment manual
Page 84 of 116

Non-words to use in treatment

1 syllable 2 syllables
doi chootee
af foekoe
toos leber
sheev doeum
ek mefoe
dach shever
peenz feether
poeuh toiler
meeth iezl
ish shaybee
whup veeder
breek zower
voo tawthee
eep jiver
reesh wooter
nie dungee
iej turmee
zien lekee
broiz juroe
thag shasoe
oit hoyter
kur neenee
froos raezl
graek hieger
choy woewuh
oos unger
faps miever
woy jawvee
awch prezhur
plown fooer
zae pire8
hob driepur
veed gower
teever
iebee

This is the proscribed list of low phonotactic probability and high neighborhood density non-words that
are recommended for repeated use throughout treatment, and should be used daily across tasks at all
levels once syllables have been introduced. Early in the program, when they are first being introduced,
you might use them for just a few minutes in the session, during a single task (e.g., repetition). As
treatment progresses, these stimuli are incorporated progressively across more tasks (e.g., stimulus
chaining, reading aloud) and for more time during each session; by the end of the treatment program,

8
This rhymes with ‘fire’. It is not written in PMT orthography format because it would be mistaken for the real
word ‘pier’.
Phonomotor Treatment manual
Page 85 of 116

these items become the primary stimuli being used in treatment. Note that tasks that involve these non-
words will also include other non-words that are not on the list; for example, in a syllable chaining task
you may start with a non-word from this list and move through a chaining sequence at arrive at another
non-word from this list (e.g., voo  voy  vee  veed). We recognized that a few of these items are
real words, but they are obscure enough that most clients do not recognize them as such. Therefore, we
have chosen to include them in the interest of having enough items available that meet the lexical
criteria for phonotactic probability and neighborhood density.
Phonomotor Treatment manual
Page 86 of 116

Real word stimuli to be used in treatment

Real word PMT orthography
plane plaen
bride bried
bruise brooz
gravy graevee
clover kloevir
speaker speekir
pie pye
fur fer
knee nee
maze maez
ditch dich
wheel weel
mop mawp
fire fier OR fieur
knob nawb
cave kaev
bird berd
jail jael
witch wich
knot nawt
meadow medoe
ivy ievee
jeans jeenz
level levul OR levl
shoulder shoeldir
ranger raenjer
heater heetir
teacher teechir
movie moovee
polo poeloe
genie jeenee
halo haeloe
father fawthir
jockey jaukee OR jawkee OR jokee
tiger tiegir
diver dievir
shower shouwir OR showir
ape aep
owl oul

This is the proscribed list of low phonotactic probability and high neighborhood density real words that
are recommended for repeated use throughout treatment, and should be used regularly across tasks at
Phonomotor Treatment manual
Page 87 of 116

all levels once syllables have been introduced. These words should be interwoven into tasks that are
based in non-words; for example, in a syllable chaining task you may start with a non-word and move
through one or more real words through the course of the task (e.g., voo  vae  vaech  aech  aep
 aef  laef). Remember that real words used in PMT are treated the same as non-words, with a focus
only on the phonological aspects of the word and not the semantic aspects. If they are written down or
represented with grapheme tiles, they should be written using PMT orthography rather than
conventional English orthography.
Phonomotor Treatment manual
Page 88 of 116

Examples of word and syllable chains

Examples of Word/Syllable Chains

*When making your own chains, consider whether ‘wider’ (easier) or ‘closer’ (harder) contrasts between
items are indicated. Think about the phoneme categories as well as voicing to determine contrast. Please
refer to the manual section Adjusting task difficulty and facilitating client success for more methods to
increase/decrease task difficulty when making chains.

New word each trial: Change 1-phoneme each trial (different position):

toos toos

sheev toob

meeth moob

reesh meeb

hob sheeb

fop shoeb

veed shoeg

dach loeg

neep lieg

zien dieg

Change 1-phoneme each trial (same position): Syllable chain:

toof vee

toob veeder

toosh kunveeder

toov kunveeshun

toop kunmooshun

toog mismooshun

tooch mismoofer

tooz mismooten

toos mooten

tood moo
Phonomotor Treatment manual
Page 89 of 116

Examples of phonotactically legal consonant clusters in English

Note that the examples are written in standard English orthography
Onset clusters Final clusters
Cluster Example Cluster Example Cluster Example
Bl- Black -dj Edge -pt Inept
Br- Brown -dz Meds -rb* Carb
Dr- Dream -ft Rift -rch Perch
Dw- Dwarf -ks Locks -rd Hard
Fl- Flow -kt Fact -rf Wharf
Fr- Fry -lb Bulb -rj Urge
Gl- Glide -lch Mulch -rk Park
Gr- Grape -ld Old -rl Girl
Kl- Clay -lf Elf -rm Farm
Kr- Crush -lk Milk -rn Burn
Kw- Quiet -lm Elm -rp Tarp
Pl- Place -lp Gulp -rs Purse
Pr- Prawn -lt Built -rsh Harsh
Sk- Skip -lv Solve -rt Cart
Skr- Scratch -mp Lamp -rth Hearth
Skw- Squid -mz Homes -rv Swerve
Sl- Sleep -nch Finch -sk Risk
Sm- Smile -nd Round -sp Wasp
Sn- Snap -nj Change -st Best
Sp- Spin -nk Think -thm Rhythm
Spl- Splash -nt Mint -ts Bets
Spr- Spring -nz Bins -vz Lives
St- Stem -ps Lips -zm Chasm
Str- Strap
Sw- Swipe
Shr- Shrug
Tr- Truth
Tw- Twin
Thr- Three
Thw- Thwart
*In some dialects, r-based clusters may not qualify as consonant clusters. The clinician should use or not
use these clusters as appropriate.

This list is based on principles of phonotactic legality outlined in Algeo (1978).

Phonomotor Treatment manual
Page 90 of 116

Examples of real words with conventional spelling in PMT orthography

This list can be helpful during orthographically-based tasks for clients with more severe aphasia. These
can be used as a quick reference if desired, but you may choose other words, as well. Note: These are
not the proscribed real words for use with all clients. These may be helpful if a client is having difficulty
with a non-word (this often occurs particularly with complex consonant clusters); you can put the non-
word into a real word context to facilitate its production, and then remove individual phonemes to
gradually move back to the non-word. Note that these words are written in standard English
orthography.

CV (including
VC CVC CCVC CVCC
rhotic vowels
bar am big mob bleed best
bee at but mom broom best
doe if dad mud brim fast
far in did mug drip hand
fee it dig not drop help
fur on fan nod drum hint
goo up fat pot flag hunt
her us fib peek flat husk
jar fit pun flip jump
nor gas rag flop just
paw get ram frog keeps
raw got rat glad kept
saw had rod grab lift
sir hot rot plan mist
toe jet rug skid must
too kit sad slam pond
lee lit sat sled romp
now lid sit sleep runt
see mad sum slim seeps
tie man ten slip shift
vow mat tip slot taps
woe smog test
snap went
snip
snug
stop
swam
sweep
swim
swoop
trim
tweed
twig
Phonomotor Treatment manual
Page 91 of 116

Phonemes with no features shared (differ on 3 features) – Easiest to differentiate

BASE No shared features

b s, sh, f, th, ch, h
p z, zh, v, th, j, n, w, r, l
d sh, f, th, h
t th, v, zh, m, w, r
g h, ch, th, sh,s, f
k j, th, v, zh, r, l
th b, d, g, j, m, w, r, l
th p, t, k, ch
f b, d, g, m, w, n, r, l
v p, t, k, ch
s b, g, m, r, w
z p, t, k, h
m t, k, f, th, s, sh, h
n p, k, f, th, sh, h, ch
w f, th, t, s, ch, k, h
l p, k, f, th, ch, h
r p, k, f, th, h
h b, d, g, m, w, l, r, n, j
ch b, d, g, th, z, v, m, n, w, r, l
sh b, d, g, j, l, m, n, w, r
zh p, t, k, h
j p, t, k, f, th, h
Phonomotor Treatment manual
Page 92 of 116

Phonemes with 1 feature shared (differ on 2 features) – Moderate difficulty to

differentiate

BASE 1 shared feature

b t, k, v, th, z, zh, l, n, r, j
p d, g, f, th, s, sh, h, ch
d p, k, m, w, v, th, j, zh, s
t b, g, ch, f, th, sh, h, s
g p, t, j, th, zh, z, v, l, n, m, w, r
k b, d, ch, sh, s, th, f
th v, z, zh, p, t, k, ch
th v, z, zh, b, d, g, m, w, l, n, r, j
f z, zh, th, p, t, k, ch
v th, s, sh, h, b, d, g, j, m, w, n, r, l
s v, th, z, zh, p, k, ch, d, l, n, r
z f, th, s, sh, h, b, d, g, m, w, d, l, n, r
m v, z, th, zh, l, r, d, g, j, p
n v, th, zh, b, g, j, s, t
w v, th, z, zh, n, l, d, g, j, p
l b, g, th, j, v, zh, m, w, s, t
r b, m, w, v, th, zh, j, g, s, t
h v, z, zh, p, t, k, ch
ch p, t, k, f, th, s, h, zh
sh v, z, p, t, k, j
zh v, z, b, d, g, j, l, m, n, w, r, ch
j b, d, g, v, th, z, zh, l, m, n, w, r, sh
Phonomotor Treatment manual
Page 93 of 116

Phonemes that share 2 features (differ on 1 feature) – Most difficult to differentiate

BASE 2 shared features

b d, g, m, w, p
p t, k, b
d b, g, z, l, n, r, t
t p, k, s, d
g b, d, k
k p, t, g
th f, s, sh, h, th
th v, z, zh, th
f th, s, sh, h, v
v th, z, zh, f
s f, th, sh, h, z
z v, zh, s
m n, b, w
n m, r, l, z, d
w r, m, b
l d, z, n, r
r n, l, z, d
h f, th, s, sh
ch sh, j
sh f, th, s, h, ch, zh
zh v, j, sh
j zh, ch
Phonomotor Treatment manual
Page 94 of 116

PMT Supplies Checklist

Client candidacy for PMT

Good candidates for PMT typically meet the following criteria:
- Have a phonological processing impairment as part of aphasia, as documented by the
presence of phonologic errors in a variety of linguistic contexts and/or performance on tests
of phonological processing;
- Can repeat at least some single sound stimuli with cues;
- Are willing and able to be actively engaged in the treatment;
- Have sufficient auditory comprehension skills to participate in Socratic questioning;
- Are able to self-reflect (i.e., identify errors and change behavior to fix them);
- Have adequate speech motor programming, planning, and execution skills to participate in
verbal treatment tasks;
- Have good executive function, including abstraction and set shifting.

In our experience, people who exhibit the following characteristics do not respond well to PMT:
- They do not have good buy-in, usually because they do not understand or like the concept
of a treatment using primarily nonwords;
- They are unwilling to actively engage in the treatment tasks;
- They cannot be facilitated to repeat single sounds accurately;
- They have untreated depression.

PMT has been developed specifically for, and tested with, people with aphasia due to a left
hemisphere stroke. No data are available that support its use with other etiologies; it may or may not be
appropriate to use with other diagnoses that may yield acquired language impairments, such as
traumatic brain injury or primary progressive aphasia.
Phonomotor Treatment manual
Page 95 of 116

What supplies are needed to deliver PMT? for details about these items.

Required Recommended
Small mirror Clipboard
Mouth photos Second set of mouth pictures
Voiced/voiceless icons Second set of grapheme tiles
Laminated blank vowel circle chart “Reset” button
Grapheme tiles Plastic bin for storage and transport
Wipe-off board, markers, and eraser
Small colored blocks
Colored felt squares
Blank index cards
Phonomotor Treatment manual
Page 96 of 116

Sample PMT Goals

These sample goals are meant to be used as a guide, to demonstrate how you might incorporate
PMT tasks and skills into a typical goal format for documentation. Please note that these are examples
only, and should be used only as models for building appropriate clinical goals specific to each client.

Speaking
Long term goal 1: Client will expressive increase phonological awareness to reduce the amount
and/or duration of anomic episodes in order to improve conversational efficiency.
Long term goal 2: Client will increase expressive phonological awareness to verbalize wants and
needs effectively with different conversational partners in different communicative contexts.
1. Short term goal 1: Client will accurately verbalize a phoneme when provided with a
_____ stimulus in _____ out of ____ trials and with ______ support.
2. Short term goal 2: After repeating a phoneme (consonants and vowels), client will
identify place, manner, and voice with ___% accuracy.
3. Short term goal 3: Given a written stimulus, client will accurately verbalize a real or non-
word target, as well as identify its associated phonemes and syllables, in ____ out of
____ trials with ____ support.
Listening
Long term goal: Client will increase receptive phonological awareness to improve
comprehension during everyday conversations.
1. Short term goal 1: Client will accurately identify mouth pictures (consonants and
vowels) when provided with an auditory stimulus in _____ out of ____ trials and with
______ support.
2. Short term goal 2: Given an auditory word pair and using blocks/felts as visual support,
client will identify the location of a phoneme change in a syllable chaining task with
____% accuracy.
Reading
Long term goal: Client will develop functional decoding skills to independently read and
understand everyday reading material.
1. Short term goal 1: Client will accurately associate sound to letter (and letter to sound)
correspondences on ____ out of ____ trials.
2. Short term goal 2: Client will identify letters in a field of _____ with ____% accuracy and
______ cues.
Phonomotor Treatment manual
Page 97 of 116

3. Short term goal 3: Client will accurately associate sound to letter correspondences in
order to decode stimuli (real or non-word) strings in _____ out of _____ trials.
Writing
Long term goal: Client will develop functional spelling skills to independently compose everyday
written material (e.g. filling out forms, sending texts or email, etc.).
1. Short term goal 1: Client will write letters (consonants and vowels) given verbally
produced phonemes with ____% accuracy and ______ cues.
2. Short term goal 2: Client will accurately write real or nonword stimuli from dictation in
____% opportunities.
Phonomotor Treatment manual
Page 98 of 116

Sample Data Sheets

Phonomotor Treatment manual
Page 99 of 116
Phonomotor Treatment manual
Page 100 of 116
Phonomotor Treatment manual
Page 101 of 116
Phonomotor Treatment manual
Page 102 of 116
Phonomotor Treatment manual
Page 103 of 116
Phonomotor Treatment manual
Page 104 of 116
Phonomotor Treatment manual
Page 105 of 116
Phonomotor Treatment manual
Page 106 of 116

Glossary of Terms

Agraphia – Impaired ability to write.

Alexia – Impaired reading ability.

Aphasia – An acquired impairment of access to language representations and processes that crosses all
language modalities (verbal expression, auditory comprehension, reading, and writing) and involves
impairments in attention and working memory, as well.

Blending –Taking separate phonemes and merging them into a single syllable (e.g., given b, e, f, create
the syllable bef).

Blocked trials – A series of trials of a task that all use the same stimulus or structure. For example, a
chaining task may only change the final phoneme on each successive trial (e.g., sath  saf  sab 
saj). Contrast with random trials.

Cognate pairs – Two phonemes that differ in only in the phonological feature of voicing (i.e., p/b, f/v,
t/d, k/g, th/th, s/z, sh/zh, ch/dz)

Compensatory treatments – Therapy approaches that are designed to improve a person’s ability to
communicate functionally without an expectation that fundamental language processing skills will
improve. Examples include training to use a communication notebook and using writing or drawing to
supplement verbal communication. Compare with restorative treatments.

Diphthong – A vowel that has two articulatory components blended together, moving from one to the
other (e.g., ie begins with the open-jaw posture of o and moves to the closed-jaw posture of ee)
Contrast with monophthong.

Elision – Removing a single phoneme from a syllable or word (e.g., “Say tesk without saying t “ or “Say
pimz without the z“).

Executive functioning – The collection of cognitive processes that regulate the ability to organize, plan,
implement, and complete tasks. These generally include attention, working memory, self-monitoring,
planning, initiation, and inhibition.

Grapheme – a written letter.

Interactive activation model of language – A model of language that assumes that the different types of
representations in the lexical network (i.e., semantic, lexical, and phonological) are organized in a
somewhat hierarchical manner, but that all elements at each level interact with the related elements at
the levels above and below them.

Language domains – The various sensory and cognitive processing networks that support language
function (e.g., visual, acoustic, tactile-kinesthetic, etc.).

Language modalities – The various ways to use language (e.g., verbal expression, auditory
comprehension, reading, writing).
Phonomotor Treatment manual
Page 107 of 116

Lexical network – A network of words and related language elements that a person has stored in their
memory.

Monophthong – A vowel that has only one articulatory component, so that the beginning and end of
the vowel are relatively consistent with each other. Contrast with diphthong.

Neighborhood density – A measure of how many words in a language are similar to a given word,
generally defined as words that can be made by adding, subtracting, or replacing one sound or letter in
the word. Neighborhoods can be phonological (defined in terms of speech sounds) or orthographic
(defined in terms of letters used in writing, even if the words don’t sound similar). For instance, the
word cat has high phonological neighborhood density, with many words that are phonologically similar
by the “change one sound” definition, including ‘mat’, ‘cast’, ‘cab’, ‘scat’, ‘cot’, and many others. In
contrast, the word elephant has low neighborhood density, with only a few neighbors: ‘element’ and
‘elegant’.

Non-words – Strings of sounds that are legal in the language (i.e., could be a real word), but do not have
any meaning. Examples: frip, smick, bruthbem.

Orthography – Letters used to write words.

Parsing – Taking a syllable (or syllables) and breaking it out into its constituent phonemes (e.g., given
pag, identify the phonemes p, a, and g).

Phonemes – Speech sounds that are used to create words. Phonemes are generally defined by their
phonological features.

Phoneme sequence knowledge – A person’s knowledge of the ways in which sounds in a language can
be used together. This knowledge may not be explicit (i.e., conscious), but is critical to a person’s ability
to assemble words to produce them in speech.

Phonological awareness – A person’s ability to recognize, identify, discriminate between, and

manipulate individual speech sounds outside of the context of real words.

Phonological features – Parameters that are used to distinctively define speech sounds. Each sound in a
language can be defined by its: 1) Place of articulation (i.e., what structures are active in producing the
sound); 2) Manner of articulation (i.e., how those structures move in producing the sound); and 3)
Voicing (i.e., whether the voice is turned on or off when the sound is being produced).

Phonology – A system of speech sounds and sound combinations that are permissible in a language.

Phonotactic probability – A measure of how common a sound combination is in the language. “High
phonotactic probability” means that the sound or sound combination is common in the language (e.g.,
st). “Low phonotactic probability” means that the sound or sound combination is uncommon in the
language (though still permissible; e.g., zh, as in ‘measure’, or the non-word syllable zow).

Random trials – A series of trials of a task that randomly change which element is manipulated from trial
to trial. For example, a chaining task may change the phoneme in a different position on each successive
trial (e.g., sath  saff  suff  luff). Contrast with blocked trials.
Phonomotor Treatment manual
Page 108 of 116

Restorative treatments – Therapy approaches that are designed to improve basic communication
functions and abilities, such as word retrieval, reading skills, and auditory comprehension. Examples
include Phonomotor Treatment, Semantic Feature Analysis, and Treatment of Underlying Forms.
Compare with compensatory treatments.

Rhotic vowels – Vowels produced in the context of r.

Socratic questioning – A process of asking questions to lead the listener on a path of self-discovery.

Speech motor programming/planning – The cognitive and motor processes that translate a linguistic
plan into an executable set of motor commands.

Spondee – A word with two syllables that are relatively equally stressed.

Tactile-kinesthetic – Involving information about how it feels to make a particular speech sound and
how the structures involved move.

Verbal working memory – The combination of verbal short-term memory (which is temporary and has
limited capacity) with attention that allows manipulation of linguistic information.
Phonomotor Treatment manual
Page 109 of 116

Answers and Explanations for Quiz Modules

Quiz Module #1 – Theoretical motivations for PMT

1. Phonomotor treatment is based on the idea that lexical processing occurs…

in networks that are massively interconnected, with representation resulting from the co-
activation of elements across domains.

EXPLANATION FOR Q1: PMT is based on a parallel distributed processing model of language,
which describes language as the product of a massively interconnected network.

2. PMT is rooted in the understanding that lexical processing is supported by…

the interaction of information across sensory-motor domains.

EXPLANATION FOR Q2: Consistent with a parallel distributed processing model of language, PMT
assumes that processing of words involves integrating information across language and related
modalities.

3. In PMT, various behavioral modalities are routinely…

paired between stronger and weaker modalities to facilitate improvement.

EXPLANATION FOR Q3: PMT is designed to use the interconnected nature of the distributed
networks that support language to improve skills and domains that are impaired. One way this is
done is by pairing weaker modalities with relatively preserved skills, to reinforce the weaker
knowledge.

4. The goal of PMT is to…

Build phonological awareness and phonological sequence knowledge.

EXPLANATION FOR Q4: PMT is designed to improve phonological awareness and phonological
sequence knowledge, rather than specific sounds or words. This focus on the fundamental
elements that support all language processing allows the potential for treatment to have wide-
ranging effects across language domains, structures, and lexical items, rather than just on the
items that have been trained.
Phonomotor Treatment manual
Page 110 of 116

Quiz Module #2 – The structure of the PMT program

1. PMT begins with introducing single phonemes and then moves to working with phoneme
combinations.
True

EXPLANATION FOR Q1: The overall structure of the PMT program involves beginning with
training single phonemes. Once each individual phoneme has been trained in isolation, though,
it is moved into short combinations even if there are other phonemes not yet introduced. This is
discussed in greater detail in the section on Progression through the task hierarchy.

2. Treatment focuses on production of specific sounds, and a sound is considered “mastered”

when the client produces it 100% accurately.
False

EXPLANATION FOR Q2: The focus of PMT is not on production of specific sounds; rather, it is on
awareness of sounds and how they are produced and used in language. A single phoneme is
considered “trained” when it has been explored across tasks that require identification,
discrimination, and description of the phoneme in isolation. Once the phoneme has been
trained in isolation it can be moved into combinations with vowels, even if production accuracy
is not yet very high.

3. PMT uses only non-words in treatment.

False

EXPLANATION FOR Q2: While most of PMT uses non-words, there are some carefully selected
real words that should be used later in the treatment program to help encourage linking
phonological information with semantic information. Even when real words are used, though,
the focus of treatment is always on the phonological aspects of the word; semantics are never
directly addressed.

4. Training of syllables only happens after all training of single phonemes is complete.
False

EXPLANATION FOR Q3: It is not necessary to have all phonemes introduced before beginning to
work with syllables. As a client becomes familiar with each individual phoneme, it can be moved
into syllable contexts, even if there are other phonemes still being introduced. The one
limitation on task progression is that all individual phonemes should be introduced before
moving any into multi-syllable combinations.
Phonomotor Treatment manual
Page 111 of 116

Quiz Module #3 – Socratic Questioning

1. Socratic questioning involves…
Using carefully crafted questions to guide the client through exploration of phonemes,
stimuli, and their own responses to treatment tasks.

EXPLANATION FOR Q1: One critical element of PMT is having the client think deeply about
phonology across modalities. This is achieved by using Socratic questioning to guide them
through a process of self-discovery, for both correct and incorrect responses, rather than just
giving them feedback about response accuracy and explaining things to them.

2. Socratic questioning for a single trial usually involves just a single question.
False

EXPLANATION FOR Q2: The Socratic questioning process generally involves a series of questions
to evaluate a stimulus or response. It is an interactive, iterative process, which may take several
minutes for a single treatment trial.

3. The clinician should always use the same questions for the same tasks during PMT.
False

EXPLANATION FOR Q3: Socratic questioning will change over the course of treatment, based on
the client’s current performance, the desired focus of the task (which may change from one
session - or trial - to another), and the relative balance of domain strengths over the course of
recovery. Keep in mind that Socratic questioning isn’t about teaching clients specific information
about phonemes, but is about guiding the client through a process of evaluation, self-reflection,
and assessment with the goal of increasing their awareness, knowledge, and ability to self-
monitor and self-cue.

4. Socratic questioning evolves over the course of treatment, depending on the client’s strengths,
challenges, and current level of success.
True

EXPLANATION FOR Q4: See response to Q3.

5. Socratic questioning should facilitate linking relatively stronger with relatively weaker
modalities.
True

EXPLANATION FOR Q5: One of the strengths of PMT is its pairing of stronger modalities with
those that are weaker; this pairing allows the weaker modality to become stronger. One way
modalities are paired is through the process of Socratic questioning, in which the clinician has
the client explore related aspects of the stimulus or response across modalities. For instance, if
Phonomotor Treatment manual
Page 112 of 116

visual and acoustic analysis skills are relatively strong but awareness of oral movements is poor,
Socratic questioning may first focus on what a sound production looks and sounds like when the
clinician says it, and then shift to having the client look in a mirror while s/he produces it,
describing the sound of their own production, and then describing the articulatory mouth
movements involved in producing that sound.

6. A client with impaired auditory comprehension is not a good candidate for PMT, because
auditory comprehension needs to be fully intact to participate in Socratic questioning.
False

EXPLANATION FOR Q6: It is important for clients to be able to participate in Socratic questioning
in PMT, but this is still possible even if their auditory comprehension is impaired. Auditory
information can be modified or supplemented to support the ability to participate in Socratic
questioning, even if auditory comprehension is impaired.

7. The best way to support a client’s ability to engage in the Socratic questioning process is to…
All of the above.

EXPLANATION OF Q7: The clinician should use whatever methods are effective, in any
combination, to facilitate a client’s ability to participate in Socratic questioning.
Phonomotor Treatment manual
Page 113 of 116

Quiz Module #4 – Conceptual introduction to PMT

1. PMT is intuitive enough that you can jump into treatment without any preliminary explanation.
False

EXPLANATION OF Q1: The approach taken by PMT is unusual, and likely very different from what people
with aphasia have experienced previously in their therapy. While most aphasia treatments have a fairly
obvious tie to functional communication, through working on specific client-driven vocabulary or
functional tasks, PMT’s focus on phonology in non-word contexts can be difficult to understand. Poor
understanding can lead to poor buy-in and commitment to treatment. For these reasons, we have found
it to be critical to provide a thorough explanation of the motivation for doing PMT, as described in this
manual, before beginning treatment, and returning to it as needed throughout the therapy program.

2. The explanation and drawing that represent the purpose of PMT should be provided at the start
of each treatment session.
False

EXPLANATION OF Q2: The detailed explanation of the motivation for using PMT does not need to be
presented prior to each session. In our experience, clients do well having it presented before the first
session, and then it can be referenced briefly if needed as questions or lapses in motivation occur during
treatment.

3. The explanatory drawing that represents the purpose of PMT should be created anew for each
client as you describe the purpose of the treatment, rather than having it pre-drawn.
True

EXPLANATION FOR Q3: The explanation for how and why PMT works is complex, and explaining it to
someone with aphasia adds an extra level of difficulty that the clinician must anticipate. One way to
manage this, and maximize, client understanding, is to build the explanatory drawing(s) gradually,
adding each element as you talk about it. This helps to break down the ideas into manageable pieces,
and in the process serves as multi-modality communication support for clients with impaired auditory
comprehension.
Phonomotor Treatment manual
Page 114 of 116
Phonomotor Treatment manual
Page 115 of 116

References
Algeo, J. (1978). What consonant clusters are possible. Word, 29, 206-224.
doi:10.1080/00437956.1978.11435661
Beeson, P. M., Rising, K., & Volk, J. (2003). Writing treatment for severe aphasia: who benefits? Journal
of Speech Language and Hearing Research, 46, 1038-1060.
Dell, G. S. (1986). A spreading-activation theory of retrieval in sentence production. Psychological
Review, 93, 283-321. doi:10.1037/0033-295X.93.3.283
Dell, G. S., & O'Seaghdha, P. G. (1992). Stages of lexical access in language production. Cognition, 42,
287-314.
Kendall, D. L., Oelke, M., Brookshire, C. E., & Nadeau, S. E. (2015). The Influence of Phonomotor
Treatment on Word Retrieval Abilities in 26 Individuals With Chronic Aphasia: An Open Trial.
Journal of Speech Language and Hearing Research, 58, 798-812. doi:10.1044/2015_JSLHR-L-14-
0131
Lindamood, C. H., & Lindamood, P. C. (1998). The Lindamood Phoneme Sequencing Program for Reading,
Spelling, and Speech. Austin, TX: Pro-Ed.
Martin, N., & Reilly, J. (2012). Short-term/working memory impairments in aphasia: Data, models, and
their application to aphasia rehabilitation. Aphasiology, 26, 253-257.
doi:10.1080/02687038.2011.648163
Nadeau, S. E. (2001). Phonology: A review and proposals from a connectionist perspective. Brain and
Language, 79, 511-579. doi:10.1006/brln.2001.2566
Raymer, A. M., Beeson, P., Holland, A., Kendall, D., Maher, L. M., Martin, N., . . . Gonzalez Rothi, L. J.
(2008). Translational research in aphasia: from neuroscience to neurorehabilitation. Journal of
Speech, Language, and Hearing Research, 51, S259-275. doi:10.1044/1092-4388(2008/020)
Robey, R. R. (1998). A meta-analysis of clinical outcomes in the treatment of aphasia. Journal of Speech
Language and Hearing Research, 41, 172-187.
Storkel, H. L., Armbruster, J., & Hogan, T. P. (2006). Differentiating phonotactic probability and
neighborhood density in adult word learning. Journal of Speech Language and Hearing Research,
49, 1175-1192. doi:10.1044/1092-4388(2006/085)