NEUROPSYCHOLOGICAL REHABILITATION

2019, VOL. 29, NO. 9, 1439–1463

https://doi.org/10.1080/09602011.2018.1425146

Long-Term maintenance of anomia treatment effects in

primary progressive aphasia
Aaron M. Meyera, Donna C. Tippettb, R. Scott Turnerc and Rhonda B. Friedmana
a
Center for Aphasia Research and Rehabilitation, Georgetown University Medical Center, Washington,
DC, USA; bDepartment of Neurology, Johns Hopkins University, Baltimore, MD, USA; cDepartment of
Neurology, Georgetown University Medical Center, Washington, DC, USA

ABSTRACT
This study examined the maintenance of anomia treatment effects in primary progressive
aphasia (PPA). Following baseline testing, a phonological treatment and an orthographic
treatment were administered over the course of six months. The treatment stimuli
consisted of nouns that were consistently named correctly at baseline (Prophylaxis
items) and/or nouns that were consistently named incorrectly at baseline (Remediation
items). Naming accuracy was measured at baseline, and it was measured at 1 month,
8 months, and 15 months post-treatment. The change in naming accuracy from
baseline to each post-treatment evaluation was calculated within each treatment
condition, and within a matched untrained condition. The change in naming accuracy
was then compared between the three conditions. The results of these analyses
indicate that phonological and orthographic treatments are both effective in the
Prophylaxis and Remediation of anomia in all three variants of PPA. For Prophylaxis
items, some of the effects of each treatment can persist for as long as 15 months
post-treatment. These long-term treatment effects were more robust in the
orthographic treatment condition and for participants with the semantic variant of PPA.

ARTICLE HISTORY Received 13 April 2017; Accepted 2 January 2018

KEYWORDS Treatment; Maintenance; Primary progressive aphasia; Anomia

Introduction
Primary progressive aphasia (PPA) is a clinical syndrome characterised by progressive
language impairment (Gorno-Tempini et al., 2011; Mesulam, 1982). Other aspects of
cognition, such as episodic memory and visuospatial skills, are relatively preserved
during the initial phases of the illness. Anomia is a common and early deficit in PPA
(Westbury & Bub, 1997).
Three variants of PPA have been identified: semantic, logopenic, and nonfluent/
agrammatic (Gorno-Tempini et al., 2011). The semantic variant (svPPA) presents with
impaired confrontation naming and single-word comprehension deficits. Impaired
object knowledge, surface dyslexia, or surface dysgraphia may also be present, while
repetition and speech production are typically spared. Autopsy studies have suggested

CONTACT Aaron M. Meyer [email protected] Georgetown University Medical Center,

Building D Suite 207, 4000 Reservoir Road NW, Washington, DC 20057, USA
Supplemental data for this article can be accessed https://doi.org/10.1080/09602011.2018.1425146
© 2018 Informa UK Limited, trading as Taylor & Francis Group
1440 A. M. MEYER ET AL.

that svPPA is associated with TDP-43 positive frontotemporal lobar degeneration (FTLD)
in a majority of cases (Leyton, Britton, Hodges, Halliday, & Kril, 2016; Mesulam et al.,
2014; Snowden, Neary, & Mann, 2007) and atypical Alzheimer’s disease (AD) in a min-
ority of cases (Knibb, Xuereb, Patterson, & Hodges, 2006; Mesulam et al., 2014).
The logopenic variant (lvPPA) presents with impaired single-word retrieval and
impaired repetition of sentences and phrases. Phonological speech errors may also
occur. Single-word comprehension, object knowledge, motor speech, and grammar
are typically spared. Autopsy and imaging studies have suggested that lvPPA is associ-
ated with atypical AD in a majority of cases (Leyton et al., 2016; Mesulam et al., 2014;
Rabinovici et al., 2008) and FTLD in a minority of cases (Mesulam et al., 2014).
The nonfluent/agrammatic variant (nfvPPA) presents with effortful, halting speech
with apraxia, and/or agrammatic language production. Impaired comprehension of syn-
tactically complex sentences may also be present, while single-word comprehension
and object knowledge are typically spared. Autopsy studies have suggested that
nfvPPA is associated with tau-positive inclusions in FTLD, progressive supranuclear
palsy, or corticobasal degeneration in a majority of cases (Hodges et al., 2004; Knibb
et al., 2006; Mesulam et al., 2014) and atypical AD or TDP-43 positive FTLD in a minority
of cases (Knibb et al., 2006; Mesulam et al., 2014).
Semantic deficits and semantic paraphasic errors occur in svPPA, suggesting that
anomia in this subtype results from degraded semantic representations or difficulty
accessing the phonological representation from the semantic representation (Hodges,
Patterson, & Tyler, 1994; Mesulam et al., 2009; Neary et al., 1998). In contrast, phonemic
paraphasic errors are likely to occur in lvPPA (Gorno-Tempini et al., 2008; Henry & Gorno-
Tempini, 2010) and nfvPPA (Ash et al., 2010, 2013), suggesting that the anomic deficit in
these two subtypes is at the level of the phonological representation itself.
A number of studies have investigated anomia treatment in svPPA. Treatment
approaches have included semantic, phonological, and orthographic interventions, as
well as hybrid treatments. These studies have typically found that anomia treatment
has a positive effect (see reviews by Cathery-Goulart et al., 2013; Croot, Nickels, Laur-
ence, & Manning, 2009; Jokel, Graham, Rochon, & Leonard, 2014; see also Beales, Cart-
wright, Whitworth, & Panegyres, 2016; Jokel et al., 2016; Savage, Piguet, & Hodges, 2014,
2015).
Six published studies have examined the treatment of anomia in a single participant
with lvPPA (Beales et al., 2016; Beeson et al., 2011; Croot et al., 2015; Henry et al., 2013;
Meyer, Snider, Eckmann, & Friedman, 2015; Newhart et al., 2009), and all of these
studies have found positive treatment effects. Treatment types have included combined
phonological/orthographic, semantic/orthographic, and semantic/phonological/ortho-
graphic interventions, in addition to separate phonological and orthographic treatments.
Three single or dual case studies have treated anomia for nouns in nfvPPA (Croot
et al., 2015; Jokel, Cupit, Rochon, & Leonard, 2009; Marcotte & Ansaldo, 2010), and all
of these studies have found positive treatment effects. Phonological/orthographic
and semantic treatments have been utilised in this subtype.
In summary, studies of treatment for anomia in PPA have included phonological,
orthographic, semantic, and hybrid treatments, and all appear to have met with
some degree of success within every subtype. However, with rare exceptions, different
types of treatments have not been tested against one another. Only two studies have
compared phonological and semantic treatments in a within-subjects design (Dressel
et al., 2010; Jokel et al., 2016). Jokel et al. found that both types of treatment were
 NEUROPSYCHOLOGICAL REHABILITATION 1441

effective in svPPA, and only one of the four participants showed a significantly larger
effect for semantic treatment than for phonological treatment. Dressel et al. found
that semantic treatment was initially more effective than phonological treatment in a
single participant with svPPA, but this advantage was not maintained at follow-up.
Thus, there is currently little evidence that a particular type of treatment is more effec-
tive within a given subtype of PPA. One of the goals of this study is to directly compare
different types of treatment within subjects.
In addition, it is unclear whether anomia treatment gains in PPA are maintained over
time. Studies that have examined maintenance have reported mixed results, and the
retention interval has typically been relatively short (see Cathery-Goulart et al., 2013;
Croot et al., 2009; Jokel et al., 2014). For example, other than three single- or dual-
case studies (Croot et al., 2015; Meyer, Snider, Eckmann et al., 2015; Snowden &
Neary, 2002), no published studies of anomia treatment in PPA have tested mainten-
ance beyond 6 months post-treatment.
A sufficient treatment length is one factor that may be important for the mainten-
ance of treatment gains. In a recent study, three out of four participants with svPPA
showed significantly greater maintenance of treatment gains for items that were prac-
tised for 6 weeks than for items that were practised for 3 weeks (Savage, Ballard, Piguet,
& Hodges, 2013). Furthermore, after an initial period of intensive treatment, some par-
ticipants may require additional treatment or practice sessions in order to maintain the
initial treatment gains. In a study conducted by Savage et al. (2015), nine participants
with svPPA completed an intensive 2-month anomia treatment programme, and they
were then monitored over a 6-month period. During this period, six of the participants
required additional training sessions in order to maintain at least 80% of their treatment
gains from the initial training programme.
In the current study, treatment was administered for 6 months. The first month
included two treatment sessions per week, while the subsequent 5 months included
monthly treatment sessions, as well as shorter home practice sessions that occurred
three times per week.
Two treatments were administered during the 6-month period. One treatment
focused on phonology, while the other treatment focused on orthography. In the pho-
nological treatment condition (PTC), an auditorily presented word occurs in conjunction
with the corresponding picture, and the participant repeats the word. The goal of this
treatment is to strengthen the phonological representations of the treated words,
thereby bolstering their production (see Figure 1(a)). In the orthographic treatment con-
dition (OTC), the written word occurs in conjunction with the corresponding picture,
and the participant reads the word out loud and copies it. The goal of this treatment
is to strengthen the orthographic representations of the treated words, thereby bolster-
ing the alternative, orthographic route from the semantic representation to the phono-
logical representation (see Figure 1(b)).
OTC was predicted to be more effective than PTC in svPPA. If a word’s semantic rep-
resentation is not completely degraded, and the anomia is due in part to difficulty acces-
sing the phonological representation from the semantic representation, then facilitation
of the orthographic route may circumvent the direct semantics-to-phonology route.
Furthermore, if the primary locus of the anomic deficit is the semantics-to-phonology
pathway, rather than the semantic representation itself, then OTC would also be
expected to result in stimulus generalisation (i.e., generalisation to alternative exemplars
of trained items), because an alternative picture exemplar would be expected to
1442 A. M. MEYER ET AL.

Figure 1. The ovals depict internal representations, while the rectangles depict external stimuli and outputs. Bold
arrows identify the stimuli that were paired during treatment, solid thin arrows represent the pathways that are
normally activated during confrontation naming, dashed thin arrows represent additional pathways that were
activated during treatment, and shaded ovals depict the representations that are intended to be strengthened
by treatment.

activate the trained item’s strengthened orthographic representation via the semantics-
to-orthography route (see Figure 1(b)).
PTC and OTC were predicted to have similar levels of effectiveness in lvPPA and
nfvPPA, since each treatment was expected to facilitate access to phonological rep-
resentations, either by strengthening these representations (in the case of PTC) or by
bolstering an alternative route to these representations (in the case of OTC). Further-
more, in these subtypes stimulus generalisation was predicted to result from both treat-
ments, since semantic representations were not expected to be impaired.
Naming accuracy was measured at baseline, and it was measured at 1 month, 8
months, and 15 months post-treatment. At each time point, accuracy for the selected
items was tested in four ways: (1) Oral naming accuracy for untrained and trained
items (Exemplar set 1) was measured. (2) Stimulus generalisation was examined by
having participants orally name an alternative exemplar of each untrained and
trained item (Exemplar set 2). (3) In the written naming task, participants were asked
to print the name of each picture from Exemplar set 1. (4) In the naming during
scene description task, each participant was asked to describe a series of visual
scenes, and each scene contained one of the untrained or trained items. This task
was used to assess task generalisation.

Method
Participants
Potential participants were referred by neurologists, clinical neuropsychologists, and
speech-language pathologists in the Washington, DC and Baltimore metropolitan
 NEUROPSYCHOLOGICAL REHABILITATION 1443

areas. The inclusion criteria were a clinical diagnosis of PPA, English fluency since child-
hood, at least 10 years of education, age of at least 40 years, and no history of other
neurological or psychiatric disorders. All participants had clinical MRI (magnetic reson-
ance imaging) scans or MRI scans for research purposes that ruled out causes other than
neurodegeneration.
A total of 41 individuals enrolled in the study. Three individuals were removed from
the study because their medical history and/or baseline assessment results indicated
that they did not have PPA. Eleven participants withdrew from the study before com-
pleting treatment or the post-treatment evaluation. Out of these 11, one withdrew
during the baseline assessment due to a lack of interest; one withdrew because he
did not understand the treatment tasks; one developed other health problems; one
had a spouse who developed health problems; three moved to a different state or
country; and four withdrew because they did not think the treatments were beneficial.
Twenty-seven individuals with PPA completed treatment and at least one post-treat-
ment evaluation, including 12 with lvPPA, five with svPPA, nine with nfvPPA, and one
with mixed PPA (with features of both nfvPPA and svPPA at initial presentation).1 The
latter participant was excluded from the data analyses. Demographic information is pre-
sented in Tables 1–3.
Subtyping was based on the international criteria (Gorno-Tempini et al., 2011) and
each participant’s baseline assessment results and medical history, including the
results of prior language and neuropsychological testing, which was typically completed
at the time of the clinical PPA diagnosis. In a few advanced cases, participants received a
subtype diagnosis that was based in part on prior testing, but they developed additional
deficits before the baseline assessment. For example, LV3 and LV4 both showed evi-
dence of a semantic deficit at the time of the baseline assessment (see Table 1).

Treatment stimuli
For each participant, up to 120 items were selected from a set of 294 nouns. For each
selected item, there were three different picture exemplars. Oral naming accuracy for
Exemplar set 1 was tested twice during the baseline evaluation. This exemplar set
was utilised during treatment, and naming accuracy for this set was tested during
the post-treatment assessments. Oral naming accuracy for Exemplar set 2 was tested
once at baseline. Exemplar set 2 was not utilised during treatment, but it was used to
assess stimulus generalisation during post-treatment testing. Pictures from Exemplar
set 3 were only used as foils during treatment. As determined by norming conducted
with unimpaired controls, Exemplar sets 1 and 2 have high name agreement. For
every item, at least 80% of the controls produced the target word, and this was true
for both exemplars. For Exemplar 1, the unimpaired group consisted of 24 individuals
with a mean age of 52.3 (SD = 7.0) and mean education of 15.4 years (SD = 2.4). For
Exemplar 2, the unimpaired group consisted of 24 individuals with a mean age of
51.4 (SD = 7.5) and mean education of 15.9 years (SD = 2.1).
Trained and untrained items were selected during the baseline evaluation. Each
selected item was either named correctly by the participant during all three of the base-
line oral naming tests (Prophylaxis items), or it was named incorrectly during all three of
these tests (Remediation items). All 26 participants had Prophylaxis items, and 12 of the
participants also had Remediation items (see the Appendix). All of the selected words
were read and repeated accurately at baseline. The selected items were divided into
 1444
A. M. MEYER ET AL.
Table 1. Demographic information and baseline assessment results for participants with lvPPA.
LV1 LV2 LV3 LV4 LV5 LV6 LV7 LV8 LV9 LV10 LV11 LV12 M (SD)
Age at baseline 69 66 71 88 73 68 67 67 71 70 51 69 69.2 (8.1)
Education (years) 18 18 18 16 18 18 14 18 16 19 18 18 17.4 (1.4)
Sex F F F M F M F F F M F F
Symptom duration (months) 45 67 72 16 49 39 96 46 44 56 15 46 49.3 (22.4)
Time post-diagnosis (months) 2 11 12 12 13 17 2 5 9 7 2 3 7.9 (5.1)
FBI/63 11 0 11 14 11 3 8 33 12 23 3 6 11.3 (9.1)
MoCA/30 12 20 16 5 21 18 12 20 11 3 19 22 14.9 (6.3)
BNT/60 15 46 18 32 31 34 13 37 24 7 50 27 27.8 (13.1)
P&PT, 3 Pictures/52 49 51 40 35 50 51 48 49 49 47 47 51 47.3 (4.9)
Word–picture matching/48 44 48 43 45 48 48 47 48 48 48 48 47 46.8 (1.8)
NAT/10 6 9 9 5 7 5 7 5 5 NA 3 9 6.4 (2.0)
BDAE articulatory agility/7 7 7 7 5 7 7 6 6 7 7 7 7 6.7 (0.7)
BDAE phrase length/7 7 7 7 7 7 7 7 7 7 7 7 7 7 (0)
BDAE embedded sentences/10 4 10 10 7 10 6 3 5 6 5 5 10 6.8 (2.6)
BDAE sentence repetition/10 3 7 8 4 6 7 2 4 6 2 8 9 5.5 (2.4)
Pseudoword repetition/10 7 9 NA 9 NA 0 0 6 7 6 7 10 6.1 (3.5)
Reading (irregular minus regular) 0 NA −1 0 −2 −1 −4 −5 0 -−4 0 −1 −1.6 (1.9)
Spelling (irregular minus regular) −7 NA −5 −3 0 2 NA 0 −6 −1 0 0 −2.0 (3.1)
Note: NA = not administered. SD = standard deviation.
 NEUROPSYCHOLOGICAL REHABILITATION 1445

Table 2. Demographic information and baseline assessment results for participants with svPPA.
SV1 SV2 SV3 SV4 SV5 M (SD)
Age at baseline 68 71 61 59 69 65.6 (5.3)
Education (years) 16 20 16 18 16 17.2 (1.8)
Sex F M M F M
Symptom duration (months) 264 43 46 18 96 93.4 (99.5)
Time post-diagnosis (months) 25 20 10 13 7 15.0 (7.4)
FBI/63 18 49 8 11 2 17.6 (18.5)
MoCA/30 1 20 19 12 20 14.4 (8.2)
BNT/60 10 14 11 6 5 9.2 (3.7)
P&PT, 3 Pictures/52 22 38 45 17 41 32.6 (12.3)
Word–picture matching/48 39 43 43 15 33 34.6 (11.7)
NAT /10 0 4 9 6 9 5.6 (3.8)
BDAE articulatory agility/7 7 7 7 7 7 7 (0)
BDAE phrase length/7 6 7 7 7 7 6.8 (0.4)
BDAE embedded sentences/10 1 9 10 4 8 6.4 (3.8)
BDAE sentence repetition/10 3 10 10 2 9 6.8 (4.0)
Pseudoword repetition/10 4 4 10 5 10 6.6 (3.1)
Reading (irregular minus regular) −2 −3 −4 −8 −7 −4.8 (2.6)
Spelling (irregular minus regular) NA NA −7 NA −6 −6.5 (0.7)

three sets [Phonological Treatment Condition (PTC), Orthographic Treatment Condition

(OTC), and the Untrained Condition (UC)], and they were matched across sets for fre-
quency (Baayen, Piepenbrock, & Gulikers, 1995), semantic category, and length
(number of syllables, phonemes, and letters). Possible semantic categories included
Animals (e.g., cow), Appliances (e.g., toaster), Body Parts (e.g., arm), Clothing (e.g.,
shoe), Food (e.g., banana), Furniture (e.g., desk), Musical Instruments (e.g., guitar),
Items from Nature (e.g., cloud), Common Objects (e.g., candle), People (e.g., doctor),
Structures (e.g., bridge), Tools (e.g., hammer), and Vehicles (e.g., truck).
Participants typically had 40 items per treatment condition, resulting in 80 trained
items per session. Eight participants did not have 40 critical items per treatment con-
dition because they had fewer than 120 items that met the selection criteria and
could be matched across the treatment conditions. For these eight participants (LV4,
LV7, LV8, LV9, NFV1, NFV8, NFV9, and SV2), filler items were included in order to
reach 40 items per condition (see the Appendix). The range across participants was
1–11 filler items per condition (each participant had an equal number of fillers across
treatment conditions). The fillers were selected from the items that were not consist-
ently named correctly or incorrectly at baseline, and/or the items that could not be
matched across the treatment conditions (on frequency, semantic category, or
length). The fillers were not included in the statistical analyses.

Procedure
Baseline evaluation
The baseline evaluation occurred over the course of six sessions, with one or two ses-
sions per week. During these sessions, participants completed a battery of language
and cognitive tests, including the Montreal Cognitive Assessment (MoCA) (Nasreddine
et al., 2005), the Boston Naming Test (BNT) (Kaplan, Goodglass, & Weintraub, 2001), the
Three-Picture version of the Pyramids and Palm Trees test (P&PT; Howard & Patterson,
1992), Word–picture matching (Rogers & Friedman, 2008), subject and object Wh-ques-
tions from the Northwestern Anagram Test (NAT) (Weintraub et al., 2009), selected
 1446
A. M. MEYER ET AL.
Table 3. Demographic information and baseline assessment results for participants with nfvPPA.
NFV1 NFV2 NFV3 NFV4 NFV5 NFV6 NFV7 NFV8 NFV9 M (SD)
Age at baseline 48 55 67 75 74 76 81 69 68 68.1 (10.5)
Education (years) 12 16 15 18 18 16 12 18 12 15.2 (2.6)
Sex M M M M M F F M F
Symptom duration (months) 84 59 17 26 14 36 212 96 76 68.9 (61.5)
Time post-diagnosis (months) 2 9 5 1 3 14 210 15 1 28.9 (68.1)
FIB/63 17 20 28 11 16 4 5 13 8 13.6 (7.7)
MoCA/30 17 24 27 19 26 19 16 14 21 20.3 (4.5)
BNT/60 24 46 57 53 44 45 38 15 29 39 (13.8)
P&PT, 3 Pictures/52 48 48 50 51 49 51 44 48 46 48.3 (2.3)
Word–picture matching/48 48 48 48 47 48 45 46 45 48 47 (1.3)
NAT/10 3 5 8 0 9 5 6 4 10 5.6 (3.1)
BDAE articulatory agility/7 2 4 6 6 7 6 6 4 4 5.0 (1.6)
BDAE phrase length/7 2 7 7 7 6 4 4 4 6 5.2 (1.8)
BDAE embedded sentences/10 9 10 10 7 10 7 8 4 10 8.3 (2.1)
BDAE sentence repetition/10 1 8 9 8 9 6 5 2 7 6.1 (2.9)
Pseudoword repetition/10 0 7 8 6 10 4 5 0 8 5.3 (3.5)
Reading (irregular minus regular) 2 1 0 0 −1 1 −1 −3 −3 −0.4 (1.7)
Spelling (irregular minus regular) 2 1 0 NA −1 0 1 NA NA 0.5 (1.0)
 NEUROPSYCHOLOGICAL REHABILITATION 1447

subtests from the Boston Diagnostic Aphasia Examination (BDAE) (Goodglass, Kaplan, &
Barresi, 2001), repetition of five-syllable pseudowords (Meyer, Snider, Campbell, & Fried-
man, 2015), and the reading and spelling of irregular and regular words (the reading and
spelling tasks were developed at the Center for Aphasia Research and Rehabilitation at
Georgetown University Medical Center). In addition, each participant’s caregiver com-
pleted a modified version of the Frontal Behavioral Inventory (FBI) (Kertesz, Davidson,
& Fox, 1997). The modified FBI omitted three questions: Concreteness, Verbal Apraxia,
and Alien Hand. The baseline assessment results are presented in Tables 1–3.
Individualised treatment words and matched untrained items were selected as
described above. After stimulus selection, naming accuracy for all of the selected
items from each participant’s individualised sets was also tested in two other ways:
written naming and naming during scene description. In the first task, the participant
was asked to print the name of each Exemplar 1 picture. In the second task, the partici-
pant was asked to describe a series of 120 visual scenes. Each scene consisted of a
colour photograph that contained one of the participant’s selected items, as well as
other objects that were appropriate to the scene. If the participant did not describe
the portion of the scene that contained the target item, the experimenter pointed to
this region and said, “What about over here?” If the participant produced the correct
name of the target, the item was scored as correct, unless the correct name was com-
bined with a negative particle (e.g., “That’s not an elephant”). The naming during scene
description task was used to assess task generalisation.

Treatment timeline
Following the baseline evaluation, treatment took place during the next six months. In
the first month of treatment, there were two sessions per week. Each session lasted
about 45 minutes and included both types of treatment. These sessions included a
spaced retrieval recognition task to aid in stimulus encoding [see Meyer, Tippett, and
Friedman (2018) for additional details]. Home practice sessions occurred over the sub-
sequent 5 months. During this 5-month period, participants and caregivers were
instructed to complete shorter (10 to 15 minutes) practice sessions at home three
times per week. Three individuals participated remotely (see Meyer, Getz, Brennan,
Hu, & Friedman, 2016), and the experimenter conducted all sessions with these partici-
pants, including home practice. For all participants, one treatment session was also con-
ducted by the experimenter each month to help ensure that the participant was
performing the tasks correctly and to help the participant remain engaged in the
study. Thus, participants completed a total of 13 treatment sessions, and they com-
pleted 55 home practice sessions, on average. The post-treatment evaluation began
1 month after the end of all treatment and practice sessions.

Orthographic treatment condition

In OTC, E-Prime (Psychology Software Tools) was used to present stimuli in the follow-
ing sequence: (1) Picture alone, 1.5 seconds. (2) The written word under the picture, in
one of 15 fonts, 1.5 seconds. (3) The word alone, 1.5 seconds. (4) The picture–word
combination (PWC) appeared again, and the participant was asked to read the
word aloud. (5) A beep then signalled the participant to copy the word onto a
response sheet. (6) Two recognition slides were presented in succession, with the
words “Did you see this?” and either the correct PWC or a foil (see Figure 2). Instruc-
tions specified that both the identical exemplar of the picture and the word in the
1448 A. M. MEYER ET AL.

Figure 2. Example item (diamond) from the OTC. In this example, the foil in the recognition task is an incorrect
picture exemplar, paired with the word in the correct font.

identical font had to be present for a “Yes” response. The participant responded by
saying “Yes” or “No.” The foil used for each PWC was one of the following: (1) the
correct picture paired with the written word in a second, incorrect font; (2) an incorrect
exemplar of the picture paired with the correct font; or (3) the incorrect exemplar of
the picture with the second, incorrect font. To perform this task correctly, visual
aspects of both the picture and the word must be encoded (i.e., this task cannot be
done verbally). The purpose of this task is to ensure that the participant is focusing
on both the picture and the written word.

Phonological treatment condition

This treatment was similar in design to the OTC. However, the picture was not
accompanied by its written name. Instead, it was accompanied by a string of symbols
(e.g., #$#$$) in a specific font. The symbol string, though unrelated to the picture,
was included in this condition so that the participant would perform the same task
as in the OTC, thereby keeping the conditions well matched for activity and engage-
ment. The participant was asked to look at the picture and string of symbols on the
screen as the pre-recorded name was presented auditorily, and then to repeat the
name (see Figure 3).

Order of presentation
The order of the two treatments alternated between sessions (e.g., PTC first, OTC
second). Each treated item was presented once per session. For each treatment,
there were three item presentation orders. Each set of three consecutive treatment ses-
sions cycled through the three orders, and then the cycle began again.

Home practice sessions

The home practice sessions were similar to the treatment sessions, except that the rec-
ognition tests were omitted. Each participant was given training cards and instructed to
perform these tasks three times per week with a caregiver who lived with the
participant.
 NEUROPSYCHOLOGICAL REHABILITATION 1449

Figure 3. Example item (necklace) from the PTC. In this example, the foil in the recognition task is the symbol
string in an incorrect font, paired with the correct picture exemplar.

Each card in the OTC had one picture on the front. The back of the card had the same
picture with the corresponding written word. The participant looked at the picture on
the front, turned the card over, read aloud the name of the picture, and then copied
the name on a response sheet.
Each card in the PTC had one picture on the front, while the back had the same
picture with the associated symbol string. The participant looked at the picture on
the front, and then looked at the picture and symbols on the back. The caregiver
then spoke the name of the picture, which the participant repeated.
PTC was typically administered first within a home practice session, although some
caregivers occasionally administered OTC first. The caregiver was instructed to shuffle
the cards within each treatment set between sessions. The caregiver also ensured
that the participant practised the two sets appropriately and kept a practice session
log. The experimenters collected the cards and session log at the end of the 6-month
treatment period. On average, participants with lvPPA, svPPA, or nfvPPA completed
52, 55, or 59 home practice sessions, respectively. The number of home practice sessions
was not significantly different across subgroups (all p’s > .34).

Post-treatment and follow-up evaluations

Evaluations were conducted 1 month, 8 months, and 15 months after the final treat-
ment session was completed. During these evaluations, accuracy was measured for
each task: Exemplar 1 naming, Exemplar 2 naming, written naming, and naming
during scene description. Each task occurred during a different session.

Data analysis
At each post-treatment and follow-up time point, the change in naming accuracy from
baseline was calculated for each item type (Prophylaxis or Remediation) within each
treatment condition (UC, PTC, or OTC). For each combination of task and item type, a
3 × 3 (Treatment Condition × Subtype) mixed-design analysis of variance (ANOVA)
was conducted. In the nfvPPA subgroup, only one participant (NFV8) had Remediation
items; his Remediation data were not included in the ANOVAs, but they are presented in
Supplemental Figure 1.
1450 A. M. MEYER ET AL.

When the Treatment Condition × Subtype interaction or the main effect of Treat-
ment Condition was significant, the effect was explored with post-hoc paired-samples
t tests. The main effect of Subtype was not of theoretical interest and was not explored.
For each ANOVA, Mauchly’s procedure was used to test for violations of the sphericity
assumption; Mauchly’s test was never significant.

Results
The mean change in naming accuracy for Prophylaxis items is plotted in Figure 4
(svPPA), Figure 5 (lvPPA), Figure 6 (nfvPPA), and Figure 7 (PPA).

One month post-treatmen, Prophylaxis items

Twenty-six participants completed all four of the naming tasks at this time point.

Oral naming, Prophylaxis items

The Treatment Condition × Subtype interaction was significant for the oral naming of
Exemplar 1 [F(4, 46) = 4.21, p = .005] and Exemplar 2 [F(4, 46) = 7.60, p < .001]. Compared

Figure 4. Mean change in naming accuracy for Prophylaxis items, svPPA subgroup. For this figure and those that
follow, the change in naming accuracy from baseline to each post-treatment time point is plotted in percentage
points, and the bars represent the standard error.

Figure 5. Mean change in naming accuracy for Prophylaxis items, lvPPA subgroup.

Figure 6. Mean change in naming accuracy for Prophylaxis items, nfvPPA subgroup.
 NEUROPSYCHOLOGICAL REHABILITATION 1451

Figure 7. Mean change in naming accuracy for Prophylaxis items, all participants.

with UC, the svPPA subgroup showed significantly less decline for both exemplars in
both PTC and OTC (see Table 4). The nfvPPA subgroup showed a similar pattern,
except that the effect for Exemplar 1 in PTC did not reach significance. The lvPPA sub-
group showed significantly less decline for Exemplar 1 in OTC and Exemplar 2 in PTC.
The main effect of Treatment Condition was significant for Exemplar 1 [F(2, 46) =
25.03, p < .001] and Exemplar 2 [F(2, 46) = 39.17, p < .001]. Compared with UC, there
was significantly less decline for both exemplars in both PTC and OTC (see Table 4).

Written naming, Prophylaxis items

The Treatment Condition × Subtype interaction was significant [F(4, 46) = 3.19, p = .021].
The svPPA and nfvPPA subgroups showed a significantly smaller decline in OTC, com-
pared with UC. Neither of these subgroups demonstrated a significantly smaller decline
in PTC, and the lvPPA subgroup did not show a significant difference between UC and
either treatment condition.
The main effect of Treatment Condition was also significant [F(2, 46) = 14.68,
p < .001]. Compared with UC, there was significantly less decline in both PTC and OTC.

Naming during scene description, Prophylaxis items

The Treatment Condition × Subtype interaction was significant [F(4, 46) = 5.16, p = .002].
The svPPA and nfvPPA subgroups showed a significantly smaller decline in OTC, com-
pared with UC. Neither of these subgroups demonstrated a significantly smaller decline
in PTC, and the lvPPA subgroup did not show a significant difference between UC and
either treatment condition.
The main effect of Treatment Condition was also significant [F(2, 46) = 12.90,
p < .001]. Compared with UC, there was significantly less decline in OTC, but the
decline was not significantly smaller in PTC.

Eight months post-treatment, Prophylaxis items

Oral naming, Prophylaxis items
Twenty-one participants completed the two oral naming tasks at this time point. Four
participants withdrew before this evaluation (see the Appendix), while LV7 did not par-
ticipate in this evaluation, but she returned for the evaluation at 15 months post-
treatment.
The Treatment Condition × Subtype interaction was significant for the oral naming of
Exemplar 1 [F(4, 36) = 4.33, p = .006] and Exemplar 2 [F(4, 36) = 2.86, p = .037]. For Exem-
plar 1, none of the subgroups showed a significant difference between UC and either
treatment condition (see Table 4). For Exemplar 2, the lvPPA subgroup showed
 1452
A. M. MEYER ET AL.
Table 4. UC vs. trained prophylaxis items.
Exemplar 1 Exemplar 2 Written naming Scenes
PTC OTC PTC OTC PTC OTC PTC OTC
t p η2 t p η2 t p η2 t p η2 t p η2 t p η2 t p η2 t p η2
1 month post-treatment
sv 4.86 .008 .86 3.53 .024 .76 7.05 .002 .93 5.50 .005 .88 2.08 .106 .52 5.52 .005 .88 2.25 .088 .56 3.35 .029 .74
lv 1.45 .176 .16 2.54 .027 .37 3.52 .005 .53 1.74 .110 .22 1.01 .335 .08 1.63 .132 .19 −.151 .883 .00 0.62 .551 .03
nfv 2.13 .066 .36 2.79 .023 .49 2.65 .029 .47 2.68 .028 .47 0.72 .493 .06 4.03 .004 .67 1.29 .232 .17 2.72 .026 .48
PPA 3.74 .001 .36 4.36 <.001 .43 5.14 <.001 .51 4.11 <.001 .40 2.10 .046 .15 4.11 <.001 .40 1.75 .093 .11 3.01 .006 .27
8 months post-treatment
sv 2.69 .074 .71 2.46 .091 .67 2.24 .111 .63 2.60 .080 .69 – – – – – – 3.51 .039 .80 4.86 .017 .89
lv 1.75 .118 .28 1.47 .179 .21 4.96 .001 .75 3.54 .008 .61 – – – – – – 0.38 .714 .02 0.13 .900 .00
nfv 2.10 .074 .39 −0.54 .605 .04 0.79 .456 .08 1.41 .202 .22 – – – – – – 0.11 .914 .00 1.61 .168 .34
PPA 2.99 .007 .31 2.24 .037 .20 4.06 .001 .45 3.87 .001 .43 0.36 .722 .01 3.14 .006 .37 1.90 .074 .17 2.40 .027 .24
15 months post-treatment
sv – – – – – – 4.26 .024 .86 4.89 .016 .89 – – – – – – 2.95 .060 .74 3.53 .039 .81
lv – – – – – – −0.80 .456 .10 −0.40 .701 .03 – – – – – – 0.28 .788 .01 1.65 .150 .31
nfv – – – – – – −1.47 .216 .35 1.63 .178 .40 – – – – – – 0.00 1.00 .00 0.97 .406 .24
PPA – – – – – – 0.77 .454 .04 1.67 .116 .16 0.89 .388 .05 2.99 .010 .39 1.48 .161 .13 2.48 .026 .31
Note: α = .05. Significant effects appear in bold italics.
 NEUROPSYCHOLOGICAL REHABILITATION 1453

significantly less decline in PTC and OTC, while the other subgroups did not show a sig-
nificant difference between UC and either treatment condition.
The main effect of Treatment Condition was significant for Exemplar 1 [F(2, 36) =
12.02, p < .001] and Exemplar 2 [F(2, 36) = 14.94, p < .001]. Compared with UC, there
was significantly less decline for both exemplars in both PTC and OTC (see Table 4).

Written naming, Prophylaxis items

Eighteen participants completed the written naming task at this time point (LV3 and LV4
were unable to write at this time point, and NFV7 withdrew before the task was admi-
nistered). The Treatment Condition × Subtype interaction was not significant [F(4, 30) =
0.42, p = .794], but the main effect of Treatment Condition was significant [F(2, 30) =
5.72, p = .008]. Compared with UC, the decline in naming accuracy was significantly
smaller in OTC, but the decline was not significantly smaller in PTC (see Table 4).

Naming during scene description, Prophylaxis items

Nineteen participants completed the scene description task at this time point (NFV7
withdrew before the task was administered, and NFV8 declined to complete the
task). The Treatment Condition × Subtype interaction was significant [F(4, 32) = 6.48,
p = .001]. Compared with UC, the svPPA subgroup showed significantly less decline in
both PTC and OTC. The other subgroups did not show a significant difference
between UC and either treatment condition.
The main effect of Treatment Condition was significant [F(2, 32) = 11.60, p < .001].
Compared with UC, the decline in naming accuracy was significantly smaller in OTC,
but the decline was not significantly smaller in PTC.

Fifteen months post-treatment, Prophylaxis items

Oral naming, Prophylaxis items
Sixteen participants completed the two oral naming tasks at this time point. Nine par-
ticipants withdrew before this time point (see the Appendix), while NFV6 developed
mutism before this evaluation and only completed written naming. The Treatment Con-
dition × Subtype interaction was not significant for Exemplar 1 [F(4, 26) = 1.17, p = .347],
but it was significant for Exemplar 2 [F(4, 26) = 10.88, p < .001]. Compared with UC, the
svPPA subgroup showed significantly less decline in both PTC and OTC, while the lvPPA
and nfvPPA subgroups did not show a significant difference in either treatment con-
dition (see Table 4).
The main effect of Treatment Condition was not significant for Exemplar 1 [F(2, 26) =
2.26, p = .124], but it was significant for Exemplar 2 [F(2, 26) = 10.63, p < .001]. However,
there was no significant difference between UC and PTC or UC and OTC (see Table 4).

Written naming, Prophylaxis items

Fifteen participants completed the written naming task at this time point (LV3 and LV7
were unable to complete the task). The Treatment Condition × Subtype interaction was
not significant [F(4, 24) = 1.90, p = .143]. The main effect of Treatment Condition was sig-
nificant [F(2, 24) = 5.68, p = .010]. Compared with UC, the decline in naming accuracy
was significantly smaller in OTC, but the decline was not significantly smaller in PTC
(see Table 4).
1454 A. M. MEYER ET AL.

Naming during scene description, Prophylaxis items

Fifteen participants completed the scene description task at this time point (NFV1’s
speech was severely telegraphic and he was unable to complete the task; NFV6 was
mute). The Treatment Condition × Subtype interaction was significant [F(4, 24) = 5.31,
p = .003]. Compared with UC, the svPPA subgroup showed significantly less decline in
OTC, but the effect did not reach significance for PTC. The lvPPA and nfvPPA subgroups
did not show a significant difference in either treatment condition (see Table 4).
The main effect of Treatment Condition was significant [F(2, 24) = 9.21, p = .001].
Compared with UC, the decline in naming accuracy was significantly smaller in OTC,
but the decline was not significantly smaller in PTC (see Table 4).

One month post-treatment, Remediation items

The change in naming accuracy for Remediation items is plotted in Figure 8. Eleven par-
ticipants with Remediation items and lvPPA or svPPA completed all four of the naming
tasks at 1 month post-treatment.

Oral naming, Remediation items

The Treatment Condition × Subtype interaction was not significant for the oral naming
of Exemplar 1 [F(2, 18) = 1.18, p = .332] or Exemplar 2 [F(2, 18) = 0.53, p = .600]. However,
the main effect of Treatment Condition was significant for both Exemplar 1 [F(2, 18) =
13.64, p < .001] and Exemplar 2 [F(2, 18) = 9.31, p = .002]. Compared with UC, the
increase in naming accuracy for each exemplar was significantly greater in each treat-
ment condition [PTC, Exemplar 1: t(10) = 4.65, p = .001, η 2 = .68; PTC, Exemplar 2: t(10)
= 4.78, p = .001, η 2 = .70; OTC, Exemplar 1: t(10) = 3.71, p = .004, η 2 = .58; OTC, Exemplar
2: t(10) = 3.63, p = .005, η 2 = .57].

Written naming, Remediation items

The Treatment Condition × Subtype interaction was not significant [F(2, 18) = 0.17,
p = .842], but the main effect of Treatment Condition was significant [F(2, 18) = 10.97,
p = .001]. Compared with UC, the increase in naming accuracy was significantly
greater in each treatment condition [PTC: t(10) = 3.00, p = .013, η 2 = .47; OTC: t(10) =
4.50, p = .001, η 2 = .67].

Naming during scene description, Remediation items

The Treatment Condition × Subtype interaction was not significant [F(2, 18) = 0.68,
p = .519], but the main effect of Treatment Condition was significant [F(2, 18) = 7.13,
p = .005]. Compared with UC, the increase in naming accuracy was significantly

Figure 8. Mean change in naming accuracy for Remediation items.

 NEUROPSYCHOLOGICAL REHABILITATION 1455

greater in each treatment condition [PTC: t(10) = 2.87, p = .017, η 2 = .45; OTC: t(10) =
3.11, p = .011, η 2 = .49].

Eight months post-treatment, Remediation items

Seven participants with Remediation items and lvPPA or svPPA completed all four of the
naming tasks at this time point, while LV3 completed the two oral naming tasks and
naming during scene description. SV2 and LV10 withdrew before this evaluation,
while LV7 did not participate in this evaluation but returned for the evaluation at 15
months post-treatment.

Oral naming, Remediation items

The Treatment Condition × Subtype interaction was not significant for Exemplar 1 [F(2,
12) = 0.16, p = .858] or Exemplar 2 [F(2, 12) = 0.42, p = .665]. The main effect of Treatment
Condition was significant for Exemplar 1 [F(2, 12) = 5.53, p = .020], but not for Exemplar 2
[F(2, 12) = 1.86, p = .198]. Compared with UC, the increase in naming accuracy for Exem-
plar 1 was significantly greater in OTC [t(7) = 4.50, p = .003, η 2 = .74], but not in PTC [t(7)
= 1.60, p = .154, η 2 = .27].

Written naming, Remediation items

Neither the Treatment Condition × Subtype interaction [F(2, 10) = 2.76, p = .111] nor the
main effect of Treatment Condition [F(2, 10) = 3.23, p = .083] was significant.

Naming during scene description, Remediation items

Neither the Treatment Condition × Subtype interaction [F(2, 12) = 0.03, p = .968] nor the
main effect of Treatment Condition [F(2, 12) = 2.65, p = .112] was significant.

Fifteen months post-treatment, Remediation items

Oral naming, Remediation items
Eight participants with Remediation items and lvPPA or svPPA completed the two oral
naming tasks at this time point (SV2, LV10, and LV12 withdrew before this time point).
The Treatment Condition × Subtype interaction was not significant for Exemplar 1 [F(2,
12) = 2.27, p = .146] or Exemplar 2 [F(2, 12) = 0.27, p = .770]. Furthermore, the main effect
of Treatment Condition was not significant for Exemplar 1 [F(2, 12) = 1.30, p = .309] or
Exemplar 2 [F(2, 12) = 0.20, p = .824].

Written naming, Remediation items

Six participants with Remediation items completed the written naming task at this time
point (LV3 and LV7 were unable to complete the task). Neither the Treatment Condition
× Subtype interaction [F(2, 8) = 2.13, p = .181] nor the main effect of Treatment Con-
dition [F(2, 8) = 1.54, p = .272] was significant.

Naming during scene description, Remediation items

Eight participants with Remediation items completed the scene description task at this
time point. Neither the Treatment Condition × Subtype interaction [F(2, 12) = 1.13,
p = .356] nor the main effect of Treatment Condition [F(2, 12) = 0.21, p = .814] was
significant.
1456 A. M. MEYER ET AL.

Discussion
The goal of the current study was to examine the long-term maintenance of anomia
treatment effects in PPA. Following a baseline evaluation of language and cognition,
a phonological treatment and an orthographic treatment were administered over the
course of 6 months. Naming accuracy was measured at baseline, and it was measured
at 1 month, 8 months, and 15 months post-treatment. The change in naming accuracy
from baseline to each post-treatment evaluation was calculated for the two treatment
conditions and a set of matched untrained items. The change in accuracy was then com-
pared between the three conditions.

Prophylaxis items
For the svPPA subgroup, both treatments resulted in significantly less decline in naming
accuracy for both exemplars at 1 month post-treatment. These findings indicate that
both treatments were effective in the Prophylaxis of anomia in svPPA, and that the treat-
ment effects generalised to alternative exemplars of trained items (stimulus generalis-
ation). Furthermore, in OTC the svPPA subgroup showed less decline in naming
accuracy during the scene description task, compared with UC. This effect was signifi-
cant at all post-treatment time points. In contrast, PTC only resulted in significantly
less decline for this task at 8 months post-treatment, although there was a trend
towards significance at the other time points. These findings suggest that task general-
isation occurred in both conditions, but the effect was more robust in OTC.
At 1 month post-treatment, the svPPA and nfvPPA subgroups showed a similar
pattern of significant effects, except that the nfvPPA subgroup did not show signifi-
cantly less decline for Exemplar 1 in PTC (although there was a trend towards signifi-
cance). However, in nfvPPA the significant effects for the oral naming tasks were not
maintained at subsequent time points. This pattern of results may be related to the
further progression of verbal apraxia in most of the participants with this subtype
(NFV1, NFV2, NFV3, NFV6, NFV8). For these five participants, verbal agility (from the
BDAE) declined from baseline [M = 7.0, SD = 2.5] to 8 months post-treatment [M = 4.2,
SD = 3.2; t(4) = 4.2, p = .013; NFV7 did not complete this task at 8 months]. In contrast,
verbal agility did not decline for two other participants (NFV4, NFV5; M = 7.5, SD = 0.7
at baseline; M = 8.0, SD = 0 at 8 months post-treatment), but these participants also
showed little evidence of naming decline for Exemplars 1 and 2 in any treatment con-
dition (10% or less at any time point), making it difficult to detect treatment effects for
Prophylaxis items.
At 1 month post-treatment the lvPPA subgroup showed significantly less decline for
Exemplar 1 in OTC and Exemplar 2 in PTC. In contrast, at 8 months post-treatment the
lvPPA subgroup showed significantly less decline for Exemplar 2 in both conditions, but
no significant effects for Exemplar 1. Similarly, at 15 months post-treatment the svPPA
subgroup showed significantly less decline for Exemplar 2 in both conditions, with no
significant effects for Exemplar 1. This pattern may be the result of a testing effect.
Six out of nine lvPPA participants were tested on Exemplar 1 before Exemplar 2 at 8
months post-treatment, and three out of four svPPA participants were tested on Exem-
plar 1 before Exemplar 2 at 15 months post-treatment. By providing retrieval practice,
the initial testing session may have facilitated access for treated items during the
 NEUROPSYCHOLOGICAL REHABILITATION 1457

subsequent testing session (Friedman, Sullivan, Snider, Luta, & Jones, 2017; Middleton,
Schwartz, Rawson, & Garvey, 2015; Roediger & Butler, 2011).
Written naming was a trained task in OTC, but not in PTC. At 1 month post-treatment,
the svPPA and nfvPPA subgroups showed significantly less decline in written naming
accuracy in OTC, compared with UC. Furthermore, at the PPA group level, there was sig-
nificantly less decline in both treatment conditions at this time point. However, at 8 and
15 months post-treatment there was significantly less decline in OTC, but the decline in
accuracy was not significantly different between PTC and UC. These findings suggest
that task generalisation occurred in PTC but was not maintained over time.

Remediation items
The Remediation analyses included participants with svPPA or lvPPA. For Remediation
items, there were no significant Treatment Condition × Subtype interaction effects.
However, there were significant effects at the PPA group level. Compared with UC,
both treatments resulted in significantly greater improvement in naming accuracy for
both exemplars at 1 month post-treatment. These findings indicate that both treat-
ments were effective in the Remediation of anomia, and that the treatment effects gen-
eralised to alternative exemplars of trained items (stimulus generalisation). However, at
8 months post-treatment, there was only one significant effect for the oral naming tasks.
Compared with UC, OTC resulted in significantly greater improvement in naming accu-
racy for Exemplar 1. At 15 months post-treatment, no effects were significant.
Compared with UC, both treatments resulted in significantly greater improvement in
naming accuracy during scene description at 1 month post-treatment, indicating that
task generalisation occurred in both treatment conditions. However, these effects
were not maintained at subsequent time points.
At 1 month post-treatment, both treatments resulted in significantly greater
improvement in written naming accuracy, compared with UC. Since written naming
was an untrained task in PTC, the significant effect in this condition suggests that
task generalisation occurred for this treatment. However, there were no significant
effects for the written naming task at subsequent time points.

Comparison of Prophylaxis and Remediation items

Compared with Prophylaxis items, treatment effects for Remediation items appeared to
dissipate more quickly. Some of the differences in the pattern observed across item
types can be explained by the smaller sample size for the group of participants with
Remediation items. For example, for Remediation items the difference between OTC
and UC was not significant for the naming during scene description task at 8 months
post-treatment, while there was a significant difference for Prophylaxis items, but the
effect size was actually numerically larger for Remediation items (η 2 = .38) than for Pro-
phylaxis items (η 2 = .24).
In contrast, the different patterns at 15 months post-treatment do not appear to be
related to sample size. At this time point, the difference between OTC and UC for
naming during scene description was significant for Prophylaxis items, but not for
Remediation items, and the effect size was much larger for Prophylaxis items (η 2
= .31) than for Remediation items (η 2 = .005). One might assume that this pattern of
results could be due to smaller treatment effects among participants who had
1458 A. M. MEYER ET AL.

Remediation items, since all of these participants had severe anomia (see BNT scores in
Tables 1 –3). However, when the seven participants who did not have Remediation
items (at this time point) are removed from the analysis, the effect size for Prophylaxis
items is numerically larger (η 2 = .47). These findings indicate that task generalisation
persists for a longer period of time for Prophylaxis items than for Remediation items.
Moreover, this difference between Prophylaxis and Remediation items suggests that
the enhancement of largely intact representations or connections endures longer
than the enhancement of degraded representations or connections. One explanation
for this pattern is that many of the Prophylaxis items may have had initial strengths
that were considerably higher than the threshold for successful naming performance,
while Remediation items were (by definition) originally below threshold. If Remediation
and Prophylaxis items gained similar amounts of additional strength from treatment
and then lost similar amounts following treatment, Remediation items might fall
below threshold again, while Prophylaxis items might remain above threshold
(because they had further to drop before reaching threshold).
These findings have important clinical implications. If the speech-language pathol-
ogist’s time with the patient is limited to a brief period, focusing treatment on a set
of Prophylaxis items that are highly relevant to the patient could be more beneficial
than focusing treatment on a set of Remediation items, since the treatment effects
for the former set of items are likely to persist for a longer period of time.
It is unclear why a significant task generalisation effect would occur at 15
months post-treatment, while there were no significant treatment effects at this
time point for the oral naming of Exemplar set 1. The Exemplar 1 naming task
was more similar to the treatment tasks, and it included the picture stimuli that
were used during treatment. One possibility is that naming performance in the
scene description task (which was administered in the final session) was facilitated
by a testing effect for treated items. Another possibility is that naming during scene
description was facilitated by a more visually realistic depiction of the target item.
In contrast to the other three tasks, which utilised line drawings, the scene descrip-
tion task utilised colour photographs. In any case, a trend towards less decline in
OTC was present for Exemplar 1 at 15 months post-treatment, suggesting that a
significant treatment effect might have occurred for this task if the sample size
had been larger.

The roles of subtype and treatment type

In svPPA, OTC was predicted to be more effective than PTC. However, the svPPA sub-
group showed a positive response to both treatments (although treatment effects for
the naming during scene description task appeared to be more robust in OTC). In a
study that compared phonological and semantic treatments, Jokel et al. (2016) found
that all four of their participants with svPPA responded to phonological treatment,
and only one of their participants showed a significantly greater effect for semantic
treatment than for phonological treatment. In individuals with svPPA, anomia may be
caused by degraded semantic representations, impaired lexical retrieval, or a combi-
nation of the two (Mesulam et al., 2009). Some individuals with svPPA are able to
comprehend certain words that they are unable to retrieve during confrontation
naming tasks (Jokel, Rochon, & Anderson, 2010; Jokel, Rochon, & Leonard, 2006;
Mesulam et al., 2009, 2013), which suggests that retrieval failure for these items
 NEUROPSYCHOLOGICAL REHABILITATION 1459

may result from degraded lexical–semantic connections. Phonological treatment

may strengthen these degraded lexical–semantic connections. Nevertheless, treat-
ment effects were more consistent across tasks and time points in OTC, suggesting
that a strengthened orthographic route may be more beneficial than strengthened
phonological representations or strengthened lexical–semantic connections.
In lvPPA and nfvPPA, PTC and OTC were both predicted to be effective. In lvPPA, the
results were consistent with this prediction (although there was variability across Exem-
plars at 1 month post-treatment). In nfvPPA, treatment effects were more consistent in
OTC, with significant effects in all four tasks at 1 month post-treatment. Thus, similar to
svPPA, the results suggest that individuals with nfvPPA may derive greater benefit from
a strengthened orthographic route than from strengthened phonological
representations.
Another possibility is that both treatments strengthen lexical–semantic connections.
One way to evaluate lexical–semantic impairment (i.e., a disconnection between the
concept and corresponding word) vs. conceptual–semantic impairment (i.e., difficulty
with the concept itself) is through administration of the Word and Picture versions of
the P&PT, respectively. Greater impairment on P&PT Words than on P&PT Pictures
suggests that lexical–semantic impairment is greater than conceptual impairment,
while the opposite pattern suggests that conceptual impairment is greater than
lexical–semantic impairment. In the current study, seven participants (LV1, LV3, LV4,
LV12, NFV1, NFV6, NFV8) completed both P&PT tasks at baseline. As can be seen in
Figure 9, participants who scored higher on P&PT pictures than on P&PT words
tended to show larger treatment effects for Exemplar 2 (stimulus generalisation) at 1
month post-treatment. While the correlations were not significant with only seven
participants, the trends suggest that individuals with lexical–semantic impairment
may be more likely to benefit from phonological or orthographic treatment. In contrast,
individuals with conceptual impairment may be more likely to benefit from a treatment
that is intended to strengthen semantic representations, such as semantic feature
analysis (Boyle & Coelho, 1995; Reilly, 2016). We will evaluate these hypotheses in a
future study.
At 15 months post-treatment, five participants (LV6, LV11, NFV3, NFV4, NFV5)
showed little evidence of picture-naming decline for Exemplars 1 and 2 (decline of 10
percentage points or less in every treatment condition). For participants with nfvPPA,
this pattern may be related to a later emergence of anomia for nouns (Cotelli et al.,

Figure 9. Relationship between the relative lexical–semantic impairment at baseline and the treatment effect
(trained minus untrained) for Exemplar 2 at 1 month post-treatment.
1460 A. M. MEYER ET AL.

2006; Hillis et al., 2006; Hillis, Oh, & Ken, 2004; Hillis, Tuffiash, & Caramazza, 2002; Silveri &
Ciccarelli, 2007; Thompson, Lukic, King, Mesulam, & Weintraub, 2012). While verb-
naming impairment may emerge earlier in nfvPPA, verb-naming impairment has also
been found in lvPPA and in some individuals with svPPA (Thompson et al., 2012).
Thus, Remediation and/or Prophylaxis of anomia for verbs may be beneficial in all var-
iants of PPA. Verb stimuli will be included in a future study.

Conclusions
The findings of this study indicate that phonological and orthographic treatments
are effective in the Prophylaxis and Remediation of anomia in all three subtypes of
PPA. Furthermore, for Prophylaxis items, some of the effects of each treatment can
persist for as long as 15 months post-treatment. These long-term treatment effects
were more robust in the orthographic treatment condition and for participants
with svPPA.

Note
1. Post-treatment results from 21 of the current study’s 26 participants (without a mixed diagnosis)
were included in Meyer, Faria, Tippett, Hillis, and Friedman (2017) (all but LV4, LV11, LV12, NFV8,
and NFV9). Post-treatment results from 17 of the current study’s participants were included in
Meyer, Getz, et al. (2016) (all but LV9, LV10, LV11, LV12, SV5, NFV6, NFV7, NFV8, and NFV9). In
addition, post-treatment results from nine of the current study’s participants with lvPPA (LV1-
LV9) and all of the participants with svPPA were included in Meyer, Tippett, et al. (2018).

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This study was supported by the National Institute on Deafness and Other Communication Disorders
under grant numbers R01DC011317 and R01DC011317-01AS1.

References
Ash, S., Evans, E., O’Shea, J., Powers, J., Boller, A., Weinberg, D., … Grossman, M. (2013). Differentiating
primary progressive aphasias in a brief sample of connected speech. Neurology, 81, 329–336.
Ash, S., McMillan, C., Gunawardena, D., Avants, B., Morgan, B., Khan, A., … Grossman, M. (2010). Speech
errors in progressive non-fluent aphasia. Brain and Language, 113, 13–20.
Baayen, R. H., Piepenbrock, R., & Gulikers, L. (1995). The CELEX lexical database (Release 2) [CD-ROM].
Philadelphia: Linguistic Data Consortium, University of Pennsylvania.
Beales, A., Cartwright, J., Whitworth, A., & Panegyres, P. K. (2016). Exploring generalisation processes fol-
lowing lexical retrieval intervention in primary progressive aphasia. International Journal of Speech-
Language Pathology, 18, 299–314.
Beeson, P. M., King, R. M., Bonakdarpour, B., Henry, M. L., Cho, H., & Rapcsak, S. Z. (2011). Positive effects of
language treatment for the logopenic variant of primary progressive aphasia. Journal of Molecular
Neuroscience, 45, 724–736.
Boyle, M., & Coelho, C. A. (1995). Application of semantic feature analysis as a treatment for aphasic dys-
nomia. American Journal of Speech-Language Pathology, 4, 94–98.
Cathery-Goulart, M. T., DaSilveira, A. D., Machado, T. H., Mansur, L. L., DeMattos Pimenta Parente, M. A.,
Senaha, M. L. H., … Nitrini, R. (2013). Nonpharmacological interventions for cognitive impairments
 NEUROPSYCHOLOGICAL REHABILITATION 1461

following primary progressive aphasia: A systematic review of the literature. Dementia &
Neuropsychologia, 7, 122–131.
Cotelli, M., Borroni, B., Manenti, R., Alberici, A., Calabria, M., Agosti, C., … Cappa, S. F. (2006). Action and
object naming in frontotemporal dementia, progressive supranuclear palsy, and corticobasal
degeneration. Neuropsychology, 20, 558–565.
Croot, K., Nickels, L., Laurence, F., & Manning, M. (2009). Impairment- and activity/participation-directed
interventions in progressive language impairment: Clinical and theoretical issues. Aphasiology, 23,
125–160.
Croot, K., Taylor, C., Abel, S., Jones, K., Krein, L., Hameister, I., … Nickels, L. (2015). Measuring gains in con-
nected speech following treatment for word retrieval: A study with two participants with primary pro-
gressive aphasia. Aphasiology, 29, 1265–1288.
Dressel, K., Huber, W., Frings, L., Kümmerer, D., Saur, D., Mader, I., … Abel, S. (2010). Model-oriented
naming therapy in semantic dementia: A single-case fMRI study. Aphasiology, 24, 1537–1558.
Friedman, R. B., Sullivan, K. L., Snider, S. F., Luta, G., & Jones, K. T. (2017). Leveraging the test effect to
improve maintenance of the gains achieved through cognitive rehabilitation. Neuropsychology, 31,
220–228.
Goodglass, H., Kaplan, E., & Barresi, B. (2001). Boston diagnostic aphasia examination (3rd ed.). Austin: Pro-Ed.
Gorno-Tempini, M. L., Brambati, S. M., Ginex, V., Ogar, J., Dronkers, N. F., Marcone, A., … Miller, B. L. (2008).
The logopenic/phonological variant of primary progressive aphasia. Neurology, 71, 1227–1234.
Gorno-Tempini, M. L., Hillis, A. E., Weintraub, S., Kertesz, A., Mendez, M., Cappa, S. F., … Grossman, M.
(2011). Classification of primary progressive aphasia and its variants. Neurology, 76, 1006–1014.
Henry, M. L., & Gorno-Tempini, M. L. (2010). The logopenic variant of primary progressive aphasia. Current
Opinion in Neurology, 23, 633–637.
Henry, M. L., Rising, K., DeMarco, A. T., Miller, B. L., Gorno-Tempini, M. L., & Beeson, P. M. (2013). Examining
the value of lexical retrieval treatment in primary progressive aphasia: Two positive cases. Brain and
Language, 127, 145–156.
Hillis, A. E., Heidler-Gary, J., Newhart, M., Chang, S., Ken, L., & Bak, T. H. (2006). Naming and comprehension
in primary progressive aphasia: The influence of grammatical word class. Aphasiology, 20, 246–256.
Hillis, A. E., Oh, S., & Ken, L. (2004). Deterioration of naming nouns versus verbs in primary progressive
aphasia. Annals of Neurology, 55, 268–275.
Hillis, A. E., Tuffiash, E., & Caramazza, A. (2002). Modality-specific deterioration in naming verbs in nonflu-
ent primary progressive aphasia. Journal of Cognitive Neuroscience, 14, 1099–1108.
Hodges, J. R., Davies, R. R., Xuereb, J. H., Casey, B., Broe, M., Bak, T. H., … Halliday, G. M. (2004).
Clinicopathological correlates in frontotemporal dementia. Annals of Neurology, 56, 399–406.
Hodges, J. R., Patterson, K., & Tyler, L. K. (1994). Loss of semantic memory: Implications for the modularity
of mind. Cognitive Neuropsychology, 11, 505–542.
Howard, D., & Patterson, K. (1992). The pyramids and palm trees test: A test of semantic access from words
and pictures. Bury St. Edmunds, UK: Thames Valley Test Company.
Jokel, R., Cupit, J., Rochon, E., & Leonard, C. (2009). Relearning lost vocabulary in nonfluent progressive
aphasia with MossTalk words®. Aphasiology, 23, 175–191.
Jokel, R., Graham, N. L., Rochon, E., & Leonard, C. (2014). Word retrieval therapies in primary progressive
aphasia. Aphasiology, 28, 1038–1068.
Jokel, R., Kielar, A., Anderson, N. D., Black, S. E., Rochon, E., Graham, S., … Tang-Wai David F., Tang-Wai, D.
F. (2016). Behavioural and neuroimaging changes after naming therapy for semantic variant primary
progressive aphasia. Neuropsychologia, 89, 191–216.
Jokel, R., Rochon, E., & Anderson, N. D. (2010). Errorless learning of computer-generated words in a patient
with semantic dementia. Neuropsychological Rehabilitation, 20, 16–41.
Jokel, R., Rochon, E., & Leonard, C. (2006). Treating anomia in semantic dementia: Improvement, mainten-
ance, or both? Neuropsychological Rehabilitation, 16(3), 241–256.
Kaplan, E., Goodglass, H., & Weintraub, S. (2001). Boston naming test (2nd ed.). Philadelphia, PA: Lippincott,
Williams, & Wilkins.
Kertesz, A., Davidson, W., & Fox, H. (1997). Frontal behavioral inventory: Diagnostic criteria for frontal lobe
dementia. Canadian Journal of Neurological Sciences/Journal Canadien des Sciences Neurologiques, 24,
29–36.
Knibb, J. A., Xuereb, J. H., Patterson, K., & Hodges, J. R. (2006). Clinical and pathological characterization of
progressive aphasia. Annals of Neurology, 59, 156–165.
1462 A. M. MEYER ET AL.

Leyton, C. E., Britton, A. K., Hodges, J. R., Halliday, G. M., & Kril, J. J. (2016). Distinctive pathological mech-
anisms involved in primary progressive aphasias. Neurobiology of Aging, 38, 82–92.
Marcotte, K., & Ansaldo, A. I. (2010). The neural correlates of semantic feature analysis in chronic aphasia:
Discordant patterns according to the etiology. Seminars in Speech and Language, 31, 52–63.
Mesulam, M. M. (1982). Slowly progressive aphasia without generalized dementia. Annals of Neurology,
11, 592–598.
Mesulam, M., Rogalski, E., Wieneke, C., Cobia, D., Rademaker, A., Thompson, C., … Weintraub, S. (2009).
Neurology of anomia in the semantic variant of primary progressive aphasia. Brain, 132, 2553–2565.
Mesulam, M., Weintraub, S., Rogalski, E. J., Wieneke, C., Geula, C., & Bigio, E. H. (2014). Asymmetry and het-
erogeneity of Alzheimer’s and frontotemporal pathology in primary progressive aphasia. Brain, 137,
1176–1192.
Mesulam, M., Wieneke, C., Hurley, R., Rademaker, A., Thompson, C. K., Weintraub, S., … Rogalski, E. J.
(2013). Words and objects at the tip of the left temporal lobe in primary progressive aphasia. Brain,
136, 601–618.
Meyer, A. M., Faria, A. V., Tippett, D. C., Hillis, A. E., & Friedman, R. B. (2017). The relationship between base-
line volume in temporal areas and post-treatment naming accuracy in primary progressive aphasia.
Aphasiology, 31, 1059–1077. doi:10.1080/02687038.2017.1296557
Meyer, A. M., Getz, H. R., Brennan, D., Hu, T., & Friedman, R. B. (2016). Telerehabilitation of anomia in
primary progressive aphasia. Aphasiology, 30, 483–507.
Meyer, A. M., Snider, S. F., Campbell, R. E., & Friedman, R. B. (2015). Phonological short-term memory in
logopenic variant primary progressive aphasia and mild Alzheimer’s disease. Cortex, 71, 183–189.
Meyer, A. M., Snider, S. F., Eckmann, C. B., & Friedman, R. B. (2015). Prophylactic treatments for anomia in
the logopenic variant of primary progressive aphasia: Cross-language transfer. Aphasiology, 29, 1062–
1081.
Meyer, A. M., Tippett, D. C., & Friedman, R. B. (2018). Prophylaxis and remediation of anomia in the seman-
tic and logopenic variants of primary progressive aphasia. Neuropsychological Rehabilitation, 28, 352–
368. doi:10.1080/09602011.2016.1148619
Middleton, E. L., Schwartz, M. F., Rawson, K. A., & Garvey, K. (2015). Test-enhanced learning versus errorless
learning in aphasia rehabilitation: Testing competing psychological principles. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 41, 1253–1261.
Nasreddine, Z. S., Phillips, N. A., Bédirian, V., Charbonneau, S., Whitehead, V., Collin, I., … Chertkow, H.
(2005). The Montreal cognitive assessment, MoCA: A brief screening tool For mild cognitive impair-
ment. Journal of the American Geriatrics Society, 53, 695–699.
Neary, D., Snowden, J. S., Gustafson, L., Passant, U., Stuss, D., Black, S., … Benson, D. F. (1998).
Frontotemporal lobar degeneration: A consensus on clinical diagnostic criteria. Neurology, 51,
1546–1554.
Newhart, M., Davis, C., Kannan, V., Heidler-Gary, J., Cloutman, L., & Hillis, A. E. (2009). Therapy for naming
deficits in two variants of primary progressive aphasia. Aphasiology, 23, 823–834.
Rabinovici, G. D., Jagust, W. J., Furst, A. J., Ogar, J. M., Racine, C. A., Mormino, E. C., … Gorno-Tempini, M. L.
(2008). Aβ amyloid and glucose metabolism in three variants of primary progressive aphasia. Annals of
Neurology, 64, 388–401.
Reilly, J. (2016). How to constrain and maintain a lexicon for the treatment of progressive semantic
naming deficits: Principles of item selection for formal semantic therapy. Neuropsychological
Rehabilitation, 26, 126–156.
Roediger, H. L., & Butler, A. C. (2011). The critical role of retrieval practice in long-term retention. Trends in
Cognitive Sciences, 15, 20–27.
Rogers, S. L., & Friedman, R. B. (2008). The underlying mechanisms of semantic memory loss in
Alzheimer’s disease and semantic dementia. Neuropsychologia, 46, 12–21.
Savage, S. A., Ballard, K. J., Piguet, O., & Hodges, J. R. (2013). Bringing words back to mind – improving
word production in semantic dementia. Cortex, 49, 1823–1832.
Savage, S. A., Piguet, O., & Hodges, J. R. (2014). Giving words new life: Generalisation of word retraining
outcomes in semantic dementia. Journal of Alzheimer’s Disease, 40, 309–317.
Savage, S. A., Piguet, O., & Hodges, J. R. (2015). Cognitive intervention in semantic dementia: Maintaining
words over time. Alzheimer’s Disease and Associated Disorders, 29, 55–62.
Silveri, M. C., & Ciccarelli, N. (2007). Naming of grammatical classes in frontotemporal dementias:
Linguistic and non linguistic factors contribute to noun-verb dissociation. Behavioural Neurology, 18,
197–206.
 NEUROPSYCHOLOGICAL REHABILITATION 1463

Snowden, J. S., & Neary, D. (2002). Relearning of verbal labels in semantic dementia. Neuropsychologia, 40,
1715–1728.
Snowden, J., Neary, D., & Mann, D. (2007). Frontotemporal lobar degeneration: Clinical and pathological
relationships. Acta Neuropathologica, 114, 31–38.
Thompson, C. K., Lukic, S., King, M. C., Mesulam, M. M., & Weintraub, S. (2012). Verb and noun deficits in
stroke-induced and primary progressive aphasia: The northwestern naming battery. Aphasiology, 26,
632–655.
Weintraub, S., Mesulam, M. M., Wieneke, C., Rademaker, A., Rogalski, E. J., & Thompson, C. K. (2009). The
northwestern anagram test: Measuring sentence production in primary progressive aphasia. American
Journal of Alzheimer’s Disease & Other Dementias, 24, 408–416.
Westbury, C., & Bub, D. (1997). Primary progressive aphasia: A review of 112 cases. Brain and Language, 60,
381–406.

Appendix. Number of critical items per treatment condition, average

word frequency, and number of time points
Participant Prophylaxis Remediation Critical items per condition Number of time points
SV1 10 (50) 30 (14) 40 4
SV2 8 (46) 31 (15) 39 2
SV3 19 (67) 21 (6) 40 4
SV4 12 (97) 28 (13) 40 4
SV5 9 (56) 31 (9) 40 4
LV1 26 (52) 14 (9) 40 4
LV2 40 (15) 0 40 4
LV3 20 (65) 20 (11) 40 4
LV4 32 (33) 0 32 3
LV5 23 (52) 17 (8) 40 4
LV6 40 (30) 0 40 4
LV7 18 (40) 11 (11) 29 3
LV8 34 (24) 0 34 3
LV9 34 (38) 0 34 2
LV10 8 (77) 32 (10) 40 2
LV11 40 (20) 0 40 4
LV12 30 (18) 10 (5) 40 3
NFV1 32 (43) 0 32 4
NFV2 40 (28) 0 40 4
NFV3 40 (22) 0 40 4
NFV4 40 (15) 0 40 4
NFV5 40 (17) 0 40 4
NFV6 40 (18) 0 40 4
NFV7 40 (30) 0 40 3
NFV8 27 (62) 10 (8) 37 3
NFV9 32 (29) 0 32 2
Note: Treatment Condition refers to UC, PTC, or OTC. The average word frequency (per million) is in parentheses.
For eight participants (SV2, LV4, LV7, LV8, LV9, NFV1, NFV8, and NFV9), filler items were included in order to
reach a total of 40 items per condition. LV3 did not complete written naming at 8 or 15 months post-treatment.
LV4 did not complete written naming at 8 months post-treatment. LV7 participated in the baseline, 1month
post-treatment, and 15 months post-treatment evaluations, but she did not participate in the evaluation at
8 months post-treatment, and she did not complete written naming at 15 months post-treatment. NFV1 did
not complete scene description at 15 months post-treatment. NFV6 did not complete the oral naming or
scene description tasks at 15 months post-treatment. NFV7 did not complete written naming or scene descrip-
tion at 8 months post-treatment. NFV8 did not complete scene description at 8 months post-treatment.
Copyright of Neuropsychological Rehabilitation is the property of Taylor & Francis Ltd and
its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.