“Do you have mowing the lawn?”: Improvements in word retrieval and grammar following constraint-induced language therapy in primary progressive aphasia

Background: Much recent progress has been made in developing
speech–language therapy in primary progressive aphasia (PPA).
Several treatment approaches that have shown significant effects
with people with aphasia have been adapted and re-evaluated for
PPA. Constraint-induced aphasia therapy (CIAT) is a well-evaluated
method that has yielded significant language improvements in
people with post-stroke aphasia but has not yet been evaluated
with people with PPA. Nevertheless, the combination of CIAT
features like massed practice and a motivating communicative
setting seem likely to make it a suitable tool for improving the
speech and language performance of individuals with PPA as well.
Aims: This study investigates the effectiveness of a modified CIAT
protocol on word retrieval, grammatical structure and connected
speech in two individuals with non-fluent variant PPA (nfvPPA).
Methods and procedures: Two participants with nfvPPA took part in a
9-day intensive CIAT-based group therapy with additional computerbased
home training. Stimuli were 120 photos of people performing
daily life activities, which could be described using a simple (e.g., “The
man is mowing the lawn”) or reduced (e.g., “mowing the lawn”)
sentence structure. During the treatment phase, the participants
were required to request picture cards from other group members
using spoken language only. The task difficulty was increased hierarchically
(shaped) in accordance to each participant’s performance
level.
Outcomes and results: Directly after therapy, both participants
achieved significant improvements in their noun and verb naming
accuracy and their grammatical structure for trained items. Training
effects were maintained 2 months after therapy. Moreover, generalisation
to different pictures of the same item was found for both
participants and one participant also showed improved grammatical
structure when describing untrained pictures. No significant generalisation
to untrained connected speech samples was observed for either
participant.