Boyle_CHSS_Brown_Bag_Seminar_2-25-11

Report
The Challenge of Moving from
Naming to Discourse Production
Mary Boyle, Ph.D., CCC-SLP, BC-ANCDS
Montclair State University
CHSS Brown Bag Seminar 2/25/11
Burke Rehabilitation Hospital
Montclair State University Separately Budgeted Research Awards
Graduate Assistants
Participants with Aphasia
 Organized
method of activating semantic
networks
 Based on models of lexical retrieval that
conceive of the semantic system as a
network of concepts
(Caramazza, 1997; Collins & Loftus, 1975; Gordon & Dell,
2003; Levelt, 2001; Oppenheim et al., 2010; Lombardi &
Sartori, 2007; McRae et al., 2005)
Semantic Features
red
fruit
grows on trees
APPLE
has skin
used for cider
has seeds
has a core
Semantic networks:
pear
cherry
banana
radish
lettuce
apple
celery
pie
cake
red
fruit
cherry
grows
on trees
Used for
has a
has skin cider has seeds core
APPLE
radish
(After Dell et al, 1997; Levelt, 2000, 2001)
 PWA
actively attempts retrieval of feature
for each target in each treatment session
 Therapist
• Initially plays primary role
• Gradually fades prompts, shifting burden to patient
 Emphasized
importance of using structured
procedure consistently until pt. needs
minimal/no cueing
 Hypothesis: persistent practice with
procedure would lead to more organized
word retrieval without deliberate use of
compensatory strategies

7 participants with mild or moderate
aphasia secondary to L CVA or TBI
benefited from C-SFA
Broca’s (N=2)
Anomic (N=2)
Conduction (N=1)
Wernicke’s (N=1)
fluent aphasia secondary to TBI (N=1)
•
•
•
•
•

(Boyle, 2004; Boyle & Coelho, 1995; Coelho,
McHugh, & Boyle, 2000; Conley & Coelho, 2003)
Improved production of treated nouns
Generalization to untreated nouns from
semantic categories used in treatment
Generalization to untreated nouns from
untreated semantic categories
Improvements maintained 4 weeks after
treatment ended
Those with mild aphasia did better than
those with moderate aphasia
 Macro
versus micro discourse analysis
 Proxy measures versus direct measures
 Direct measures – what do neurologically
normal adults do?
Some participants conveyed
more information during
discourse production
Some conveyed information more
efficiently in discourse
No change in overt
manifestations of word retrieval
behaviors in discourse
 PWA
strip
tells a story about wordless comic
3
single subject designs
• 52 yo female with conduction aphasia
• 61 yo male with Broca’s aphasia
• 61 yo male with Broca’s aphasia
2
60-min sessions/wk for 12 sessions
 Results:
• Improved confrontation naming for treated Ns
• Generalized to untreated Ns (including
semantically unrelated)
• Fewer word-finding behaviors in discourse
• Participants with Broca’s aphasia improved in
informativeness of discourse
• Naïve listeners only judged one of the 3
participants to be better at word retrieval after
treatment
 Test
of Word Finding in Discourse
 German, 1991; Boyle, 2004
Verbal Paraphasia
Reformulations
Initial sounds
Empty/Indefinite Words
Phonemic Paraphasia
Time Fillers
Neologism
Delays
Repetition
Comments
 Elicitation
procedures described by
Nicholas & Brookshire (1993)
3
sessions, 2 to 7 days apart
 Audio
recorded
 Transcript analysis (German, 1991):
• Segmented into T-units
• Total T-units calculated
 Percentage
of T-units with evidence of ANY
word-finding behavior (%TWFB)
 Percentage
of T-units with each specific
word-finding behavior
 Global
measure of WFB in discourse,
%TWFD, stable across sessions
 Distribution
of specific WFBs changed
across sessions for most participants
 More
like everyday discourse for adults
 Word retrieval changes in normal aging
• More empty words
• Longer pauses
• More incorrect object labels
• Hedging about names
 Compared
10 neurologically normal
adults to 3 PWA retelling two episodes
from “I Love Lucy”
 Results
• Empty words and delays were not most common
word-finding behaviors in neurologically normal
adults; reformulations and repetitions were
• Reformulations and repetitions also most
common for PWA
• NN adults rarely produced word-finding
behaviors in fragments
• For PWA, fragments always associated with
word-finding behaviors

similar documents