THE EFFECTS OF GRAPHIC TOPIC SETTERS ON CONVERSATIONS WITH A SEVERELY APHASIC COMMUNICATOR

1
THE EFFECTS OF GRAPHIC TOPIC SETTERS ON
CONVERSATIONS WITH A SEVERELY APHASIC COMMUNICATOR

• Kathryn L. Garrett, Ph.D., CCC-SLP
• Duquesne University, Pittsburgh, PA
• Christine Huth, MS CFY
• Allegheny Intermediate Unit, Pittsburgh, PA-

2
Abstract

• In this single subject experiment, a
  commu-nicator with severe expressive aphasia
  conversed with 2 partners about a personal event
  and a news event in baseline, graphic
  topic-setter, and instructional conditions.
  Dependent variables included exchanges per
  topic, participant initiations, successful
  exchanges, modality, and elapsed topic time.
  Qualitative ratings were also obtained. Results
  showed increases in the number of turns, length
  of time, success-fulness, and ability of the
  participant to initiate/add novel info in the
  contextual con-ditions. Topic and Partner effects
  were noted.

3
Background/Review of the Lit

• Previous research showed that communicators with
  severe aphasia can participate more fully in
  interactions using the Written Choice
  Conversational Strategy (Garrett Beukelman,
  1995). However, this strategy does not allow
  communicators to introduce novel, semantically
  specific information to a conversational
  interaction
• Tangible context (visual representations of
  topics, remnants) have been used to stimulate the
  aphasic communicators to convey novel topical
  information (Beukelman, Yorkston, Dowden, 1985
  Weiss Ho, 1997).
• However, little specific data are available to
  analyze the impact of these external topical
  aids on the communication interactions of
  severely aphasic individuals

4
Purposes of this Study

• To investigate how graphic contextual information
  affects the ability of an individual with severe
  aphasia to initiate and sustain conversation in
  the following conditions
• 1) Baseline vs. Context vs. ContextInstruction
• 2) across 2 communication partners
• 3) Personal Event vs. Current Event

5
Participant Demographics - S.D.

• Male, 72 years old, married
• 1 year post thrombotic left CVA
• Diagnosis of moderate-to-severe Brocas aphasia
  with coexisting Apraxia of Speech
• Western Aphasia Battery Scores (Kertesz
  McCabe, 1982)
• Spontaneous Speech.. 3.5/20
• Comprehension .8.35/10
• Repetition.. 1.2/10
• Naming.. 2/10
• Overall WAB Aphasia Quotient 30/100

6
Methods

• Design
• Modified Single Subject (ABAC) Sequential Design
• Partner 1 2 baseline sessions (A1) 4 graphic
  context sessions (B1) 2 baseline sessions
  (A2) 8 sessions total
• Partner 2 3 baseline sessions (A1) 4
  graphic context sessions (B1) 3 baseline
  sessions (A2) 4 instructional sessions
  (C1) 2 graphic context sessions (B2) 16
  sessions total

7
Methods

• Data Collection Procedures
• Topic Selection personal event topic obtained
  from interview with spouse prior to session
  current event from 1st or 2nd headliner story in
  primary newspaper sections.
• Topic Representation key word outlines prepared
  for context conditions (see below) in accordance
  with pre-set criteria (6-8 lines, large print,
  declarative phrases, etc.)
• General Instructions Talk for as long as you
  want about . Or .. (8 minute cap). Topic
  presentation was randomly ordered. Each
  interaction was videofilmed.
• Instruction Condition Rules for conversation
  were introduced (see below) then participant
  rehearsed the conversation with the primary
  investigator prior to data collection. Feedback
  provided to communicator on opportunities for
  initiation, adequacy of message, etc.

8
Sample Graphic Topic Setters

• Personal Event Current Event

9
Sample Topics

• Current Events
• Womens World Cup Soccer Championship
• Pirates Baseball Game
• JFK Jr. Plane Crash
• Kennedy Funeral
• Drought Emergency in PA
• John Deere Golf Tournament
• Bonfire Log Collapse at Texas A M

• Personal Events
• Lunch with Daughter
• Playing Bingo
• Wedding of a Family Friend
• Bowling
• Visit to Grandkids
• Aphasia Support Group Meeting
• Thanksgiving Dinner
• Christmas Shopping with Wife Daughter

10
Sample Instructional Chart
11
Methods Continued

• Data Analysis Procedures/Dependent Vars.
• Tapes were transcribed and coded for
• Turns
• Exchanges (clusters of turns related to a single
  idea)
• Initiations (who initiated - partner or
  participant?)
• Successfulness (was idea conveyed - fully or
  partially?)
• Modality (including referential gestures to topic
  setter)
• Elapsed time per topic
• Perceptual Data were also collected on S.D.s and
  partners
• Own Competence
• Partners Competence
• Comfort with Communication Technique
• Enjoyment of Communication Technique.

12
Data Summarization

• Data were initially plotted by individual
  sessions in accordance with single subject
  analysis techniques
• However, inherent variability (perhaps because of
  the uniqueness of each topic) caused any
  potential effect to be obscured
• To compensate for variability, quantitative data
  on turns, exchanges, initiations, etc. were then
  averaged by condition/topic and graphed (see
  display)
• Descriptive statistics were calculated (see
  handout)
• Inferential (nonparametric analysis) will follow,
  although independence of data is an issue

13
Results (Overview)

• Condition Effects
• Participant Turns and Total Turns increased
  incrementally from baseline to context to
  instructional conditions across both partners.
• Initiations approx. doubled for current event
  context and contextinstruction conditions across
  both partners
• Less of an effect seen with personal event
  (although increases in initiations were noted)
• S.D. was much more successful when context was
  available during interactions with partner 1
  however, increases were very slight for partner 2

14
Results (Overview cont.)

• Partner Effects
• Experimental observations that Partner 1 was a
  less skilled interpreter were supported
  empirically -- there was more of a tx effect with
  regard to turns and successfulness with Partner 1
  than Partner 2
• However, context allowed S.D. to initiate more
  readily, which in turn changed the participants
  role in the interactions with both partners.

15
Results (Overview cont.)

• Topic Effects
• Length of interaction, as measured in seconds
  and/or number of turns, was typically greater for
  the personal event.
• However, when context was available, there was a
  more pronounced increase in initiations for the
  current event.

16
Summary and Conclusions

• S.D. quickly learned to utilize contextual
  information to initiate or add information to a
  topical conversation
• By pointing to semantically specific information,
  he extended topical interactions across both
  partners as measured by increases in turns and
  time.
• The nature of the interactions changed - from
  clinician-dominated to shared roles as initiator.
• Interaction effects were observed between
  partners, topics, and condition.
• Further quantitative analysis is warranted to
  determine
• the significance of the effects
• Whether meaningful interactions exist among the
  variables (e.g., referential acts,
  successfulness, contingent partner responses)

17
Clinical Application/Future Directions

• The presence of visual or tangible context
  appears to immediately change the nature of
  interactions between people with severe
  expressive aphasia and their communication
  partners.
• Interactions are characterized by increased
  quantity of communication turns and exchanges,
  communicator initiations and overall
  successfulness of the exchange.
• Teaching communicators to reference specific
  information within some form of visual or
  tangible context (e.g., newspaper article,
  remnant) could be an easily implemented clinical
  intervention strategy for transitional (from
  partner dependent to self-initiating)
  communicators.

18
Acknowledgements

• This investigation was supported, in part, by an
  ASHA 1998 Research Grant for New Investigators
• The authors would also like to acknowledge the
  efforts of the following individuals the
  participant and his spouse graduate students
  Katrina Dunnewold and Jeannine Christie
  (experimental partners),and Jen Martonik
  (graduate research assistant/graph-maker
  extraordinaire)