Quantitative Differences in Aphasia Interactions with Visual Scenes AAC Displays

1
Quantitative Differences in Aphasia Interactions
with Visual Scenes AAC Displays
ASHA Convention, Poster Session November 16,
2007 Boston, MA

• Jennifer M. Seale, M.S. CFY-SLP
• Institute on Disabilities at Temple University,
  Philadelphia, PA
• Kathryn L. Garrett, PhD CCC-SLP
• Duquesne University, Pittsburgh, PA
• Laura C. Figley, M.S. CFY-SLP
• The Watson Institute, Pittsburgh, PA

This project was completed as 1st authors
Masters Thesis at Duquesne University
2
Background

• Visual symbols may be more comprehensible to
  people with severe aphasia than spoken language
  symbols (Beukelman, et al., 2005 Garrett,
  Huth, 2002 Stuart, S. 2000)
• Despite visual strengths, people with severe
  aphasia are not always successful at using
  visually-based AAC strategies to communicate
  (Jacobs, et al., 2004 Kraat, 1990)

3
Purpose

• Do people with aphasia converse more successfully
  and initiate more when conversational topics are
• Not represented visually on AAC devices
• Represented photographically on Visual Scene
  Displays (VSDs)
• Represented with de-contextualized line drawings
  found on Traditional Symbol Grid Displays (TGDs)

4
Research Questions

• Does the number of communication acts per
  exchange differ across 3 display conditions?
  (index of breakdowns)
• What percentage of messages are successfully
  conveyed in a conversational story telling task
  across 3 conditions?
• How do the communication roles (initiations/respon
  ses) shift across 3 conditions?
• What are the communicative functions of
  conversational acts generated across 3 conditions?

5
Participant Descriptions
Primary Participants

• 3 men ages 34-66 years
• 3.5 years post onset at time of study
• Severe Expressive Aphasia
• Aphasia Quotients
• 58.1/100 45.6/100 21.8/100
• Passed cognitive screening

Peer Conversational Partners

• 2 females and 1 male -- ages 21-65 years
• within 15 years of Primary participants age
• Passed CLQT (Helm-Estabrooks, 2001)

1 Person with Aphasia 1 Peer 1 Dyad
6
Design

• Multiple single-subject, comparative condition
• Repeated across 3 participant dyads
• 3 experimental conditions per dyad
• 2 sessions per condition and story --
  counterbalanced
• Total 6 sessions per dyad (n18 total)
• Task Personal stories shared in conversation
• Hobbies, Cars, 4-Wheeling, Jamboree,
  Family Vacation

• No Display
• Visual Scenes Display (VSD)
• Traditional Grid Display (TGD)

7
Data Collection

• 10 Minute Conversations
• Video-taped
• Transcribed and Coded for Dep Vars
• Coding reliability 85 Segmentation
  reliability 90
• Dependent Variables
• Frequency of exchanges and acts/exchange
• Success of message exchange
• Communicative Roles
• Communicative Functions (measured at the act
  level)

8
Results Exchanges and Acts
Dyad 3
9
Exchange and Act Patterns

• The fewest number of exchanges and acts occurred
  with VSD for Dyads 2 3
• Theory Decreased need for people with more
  severe aphasia to add bits of info when story
  was conveyed with VSD
• Acts per exchange in Dyad 2 decreased by approx.
  1 (of 6 total) when SGD was present
• Theory Perhaps indicating Participant and
  Partner 2 needed less negotiation to convey
  ideas/had fewer breakdowns with both visual
  displays
• No differences in acts/exchanges for Dyad 1

10
Success Rating Examples

• 3 No partner interpretation
• 2 Partial interpretation
• 1 Inadequate information

11
Results Success Ratings
12
Success Patterns

• Exchanges were more successful (3-rating) with
  VSDs compared to no display for all 3 dyads
• Greatest increase for Dyad 1 (most verbal
  participant)
• Smaller increase for Dyads 2 and 3 (less verbal)
• Dyad 3 (most nonverbal) showed equal increase in
  success for both visual display conditions
• Fewer exchanges required partner interpretation
  (2-rating) with VSDs for all 3 dyads

13
Results Primary Participant Communication Roles
14
Primary Participant Role Patterns

• Participants were generally responsive
  communicators in all interactions, initiating
  less than 30 of exchanges across conditions.
• However, participants in Dyads 1 3 demonstrated
  an observable increase ( 15) in initiated
  exchanges when using TGDs
• Consequently, the number of responsive roles per
  exchange observably decreased
• To a lesser degree, this pattern occurred with
  VSDs

15
Results Peer Communication Partner Roles
16
PCP Communicative Role Patterns

• Peers maintained a dominant role in all
  interaction initiating gt 60 of exchange across
  conditions
• However, partners from Dyads 1 3 did decrease
  the number of initiated exchanges and increased
  responsiveness when an SGD was present
• No differences for partner in Dyad 2
• Partner maintained highly dominant role across
  conditions (gt80)

17
Results Communicative Functions for Primary
Participants
18
Primary Participant Communicative Function
Patterns

• Participant in Dyad 2 continuously referenced
  elements of VSDs to establish joint attention
• Other 2 participants did not
• Preliminary semantic analysis showed only a
  slight increase for Dyad 1
• Further analysis required to count semantic
  elements conveyed with VSD and TGDs
• Slight decrease in filler/confirmatory responses
  for Dyads 1 and 3 in visual display conditions

19
Results Communicative Functions for Peer
Communication Partners
PCP from Dyad 1 PCP from Dyad 2
PCP from Dyad 3
20
PCP Communicative Function Patterns

• Variable results across dyads and conditions
• Dyad 1
• Requested less info with both visual displays
• Requested less clarification with VSD but more
  with TGD
• Fewer confirmations with TGD
• Dyad 2
• Requested more info with both visual displays
• Requested more clarification with both visual
  displays
• Offered less of his own information with both
  visual displays
• Dyad 3
• Requested less info with both visual displays
• Requested less clarification with both visual
  displays
• Noticeably increased confirmations with both
  visual displays

21
Unexpected finding

• Data for Dyads 1 3 imply that peers may also be
  using the externally represented messages to
  comprehend conversational meaning
• Reduced need to ask questions, clarify, negotiate
  meaning in VSD and TGD conditions
• Exception increased clarification for TGD in
  Dyad 1
• The spoken messages from the SGD displays were
  equally shared by participant with aphasia and
  partners
• Reduced the need to question person with aphasia
• Displays may not have stimulated additional
  conversation as anticipated
• Note Dyad 2 maintained a unique pattern of
  results across most variables
• Dominant partner style may have influenced data

22
Preliminary Modality Analyses
23
Results Changes in Modality Use

• Participant 1 (most verbal)
• Increased verbal output by 10-20 with TGD
  compared to no display and VSD, respectively
• Jargon increased ( 15) with TGD compared to
  VSD
• Use of jargon slightly decreased (10) with TGD
  compared to no display
• Symbolic gestures decreased by 5-10 with VSD
  compared to no display and TGD, respectively
• Participant 2 (second most verbal)
• Verbal output decreased by 10 with TGD
  compared to VSD no display
• Jargon increased by 5-10 with VSD TGD
  (respectively) compared to no display
• Symbolic gestures decreased by 20 with VSD
  compared to no display
• Used SGD twice as much with VSD compared to TGD
• Participant 3 (least verbal)
• Verbal output decreased by 2.5 with VSD TGD
  compared to no display
• Use of jargon decreased by 10 with VSD and TGD
  compared to no display
• Symbolic gestures decreased by 20 with VSD
  compared to no display
• 12 with TGD

high standard deviations associated with this
data Further modality analysis in conjunction
with specific semantic information data is
warranted.
24
Overall Summary of Results

• Exchange success increased noticeably ( 10)
  when SGD displays were present compared to no
  display
• Message exchanges were most successful with VSDs
  for 2 of 3 dyads
• TGD and VSD were equally successful for Dyad 3
• Unexpectedly, people with aphasia did not
  demonstrate a large shift from a responder to
  initiator role with SGDs
• However, participants with aphasia in 2 of the 3
  dyads did demonstrate slightly more initiations
  with TGDs compared to VSDs and no display

25
Theoretical Implications

• Partners may take advantage of the abundance of
  contextual info when symbols representing
  messages are transparent (e.g., photos)
• Reduced cognitive processing in participants with
  aphasia may allow partners to jump in and
  initiate exchanges based on the context of photos
• Partners were more dependent upon individuals
  with aphasia to activate and initiate the
  information exchange in TGD
• Allowing primary participants time to process and
  initiate exchanges
• The context gleaned from the TGD wasnt enough
  to allow partners to dominate, and instead
  facilitated the discussion

26
Clinical Implications and Conclusions

• Collaboration between PCPs and participants with
  aphasia may be necessary to achieve successful
  communication
• However, partners may continue to dominate
  conversation when given too much info through
  transparent, symbolically represented messages
  (e.g., VSDs)
• Therefore, this study suggests a combination
  display may be best
• Providing enough context to start a conversation
  for both peer and primary participants through
  photos
• Yet maintaining a degree of message-secrecy
  with individual line drawings that represent a
  single idea

27
Limitations

• High standard deviations for the following
  variables
• The frequency of communicative exchanges, acts
  and functions
• Variability was not cancelled out with only two
  data points (conversations) for each condition
• Participants with aphasia had varying interest in
  story topics
• This influenced frequency of acts, exchanges,
  intiations/responsiveness, communicative act
  function and (perhaps most importantly) duration
  of conversation
• Within-group variability occurred in
  communication performance for all dyads and
  individual participants
• One dyad (2) demonstrated unique interpersonal
  dynamics
• Investigator presence during dyad conversations
  to collect data (Hawthorne effect?)

28
Directions for Future Research

• Larger participant pools and more data points
  (conversations) per condition
• Either use novel topics for each session, or
  replicate all topics in all conditions
• PCPs should retell story after conversations
• So amount of information conveyed by participant
  with aphasia can be measured more precisely
• Use friends or relatives as partners instead of
  novel peers
• Use remote video-taping
• Analyze relationship of quantitative to
  perceptual (Figley, 2007) data

• Acknowledgements
• Participants and their families.
• Dr. Garrett, Chair Committee ,
• Dr. Chen Dr. Fromm, Committee MembersLaura
  Figley, M.S., related study investigator
• Katie Huwar, M.S., 3rd party data collector
• Dynavox Technologies

29
Selected references

• Beukelman, D., Fager, S., Carlton J. (2005).
  R2-B Contextual Scenes for Adults With Aphasia.
  www.aac-rerc.com/pages/projects/R2.htm.
• Chapey, R. (1983). Language-based cognitive
  abilities in adult aphasia Rationale for
  intervention. Journal of Communication Disorders,
  16, 405-424.
• Garrett, K. Huth, C. (2002). The impact of
  graphic contextual information and instruction on
  the conversational behaviours of a person with
  severe aphasia. Aphasiology, 16, 523-536.
• Garrod, Simon (1986). Language comprehension in
  context a psychological perspective. Applied
  Linguistics, 7 (3), 226-238.
• Hengst, J. A. (2003). Collaborative referencing
  between individuals with aphasia and routine
  communication partners. Journal of Speech,
  Language and Hearing Research, 46, 831-848.
• Light, J., Dattilo, J., English, J. Gutierrez,
  L., Hartz, J. (1992). Instructing facilitators
  to support the communication of people who use
  augmentative communication systems. Journal of
  Speech and Hearing Research, 35, 865-875.
• Pierce, R. S., Beekman, L. A. (1985). Effects
  of linguistic and extralinguistic context on
  semantic and syntactic processing in aphasia.
  Journal of Speech and Hearing Research, 28,
  (250-254).
• Yorkston, K.M., Beukelman, D.R., Flowers, C.R.
  (1980). Efficiency of information Exchange
  between aphasic speakers and communication
  partners. Proceedings of the Clinical
  Aphasiology Conference. Minneapolis BRK
  Publishers.