|
Project R3-1: Learning demands for
2-year-old children of three different approaches to the organization
and presentation of language concepts in AAC technologies.
This study investigated the learning demands of three different
approaches to the organization of language in AAC systems: vocabulary
presented in a grid organized taxonomically (i.e., concepts
grouped in categories such as people, actions, places); vocabulary
presented in a grid organized schematically (i.e., concepts
grouped in events such as snack time, circle time); and vocabulary
presented in an integrated scene organized schematically (i.e.,
concepts embedded under "hotspots" in an integrated
scene). The children were introduced to 12 target vocabulary
items in a series of four learning sessions. Each session was
structured around a play context involving a birthday party
for a teddy bear (Bobby) who could not talk and used an AAC
system to communicate. In each session, children were presented
with a play situation and asked to help the teddy bear locate
target vocabulary on the AAC system (e.g., Bobby wants to call
his Mommy. Find "Mommy.") Feedback was provided for
all incorrect selections during the learning sessions. Data
were collected on the accuracy of the children's responses.
After three learning sessions, the children participated in
a generalization session in which 12 novel words were introduced.
As with the learning sessions, children were asked to locate
these target concepts in the context of a play scenario.
A total of 30 typically developing 2-year-old children participated
in the study. Data analysis was completed on the accuracy of
the children's responses using a mixed ANOVA; error analyses
were completed to derive insights into the cognitive processing
demands.
Project R3-2: Learning demands for
3-year-old children of four different approaches to the organization
and presentation of language concepts in AAC technologies.
This study investigated the learning demands for 3-year-old
children of four different approaches to the organization of
language in AAC systems - taxonomic grid, schematic grid, schematic
scene organizations, and a variation on the schematic grid condition.
The traditional schematic grid condition uses an index page
with a single symbol representing each of the dynamic pages
of vocabulary available. This approach fails to make the design
of the system transparent to young children since the pages
are "hidden" to the user when the system is on the
index page. To make the design more transparent to young children
and reduce the memory demands of the system, screen shots of
the actual pages were used on the index page to represent the
vocabulary pages. In this study, the same procedures as in R3-1
were used except the children were introduced to 18 target vocabulary
items (instead of 12) in a series of four learning and testing
sessions. Data collection and analysis was recently completed.
Project R3-3: Learning demands for
four and five-year-old children of four different approaches
to the organization and presentation of language concepts in
AAC technologies. This study investigated the learning
demands for 4- and 5-year old children of four different approaches
to the organization of language in AAC systems: taxonomic grid,
schematic grid, schematic scene organizations, and iconic encoding
or Minspeak. Procedures were similar to Projects R3-1 and R3-2
except that the children were introduced to 24 items (4-year-olds)
or 30 items (5-year-olds) in a series of 4 learning sessions.
Data collection and analyses is complete for 80 children over
4 learning sessions and a generalization session.
This study was recently expanded to include a spin off study
designed to compare the performance of 5-year-old children using
iconic encoding with icon prediction to their use of iconic
encoding without icon prediction. This follow up study is intended
to provide a measure of the impact of icon prediction on easing
the learning demands of iconic encoding systems for young children.
Data collection and analyses were recently completed.
Project R3-4: Investigation of the
semantic organization patterns of young children. Results
of projects 1-3 suggest that young children have significant
difficulty learning the layouts and organizations of current
AAC technologies. This project investigated how young children
organize language concepts to better understand the gap between
the organizations used in AAC technologies currently and those
of young children. This study employed a graphic symbol arrangement
task to investigate how 20 children between the ages of 4;0
and 5;11 organized a variety of linguistic concepts. The children
were asked to organize 42 pictures depicting various familiar
vocabulary items including nouns, verbs, descriptors, prepositions,
pronouns, and question words. During and after the sorting task,
the children were asked to talk about the rationales for their
arrangements. Data collection and analysis are complete.
Project R3-5: The identification
of desired features for auditory scanning technology.
We have worked with rehabilitation professionals and AAC product
developers to design and initiate a research project to investigate
design issues in the development of effective auditory scanning
technologies. Twelve participants, including both experienced
clinicians and product developers with major AAC manufacturers,
contributed to in an internet-based discussion to identify the
learning demands of auditory scanning technologies and to determine
key design features for these technologies. Data analysis is
currently in progress. We anticipate that this project will
result in significant new recommendations for the development
of auditory scanning.
Project R3-6: Children's designs
for AAC technologies (an exploration of participatory design).
Five 7-year-old children without disabilities (3 boys
and 2 girls) and two young adults who were not familiar with
AAC were recruited to participate in designing an AAC system
for children. The group was asked to design ways that would
help a preschooler with significant physical disabilities to
communicate using low-tech prototyping materials (e.g., chalk,
blackboard, scissors, clay, tape, poster board, paper, pencils,
cardboard boxes, rubber bands, construction paper, markers,
crayons, glue, and Legos). Data collection and analysis are
complete. The interactions during the design and presentation
phases were videotaped, transcribed, and coded to identify design
features. The notes, drawings, and the actual prototypes constructed
by the participants were also analyzed according to their design
features. This project explored a very interesting methodology
that may have important applications in the field for future
research and development - participatory design. In this methodology,
end users are involved early in the design process. Having children
act as reviewers of AAC technologies after they are developed
is "too little, too late." Participatory design includes
the end-users of the technology working collaboratively with
researchers from the very beginning to develop low-tech prototypes
and implement them as technologies.
Project R3-7: Improving the appeal
of AAC technologies for young children: Lessons from the toy
manufacturers. Observations suggest that the current
generation of AAC technologies do not attract young children
or sustain their interest over long periods of time. Results
from project R3-6 suggest that children value systems that are
highly colorful and decorative, are "fun", and incorporate
play activities. Toy manufacturers have been successful in designing
products that attract young children and that sustain interest
over a long period. A new study was initiated to determine strategies
to increase the appeal of AAC technologies for young children.
The goals of the study were: (1) to conduct a systematic review
of award winning toys to determine features that may be used
to attract young children and sustain their interest; (2) to
compare these features to those of existing AAC technologies;
and (3) to consider ways to incorporate features into AAC technologies
to increase the appeal for young children. Award winning toys
were selected for each age group and were described according
to a feature analysis (e.g., color, shape, moveable parts).
Current AAC technologies were also described using the feature
analysis. Results were summarized across popular toys for each
age group and across AAC technologies. Features were compared
across toys and AAC technologies. Focus groups were held to
brainstorm application of toy features to AAC technologies to
enhance appeal and use by young children. Data analysis is currently
in progress.
Project R3-8: Graphic representations
of language concepts of children from diverse cultural backgrounds.
For young children who are preliterate, the vocabulary stored
in AAC technologies must be depicted using graphic representations.
Most current AAC symbol sets use the ideas of white, middle-class
adults to represent language. These ideas may not be meaningful
to young children, especially those from different cultural
backgrounds. The goals of this study are to (1) investigate
children's graphic representations (drawings) of early emerging
language concepts; (2) compare these representations to those
of current AAC symbols; and (3) investigate children's understanding
of current AAC symbols. Fifty typically-developing children
from diverse cultural backgrounds were recruited: 10 African
American children; 10 Hispanic children; 10 white children;
10 children from families who had recently immigrated from Russia;
and 10 children from families who were from India. The children
were provided with paper and markers and were asked to draw
10 early emerging language concepts (e.g., more, all gone, what?).
They were then asked to describe their drawings. Finally the
children were shown AAC symbols representing early emerging
language concepts and asked to name the symbols. Data collection
is currently in progress.
Project R3-9. Applications of personalized
integrated scene organizations with young children with significant
disabilities. Results of project R3-1 suggested that
young typically-developing children performed more accurately
using integrated scene layouts in AAC technologies than the
grid layouts that have been used traditionally. Integrated scene
layouts embed language in meaningful contexts replicating the
way that young children learn language. The goal of this study
is to investigate the use of integrated scene layouts as realized
in personalized digital photos of young children with significant
disabilities who require AAC. To date, 3 children with significant
speech impairments (ages 2-4) have been recruited. Prototypes
of systems using personalized digital photos with vocabulary
stored under hot spots are currently being developed for these
children.
|
|
Project R3-1
- The children performed with low levels
of accuracy with all systems. Children were more accurate
using the schematic scene layout (average 4.1 correct answers)
than both the taxonomic grid layout (average 1.6 correct answers)
and the schematic grid layout (average 1.9 correct answers).
- Over the 4 learning sessions, there
was a very slight increase in the average gain in vocabulary
items.
- The children learned more concrete vocabulary
than abstract across all systems.
- There was no evidence of generalization
of system organization to novel vocabulary.
Project R3-2
- The children performed with low levels
of accuracy with all systems. The averages were 5.7, 7.4,
8.9 items correct out of 18 for the schematic grid, schematic
grid with screen shots menu page, and schematic scene conditions
respectively.
- The average gains from session 1 to
4 were 3.4, 5.3, and 7.0 items for the schematic grid, schematic
grid with screen shots menu, and schematic scene conditions
respectively.
- Children learned more concrete vocabulary
than abstract across all systems and sessions.
- There was some evidence of generalization
to new items, although gains were minimal.
Project R3-3
- Both the 4-year-old and 5-year old children
were more accurate locating target vocabulary in the three
dynamic display conditions (taxonomic grid, schematic grid,
schematic scene) than in the iconic encoding technique; results
were statistically significant.
- On initial exposure to the systems (session
1), the children averaged 0% and 3% accuracy for the 4- and
5-year-olds respectively with the iconic encoding technique;
the children averaged 26% and 33-41% accuracy for the 4- and
5-year-olds respectively across dynamic display systems.
- All but one child showed gains in accuracy
across the learning sessions. The children's rate of learning
with the iconic encoding technique was significantly slower
than with the three dynamic display systems.
- The children were more accurate learning
concrete vocabulary than abstract vocabulary. The advantage
for concrete vocabulary was less pronounced in the iconic
encoding technique than in the dynamic display techniques.
- There was evidence that the children
generalized knowledge of system organization to facilitate
learning of novel vocabulary. However, the evidence of generalization
was not great. The children's performance in the generalization
session exceeded that in the initial learning session by an
average of 0.2 - 4.2 items across the four systems.
Project R3-4
- The children performed with low levels
of accuracy with both systems.
- The averages after four learning sessions
were 4.6 and 6.6 items correct out of 30 for the iconic encoding
without prediction and with prediction respectively.
- Children showed some evidence of generalization
in the iconic prediction condition but the gains were modest
(average of 4.7 items); children did not show evidence of
generalization in the condition without prediction (average
of 0.4 items)
- Children performed more accurately with
the three dynamic display conditions (see R3-3 Part 1) than
either of the iconic encoding techniques (with or without
prediction).
Project R3-5
Data analysis is in progress.
Project R3-6
- Systems were highly colorful and decorative.
- Systems were fun incorporating many play
activities.
- Systems integrated numerous functions
e.g., play, communication.
Project R3-7
Data analysis is in progress.
Project R3-8
Data analysis is in progress.
Project R3-9
Project is in progress.
|
