
Designers: Muralikumar Krishnamurthi
and Kalpana Muralikumar
Client Coordinators: Patricia Nizio,
Detroit Institute for Children
Supervisor: Robert Erlandson, Ph.D.
Department of Electrical and Computer
Engineering
Wayne State University, Detroit, MI 48202
INTRODUCTION
The Detroit Institute for Children (DIC)
provides disabled children in the Metro Detroit area with specialized medical,
dental, and rehabilitative care as well as high-tech adaptive equipment.
Therapists and teachers at the DIC have difficulty assessing the language
and mathematics competencies of severely multiply impaired children because
standardized assessment tests typically require (1) generalization of skills
beyond the scope of the assessment tests, (2) generalizing from concrete
examples to abstracted drawings, (3) verbalizing answers, and (4) pointing
in some way to respond. Such requirements are beyond the physical and motor
capacities of many DIC clients. Therefore, a multi-purpose, multimedia-
based computer system for early language and mathematics knowledge/skills
has been designed to allow staff to evaluate such skills in this population.
Field testing of an earlier version of this program established the efficacy of the concept, but highlighted several areas for improvement and prompted DIC staff to request additional functions. Inclusion of the new functions resulted in a drastic redesign of the original program. From a user’s perspective the program should look and feel like the original, but includes the requested performance enhancements and functional additions.
SUMMARY OF IMPACT
The children using the Multimedia Assessment
Tool (MAT) at the Detroit Institute for Children are primarily between
the ages of five and fifteen years and have severely limited physical capabilities
due to Cerebral Palsy and Quadriplegia. Due to these disabilities, most
of these students cannot sit independently, walk, or use their hands to
communicate. The children can, however, use MAT to demonstrate what they
know of a particular vocabulary group, such as "animals" or "objects,"
by simply activating a switch to select pictures scanned across a screen.
A second task has been added to this version of the program which involves
short term memory and an understanding of logical connectives such as “AND.”
For example, a child might be asked to select a cow and a horse and a bird.
These activities allow staff to determine each child's level of receptive
language development and provides the children with the opportunity to
demonstrate their are capabilities given the appropriate forms of assistive
technology. MAT also allows the children to practice the motor skills associated
with listening, watching, and maintaining attention, all of which are necessary
for operating computers and improving communication skills.
The original field testing was so successful that DIC will be adding these activities to their collection of assessments and a record of client performance will be included in the client’s treatment files. Based on the field testing experience a new report generating system was designed and developed. This activity was a second semester’s project for the same student design group.
TECHNICAL DESCRIPTION
MAT offers two types of task activities
each looking at different language and cognitive processing skills. In
one case the child is simply asked to pick an object or an animal from
the pictures presented on the monitor. In the second group of tasks the
child is asked to make selections based on increasing memory demands and
logical connections. The task implemented in the current version of the
program uses an “AND” connective. For example, the system asks the child
to pick the square and the circle, or perhaps, the cow and the bird and
the lion. The program supports up to six pictures on the screen and
five connectives. The new design allows the simple addition of additional
tasks. In addition to “AND,” therapists want to use the “OR” connective
and combinations of connectives.
MAT can be run in two different modes: learning mode and assessment mode. In the learning mode, the student's learning process is reinforced by repeated voice prompts and the display of the correct picture in the case of a wrong selection by a student. In the assessment mode, software does not correct the student in the case of an incorrect selection. Rather, the system records all details of the selections made, the time taken to make each selection, and which picture was selected. Data collected during the assessment mode can then be tabulated and displayed in graphical form by MAT’s multi-functional report generator.
For example, the example depicted in Figure 1 below may be used for two different activities. In both cases the program displays four different animals from which the child will be required to make a selection. In one activity the child will receive a verbal prompt to select one particular animal from the set. For the second activity the child is asked to select two or more specific animals. For example, select the horse and elephant, or select the cat, and the horse and the elephant.
Since many of the children cannot use a mouse, the program supports a visual scanning mechanism for switch users. The child makes a selection with a single switch or a mouse when their choice is highlighted. In the learning mode, if their selection is correct, the program will move to the next exercise. If the choice is incorrect, the same exercise will be repeated. In the assessment mode the program simply moves on to the next exercise and records the child’s responses. In all activities verbal prompting is provided to maintain the child's attention and to guide the child through each exercise.
Figure 1: A display of selectable items
The number of pictures displayed in the
program depends on the level of difficulty specified by the staff. As the
level
of difficulty decreases, the student is
required to choose among a fewer number of pictures. The system also allows
staff to create user-defined session settings to meet the needs of individual
students. These settings include the color of the highlight scan, width
of the highlight scan, image scan time, image delay time, and the number
of scans over each image before session termination.
Figure 2: Session settings screen for setup of individual student settings
In addition to the report generator, MAT maintains a database of student and staff information. This database offers controlled access to student information, which is available via password only to staff members assigned to that student. While preparing MAT sessions for their assigned students, staff may record new audio prompts, overwrite existing audio prompts, and play existing audio prompts. This feature also contains a preview of the images associated with each audio file, allowing staff to preview the MAT settings before starting the session.
MAT is implemented in Visual Basic 5.0. To function optimally, MAT requires a Pentium-grade CPU with Windows 95, 98,2000 or NT, a color monitor, a single-switch (if a mouse will not be used), and computer speakers. If staff wish to add or replace photos in the system, a digital camera or scanning device will also be necessary.