Collection: Early Childhood

The Baby-Babble-Blanket, Technology for Early Intervention

REFERENCE: Fell, Harriet, J. and Ferrier, Linda, J. (1994). The Baby-Babble-Blanket, Technology for Early Intervention.

The BBB (patent #5260869) is a multiple micro-switch-activated pad serving as input to a Macintosh computer and our software. The software allows digitized output of speech, babbles, words, sentences, music or environmental sounds, to be accessed by the switches. The BBB includes a data collection system that uses a single-case study design to evaluate the responses of infants who are otherwise difficult to test.

FIG. 1. BBB components with a user

Specification of the BBB

The "Baby Babble Blanket," a Macintosh based system (See Fig. 1.) includes the following components:

1. An input device, a blanket the infant lies on, that includes 12 switches. This is the Powerpad, a device developed by Nintendo for computer games (but regrettably no longer produced.) It is a plastic, waterproof blanket that includes large pressure sensitive switches ten inches apart. The blanket consists of two layers of polymer plastic separated by urethane sponge material to inhibit inadvertent contact of the outer layers unless they are overtly pressed together. One inch holes in the sponge allow contact under pressure. The inner surfaces of the polymer are imprinted with a conductive material. The conductive material pads are brought together to a single location so that they can be connected to an output connector. The device is then hooked up to the computer via a standard cable connection. Pilot investigation showed that the pad can be activated by infants weighing less than twenty pounds.
2. A Macintosh computer control system that connects with the Powerpad via a Gold Brick interface box (Transfinite Systems Inc.), and is commercially available.
3. The Hypercard software we have developed that carries out the following functions: selection of particular switches to be activated by the infant, depending on the particular pattern of motor movement available to the infant and the positioning of the infant on the blanket; assignment of sounds to particular switches; collection of data on the infant's switch activations.
4. A graphing and data analysis program that summarizes the number and type of switch activations over time in a variety of formats, graphs them, and provides summary statistics.
5. Customized sound selection: Digitized sounds are entered into the Macintosh computer using the MacRecorder (Farallon, Inc.) or the Macintosh built in microphone. Sounds can be spoken directly into the microphone or entered from recordings through a line-in from a tape recorder. These can then be stored in a sound library. Alternatively, sounds can be selected from libraries of sounds supplied with the software. These include babbles at different developmental levels, early words, environmental sounds, and short sequences of music. The sounds can then be allocated to particular Powerpad switches.

Single-Case Study Software

Single-case study experimental designs are particularly suited to intervention with severely impaired and multiply handicapped children since they provide the freedom to conduct studies on small groups of heterogeneous subjects. The children we are studying are heterogeneous, since they are diagnosed at different ages, and have different levels of motor and cognitive ability.

A requirement of single-case study experimental design is that the recorded data must consist of observable behavioral events. This frequently presents difficulty with children with motor and cognitive deficits whose responses often appear random or difficult to read. With the use of the BBB, the observable events are switch activations, that can be counted and tabulated by the program. While it may be assumed that some switch activations by infants with motor problems are the result of involuntary muscular activity, we expect that, with repeated exposure to the device, most infants will have sufficient motor control to respond to the sound output with a change in the frequency of switch activations. Through these changes in switch activations across time and in response to different sounds, we can evaluate how responsive the infant is to sound. We can track how quickly s/he learns to activate switches to hear the sounds, and whether s/he prefers one sound to another.

The fact that the software will record the switch activations has many experimental advantages: It eliminates the error of a human recorder; and it can record many data rapidly across many sessions. By examining the patterning of switch activations across time, in multiple sessions and with different reinforcers, we have a clear record of each child's ability to learn. Unlike large group designs that rely on statistical analysis of data, single subject experimental case study designs utilize visualization of data, usually displayed in a line or bar graph format. One advantage is that statistically unsophisticated parents, teachers, or clinicians can determine by the slopes of the data in different phases, baseline versus treatment, whether the child is responding to the system.


Sample recorded data

Design and Data Collection Formats

The program at this point allows data collected in a variety of formats. These include:

1. Baseline or no sound
2. All sound-bound (A single sound is assigned to each switch.)
3. All sound-random ( A set of 12 sounds is stored under each switch. Each switch activation selects a sound at random from the set.)
4. ABA (No sound, sound, no sound)
5. Totally random (sounds are randomly selected from sets but are unconnected to switch activations. This format allows us to test whether infants show an increase in movement in response to the noise that is to ounrelated to their ability to cause the sounds ccur.
6. Single switch (Only one switch results in sound, but data are collected on all switches.)
7. Half and half (Six switches on half the blanket produce sound but data are collected on all switches.)

The information above is an exerpt from a paper that will be presented at ASSETS'94, November 1, 1994. We are in the third year of field-testing the Baby-Babble-Blanket. We are looking for an infant, five months old or younger, with motor problems, reduced vocalizations but normal cognitive ability to study in the home, in the Boston area.

*Partial support for this work was provided by Department of Education grants: NIDRR # H133c20076 and OSEP, OSERS # H180E20032.

CONTACT(S): Harriet J. Fell, Ph.D., College of Computer Science, Northeastern University, Boston, MA 02115
Linda J. Ferrier, Ph.D., Dept. of Speech Language Path. & Aud., Northeastern University, Boston, MA 02115

[Eary Childhood Table of Contents]

[ Home | Library | Videos | Tour | Spotlight | Workshops | Links ]

ŠEducation Development Center, Inc.