Speech Prostheses as a Metaphor for Verbal Short-term Memory
Gary W. Strong is Program Director for Interactive and
Intermedia Technologies at the National Science Foundation, and
Coordinator for the NSF on the National Science and Technology Council
Committee on Computing, Information, and Communication.
Lap-top computers with touch-screens and speech synthesizers
can be effective prostheses for people who are unable to speak.
Observations of children who have never been able to speak due to
physical disability indicate that their synthetically-produced speech
is slower than real time speech due to the complexity of language
selection from a touch-screen and that their sentences are initially
"telegraphic" in grammatical style. Nevertheless, synthetic speech is
better than no speech in nearly all situations. But, even more
important, remarkable serendipitous benefits have been observed,
benefits never anticipated, let alone predicted, by speech therapists.
Speech prostheses through dedicated programs on laptops are
given some people who are born without the capacity to speak in order
to enable them to converse and express their needs. However,
observation shows that sometimes in the beginning an inordinate amount
of time is spent by users playing with the device by themselves rather
than employing it to communicate with others. However, long-term
observation suggests that this activity is more than merely play. The
way the device is employed resembles an outward expression of verbal
short-term or working memory. The user selects words and constructions
for himself or herself rather than for others in a way that suggests a
rehearsal or recall instead of an intent to communicate with others.
If this is true, the computer speech prosthesis may be a standing-in
for a verbal short-term memory that was never developed because the
users are congenitally without speech.
If it is true that such users do not have a verbal short-term
memory, they will be unable to internally speak to themselves in a way
that normal speakers take for granted. Without this ability, it is
possible that the sense of self is impoverished in comparison to that
of normal speakers. Therefore, initial behavior with a speech
prosthesis may be that associated with discovering for the first time
that one has something to say and that it can be sculpted before
communication takes place. In this way, the device serves as not only
a metaphor for the internal speech process, but as an actual external
replacement for an important short-term working memory process that
plays a role in the developing sense of self.
Strong, G.W. (1994) Augmentation of self-reflective activity for
non-speakers. Proceedings of IWHIT94, International Workshop on Human
Interface Technology 94, Aizu-Wakamatsu, Japan: The University of
Aizu, Pp. 96-99.
Can be downloaded from URL:
Edwards, A. D. N., (Ed.) (1995). Extra-Ordinary Human-Computer
Interaction: Interfaces for Users with Disabilities. Cambridge Series
on Human-Computer Interaction. New York, Cambridge University Press.
Described at URL:
General description of Short-term (Working) Memory and reference at
General reference on assistive technology Ried, S.; Strong, G.W.;
Wright, L.; Wood, A.; Goldman, A. & Bogen, D. (1995) Computers,
assistive devices, and augmentative communication aids: Technology for
social inclusion. Journal of Head Trauma Rehabilitation 10(5): 80-90.
Pictures of the Dynavox and downloadable demonstration software are
available at URL:
Picture of a new competitor product at URL:
Description of a competitor for traditional computer platforms at URL:
Report calling for interface research for people with disabilities.
More Than Screen Deep: Toward an Every-Citizen Interface to the
Nation's Information Infrastructure. NATIONAL ACADEMY PRESS,
Washington, D.C. 1997.
More than Screen Deep