Stanford Research Institute
Dr L R Pinneo and Mr D J Hall
Published:
August 1975
Published by:
U.S. Department of Commerce
National Technical Information Service
www.NtiS.gov...
NTIS document ref ADA017405
Page 1, second paragraph of SUMMARY
"The research was predicated on existing evidence that
verbal ideas or thoughts are subvocally represented in
the muscles of the of the vocal apparatus. If the
patterns of this muscle activity are at all similar to
those involved in normal overt speech, a reasonable
assumption is that electrical activity of the brain
during verbal thinking may be similar to that during
overt speech."
Page3, second paragraph, numered (5)
"It was determined that, if all the sources of error
could be eliminated, significant gains in correct word
classification using biological responses would be
achieved (perhaps approaching 90% or better)."
Page 3, penultimate paragraph
"EEG responses for covert speech mimicked those of
overt speech for the same subject, electrode and
spoken word. When sources of error were reduced as
much as possible, correct computer classification
rates ranged from 52 to 72%, which was significant at
p < 0.001.
We conclude that both overt and covert
speech can be identified by computer classification of
electrophysiological responses and that a practical
biocybernetic communication system is feasible,
provided that sources of error can be removed."