OUTLINE CLASSIFICATION OF POLYGRAPH RESPONSES.

Abstract

The development of interfacing instrumentation and software programs for implementing a necessary increase of the data entry rate in the general purpose digital computer, IBM 360-50, via a 2701 terminal is discussed. The data generated by automatic analysis of the psychophysiological measures require that this critical transmission link between the special purpose analyzers and the decision logic in the general purpose machine be improved markedly over the current state of the art. Interfacing hardware has been developed, fabricated and checked out and software programs have been written and debugged which allow an entry rate of 800 bites (6400 bits) per second which is more than adequate for logging four channels of period processed EEG and four channels of the lower frequency psychophysiological measures such as respiration, plethysmogram, etc. The automated polygraph decisions which have been employed fall into the two logical approaches of 'pattern recognition' programs such as discriminant analysis and nearest neighbor non-parametric techniques. Both types of logic yield decisions which approximate the efficiency of professional polygraph operators. Results of this study demonstrate that online feedback of automated polygraph decisions can be accomplished with appropriate interfacing to the IBM 360-50 and forces the conclusion that such online feedback would improve the efficiency of polygraph operations. (Author)

Document Details

Document Type
Technical Report
Publication Date
Oct 17, 1969
Accession Number
AD0695432

Entities

People

  • Neil R. Burch

Tags

Communities of Interest

  • Human Systems

DTIC Thesaurus Topics

  • Analyzers
  • Automatic
  • Classification
  • Computers
  • Digital Computers
  • Discriminant Analysis
  • Efficiency
  • Feedback
  • Frequency
  • Instrumentation
  • Measuring Instruments
  • Pattern Recognition
  • Recognition
  • Respiration
  • Terminals

Readers

  • Brain and Cognitive Science; Experimental Psychology; Cognitive Neuroscience
  • Computer Science/Computer Engineering/Data Science/Digital Signal Processing.
  • Regression Analysis.

Technology Areas

  • AI & ML