P. I. TIME REQUIREMENTS FOR PSYCHOPHYSICS EXPERIMENTS

• Document Type: CREST
• Collection: NGA Records (Formerly NIMA)
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78B05171A000300030009-5
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 2
• Document Creation Date: December 28, 2016
• Document Release Date: July 30, 2003
• Sequence Number: 9
• Publication Date: October 9, 1969
• Content Type: MEMO

STAT
Approved For Release 2003/08/05 : CIA-RDP78BO5171A000300030009-5
9 October 1969
STAT
MEMORANDUM
Subj: P. I. Time Requirements for
Psychophysics Experiments
The purpose of this memorandum is to provide estimates of
time requirements for photointerpreters to be used in the information
extraction experiments. Our preliminary study indicates that approxi-
mately one hour of observer time is required in order to obtain one
curve of probability of correct decision versus signal-to-noise ratio.
Such a curve deals with one type of degradation, one quantitative
amount of that type of degradation and, of course, for one set of ob-
ject alternatives. If we involve two different types of degradations,
and five levels of degradation then we have ten hours required to
obtain a complete family of curves for one observer and one set of
object alternatives. I would like to start with alphanumerics in the
first object set for exactly the same reasons that prompt eye charts
in vision testing and bar charts in photographic testing. I estimate
that probably three other sets will be requ?r.. l in attempting to find
object sets which are realistic. This means we end up with 4 x 10
or 40 hours of observing time for a complete set on one P. I, This
40 hours is, of course, spread out over one-year's time.
My personal desire would be to have two P. I. 's run the
complete set. If very large individual differences become apparent,
the experimental plan would have to be reviewed. Such was not the
case for our local observers and I would not expect it to be true for
the P. I. Is.
Declass Review by
NIMA/DOD
Approved For Release 2003/08/05 : CIA-RDP78BO5171A000300030009-5
 Approved For Release 2003/08/05 :.C2A..RDP78B05171A000300030009-5
By the time that we have progressed to the third object set
I think it would be highly desirable to involve three or four more
P. I. 's in a reduced run, say five hours each. This would allow
additional observation of individual differences and would also in-
doctrinate these people in the information extraction concepts of
the experiment.
In summary then:
2 P. I. 's - 40 hours each (over 1 year)
4 P. I. 's - 5 hours each
Total - 100 P. I. Hours
These estimates are made without knowledge of the impli-
cations of such demand on P. I. 's time and are, of course, subject
to negotiation.
JLH:pw
Approved For Release 2003/08/05 : CIA-RDP78BO5171A000300030009-5