|
Research into Visual Perception conducted by Johnson et.al.
Following the failure of the early investigators into visual perception
of simple two-dimensional shapes to reach conclusions consistent with
each other, and to explain their findings systematically,
experiments of the same form were still being undertaken many
years after Helson
and Fehrer’s account of the rôle of form in perception.
See for example, work by Blackwell, Casperson,
and others..
Lie’s 1979 paper (details at bottom of this page) describing
a comparison of ‘area-threshold
functions for the detection and resolution of single targets presented
at various retinal eccentricities’ is an extension of the earlier
investigations conducted by Aulhorn and Johnson
et.al..
Although Lie’s investigations were based on these earlier studies,
the procedure used was different in three respect:
First, the
circles and squares used by the previous investigators were replaced
by square
targets placed in one of the two positions illustrated below, the orientation
being selected at random on each occasion.
The test targets used by Lie (1979).
Second, both thresholds were determined by a stepwise
variation of target size for fixed levels of target contrast. Target
size was varied
until either the detection or the resolution threshold was reached.
Third, the viewing time was unlimited.
Only one (normal-sighted
and extensively trained) observer participated under all experimental
conditions. In
addition, ‘control
data’ were collected from three (less trained) observers. As in Aulhorn
and Johnson et.al.’s investigations, observers viewed the targets using
only one eye. The shape of the resulting area-threshold function for detection was
found to differ from that for recognition. Lie identified two important
differences between the detection and resolution curves. These were that
as the stimulus size is decreased below a certain minimum, resolution
thresholds cannot be obtained, though a corresponding limitation was not
apparent for detection thresholds. In addition, the number of linear segments
having different slopes which must be fitted to the data points to achieve
a good fit to the data was larger for the resolution curves than for the
detection curves.
Lie considered the implications of his findings in terms of the accepted theory
of spatial summation within retinal receptive fields (as described by Johnson
et.al.). He stated that this criterion theory may explain why resolution thresholds
cannot
be
obtained
when
the stimulus area is decreased below a certain minimal size. However, the other
shape differences between the detection and resolution curves, such as different
number of ‘slope shifts’ (i.e. the linear segments mentioned above),
suggest that the relationship between the detection and resolution properties
of the visual system is more complex than the criterion theory implies.
It
is suggested that these ‘slope shifts’ indicate transitions from
one receptive field size to another.
For further information about this work see:
I.Lie, "Visual Detection and Resolution as a function of Retinal
Focus",
Vision Res. Vol.20, pp.967-974 (1980).
This Section consists of short summaries of historial research and theories
into human visual perception of simple two-dimensional objects (these
are extracted and summarised from a Ph.D. Thesis [67] dated
1996).
For more general information about the human visual system see
the sections about:
The
Eye; Parts
of Eye; Visual
Disorders;
Ophthalmological
Procedures.
|
