, V.
.flfl ..

‘1} 05‘ . . I \l
1......) <1. .i ‘ . .. 1 f1 ,

$21!: », . 43!...K .

d. E: 32‘... . 1

(A 3.3.1.3.:

 

 

guess
I...

This is to certifg that the

thesis entitled

THE ORGANIZATION OF THE VISUAL RESPONSE

presented by

Frank Ray Wilkinson

has been accepted towards fulfillment
of the requirements for

Ph, D, degree mm

 

Major profeM

Date Ma 3. 1955

0-169

 

 

THE ORGANIZATION OF THE VISUAL RESPONSE

By
Frank Ray Wilkinson

AN ABSTRACT
Submitted to the School of Graduate Studies of Michigan
State College of Agriculture and Applied Science
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Psychology

1955

Approved $4M «it: '

‘77 -f'
, 3/ 1;:
(I

 

 

ABSTRACT

The current tendency is to interpret visual phenomena as the
end products of a seriestof neural events occurring in the retina,
optic pathways and optic cortex of the organism. One theory as to
the functioning of these neural elements has been called the
"altemation-of-response" theory. This theory assumes that the
optic tract consists of a number of parallel pathways which cooperate
in transmitting information from.the retina to the Optic cortex.
In.the perception of flicker, when the pulses are delivered at rapid
rates, these pathways alternate in their responses to the individual
pulses, so that each pulse is responded to, but any single pathway
is not responding to every successive pulse of stimulation. In
support of this theory it has been shown that a pulsating stimulus
gives rise in the cortex.to bursts of'nervous activity which do not
follow the stimulus either in terms of frequency or intensity at
first, but which later come to follow the stimulus pattern closely.
This initial period of somewhat erratic fluctuatim of response
has been called the period of "reorganization”, in reference to the
”sorting out " of the pathways into functional, harmonic groups. The
present study is an attempt to investigate the perceptual concomitants,
if any, of this period of "reorganizatim" by presenting to the eye
of the observer short trains of pulses,the lengths of which in terms
of number of pulses are well within the period of "reorganization".
In this situation, the Observer reported the number and character of

flashes seen. The results indicate that there exists at the

 

 

 

beginning of stimulation, a correSponding perceptual organization
with a number of characteristics in common with the period of neural
"reorganization" mentioned above. Certain features of the results
are seen to be clearly in line with the hypothetical events postulated
by the "alternation-of-response" theory, and the theory is extended

to account for flicker fusion as a special case of steady vision.

{THE ORGANIZNTION OF THE VISUAL RESPONSE

By
Frank Ray Wilkinson

A THESIS

Submitted to the School of Graduate Studies of Michigan
State College of Agriculture and Applied Science
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Psychology
1955

THESIS \‘

 

The author wishes to express his
appreciation to the members of his
committee ani especially to Dr.

S. H. Bartley for his encouraging
help. Thanks are also due to
Mrs. F. Wilkinson for her help in
preparing the manuscript.

359310

Chapter

I.

II.

III.

V.

TABLE OF CONTENTS

THE PROBLEM

Historical and theoretical basis
and statement of the problem.
APPARATUS AND METHOD
Description of the apparatus and
discussion of the method of
recording data and of the choice
of observer.

RESULTS

Presentation and description of
the results with some interpretive
remarks.

DISCUSSION

Interpretation and evaluation of

the results in terms of the
theoretical position of the problem.

SUMMARY.AND CONCLUSIONS

REFERENCES

APPENDIX

GLOSSARY

Page

ll

19

34

50

57

 

Figure

LIST OF FIGURES

Schematic drawing of the apparatus.

Schematic representation of the
perceptual results.

same

same

same

same

same

same

same

same

Frequency for fusion of trains of
pulses of varying lengths and
intensities.

Frequency for fusion of trains of
pulses of varying intensities

and lengths.

Frequency at which the second pulse

of the train begins to fail to
arouse an effect equal to the rest.

.0.

0..

Page

14

20

22
23
24
25
26
27
28

29

29

30

Chapter I
THE PROBLEM

In the past, various attempts haVe been made to account for
visual perception in terms of the lawful operations of the Mind
which was supposed to be manifest in conscious experience. Thus,
Nundt and his students, notably Titchener, set about the analysis
of the content of Mind in terms of the elements of sensation of
which it was supposed to be composed (16). The brightness of a
light, its color, its shape, etc., were assumed to become amalgamated
under the laws of mental activity into the perception of, for example,
the flame of a candle. The study of such elements of sensation and
their resulting amalgamation into percepts was expected to reveal the
laws and principles upon which the processes of Mind proceeded. The
theoretical nature of this approach would seem to presuppose the
existence of some entity to be active in the formation of these per-
cepts from the raw material of sensation, though this view of Mind

was denied.

Since then, this view of visual perception has gradually been re-
placed by one in which perception is defined as the end product of a
series of neural events occurring in the organism and normally initiated
by the action of light on the eye (18). Nevertheless, the experience
of perceiving is vastly different in quality from the conceptualiza-
tion of an intricate pattern of neuron discharges in the cortex and

elsewhere which is assumed to be the physiological event involved in

Page 2
seeing. At present, it appears difficult, if not impossible, to
comprehend the phenomena of perception in terms of nervous activity.
Atfibest, only apparent relationships (natural correlations) between

nervous events and experiential phenomena may be described.

These relationships are assumed when a given stimulus situation
regularly gives rise to a.) a given perceptual phenomenon with certain
quantitative characteristics on the one hand, and b.) a given.nervous
event, as deduced from.changes in electrical potential, etc., in nerve
tissue, with the same or similar quantitative features, on the other.
thus, it has been shown that a pulse-“- or light of a certain duration
and intensity directed into the eye will give rise a.) to the per-
ception of two flashes of light (10) and b.) to a duplex nervous
discharge along the Optic nerve (8). These results lead us to
believe that the duplex discharge bears a direct relationship tO the
double flash. It is also known (A) that trains Of pulses of light
Of a given frequency will result in either the shifting Of phase,
and, within.limits, the "tuning" of frequency of the spontaneous
fluctuations of potential in the optic cortex (alpha waves), or in
substituting for these fluctuations others which are highly similar,
but peripherally generated. Trains of light pulses Of these same
frequencies have also been shown (3, 9) to give rise to the perception
Of flicker at its maximum brightness for a given intensity of stimulus
source. Again, the assumption is that there is a relationShip,

perhaps causal, between this cortical "driving" and the "enhanced"

 

* This and other technical terms are defined in the Glossary.

Page 3
brightness (3). Of course, these natural correlations are not found
to be perfect when reduced to quantitative data. The discrepancy,
however, can easily be ascribed to the complexity of the organism
and to experimental error, and the examination of these supposed
relationships would seem.to be a fruitful line of research in.the
field of vision since it is in line with the prevailing definition
of perception and leads to an integration of knowledge on the phony

omenal and the nervous levels.

Nevertheless, the evidence shows that perception is not a simple
cepying of the stimulus event. The perception of a dim.disc of
light may bear little resemblance to the stimulus configuration
(5, rage 6). The two-flash phenomenon (8, 10) is similar evidence
that the nervous system.actively engages in setting up the perception
according to definite laws of its own with the characteristics of the
stimulus acting as limiting or guiding factors. Thus, what is seen
nmw'depend upon some resulting unified pattern of activity emerging
from.the interaction of what is going on in the nervous system.and
the stimulus input. In this reapect it is possible to talk about

the organization of the percept as a process in time.

One phenomenon which can be explained in these terms has been
observed by Basler (11). He has shown that with pulses of light of
equal duration and separation, a train of two pulses will fuse
(appear as a single on—off percept) at a lower frequency than will

a train consisting of an indefinite number of pulses of equal duration

Page 4
and separation. In terms of a simple persistence-of—vision theory
based upon brightness times duration of pulse, Basler's results
should not have occurred, since in any instance, the individual
pulse of light is equal in duration and intensity to every other
pulse. According to a copy theory of vision, the pulse should thus
give rise to the same effect as any other of equal brightness and
duration. If fusion is taken to result when the effects of one pulse
persist until the next appears, fusion should occur at the same
frequency regardless of the length of the train. To account for
these results it is necessary to assume that the state of the visual
mechanism is altered by the preceding pulse, so that a succeeding one
does not have the same effect. In other words, the flicker fusion
phenomenon depends in part upon the duration of exposure to the
pulsating stimulus. If this were all, these results might easily
be explained in terms of conditions of the receptor (photochemical
theory). But Bartley and Wilkinson have shown (9) that brightness
"enhancement", for example, has characteristics which are directly
attributable to nervous activity farther along the visual pathway.
This and other evidence (5, pp. 313-321; 6, 12) lead us to suSpect
the operations of other than photochemical principles in thedetermining

of Basler’s phenomenon.

Thus, at least one alternative theory regarding the origin of
Basler's data may be offered. Bartley (1) has shown that in the
optic cortex of the rabbit, repetitious electrical stimulation of

the stump of the optic nerve at a medium intensity results in a

Page 5
complex response which at first does not directly follow the pattern
of stimulation imposed. On the contrary, while the first stimulus
pulse gives rise to a large cortical response, the second may not
be responded to at all. Responses may then appear to the third and
fourth pulses, but not to the fifth. These responses vary somewhat
randomly in magnitude, with the first generally larger than any of
the others. At length, however, as the stimulation is maintained,
the cortical activity settles down to a simple following of the
pattern of stimulation with each pulse of light giving rise to a
corresponding undulation in the record of cortical potential. These
undulations are all of similar magnitude and are all smaller than
the original activity which the stimulus aroused at the onset of

stimulation.

These data have been interpreted to mean that a burst of stimu-
lation occurring when the system.is at rest will find a certain
number of the higher elements1 in the visual pathway in a state of
high excitability and that the number that can be excited is
dependent upon the strength of the stimulus. Following their
response, these elements become temporarily non—stimulable or less
susceptible to further stimulation of equal intensity so that the

second burst of stimulation finds few or none prepared to respond.

 

1. These elements are not to be thought of as single neurons, but
rather some sort of a grouping of them related in a way causing
the channel as a functional unit to possess a much longer period
of "recovery" than is characteristic of a single neuron.

 

Page 6
The third burst of stimulation may or may not find some of these
elements prepared to respond, but generally will find some that have
returned to a state of excitability, and so on. Sines it has been
shown (12) that there is a definite excitatory rhythm of these
elements which is apparently spontaneous, stimulation catching a
given group of elements in some phase of this excitability cycle
has a certain probability of eliciting a reaponse. Those elements
which do respond are thus Selected out of the total as a functioning
unitary group. As the process is continued, other such groups come
to be organized so that eventually the cortex, through this division
into functional units, can respond to individual stimuli pulses at
a higher frequency than that to which any one element is capable
of responding. Of course, there is no reason to suppose that any
one element responds only with its own group; 1.9., that group with
which it first responded. If element A, for example, should have
a "recovery" period of shorter duration than element B, it may
respond with group B now, with group_A next, then be ready to respond
again with group B as the group becomes active again. This is to
say that there is no necessary reason to suppose that the groups are
anatomically distinct in number. It is only necessary to assume q
that, functionally, the cortex has imposed upon it, by the stimulus
train, a pattern of pulsating activity in which the individual
elements are organized into harmonic relationships. Thus the
initial pulse of light probably activates the largest number of
these elements catching all those at their maximum level of excit—

story potential and also activating a number approaching or just

Page 7
leaving their peak of excitatory potential. The second pulse then
may find none or an insufficient number of elements ready to respond.
The third may catch some slower elements not in phase for the first
pulse and some faster ones that responded to the first pulse. The
fourth may catch some slower ones that responded to the first pulse
but were not sufficiently recovered for the third. The fifth may
find again that none '15 prepared to respond, but that many are
closer to threshold for response than they were for the second pulse,
etc. Since the length of the period of "recoveryn of the elements
might depend in part upon the period of their excitatory cycle in
which the successful stimulus finds them, it would be longer or
shorter in the different elements after stimulation. And, of course,
we have not taken into account the effects of local excitatory
potentials which might act to enhance the excitability of an element
so that it becomes stimulable ahead of schedule, so to speak.
Analogous inhibitory activities may also be present. Thus, the
whole pattern of activity could be and probably is extremely complex,
but the result is an "organization" of the cortical elements into
functional groups which Bartley has called an "alternation-of-response"
pattern, and which is similar in theory to the nervous activity of

the auditory pathways proposed in the "volley" theory of hearing (19).

In the case of Baeler's results, it is obvious that, assuming
the above theory to be true, as the frequency of the two pulses of
stimulation increases from that giving rise to two isolated bursts

of stimulation, the state is rapidly reached in which the second

Page 8
pulse is not finding any elements ready to respond. Since the
train of stimulation ends at this point, the whole perceptual result
is a single flash of light -- fusion of the two appears to have
occurred. Actually, the result would not be fusion, but a case of
the second stimulus being ineffective. With a longer train of
pulses, the second pulse in the series may still be ineffective, but
the lack of response is swallowed up in the over-all continuing
response which is perceived as flicker. Fusion in this latter case
(with stimuli in trains of indefinite length) must be explained in

some other fashion.

Assuming the fialternation—of—response" theory, the question
arises as to whether there are any perceptual concommitants of this
initial period of reorganization of the cortical elements. Such
concommitants might logically be expected. It is possible, for
example, that a single flash followed by a dark interval and then
flicker in that order might be seen. The flicker would correspond
to the regular nervous response which sets in after the sorting—out
process is completed. Or it is possible that a few flashes dis-
similar in duration or brightness would appear to be followed by
regular flicker. Actually, any prediction is impossible on the
basis of present knowledge since individuals differ from each other
and from moment to moment in what happens in the cortex at the onset
of a train of pulsating stimulation, but the factors controlling
this difference are unknown. For example, Bartley has shown (1)

that bursts of stimulation arriving at various points in the cycle

Page 9
of the alpha rhythm may, and do, summate, inhibit or not affect the
succeeding wave. But since it is believed that the spontaneous
fluctuations of excitatory state in the visual pathways are closely
coordinated with the alpha rhythm, if not another manifestation of it,
it becomes obvious that some perceptual events corresponding to the

period of reorganization may be expected.

In light of the preceding theory, therefore, the problem with
which we shall be concerned in this study is the observation and
analysis of perceptual events assumed in a general way to exist
concomitant. to the period of neural reorganization at the beginning
of a train of pulses. We hope to find phenomena similar to that
found by Basler, but more complex, since we shall be using a greater
number of variables, and to be able to develop a similar way of
explaining those phenomena in relating them to, and thus, adding
substance to the "alternation-of-response" theory. Since the
period of neural reorganization during which we expect these visual
phenomena to exist endures only through the first few pulses inla
train of pulses, it will be necessary to limit the attention of the
observer to these first few pulses of the train. Otherwise, it is
likely that the observer will not be able to sort out those effects
attributable to the period of reorganization from those elicited
further along in the train (flicker). This inability is due to the
tendency of the individual to perceive the whole of something and
not the temporal sequence of changes which composes it. Thus, here
the observer may well "see" flicker, but not its onset or the relative

appearance of the first few undulations. Therefore, the problem

Page 10
involves limiting perception to these first pulses of the train.
By doing so, the subject will be enabled to report phenomena limited
to that period. The simplest way to do this is to present only the

first pulses for observation.

In general, then, we deduce from the "alternation-of-response"
theory and the observed neural events at the beginning of a train
of pulses, that corresponding visual phenomena exist, and may be
observed if sought under the proper circumstances. Our problem
is to observe and describe them, and to relate them, if found, to

the theory.

Chapter II
APPARATUS AND METHOD

The problem presented above requires that the observer report the
appearance of the first few pulses of a train of pulses. As has been
indicated, this is likely to be difficult to do. The observer will
not be able to separate the effects of these first pulses from those
of the later ones which are producing a regular flicker percept.
Therefore, the problem requires that the first pulses be isolated so
that the effects may be studied without complication from later pulses.
This means that pulses of light must be controlled in terms of dura-
tion, separation, intensity and numbgzg. The following apparatus

was designed to accomplish this controlled stimulation.

Light from a 32-candlepower automobile headlamp housed in a
lightproof housing was collimated and then brought to a focus. At
the focal point the beam of light was chopped by a sectored disc, the
primary disc, which had two openings of five degrees each, the openings
located exactly opposite each other on the edge of the disc. Thus,
light was passed in the ratio of one to thirty-six; that is, for
every unit time of light passed, there were thirty-six units of dark-
ness. Proceeding past this point, the diverging light rays were

caught and converged by another lens to pass through a circular optical

 

2. Ideally, some provision should be made for simultaneous observa-
tion of the alpha rhythms so that the temporal relationships of
the stimulus pulses and the fluctuations in cortical potential
might be known and predictions of response he made more accurate.
However, the equipment necessary to do this was not available.

Page 12
wedge within a very small area of the wedge surface. The light then
passed through an eyepiece consisting of an artificial pupil of approx-
imately three millimeters diameter. When the eye was placed before
this artificial pupil the surface of the last mentioned lens was seen,
appearing as an evenly illuminated, circular patch of light of four
degrees and eight minutes of visual angle. An adjustable chin rest
was provided to help hold the head position steady. The foregoing
was the basic arrangement for the production of evenly spaced pulses
of light. The intensity of the pulses was continuously adjustable
over a range of approximately four log units by the use of the
circular wedge and a neutral density gelatine filter of 10% trans-
mission which could be inserted at will in the path of the light.

The duration of the pulses was controlled by varying the speed of a
universal, series-wound motor which drove the sectored disc. This
speed variation was accomplished by the uso of a variac variable
transformer in the motor circuit. The speed of rotation of the
primary disc was read directly from a Weston electric tachometer con-
sisting of a magneto, Model 44, coupled directly to the shaft of the
disc and a specially calibrated Weston voltmeter which read directly
in RPM. For speeds higher than those given by the Weston voltmeter,
a Simpson DC voltmeter with a 25-volt range was switched into the
circuit of the magneto. This voltmeter was calibrated by plotting
the voltage given against the RPM from the Weston voltmeter for
selected speeds over the overlapping speed range of the two instru-
ments, the plot, which was a straight line, then being extended to

provide values beyond those obtainable with the Weston voltmeter.

Page 13
The number of pulses in the train-was controlled by a larger,

secondary disc which extended over the surface of the primary diso
and which was geared so as to rotate at 1/90th the speed of the
primary disc. This secondary disc had an opening coinciding in
linear space with that of the smaller disc when the two were aligned.
The opening was variable in size. The two together operated to
produce trains of pulses of light of given length. Thus, an opening
of two degrees in the secondary disc moving past the point of focus
of the light at l/90th the speed of the primary disc allowed one
pulse produced by the primary disc to pass through since during the
time the open sector was passing the focal point the primary disc
had rotated through 180 degrees, one of its open sectors passing the
focal point once. An opening of four degrees would allow two pulses
to pass; one of six degrees, three pulses, etc. A hand-operated
baffle was also introduced into the path of the light so that the
discs could be left to revolve continuously without any light pulses
being passed until the baffle was pulled aside. Thus, a single
presentation of the train of pulses could be given or a regularly
repeated presentation which depended in its timing upon the rate of

pulsation being employed at the moment.

The whole apparatus was arranged with dual controls so that
either the observer could operate it without moving from the point of
observation or it could be operated by an experimenter from a point
of vantage behind the light housing. The details of the setup can

be seen schematically in Figure l.

 

Page 14

'A ---’

a)

 

   

 

 

IZy/wt

De z‘ac'Z of DL'JCJ

Fig. 1. Schematic drawing of the apparatus.

The numbered items are as follows: 1. Light house; 2. light
source; 3. Variac variable transformer; 4. first collimating
lens; 5. converging lens; 6. baffle; 7. secondary disc(s); 8.
primary disc; 9. second collimating (converging) lens; 10.
screen; 11. circular optical wedge; 12. Artificial pupil; 13.
reflecting glass for fixation point; 14. lamp housing for red
fixation point light; 15. gear housing; 16. Weston electric
tachometer magneto, Model 44; 1?. universal, series-wound motor;
18. Variac variable transformer (speed control); 19. Weston
voltmeter calibrated in RPM; 20. Simpson DC voltmeter.

Page 15
It should be noted that the apparatus was so designed that the
intensity of the pulses of light, their duration and the number of
pulses in the train were independently variable. However, in order
to vary the number of pulses in the train, it was necessary to stop
the apparatus and adjust the opening of the secondary disc. The

other variables were adjustable with the apparatus in operation.

In addition to the apparatus proper as described, it should be
added that a fixation point was introduced by inserting a thin piece
of clear glass into the eye-piece just behind the artificial pupil
and at an angle with the line of vision. To one side was located a
second light housing which held a 7% watt red bulb behind a small
opening. In the dark—room conditions, this small opening was
reflected upon the glass in the line of sight providing a red dot.
Actually, the dot was doubled due to reflection from both front and
rear surfaces of the glass. This doubling was of value, however,
since slight adjustments of the head were easily perceptible by
changes in the relative positions of the two dots, and with a little
learning it was possible to fixate and maintain the eye in position
quite accurately without a rigid head brace. It should also be
noted that the light source of the apparatus proper was powered for
some of the experiment with a standard automobile battery and for the
rest by regular AC house current stepped down to 6 volts with a Variac
transformer. In the latter case, it was decided that fluctuations
in line voltage were not of sufficient importance to necessitate the

use of a voltage stabilizer. .All the data to be presented here were

Page 16
gathered using the line current as source of supply except for certain
supplementary and/or preliminary data which will be reported in the

appendix.

Actual observations were made as follows: the experimenter,
acting as his own observer, adjusted the equipment to produce trains
of pulses of given lengths and the wedge was set at a predetermined
intensity. The observer then sat in darkness for approximately ten
minutes. This procedure was utilized to escape any effects of light
adaptation that might obscure the perception of the pulses of light
to be used. However, inasmuch as the intensities were quite definitely
far above threshold, complete dark adaptation was not considered
necessary. After a period of approximately ten minutes, the light
source was turned on and the motor started. The observer fixated the
fixation point and when ready, pulled aside the baffle. At a certain
fixed amount of angular turn before the first pulse of light was
passed, a click sounded to act as a warning signal. Following this,
the train of pulses appeared. The observer then immediately recorded
the RPM of the primary disc (thus, the rate of pulsation) and his
report of the percept. Care was taken to see to it that the disc was
running at a constant speed before and after the observation and the
recording of the response. Following the recording of the response,
which was done diagrammatically in pencil on cards under a red light,
the observer selected at random another rate of pulsation by adjusting
the motor variac and the process was repeated. This process was

continued until the whole range of pulsation rates from those giving

Page 17
isolated, discrete flashes to those giving complete fusion of the
train was covered. Usually, only one such range was observed in any
one session. Once, however, two different train lengths were thus
observed during one evening. All observations were made at night to
insure dark—room conditions, the periods of observation extending
over 15 days. During each session observations were made of the
selected train length at each of four different intensity levels.
These proceeded from weakest to strongest during the evening. In
all, observations were made of four different train lengths, from two
to five pulses, at four different intensities over a wide range of
pulse rates depending upon the train length. The train lengths were
varied randomly from session to session and in one case were inadver-
tently repeated, thus offering a check on the objectivity and stability
of the reported phenomena. This repetition occurred because of an
oversight, and was not disCOVered until later. In addition to these
observations which were all made by the same observer, and which are
the ones to be reported here in detail, observations were also made
by four other observers with the writer serving as experimenter and
recording the spoken report of the observation at various intensities
and train lengths. In no case did any one of these other observers
duplicate the whole of the procedure of the writer. Enough data
were gathered from these observers, however, to substantiate the main
details of the writer's observations, and at the same time to indicate
that the perceptual responses of different people in this situation
may vary in detail from time to time. These data are to be found in

the appendix.

 

Page 18

Because of the subjective nature of the results, it is not possible
to present statistical analyses or even, in some cases, quantitative
data. The attempt was to report the number of flashes seen and a
qualitative evaluation of their relative brightness, duration, etc.
However, it was not attempted to reduce this last to any kind of
quantitative statement - indeed, the nature of the situation made it
impossible to do so. Therefore, the results were diagrammed in
schematic form and presented along with verbal reports of the

phenomena.

Chapter III
RESULTS

The results may be seen in Figures 2a, 2b, 2c, 2d, 3a, 3b, 3c,
3d and 4. In these figures, the perception is represented schematic-
ally, each of the undulations representing a change of brightness
with a baseline of no brightness, and increasing height representing
increasing brightness. From these results, and as abstracted in
Figure 5, it can be seen that, in general, the fusion point of the
train is higher the greater the length of the train3. On the other
hand, the fusion point bears no simple relationship to the intensity
of the light at the train lengths used here (Figure 6). Other studies
have shown (13) that the off for flicker is definitely a function of
intensity. The present data can be explained only if it is assumed
that with the train length used (within the period of reorganization),
the usual conditions for flicker do not operate and that fusion here
does not represent the same thing that it does in the case of fusion
of an indefinitely continued train. We will discuss this more in

detail later.

It can also be seen that in many cases, as the frequency of the

train increases, some of the pulses tend to be ineffective in producing

 

3. The fusion point is defined as the frequency of pulses per second
at which the train is seen as a single flash, or cannot be dis-
tinguished from a single flash. For definitions of other terms,
see Glossary.

Na. PulJcJ in train

Page 20

 

 

”(Q/"C- ; u j
} .1I1l____. .AKALAL__ .J.hu“L1.
t -alhlL___. -1IJIJI1L. .4anuanhns
5’ L .m— M A1144."
6 .m— A— M .AAAA—
7 _AMI______ .nlLAIl___. nanhannu.
5 -4llL————— .1l4uln__. .1Luaann_
) _‘IL_____. .Auhuu___ .J.IIL___. .Ahna-n_
IO
41 L— _-_ ‘—
42 L— ‘—
I} ‘— .m—
14 .L— _-__
If A— L.
4‘ .4llL_____. _AIIIL____
47 A—___.
46 .‘___ A— ‘—
4, L— A___
20 #— ‘___ A—
21 ddh___———
22 L—
210 A—_
25 L.

 

Fig. 23. Schematic representation of the perceptual results.

Intensity:

.0015 e/ftZ.

'2 means the percept is doubtful; .1 means the
percept is very definite; 1. second pulse unimportant.

 

-a—pm-*—~....y —' '“e vs— -

Page 21

No. Pa (.5 (J in (rain

 

free/rec. 2 3 1. 5
2 AF—
} .J.J-L____. .1l1I1l___ .nlul1L1L__
‘0 4A— _A_“_ Jul—
f .4..— M___ M!
s L _A..A_ .444; ML
7 L 41“.
8 4L— 4“_ A“—
? .Jnn______. ._AL.‘hIL__
10 _AL_ AA— AA—
41 A— u_. A—
12 ‘—
4} _A__ _A.A__ .4“—
1‘r _“L_ .4“.—
‘Lf .JIL______. _ALIL____.
46 _‘.__. .“— _L__ _‘__
47 A— A—
15 .J‘I‘L____. _AI|_____..nn-n_____
47 L— _‘__
2.0 L— ‘—
21 #— _‘—
22
Z} .4-L______ .n‘IL___L_. _‘-I-L___.
24 _-__
2; _A___
Z6
27 A—_ .‘____ _‘____
26
27 _‘___
}0 L—

 

Fig. 2b. Schematic representation of the perceptual results.
Intensity: .015 c/ftz.

? means the percept is doubtful; 1 means the
percept is very definite; 1. second pulse unimportant.

 

*“M‘Wy—D-“fifi—'o_—» _ n

 

‘ .

f req/feq

No.“ False: in train

Page 22

 

.5 .1LALALAL_

4 - .nlulnln__ _ALAuLlulJ
I L. _A_Al_ .m—

‘ —‘-A— Am— w
7 .JIuIL_____ ._A|.A|AL__. .4n1unAL_i
8 L M
9 -‘—— AL

40 L— _A.L_ 44L

41 ‘4“; ‘9‘). . l-

12 A ‘49.)

13 -a‘|~—————— - _‘.JIIIL__
14 ..‘II5_____ _A|.1L 4|...

’5 A——' A— L
16 4—? 4.4—. _‘_

’7 -4‘.L—————— —AIl|—————- -4‘IIL————- ~1I“L————-
15 A— ‘—

17 .4IL______ ..‘II______

20 L— _A___ 4‘.—
2” L.—

22 n

23 A__ A ... 2
24 A

2f A... ‘ ...

24

27' ndlh______ .nlIL____

25 g ‘

Z7 4-.

)0

J" L
)z

)5 _L__
)4

57

34 .

37 ‘1];

 

,«o— --

Fig. 2c. Schematic representation of the perceptual results.

Intensity: .15 c/ftz.

7 means the p ercept is doubtful;
1 means the percept is very definite.

Page 23

g No. Pulses in train

 

i freqflec. 2 3 A. j

I 5 AL—

I a. #— _AAA_ AAAL M
.5 .ALJIIIMI_.
6 _A_A__ _AA_ ...A“. ”L
7 .ILJIIK___ .1InlllL__
a L m .m_

9 _A_ JL _M_ _‘_
1O -4IIL_____ .nlL‘dl____ .munuIIL__

11 _A___ J‘— M;
12 _‘__
13 _A.____ t #—

1~ JA— _A.A__ A.—
15 _All____a_ .4lu|AL_1_

16 _‘_2_
17 M; g
18 ‘—

19 ..IL______ ~"%?'9-- ..AIL______
21 .Aln______ _Alla_____

22 L
2; L

24 A—

25’ .1IL______ .elIIL_____
26

27

25 L—_
29

}o

)1 L—
)2

}} L—
3}

 

-_--—_— ...-.....-“ “...... “Ho—...“

Fig. 2d. Schematic representation of the perceptual results.
Intensity: 1.5 c/ftz.

? means the percept is doubtful;
1 means the percept is very definite.

l

Intensities m c/ft”

Page 24

 

freq./scc p.00 If .07)“ .1} 1-}

2 AA—

} .AIJIn_____ .4‘L1.L____

4 .nlle_____ ,nlLAI_____ .nlLAI_____

6 .4;— 44__ _AL_ _AA__

7 _A__ L _AA_

7 .JI.-______ .JIII______. .1IIL_____ _A|l-_____
10 ..‘In_____ .nlll______ .n-lIL____.
11 _L____ _A___
12 n‘IIL____. .nlIL_____

1} _AIL__.___ .n‘|n_____. _Alh______
74 .1Il-_____
15 _A___ L— AJ—
16 .‘In______. .4IIL_Z___.
77 ..A.___ L_ _L__.
18 .nlIL______

 

Fig. 3a. Schematic representation of the perceptual results.

Train length: 2.

? means the percept is doubtful;
means the percept is very definite.

Page 25

 

Intensities tn c/ft.‘
" " freq/sec. .001; . 015 . 7; 1.;

5 .J|ML‘L___. .1L‘blu___

k _1|1Nl____. .1L‘LAL___ .1LlnlL...
5 AL AL; 4mg

6 AL— “— AM_ _‘.A__
7 AL— _A“___ 4.“—
6

9 .nlulln___ _1L4uL____
‘10 .J...L_____ .nlLAIn___. .Jlllll____

11

12 .4I41AL... .JI4IIL___
1» .JI‘IL____
'15 .nlla_____
16 .JIIIn____ .ellIL_____ .JIMI_____

77 a. .J“L_____ .nlln_____
18 _‘—_ A—
17 L— L— _‘___ _‘___
‘LO L— _A____
21 _-_._ A— _A__ A___
22

z; .4In______ .n4IL______ .nalL_____

2k _‘.___

2)” _A___

26

27 A—_ L—

 

 

Fig. 3b. Schematic representation of the perceptual results.
Train length: 3.

1. Sec ond pulse as bright as first, but "unimportant" - count as 2
if not careful.

2. It may be that the "on" reaponse (for the first pulse) and the "off"
response (for the third)'makes these appear to stand out.

 

Page 26

 

Intensities in c/ft‘
freq/rec. .0015 .01; .1; 1-}

3 ML _AAAA_ AMA—

t M M . JAIL—’-
; JIéIMI- _ .mldllll__. .

6 .AAA.__ “J. ..LAAA—
7 .Ahulll___.

5 AL “_ J-AL—
) L .m— .4‘_ 4L.
10 #A— .AA‘_
11 L ...AL_ w— A‘—
-12 .AIIIL____

1} .1I‘hlL... _4|4|______
14» .‘III_____ .nllln_____ _AIIL_____ .JL‘L_____.
1; L

16' .JIIL______ _AIII.____

17 .1IIL_____. .annlL__£__
16 .J‘-L_____. .nlIL______ .4‘.L_____. _4|@-i____
1L7 _AIL______ .4III_____ _AII______ .4-4agL___
2.0 L A— L. L—
22: _AIIL_____ .4IIL_____. .4-IL______

z} .4-‘L_____. .4-nn_____ _AI|______
24L _AIL______ ...-...... .nlln______
25' _AIL_JE__. .4lI______ .nlln_____.
26 _‘.____

27 A...— A—

25 4—:—

27 4—..—

 

 

Fig. 30. Schematic representation of the perceptual results.
Train length: 4.
3 means the percept is very definite.

1. Sometimes gamma movement confuses the perception here.
2. Pronounced "on" and "off" responses.

 

Page 27

 

Intensities in Vft"
freq/sec. .001; . o1; . 1; 1.;

5 .nIlL‘Ul‘L_ .JIAUIIUI__ ..‘ut‘ul‘L.
’— . n. 4. In“! I n“
.6 A “1 M w
7 m .ml #441. w_
5 M u— M
9 .‘L A
10 _AH__
17 L A. A
72. L— L
13 .m— 4*
7‘? “— _A“_
15 “.— A.
16 ‘— LL.
77 .444__ L _‘_.'_
19 L— _
11
22 2 _‘-.IL____
7.) _‘__ ‘— _A“_'_
25
z; ..— 444_ _A___
16
27 .nlll______
25 .4.‘-L____ ..All______
2)
)2.
)1:
)5
‘}6 .4‘-L_____ _AIL______
)7 ..‘____

 

Fig. 3d. Schematic representation of the perceptual results.
Train length: 5.

7 means the percept is doubtful; J means the
percept is very definite; 1. second pulse "unimportant".

 

 

Page 28

 

freq. sec. Aug. 1) Aug. 27

} _LAAA_ JAAL
4 .1.“__

5' _4llllhllill___ ..AILllnlIIl___
6 .41I4ulIAIn____

7 _A_AA___ 44A—
8 J.“—

9 ‘6‘“ .m—
1 O

1 7 _n‘|LAII|n____
11 4.1L

7.3

M M A—
16 ‘— ‘—
7‘7 .4‘.‘|L_______.
78 Am— A—
20 ‘— _‘__
2‘1 .
23 _A_ A__
2x#
2; A—
2.6
2f7 _nAIIL________.
25 L... _‘.___
2T9 . ..II|L___._____—- fall.k——————————-

 

 

 

Fig. 4. Schematic representation of the perceptual results:

same subject on different dayS.

Intensity: .15 c/ftz; train length: 4.

Page 29

  

panes/Jet. Intensity (era!
34- — .1
fl— 2’

y‘ —' I

,U1-i
fl —
)0 — ° .01)r
zo —- ° 1’

 

 

l I I
10 } 1’
length of train

 

 

H-
Fig. 5. Frequency for fusion of trains of pulses of varying
lengths and intensities. LDR: 1/36.

i fwlo/g [12¢ train length

——-0
}h_ M;
)3—
)0—

m _

u — °/\/}_4
a —— __________...-——---o

a _ 0/ \L}
m —— °_______————°

16 — 0/

1" — 44.2
. _J_ n I ,1

.005' .09’ .4“
log Intensity m c/fz“

 

 

 

———‘

Fig. 6. Frequency for fusion of trains of pulses of varying
intensities and lengths. LDR: 1/36.

1

 

 

Page 30

 

 

frag/sec. intensity (mi/2 [myth
w ——
14 —
7. _ \
IZ — \so/L 2
70——
Z‘ ——°*~~;;: #42::
—- /o o o._~~ .3‘,-o———"61'
. _ \/¢ :93, :7 '°“ 7
Z I I l l I I I
J 1‘ 5 .004)” -07)' '7)’ 1-;
train length log. Intenslty in c/ftz
_____ .- i.“

Fig. 7. Frequency at which the second pulse of the train begins

to fail to arouse an effect equal to the rest.

Page 31

a flash. Thus, with a train of four pulses, for example, it is not
at all unusual to find a definite impression of a triple or duple
sensation. The point at which the two—pulse train becomes fused
might be taken to be the critical frequency for the disappearance of
the second pulse insofar as effect is concerned. However, that this
is not the case is shown in Figure 7, which gives the frequency at
which the second pulse begins to lose effect. It is evident that
for our conditions we have the second pulse failing in effect in
longer trains long before the frequency reaches that of fusion of the
two-pulse train. This phenomenon will be explored in the discussion

chapter.

A few additional points may be noted with respect to these data:
the fusion points as given are not exact. There is some fluctuation
of fusion point in the data, and those cited in the tables and graphs
represent an average arrived at by inspection. Also, the schematic
appearances given in Figures 2a, 2b, etc., in some cases are not con-
gruent for a given frequency of pulsation. This apparent inconsistency
can be traced to the fact that it is highly unlikely that the initial
pulse came at the same place in the cycle of excitability of the
neural elements for a given frequency upon repetition. Thus, it is
not expected that the same results would necessarily follow each time
in the neural record. That they do not follow in the perceptual
one is simply a reflection of this variability of neural response.
Nevertheless, the general trend is clear. As the frequency is

increased, the resulting visual response becomes shorter and less

Page 32
regular. Fluctuations appear which have no correlate in the stimulus.
Eventually, the brief stimulus train tends to lose all evidence of
temporal discontinuity and a single flash, of longer or shorter dura-
tion appears. In this process, flicker may or may not make its
appearance prior to fusion and if it appears can be observed to fade
away before fusion takes place. (It tends to appear more in connec-
tion with longer stimuli trains.) It is easily observable that the

flicker does not increase in frequency, but simply becomes fainter (2).

If the attention he directed toward the dark intervals between
pulses, temporal variabilities are observable. Frequently, even when
the correct nmmber of flashes could be counted, the first interval
would be more pronounced than the later ones. This effect may be
attributable to a more pronounced "off" response following a primary

burst of stimulation.

One curious phenomenon should be mentioned. The observer sometimes
deliberately made the attempt to count the flashes. ,At certain fre-
quencies, which varied with the train length, intensity3and from.time
to time, it was possible to count the number of pulses in spite of the
fact that a strong impression of some lesser number would persist.

For example, with a train of four pulses it would be possible to count
four flashes, but the second would be nearly at the threshold of
awareness and the general impression would be triple. It was as if
the second pulse could be sensed but was simply ignored or rejected
in the assay of the over-all impression of the short burst of pulses.

It is possible that the ability to count these pulses was spurious

Page 33
and that the counts arose from a rhythm fallen into or was simply
given by the knowledge of the number of pulses being presented. Never-
theless, the general impression of the second pulse as outlined above

seemed very real.

It was extremely difficult to avoid confusion with certain other
visual phenomena which could be observed. For example, in some cases,
noted in the schematic representations of the data, the "on" effect
at the beginning of the train and the "off" effect at the close may
have been confused with an enhanced brightness for these pulses. In
other cases, also noted, gamma movement may have been misinterpreted.
In many cases, the impression of number of flashes was nothing more
than that, the actual number being uncountable because of the high
frequency. There was, however, no doubting the differences from flash
to flash in a given train. (Those observations which were especially

definite are marked with an exclamation point in the figures.)

The observer had much practice observing before these data were
recorded. Actually, there was not much change from early observations
to later ones (see Figure 4), or even from the very first practice
observations made and these later made and reported in the body of
this report. (Compare Appendix A with Figures 2a, 2b, etc.) In the
case of Figure 4, it is easily seen that with an interval of two weeks
between observations, the differences are trivial for a given train
length and intensity. It should be emphasized again that this dupli-
cation of data occurred spontaneously and was not discovered until
after the data had been gathered. Thus, it is not a matter of pur-

poseful reproduction of data.

Chapter Iv
DISCUSSION

The primary point which is apparent from.the results is that with
the short trains of light pulses used, the perceptual response may not
bear a one-to-one relationship to the stimulus. It is, of course,
to be expected that the perceptual response may’differ from.the char-
acteristics of the stimulus in.many situations. Here, however,
present theory would lead to the expectation that as the frequenoy
of pulsation increases for a given short train of pulses, there.would
be a gradual and perhaps somewhat imperceptible change from.isolated
pulses of light to flicker which would then become fainter and faster
until it disappeared, leaving a single, homogeneous flash. Such a
result would be expected if it were postulated that the visual response
were determined by the photochemistry of the receptor cells of the eye.
Hecht has outlined what in its essentials remains the theory that
vision is determined qualitatively by photochemistry (15). This
theory is still widely accepted. According to these formulations,

a pulse of light causes a decomposition of the photochemical substance
according to its intensity, duration and the concentration of the sub_
stance. Immediately, there begins a regeneration of the substance
which proceeds until complete regeneration.has occurred. As the fre-
quency of pulsation increases, the amount of the substance decomposed
lessens, due to shorter duration of exposure. The rate of regeneration
is thus indirectly affected through the concentration of breakdown

products present from.moment to moment. It is hypothesized that

Page 35
flidher results when these oscillations reach a certain minimum.and
fusion occurs when they reach a still smaller minimum. At any pulsa-
tion rate this system.demands a one-to-one relationship between.pulses
and perceived flashes. Our results definitely show that this simple
process based upon the photOChemical reaction as outlined by Hecht
cannot apply, since the visual system.is found to fail to follow the
pulsation even at slow rates. The disappearance of some of the
stimuli pulsations in experience cannot be accounted for by photochemr
istry, nor can the arbitrary and somewhat erratic nature of the percept-
ual response in our situation. The two together are sufficient'to
refute the photochemical theory as the simple direct determinant of
the perceptual end result. Of course, we do not deny that photo-
chemistry is involved - only that it does not alone determine the

nature of the percept.

In place of this theory, there has been proposed an alternative
one depending upon the action of the nervous elements making up the
higher regions of the visual pathway ( l, 7, 8). This theory has
already been outlined somewhat in the introductory chapter of this
study; In brief, these elements, which may be tentatively identified
as the terminations of the visual pathways in the cortex and their
downward extensions for an indefinite distance, exhibit spontaneous
fluctuations in excitability which seem to be correlated with the alpha
rhythm, A pulse of light sufficient to excite a receptor in the eye
gives rise to a burst of nervous impulses which, upon arriving at these
elements in the pathway, may or may not succeed in traversing them

depending upon just where in their excitability cycle the burst arrives.

Page 36
If the burst of impulses succeeds in traversing them, the elements
become resistant to further stimulation for a time. As the frequency
of stimulation increases in the eye, the various pathways, dominated
and controlled as to their ultimate effects in.producing visual
responses by the higher elements, tend to become "sorted out", some
conveying the effects of one pulsation and some of the next, so that
each pulsation is represented in cortical activity but not by the same
element. The initial stimulus pulse would find the largest number
of these pathways capable of transmitting an effect to the cortical
centers where vision is elaborated. Subsequent ones would find less
of them.capable of this activity until eventually, as the "sorting out"
process is completed, a certain minimum.number of the pathways would
be conveying the effects of each pulse of stimulation depending upon
the rate. During the period of "sorting out", however, there would
occur arbitrary and erratic fluctuations in.the number of pathways
responding together to a given pulsation. If it is assumed that the
brightness of the visual response depends upon the total number of
impulses arriving at the cortex for a given pulsation, as has been
assumed, for example, by Crozier, in a somewhat different context (14),
then differences in the number of pathways responding would effect this
total, and, thus, differences in brightness of the flaShes correspond-
ing to the different pulsations would be expected, with some pulsations
failing to give rise to any response at all. As the frequency of
pulsation is increased, flicker would be expected to result as a
response if the train of stimulation were sufficiently long. In the
short train, however, occurring wholly within.the period of reorganiz-

ation or "sorting out", a flicker-like response might or might not occur.

Page 37
In the special case of two pulses, we would expect that the second
pulse would drop out in effect as the frequency rose to that point
where the second stimulus was occurring before sufficient of the
pathways were recovered to reSpond a second time. we could not call
this result fusion in the ordinary use of the term, since the two
would not have coalesced, and the visual reSponse would be simply a
response to one pulse. With longer trains, we would expect the
second pulse to become ineffectual, thus reducing the number of observed
pulsations below the number of stimuli. Farther along in the train
we might also expect other pulsations to disappear in effect in the
same way. Depending upon where in the excitability qycle the initial
burst of impulses arrived at the higher elements, we would expect the
initial flash to be brighter and subsequent flashes to be less bright
in a somewhat unpredictable order as more or less of the pathways were
able to reSpond each time. In these longer trains that are yet within
the reorganization period, fusion would eventually be expected to occur,
of course. we would expect the pulsation frequency for this type of

fusion to be higher than that for fusion of a two-pulse train.

This recapitulation of the theoretical position has been made in
order that it may be clearly seen that the expectations are all found
to appear in the results. The diagrams of the perceptual reSponses
Show that the flashes did vary in brightness. ,At certain pulsation
rates the second pulse regularly failed to give rise to a corresponding
flash and, moreover, it was possible to observe its decrease in effect-

iveness as the pulse rate became higher. This led to the observation

Page 38
of fewer flashes than there were stimulus pulses, as predicted. More-
over, at higher frequencies other pulses dropped outt as well, thus
still further decreasing the number of observed flashes. The data
are not sufficiently accurate to warrant the definite statement that
this further reduction occurred at double the frequency of the first
such reduction, though such might be predicted if the argument regarding
the disappearance of the second flash be extended. Occasionally,
flicker appeared, and this response occurred with greater frequency
in connection with the longer trains. Almost always the first flash
was as bright or brighter than any of the others. Those few negative
instances may well indicate coincidental examples of just the proper
juxtaposition in time of the first pulse and an alpha wave, so that
the pulse is rendered relatively ineffective ( 1). Finally, it is
very clear that there is a marked separation in frequency rate for
fusion of a two-pulse train and fusion of a 3, 4 or 5-pulse train.

The frequency differences among the latter are rather small, and the

indication is that the data may represent different types of "fusion".

There are other aspects of the results that were not so readily
predictable from the theoretical position, however. we shall now turn
to them, As shown in Figure 7, the general point at which the second
pulse begins to be ineffective in producing a response is much higher
for a train of two pulses than for longer trains used. Perceptually,
it would seem that the third pulse somehow assists in suppressing the
effect of the second. Perhaps this is a retinal function mediated

by delayed reactions in the horizontal and amacrine cells. It seems

Page 39
more likely, however, that it is cortical. . If the second pulse should
come at the proper place in the cycle of spontaneous rhythm of the
cortex, the effects might be delayed until the next peak of excita-
bility appears, at which time a larger and perhaps functionally'dif-
ferent (in terms of visual effect) change in potential (alpha waves)
would give rise to a visual response. iA third pulse, however, might
coincide with this peak of excitability, and thus,its effect would
tend to merge with that of the second pulse, rendering the second
inoperative in producing a visual effect. The early dropping out
of effect in longer trains of the second pulse may thus be more a
matter of its being delayed and then merged with the third pulse than
a simple failure to produce any nervous effect. An alternative
explanation might be in terms of a breakdown of the analytical feature
of the observation process as the train length increases. However,
the discontinuity between the two-pulse train and the three-or-more-
pulse trains and the high similarity between the three, four—and-five-
pulse trains in this regard (the frequency at which the second pulse
begins to fail of effect) would seem to indicate that the explanation
must be sought in the effects of the third pulse, not simply in those
of a longer train. In the latter case, the frequency of second-pulse
failure would be expected to go up continuously at least over the first
four or five increases in train length. But it apparently does not.
There is a sharp division between the two-pulse train and longer trains.
This perceptual phenomenon seems to be clear-cut. Figure 7 also
shows a slight trend toward a rise in frequency at which the second

pulse becomes ineffective with an increase in illumination, which is

Page 40
what we shouldeexpect. Apparently, however, the length of the train
has little effect. A weakness here is that in both these cases we
cannot assume anything about the statistical significance of the

differences involved because of too few cases.

In the matter of fusion, it has earlier been pointed out that
the fusion point here is largely determined by the length of train,
and not by the intensity of the illumination. This may not be quite
true for the five- pulse train where the curve seems to rise with the
increasing intensity. ‘We'will come back to this in a moment. It
was suggested that in the case of the two-pulse train the teerfusion"
is inappropriate. The suggestion was made that, in effect, we have
the failure of the second pulse to arouse a response because it comes
within the pseudo-refractory period established by the first pulse,
and thus does not find enough pathways ready to respond to be effective
perceptually; The term."fusion" ordinarily seems to be used to refer
to a kind of temporal resolving power of the cortex or the retina (29).
Actually, it has been pointed out (17) that with all'the work that has
been done on flicker fusion, there is as yet no completely satisfactory
theory to account for the phenomenon. In terms of the alternation
of response theory, fusion may be said to refer to the fact that the
pathways have been spread so thinhy throughout the temporal pattern
of stimulation that very few, relatively, are responding to any one
pulse. This means that at the cortex, stimulation is occurring
diffusely (in the spatial sense) to each pulse, and that the effects

from.pulse to pulse are arriving so rapidly that to all intents the

Page Al
effect is the same as with steady illumination, in.which as.many elements
are being activated at any instant as are going into a state of "recovb
ery". The theory thus accounts for perception of steady illumination
by assuming the cortex to operate with a very rapid on-off action with
fusion as such representing simply a lower limit to continuity in
visual experience. This making of something continuous out of some-
thing essentially discontinuous might be assumed to be a principle of
primitive organization inherent in the structures of the organism
giving rise to experience. Thus, we approach here a Gestaltist type

of closure.

If some such explanation of the phenomenon of fusion Should be
found to be the correct one, then "fusion" as we have found it in our
records would represent a special case. 'With the short trains used
here, intensity is found not to be a determining factor (within the
limits of intensity used). This is so because with increasing train
lengths within the period of reorganization the intensity of the light
does not play a determining role in establishing the brightness of the
experienced response, since the number of elements capable of respond—
ing would fluctuate in part with intensity, but also in part indepen-
dently of it so long as reorganization is going on. Therefore, with
increasing train lengths, the effect of frequency on failure of pulses
to elicit a response would shorten the total response. This is shown
in the results where, with increasing frequency, the number of per-
ceived flashes decreased. The first pulse arousing a certain number

of elements leaves the remainder insufficient to arouse further

Page 1.2
effect when the train is short. As the train lengthens, the prob-
abilities become greater that succeeding pulses will find acitable
pathways awaiting stimulation. Thus, in order for fusion to tales place
when the recovery period of the elements remains fairly constant,
frequency of pulsation would have to go up. In this view, intensity
is seen to play a minor role. With increasing intensity, especially
at the longer train lengths, we would expect fusion frequency to go
higher since a greater intensity would stimulate more elements at
each pulsation by bringing to response more elements in the early
stages of their response cycle. The effect of intensity upon fusion
frequmcy would not be as strong as that of train length when we
are working with short trains within the reorganization period. These

theoretical considerations are in line with the results.

At the highest intensity, there is a tendency for fusion frequency
to go down. Exactly what this means, it is difficult to say. It
may be a retinal effect in which after-images at the higher intensity
levels tend to obscure the vestiges of flicker. It might represent
some perceptual function dependent upon the train length except
that the result was not observed with lower intensities. It is
interesting to note that the rate at which the fusion point increases
with increasing intensity is different for the five-pulse train than
it is with the other, shorter trains . As the train length becomes
longer, more of the sorting-out or reorganizing of the pathways has
been accanplished, and, therefore, the more dependent the fusion point

becanes upon intensity of illumination. It would be expected

Page 43
that the fusion curve would appear more and more similar to that with
a continuous train. On the other hand, the rate at which the fusion
frequency increases with train length is markedly different for the
higher intensity than for the others. Actually, the two effects
are facets of the same phenomenon. Thus it may be that as the train
length increases, intensity comes to play a larger and larger role in
determining the fusion frequency and that a train length of five may
represent a sort of threshold of the effect of intensity'which has
heretofore (in shorter lengths) been somewhat inoperative. This
would be the case if the duration of the individual pulse and the
intensity were reciprocals in the short train lengths with the
reciprocity breaking down at longer train lengths after reorganiza-
tion has been completed. Before anything definite can be said about

this aspect of the results, however, more work needs to be done.

Chapter V

SUMMARY AND CONCLUSIONS

This study was undertaken with a view toward exploring the per-
ceptual events elicited by a short train of light pulses and comparing
the results with theoretical expectations based upon recordings from
the optic cortex and the assumption that these recordings are highly
correlated with perceptual events. This assumption is in keeping with
other similar studies in which it was shown that there are certain
natural correlations between perceptual responses and nervous act—
ivity in the visual structures (7, 9, 10). In order to accomplish
this purpose, apparatus was designed which permitted the stimulation
of the eye by trains of light pulses of any number from one to five
inclusive. The intensity of these pulses could be varied at will
and the rate of pulsation was also variable with the light-dark ratio
held constant at one/thirty—sixth. One observer, the author, was used
to gather the data from which the conclusions are to be drawn, but other
observers, three in all, submitted evidence that the data reported in
the body of the work are not idiosyncratic in general outline, though
somewhat so in detail. (These data may be found in the appendix.)
This individuality of detail in the results is perfectly understandahle
and acceptable in the light of the nervous apparatus involved in vision
and the limitations of the experimental situation. It has been
pointed out in Chapter II that without some means of recording moment
to moment changes in the excitability of the visual pathways (or at

least recording the alpha rhythm of the optic cortex which seems to

Page as
be closely related to this excitability change) it was impossible
to stimulate the system in exactly the same phase of the cycle frcm
time to time, thus naking for sane variability of the neural activity
of the visual system. The results here given are variable and
differ from time to time, within certain limits, and from person to
person. Such variance might be expected if these perceptual events
are actually based upon the nervous activity which has already been
seen to vary in like manner. In view of this situation, it was felt
that the securing of data from other observers over and above that which
was done would add nothing to the meaning or accuracy of the conclu-

sions which are to be drawn from this work.

The procedure used in securing these data consisted of stimulating
the eye of the observer randomly in regard to length of train and
intensity over a period of several sessiom until the different combin-
ations of intensity and train length had been covered. The observer
adjusted the apparatus and thm activated it when ready, and recorded
the perceptual results verbally ard by diagram immediately after each
observation. The results were then compared and contrasted, and
certain reference points such as fusion of the train and point at which
the second pulse began to be ineffective in eliciting a concomitant
response were abstracted from the data in terms of frequency of pulsa-
tion. These variables were graphed for examination as to trends and
relationships. From this procedure we can conclude the following:

(1) Theories of visual resp0nse based upon photochemistry as
the determining factor are not satisfactory to account for the results

obtained. The variability of perceptual response and the general

Page 46
features such as the failure of the second pulse and the dependency of
fusion upon length of train more than upon intensity of light are in-
compatible with the theory that photochemistry determines the visual
response sinCe the photochemical theory cannot account for them. It
should be mentioned that the term "photochemical theory" refers to the
theory that photochemical events determine the quantitative and qual-
itative aspects of perception. (See pg. 34.)

On the other hand, these data can successfully be integrated into
Bartley's "alternation-of-response" theory of visual functioning thus
providing strong support for the theory's validity. The fact is that
all but a few of the data found in this study could be predicted from
the "alternation-cf-response" theory. And those few data whichvrere
not so readily predictable can easily be explained by making reasonable
and simple assumptions in line with the theory. For example, assuming
that the visual response is given by the nervous response in its gen-
eral details in this simple situation, which has already been done in
other studies before cited, both the variability and the general na-
ture of the progressive change in the visual response with increasing
frequency of pulsation that was found are in direct agreement with the
results found from recording action potentials from the optic cortex.
The "alternation-of—response" theory accounts for the cortical action
record results. It can thus be said to account for these results as
well. It has been demonstrated that the results with regard to
fusion of trains of different lengths can also be satisfactorily ex-
plained in terms of the theory.

The foregoing conclusion points to the fact that the visual res—

Page A7
pause in the simple situation reported here depends upon the operation
of a number of parallel circuits from retina to cortex which cooperate
in the transmission of infonnation concerning the rate of pulsation
of the stimulus. These circuits are known to fluctuate cyclically
in excitability which results in their "reorganization" by the stimulus
at the beginning of the visual process (at the beginning of stimula-
tion). It is this reorganization which is observed experientially as
the variable and somewhat erratic fluctuation.in.number of flashes and
intensity of these flashes.

(2) Out of the above conclusions it is suggested that, in line
with the theory, fusion in flicker occurs when.the pathways responding
to any cne pulsation become too thinly spread out in space on the cortex
and so little different in their time of response from those reporting
other pulses that the cortical conditions approach those of continuous
vision so that the perceptual result is one of continuity; This makes
fusion a cortical affair. It also suggests that continuous vision
is in.the nature of an ultimate extension of this principle of alterna—
tion-of-response of the visual pathways and that the cortex acts to
fuse this rapid alternation in the ordinary process of’seeing, perhaps
as- an inherent characteristic of its structure.

(3) In addition, the study indicates that these sorts of phenomena
are really subject to investigation, and that the attempts to demonstrate

relationships between perceptual and nervous events is not futile.

9.

10.

11.

12.

13.

REFERENCES

Bartley, S. H. Temporal and spatial summation of extrinsic
impulses with the intrinsic activity of the cortex. £,_§gll.
and Comp. Physiol., 1936, a, 41-62.

Bartley, S. H. Subjective flicker rate with relation to
critical flicker frequency. JI Exp, Psych,, 1938, 2g, 388-394.

Bartley, S. H. A central mechanism in brightness enhancement.
Proc. Soc. fixper, Biol, and Med., 1938, 3a, 535-536.

Bartley, S. H. Some effects of intermittent photic stimulation.
M, 1939, .25, 462-480.

Bartley, S. H. Vision, 3 Study 9; itg Basis. van Nostrand,
New York: 1941.

Bartley, S. H. Visual sensation and its dependence upon the
neurophysiology of the optic pathway. Biol,I Sympogia, 1942,
é, 87-106.

Bartley, S. H. The feature of the optic nerve discharge under-
lying recurrent vision. J. Exp, Psygh., 1942, 39, 125-135.

Bartley, S. H. and G. H° Bishop. Optic nerve response to
retinal stimulation in the rabbit. Prgg. Soc, Expgr, Biol, and
Midas, 1940: .44: 39'41-

Bartley, S. H., and F. Wilkinson. Brightness enhancement when
entoptic stray light is held constant. J, Psych,, 1952, 33,
301-305.

Bartley, S. H., and F. R. Wilkinson. Certain factors in pro-
ducing complexity of response to a single pulse of light. l;
m, 1953, 25, 299-3060

Basler, A. fiber die Verschmelzung vnn zwei nacheinander erfol—

genden Lichtreizen. Pflfiger'g Agchiv g, d. neg. Physigln 1912,
m, 245-251.

Bishop, G. H. Cyclic changes in excitability of the optic path-
way of the rabbit. Am. J, Physiol., 1933, 193, 213-224.

Crozier, W. J., and E. Wolf. Theory and measurement of visual
mechanisms: I and II. JI GenI Physigl., 1940—41, 24, 505—.

14.

15.

16.

17.

18.

19.

20.

Page 49

Crozier, W. J. , and E. Wolf. Theory and measurement of visual
mechanisms: IV. J Gen si , 1941-42, 2i, 89-109.

Hecht, S. Vision II, The nature of the photoreceptor process.

in A @999; of General Megmnta; Psychology, C. I‘Jm‘chison,
ed. Clark University Press, Worcester: 1931..

Heidbreder, E. 345123 W. Appleton-Century, New
York: 1933.
Land-is. c. a Ann—Land 24.12am at F_iiL_ker Elaine Elna-

m, ugh-ma. Armed Forces - Nat. Research Coun., Vision
Committee, Ann Arbor, 1953.

Wm. N- In W. mwamw.
2nd ed. Houghton Mifflin 00., Boston: 1951.

Never, E. G. , and C. W. Bray. The perception of low tones and
the resonance-volley theory. J , Pgmh" 1936, 3, 101-114.

Woodworth, R. 8., and H. Schlosberg. W W,
rev. ed. Holt, New York: 1951...

APPEND IX

Page 51

 

Observers
frag/J .S.H.B. JM. FR. W.

2 _‘A___ 4L—
} _A_A.A__ A— 44——
4 .m— _M_ L—
; _AI‘.L______.
é _AIIL_______. _AIAI._______
7 A—
8 .A‘II________
7 .4‘II'IIIIL___ .4llll________

1 0 L— _‘—_

1 1 L___L A—

1 Z ._AIIL_______

7 }

1 I. - ‘

4 5

7 4 .4‘..._......

4 7

1‘6 _1|_________

1 J

2 0’ ..AIII|.______

2 1

2.2

2 } F

 

Schematic representation of the perceptual response for three
different observers. 2
LDR: 1/36; Intensity: .015 c/ft ; Train length: 2.

 

flay/yer.

Schematic representation of the perceptual response for two

Page 52

 

Observers

J/‘Z 5/1144
1 .41“— .4111—
Z _AAJ.L_ 4“,}—
5 .444... 444—
4 .4344— '
2 4M— 4‘4
7 _A_A__ .411—
(3
7 JA—
70 ‘— _A‘__
‘77 ..AI‘AI_____.
42 ‘—
4} —4“—
14
1f .4—
16 fl
47 .4— L
’76
7? A—
20 ‘
Z7 ‘
22
23
at
a)” —4—-
24
1’7 - o

 

different observers.

LDR: 1/36;

Intensity: .015 c/ftz; Train length: 4.

Page 53

 

Observes
free/wee. R. R. ERW.
1 AF. .444—
2 _A_AA_ 44L—
} J4“— _‘¢‘__
4 441——
5
z, A _ L
7 A— .11—
e _L_
9 ‘—
1 0
1 1 A—
12! ._AII_________
1}
14
12" F

 

Schematic representation of the perceptual response for two

different observers.
LDR: 1/36;

Intensity: 1.5 c/ftz; Train length: 3.

Page 54

 

 

 

 

 

Observers
freqflec. B. P}. LAN. Raw.
1 ..JIKJ‘LJIL_____
2 #— _“‘ ...
4 44L— 4‘4—
)“ .4“— 44
4 z M—
7 2 .414—
5 LL—
7 4442.—
1a
14
12
7}
14
1;
16
17
1a
47
20
£1
22 I g

Schematic representation of the perceptual response for three
different observers. 2
LDR: 1/36; Intensity: .015 c/ft ; Train length: 3.

Page 55

~ Observers
freq/rec. .S.H.B. E R W.

.JIILAICILA|__.
._“lAL‘.J‘L___
..AI.LAI|______.

L.
..AIIIIIII|____

_iAII|________.

17' ..IIIII________
..AII|_________
.iAIIL_________

 

 

 

 

’ “ ‘ M f W E

 

Schematic representation of the perceptual reSponse for two

different observers.
LDR: 1/36; Intensity: 1.5 c/ftz; Train length: 4.

Page 56

 

 

Intemlty Zn c/fz‘Z
fire/r .001; .0024 1;
4 _‘_“.__
2 Li—
} ‘lg .?
4 ‘ 4 L J‘—
j’ - ‘—
7
<5
7 L
40 . _‘__ #—
4 1-
1 2 h
4 5
‘14 _AI-._____ iAIII-n____
4 f
4 4
4 7
45 _4—
47
20
Z1
21 _‘—___
7-}
21+
25
26
a7
39/ _A_
}f L—
:2:
42, ‘
(no fusion found)

Schematic representation of the perceptual reaponse for observer
at three different intensities.

S OHIB.

LDR: 1/36;

Train length: 2.

 

 

GIDSSARY

The following terms havetaeen used in this study in a somewhat
technical sense and are, therefore, defined here for the benefit of

the reader.

cff (critical flicker frequency): the frequency per second of
a pulsating light Just sufficient to make the light

appear steady.

enhancement: the increase in brightness of a flickering light
over that of a steady light of equal intensity that

occurs at frequencies centering around lO/sec.
frequency: number of pulses per second.
flash: see pulse.

fusion: the condition of a pulsating light such that it

appears steady.

pulse: one unit of stimulation, discrete in time; the stimw

ulus term corresponding to the sensation tenm "fladi".
train; a series of pulses.

train length: the number of pulses in a train.

 

 

a w e ... u. . .c . .. ....tullxw . er ,.
. . ex}. ...I .511: .. :1 I11!
.I

 

 

 

mum ust 0W

IRIS '58
Aug 5 §
{Li :3")
6 Jun 59.

 

;_.,..»:..:

 

1|tilWW[LlILWflfflfLUil‘leflllljlfliflHlfl“