University of Southern California Dissertations and Theses

Age Differences In Serial Reaction Time As A Function Of Stimulus Complexity Under Conditions Of Noise And Muscular Tension

(USC Thesis Other)

Age Differences In Serial Reaction Time As A Function Of Stimulus Complexity Under Conditions Of Noise And Muscular Tension

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been
microfilmed exactly as received
70-8527
JEFFREY, Dwight Wendell, 1940-
AGE DIFFERENCES IN SERIAL REACTION
TIME AS A FUNCTION OF STIMULUS
COMPLEXITY UNDER CONDITIONS OF NOISE
AND MUSCULAR TENSION.
University of Southern California, Ph.D., 1969
Psychology, experimental
University Microfilms, Inc., Ann Arbor, Michigan
AGE DIFFERENCES IN SERIAL REACTION TIME AS A FUNCTION
OF STIMULUS COMPLEXITY UNDER CONDITIONS
OF NOISE AND MUSCULAR TENSION
by
Dwight Wendell Jeffrey
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Psychology)
August 1969
UNIVERSITY O F SOUTHERN CALIFORNIA
TH E GRADUATE SC H O O L
U NIV ERSITY PARK
LOS A N G ELES. C A LIFO R N IA S 0 0 0 7
This dissertation, written by
Dwight Wendell Jeffrey
under the direction of Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Gradu
ate School, in partial fulfillment of require
ments for the degree of
D O C T O R O F P H I L O S O P H Y
Dean
Date„ August 1969
DISSERTATION COMMITTEE
Chairman
ACKNOWLEDGMENTS
This research was conducted while I was an N.I.C.H.D.
(HD00157-04) trainee in psychology at the Gerontology
Center at the University of Southern California. I would
like to express my appreciation to the Gerontology Center
and its training program for support in providing the
necessary laboratory facilities and equipment to conduct
this research.
I would like to express my appreciation to my guidance
committee. I would also like to thank my dissertation
committee for their valuable comments and suggestions in
the formulation and completion of this dissertation
research (Drs. J.E. Birren, N. Cliff, V. Garwood, and J.
Szafran). I owe a special debt of gratitude to my
chairman, Dr. James E. Birren for his assistance and
guidance during the course of my graduate studies at the
Gerontology Center and for providing equipment and
consultation for the dissertation research.
Acknowledgments must also be extended to the many
individuals who contributed to the organization of the
research effort. In particular I would like to thank
|Dr. V. Garwood for the loan of the noise generator and
audiometer, John Parnell for calibration of the noise
generator, Dr. R. Frommer for direction of maintenance of
ii
I __________________________________________________________________
the PSYCHOMET, Forrest Young for assistance in programming
and computer analysis, Dr. Tima Clare for help in obtaining
elderly subjects, and Dr. J. Szafran for critical evalua
tion of the dissertation manuscript.
Finally I wish to express my greatest debt to my wife,
Shirley, for her encouragement and sacrifices during my
graduate carreer. I would like to express appreciation to
her for many kinds of help generously and patiently given
and for her thoughtful and perceptive counsel. The price
a wife pays for a doctoral candidate's degree is high.
She must put up with his obsession, the pitfalls and
calamities, and the unpredictable ups and downs. It is
with great love and appreciation that I dedicate this
dissertation to my wife, Shirley.
iii
TABLE OF CONTENTS
Page
LIST OF TABLES..................................... V
LIST OF FIGURES............................... vi
Chapter
I. INTRODUCTION ....................... 1
II. DEVELOPMENT OF RESEARCH PROBLEM:
LITERATURE REVIEW .................. 7
III. EXPLORATORY INVESTIGATIONS OF
ACTIVATION........................ 29
IV. FORMULATION OF EXPERIMENTAL
HYPOTHESES........................ 50
V. M E T H O D ................................. 55
VI. RESULTS................................. 64
VII. DISCUSSION .............................76
VIII. SUMMARY AND CONCLUSIONS................92
REFERENCES ....................................... 96
APPENDICES ...................................... 106
iv
LIST OF TABLES
Table Page
I. Mean serial reaction time as a function
of task complexity and sensory conditions
for young and elderly subjects ............... 65
II. Median serial reaction time as a function
of task complexity and sensory conditions
for young and elderly subjects ............... 67
III. Mean serial reaction time as a function
of age, task complexity and presentation
order ........................................ 75
v
Figure
I.
II.
III.
IV.
TABLE OF FIGURES
Page
Mean serial reaction time for
pilot series A and B as a
function of task complexity
and sensory conditions for
young and elderly subjects........... 46
Mean serial reaction time as
a function of task complexity
and sensory conditions for
young and elderly subjects........... 69
Mean serial reaction time as
a function of presentation
order for young and elderly
subjects................................71
Mean serial reaction time
as a function of presentation
order and sensory conditions
for young and elderly subjects . . . 73
VI
1
CHAPTER I
INTRODUCTION
This research is concerned with testing certain
hypotheses about the role of the central nervous system in
the slowing of behavior that occurs with age. The research
design involves the comparison of serial reaction times of
young and elderly human subjects for five levels of task
complexity, under conditions of auditory white noise and
induced muscular tension. The data are discussed in rela
tion to a neural noise and an activation hypothesis of the
functioning of the aging nervous system.
Studies using measures of latency of response as a
dependent variable report a decrease in speed as age
advances past forty (Welford, 1959). This slowing of
behavior is readily observable in a wide variety of every
day situations as well as in the experimental laboratory.
The appearance of a general decline in response latency
with advancing age is widely reported in the study of the
aging organism and because it appears to be a consistent
age-related change is worthy of close examination. This
introduction will discuss age-related changes in response
latencies as they may reflect the state of the aging
2
nervous system. In Chapter II the experimental literature
concerned with behavioral slowing with age will be re
viewed and other variables will be considered as they
relate to response latencies. After these variables have
been examined, two major explanations for behavioral
slowing with age will be presented.
The time taken for nervous impulses, along the affer
ent and efferent nerves forms a real although small part
of a subject's total reaction time (Helmholtz, 1850; 1867;
Magladery et. a_l. , 1951) . If the velocity of nervous
impulses is reduced with age, this reduction could account
for part of the total increase in reaction times with age.
However, studies of conduction velocity have shown little
or no reduction in the speed of nerve conduction with age
(Birren, and Wall, 1956; Norris, Shock, and Wagman, 1953;
Sommer, 1941; Wagman and Lesse, 1952). This would suggest
that conduction velocity is not a large component of
changes in response speed and that the slowing of behavior
with age may be more readily explained in terms of reduc
tion in sensory input or changes in the central nervous
system.
A number of changes with age occur in the structure
and function of the peripheral sense organs, which, by
virtue of reducing the effective intensity of stimuli,
would tend to slow responses. The well-known limitations
3
on the sensory pathways, however, are not sufficient by
themselves to account for a major part of behavioral slow
ing with age (Wallace, 1956). The facts point to changes
in the central nervous system as the major factor in
increased latencies and to the increased time taken or
required there in mediating the input of stimulation and
output of behavior.
A number of indices of central nervous system function
demonstrate a lowering of nervous system activation with
increasing age. Indicators showing lower levels of acti
vation in elderly than young subjects have included EEG
(Obrist, 1954; Mundy-Castle, Hurst, Beerstecher, and
Prinsloo, 1954), heart rate (Malmo and Shagass, 1949), and
galvanic skin response (Botwinick and Kornetsky, 1960).
Studies of the relation of activation to the performance
of tasks involving time-stress have demonstrated that in
young subjects, increasing the level of activation using
induced muscular tension, ambient stimulation, and exercise,
serves to reduce reaction times, lower sensory thresholds,
and improve vigilance performance. The general interpre
tation of results of studies demonstrating that level of
activation affects performance has been that activity,
activation, and autonomic activity are related to a stream
of impulses from the reticular formation impinging on the
cortex, and that these impulses have a facilitating effect,
4
making the cortical cells more ready to fire. Thus,
sensory thresh.o_l.ds are lowered and action becomes quicker
and more vigorous (Morruzzi and Magoun, 1949; Lindsley,
Bowden, and Magoun, 1949; Fuster, 1958). It would be
expected that the older nervous system, while at a lower
level of activity than the young nervous system, would be
affected in a similar fashion by increasing the level of
activation of the central nervous system.
Much, if not all, experimental findings point to a
general limiting "speed factor" which develops with in
creasing age. This speed factor reflects primarily those
aspects of behavior which are mediated by the central
nervous system. To investigate further this factor,
Birren, Riegel, and Morrison (1962) conducted an experiment
in which stimulus complexity was varied while holding
constant the response task of quickness of button pressing.
Two subject groups were used, 23 older subjects (ages 60
to 80 years) and 30 younger subjects (18 to 30 years).
Under all stimulus conditions used, the older subjects were
slower than the younger subjects. This consistency could
have been predicted from previous research. However,
when response latencies to the various stimuli were inter
correlated for each group, it was found that the data were
more highly intercorrelated among the older subjects, indi
cating that elderly subjects respond with characteristic
5
slowness. In the younger group a single "speed" factor
accounted for 29 per cent of the variance as compared with
43 per cent in the older group. Thus, the young subjects'
reaction would appear to be more task-specific. How
quickly young people respond in simple tasks has only a
small relation to how quickly they respond when more
complex associative responses are required.
Specific problem area defined
The purpose of this study is to test the explanatory
utility of an activation or a neural noise hypothesis of
the functioning of the aging central nervous system. Speed
of responding under conditions designed to increase "acti
vation" will be compared to speed of responding under a
control or "normal" state of central nervous system activa
tion. The effect of increased activation versus a control
level of activation will be compared for young and elderly
subjects. Differences in mean serial reaction times for
young and elderly subjects will be employed as measures of
differences in the speed of behavior with age. Once these
differences are established, the relationship of varying
levels of task complexity to age can be evaluated. If
there is an interaction effect of age, task complexity, and
I
central nervous system activation on serial reaction times,
it will be analyzed within the framework of the concepts of
activation and neural noise.
6
In order to place empirical relations of psychomotor
speed, age, and the central nervous system in perspective,
in Chapter II, a systematic review has been made of those
significant topics that appear to have relevance to the
research problem.
7
CHAPTER II
DEVELOPMENT OF RESEARCH PROBLEM: LITERATURE REVIEW
Simple reaction time
Simple reaction time is defined as the time between
the onset of a signal and the beginning of a responding
movement. The time required for the central processes
that take place is dependent upon the nature of the stimuli
and the required response. With greater complexity of
either of these factors the length of the reaction time
is increased, the usual interpretation being that the
central processes are correspondingly more complex. Many
factors have been shown to affect the length of the reac
tion time. Some of these factors are: the magnitude of
the stimuli, the summation effect of two or more low-
intensity stimuli presented simultaneously, the subject's
set or expectation, the effect of practice, the number
of alternative stimuli or responses, the sense organ
stimulated, neuromuscular response time, and organic
factors such as body temperature or pulse rate (Woodworth,
1938). In this discussion interest is directed toward
the effect of age on reaction time, and age would be
another example of an organic factor.
8
Simple reaction time appears to reach its minimum at
approximately the age of eighteen and progressively
lengthens with age from about forty years (Koga and Morant,
1923) . A number of studies of age and simple reaction time
have measured reactions to both auditory and visual stimuli
(Beilis, 1933; Cesa-Bianchi, 1955; Koga and Morant, 1923;
Miles, 1931). These studies indicate that auditory
reaction times are slightly faster than visual reaction
times at all ages, although both slow with age.
There appears to be no evidence of a differential
decline with age for reaction times to auditory and visual
stimuli. This suggests that the decline in reaction time
with age is not due to changes in peripheral mechanisms.
In one of the studies of auditory and visual reaction time
mentioned above (Koga and Morant, 1923), measurements were
also made of auditory and visual sensitivity. In examin
ing these data the question arises as to whether decline
in reaction time to signals in these two sensory modes
results from a deterioration of sensory acuity or changes
at a higher level of the central nervous system. The
investigators found that correlations were higher between
auditory and visual reaction times in their subjects than
between auditory and visual sensitivity. Their evidence
suggests that the slowness in reaction time is not directly
determined by decrements in the sensitivity of sense
organs, but that the basis for the slowing of response
speed with age is more a function of common processes with
in the central nervous system, rather than of the specific
peripheral sense organs. i
Individual differences in reaction times increase
I
markedly with age, both in terms of variability between
subjects within an age group and among individual responses
given by the same subject (Obrist, 1953). A possible
reason for greater individual differences among older sub- j
/
jects is that individuals' functional capacities decline atj
different rates, so that the population of persons of a
given chronological age will necessarily include people of
widely varying functional capacities. It should be empha
sized that behavioral slowing is not a simple chronological
function. Talland (1965) compared the variability of
simple reaction times in three different studies. Reaction
times in both old (ages 77 to 89) and young (ages not spec
ified) groups varied to a significantly larger extent
between than within the eighteen to twenty subjects per
group. The old and young groups differed significantly in
their simple reaction times, and in the three studies, 20,
22, and 26 per cent of the older subjects' mean scores
were at or below the group mean of the young subjects.
Two of the elderly subjects also had individual mean reac
tion times shorter than the younger group, but a third of
10
them did not score below that level once in 20 or 40 trials.
These results indicate that although generally, older sub
jects respond more slowly than their younger counterparts, j
the mean response latencies of some older subjects will be j
very similar to those for the younger subjects. j
!
Speed and complexity
As we move from simple reaction time to more complex
responses demanding choice and discrimination, one finds
that these more complex responses require more time.
Goldfarb in 1941 studied simple and complex reaction in
five age groups of male subjects. The youngest subjects
were aged eighteen to 24 and the oldest 55 to 64 years of
age. Goldfarb found an increase in simple reaction time
from the youngest to the oldest group of .011 seconds.
The difference for a two-choice reaction time task was
.057 seconds; and for a five-choice reaction task, .066
seconds. In all age groups, the choice reaction times were
longer than the simple reaction times. If the slowing of
reactions reflected only the execution of the motor re
sponse, it would be expected that elderly and young sub
jects would respond at more nearly the same speed in the
more complex situations. These results do not indicate
that this is so, and whatever central factors are involved
in simple reaction time appear to assume greater importance
as the task becomes more complex.
11
In more complex tasks speed of responding may be
broken into at least two parts. The first of these is the
motor response.;; the second part involves the cognitive or
associative processes required in choosing the correct of
two or more responses available. In general, age differ
ences in speed appear to increase as the task requires more
consideration of the stimulus before a response is given.
In the Birren, Riegel, and Morrison (1962) study, this
second factor was examined by measuring response time to
various kinds and complexities of stimuli. The subjects
had to respond to stimuli which required choice and dis
crimination. The age differences in response time were
greater for the more complex stimuli, and the increase in
time required for the more complex stimuli was proportional
to the time required for the simple stimuli.
This review of the literature indicates that slowing
with age is a well-established fact and is characteristic
of many different aspects of behavior. In particular, the
evidence indicates that all behavior mediated by the
central nervous system tends to be slow in the aging organ
ism. Of considerable psychological significance is the
implication of increased slowing of all voluntary responses.
The older organism appears to require additional time to
integrate information from the outside world and to initi
ate the appropriate activity. This is particularly true in
12
complex response situations or when the stimulus is unclear
or ambiguous. The more intricate the task, and the more
judgment demanded, the greater is the slowing of any re
sponse in the older organism.
Central mechanisms and speed of behavior
A great variety of anatomical changes in the healthy
nervous system with advancing age have been described. One
of the best documented morphological changes occurring in
aging is the decrease in the size and number of nerve cells
(Brody, 19 55) . Other changes which might be involved are
physical-chemical changes at the synapses that limit
transmission speed, and a lowering of subliminal excitation
of the nervous system resulting from changes in subcortical
centers (Birren, 1965). It is not clear, however, whether
behavioral slowing with age is a result of general and
"normal" changes in the biological properties of the
nervous system or is the cumulative result of the effects
of disease. However, in the diseased organism, central
nervous system changes contributing to slowing of behavior
are known to occur. Studies have demonstrated that speed
of reaction reliably discriminates between brain damaged
and non-brain damaged individuals (King, 1965; Talland,
1965) , cardiovascular from non-cardiovascular patients
(Simonson, 1965; Spieth, 1965; Szafran, 1965), between
schizophrenics and normals (Talland, 1965; Venables, 1965),
13
and between extremely healthy elderly subjects and randomly
selected elderly subjects (Birren et. al., 1963).
Muscular tension
Muscular tension has been shown to affect measures of
functioning of the central nervous system and speed of
behavior. When muscular tension is voluntarily induced by
the instructions or by the apparatus used, reaction times
are affected (Davis, 1940; Freeman, 1933; Freeman and . .
Kendall, 19 40). Freeman and Kendall (19 40) studied the
effect of muscular tension and preparatory intervals on
reaction time. Their findings indicated that both heavy
and light tension induced at certain intervals prior to
stimuli presentation significantly shortened reaction times.
Muscular tension has been the dependent variable in
learning experiments more often than in studies of reaction
time. Although learning per se is not involved in the
research to be reported, it is of some interest because
learning studies have more thoroughly investigated muscular
tension as a variable. In a study by Courts (19 39) sub
jects were required to squeeze a hand dynamometer while
learning three-letter nonsense syllables. Prior to the
learning trials each subject squeezed the dynamometer as
hard as he could for 30 seconds. The reading at 30 seconds
was used as the subject's maximum grip. Lists of syllables
were memorized under seven conditions, while the subject
14
was squeezing the dynamometer at 0, 1/8, 1/4, 3/8, 1/2,
5/8, and 3/4 of his maximum grip. The results indicated
that learning increased with the amount of tension, from j
a mean retention score of 55 with no tension, up to a mean
score of 6 4 with 1/4 maximum grip. Greater tensions were j
accompanied by reduced learning until at 3/4 maximum grip
the mean score was 51, which was below that recorded with
i
0 or no tension. j
Another experimenter (Bills, 1927) conducted a series
of four experiments to assess the effect of muscular
tension on learning, and on practice and fatigue. During
learning, subjects squeezed two hand dynamometers. The
tension condition required the subject "to squeeze an j
amount of pressure that could be maintained without annoy
ance." The tasks involved learning nonsense syllables and
paired associates, adding columns of 20 figures, and read
ing letters on cards. The results indicated that the mus
cular tension used increased the efficiency of these tasks
and that the amount differed with individuals. The effect
of practice was to enhance increased efficiency under
tension on tasks emphasizing speed. The results with
fatigue are interesting in terms of potential age differ
ences. The added efficiency under tension tended to in
crease as the subject grew more fatigued, particularly on
tasks involving time-stress. Older subjects and younger
" 15 '
subjects may be differentially affected by fatigue with
corresponding changes in their performance. However, the
subjects used in these studies were all young adults and !
no data are available for older subjects.
Theoretical consideration of the aging central nervous
system
There are two major current explanations for slowing
with age of behavior mediated by the central nervous system:
One is the neural noise hypothesis, and the other is the j
activation hypothesis. Slowing of response time with
increasing age has been theorized by Crossman and Szafran
(1956) to be the result of an increasing level of internal
random "neural noise" in the brain, which tends to reduce i
the signal-to-noise ratio in the central nervous system.
The theory assumes the central nervous system to be anal
ogous to a communication system with a noise level that
increases with age. This hypothesis of an increase in
neural noise with age is not a development from neuro
physiology but employs the statistical theory of visual and
auditory signal detection as developed by Tanner and Swets
(1954). In physiological terms, according to Crossman and
Szafran, the higher noise level might be the result of an
increased rate of spontaneous firings of neurons, or an
increased likelihood for neighboring neurons to excite
randomly one another. In terms of the general properties
16
of communication systems, the increasing level of neural
noise would be reflected in a reduction of the signal-to-
noise ratio in the brain. In fact, the crucial character
istics of the neural noise hypothesis have yet to be
formalized experimentally and tested.
Proponents of the neural noise explanation of age
changes in performance (Crossman and Szafran, 1956;
Gregory, 1955; 1957; Szafran, 1965; 1968) hypothesize that
slowing of responses with age is one of the results of an
increased level of internal random discharges, prior to
any sensory input, which creates an increase in the level
of background noise in the older central nervous system or
as in the case of audition a reduction in the strength of
a signal input. Either an increase in background noise or
a decrease in signal strength would have the effect of
lowering the signal-to-noise rafio. Given a lower signal-
to-noise ratio the older brain must accumulate data over
an appreciable interval of time, to distinguish the signal
from background noise present in the system, and this
period of time will become longer as the ratio of signal
strength to background noise becomes smaller. At the
present time, the validity or usefulness of a neural noise
explanation of the aging nervous system is not clear.
Adequate supporting experimentation has not been reported,
and the concept has not been stated in operational terms.
17
In terms of behavior, the neural noise hypothesis
implies that conditions which would lower the signal-to-
noise ratio in the nervous system would tend to result in
a slowing of performance. If the level of internal noise
in the system can be raised by externally produced noise
this explanation would predict that response speed would
be slower than in the absence of such additional stimula
tion. For instance, it would be expected that the older
nervous system, postulated to be more "noisy", when
presented with additional stimulation, such as auditory
white noise or induced muscular tension, would have the
signal-to-noise ratio lowered still further with corre
sponding slowing of response speed. According to Szafran
(1968), in the auditory system, hearing loss in the older
individual may be due to an increased level of random noise
associated with damage to the hair cells. This would
result in a lower signal-to-noise ratio and slower response
speed to an auditory stimulus. If the older central nerv
ous system has a lower signal-to-noise ratio than the
young nervous system, then, according to the neural noise
hypothesis, lowering the signal-to-noise ratio experimenr-
tally in a young nervous system would make the young nerv
ous system functionally more similar to the older nervous
system. Both the young subject and the elderly subject
would under these conditions be expected to respond more
18
i slowly in a simple reaction time task than under control
I conditions.
j By contrast, the activation hypothesis of the aging
central nervous system would predict the opposite result.
!An activation hypothesis would assume that the central
:nervous system, through loss of functional neural cells
!with age (Brody, 1955; Critchley, 1942) and associated
!
i sensory input, is at a lower level of activation than the
!younger nervous system. A reduction in the number of
;functional nerve cells in the brain is believed to make
;the older nervous system less sensitive and less responsive
than it otherwise would be. Being less sensitive and less
|activated by any signal, the older nervous system's ability
jto respond quickly to an incoming signal is reduced by
'comparison with that of the younger, more aroused system
!(Birren, 1960). Given a signal, the older nervous system
!mast.integrate data over a longer period of time than the
;younger system, and consequently the older nervous system
jresponds with increased latency (Welford, 1959).
| States of lowered activation in the nervous system
'have not been systematically manipulated. Experimental
|
I studies using vigilance tasks have demonstrated that varied
i
(environmental stimulation such as music, can enhance visual
(vigilance performance, and that similarly, auditory vigi-
ilance is improved when subjects are allowed to look at
19
pictures (McGrath, 1960; McGrath and Hutcher, 1961). In
another area of investigation, as mentioned earlier, Bills
(1927), Courts (1939) , and Freeman (1931; 1933) have found
that performance in learning tasks is improved if tension,
by the use of a handgrip, is induced in irrelevant muscles
of the body. Additional evidence has been reported by
Pinneo (1961) in a tracking task experiment in which five
increasing levels of induced muscular tension by handgrip
were found to produce corresponding increments in the
level of central nervous system activation as measured by
palmar conductance, heart rate, respiration rate, frontal
and occipital EEGs and EMGs from active and passive
muscles. In another study (Sjoberg, 1968) the level of
physical work on a bicycle ergometer was examined in rela
tion to heart rate and performance on a choice reaction
task. Using heart rate as an index of activation, mean
choice reaction times were found to have an inverted
U-shaped relationship to the five levels of activation.
That is, mean choice reaction times were lowest at a
medium level of activation and progressively longer with
both increasing and decreasing levels of activation.
The activation explanation of behavioral slowing with
age would predict that response speed should be faster in
the elderly subject, if in some way the level of activation
of the nervous system could be raised. For instance, it
20
would be expected that the older nervous system, postulated
to be less activated, when presented with additional stimu
lation such as auditory white noise or induced muscular
tension, would have its level of activation increased and
consequently would show faster reaction times to a given
signal. If the older central nervous system tends to be a
less activated system than the younger nervous system, then,
according to the activation hypothesis, raising the level
of activation in the older nervous system experimentally
would make the old nervous system functionally more similar
to the young nervous system. Both young and elderly sub
jects under conditions of experimentally induced high
levels of activation would be expected to respond more
quickly in a reaction time task than under control condi
tions .
It is the purpose of this study to secure experimental
data to test the activation hypothesis and the neural noise
hypothesis of the functioning of the aging central nervous
system.
Historical background of the concepts of activation and
neural noise
The nature of the questions asked and of the approach
taken in the research to be presented can best be under
stood after a brief consideration of the historical back
ground of the concepts of activation and neural noise.
21
Although these two concepts are both being used here as
possible explanations for age changes in the functioning
of the central nervous system, the concepts have been j
derived from different sources. The concept of neural j
l
noise has its origin in Shannon's communication theory
(Shannon, 1948) and the decision making theory of visual
detection developed by Tanner and Swets (1954). The con
cept of activation has its origin in studies of the rela
tionship between emotion and physiological indices (Duffy, i
1957, 1962; Hebb, 1955; Lindsley, 1951, 1957; Malmo, 1959),
and studies of the function of the ascending reticular ac
tivating system of the central nervous system (Morruzzi
and Magoun, 1949; Lindsley, Bowden, and Magoun, 1949).
The concept of neural noise can be historically traced
to its origin in early work by physical scientists con
cerned with the concept of noise in communication engineer
ing. In particular, Shannon and Weaver (1949) developed
the notions of noise, and of signal-to-noise ratio as a
framework for the analysis and evaluation of communication
systems. The first use of the concept of noise in the
study of aging was by Birren, Allen, and Landau (1954) in
a study of addition speed and accuracy in relation to age.
They alluded to the possibility of using noise as an expla
nation for declining performance with age in solving addi
tion problems.
22
The concept of neural noise originated in studies of
sensory threshold determination, and signal detection
theory. In studies of sensory threshold, the slight vari
ation of sensory responses near threshold suggests that
sensitivity of sensory systems is limited by the presence
of moment to moment changes in neural excitability. These
moment to moment changes in neural excitability have been
!
described as random energy fluctions or noise (Fitzhugh, j
1957). In the measurement of sensory thresholds, the task
of the observer is to detect the presence of a signal
against a background of such random energy fluctuations or
noise.
I
Signal detection theory as presented by Tanner and |
i
Swets (1954) postulates that, at any moment, the observer j
faced with need for making a decision concerning the sum
total of neural excitation, establishes a criterion value
to distinguish whether at a given instant, the level of
neural excitation is sufficient to give rise to the decision
of "signal present" or is below that criterion level and
justifies only the decision of "signal absent". Eor any
criterion value the observer will score a proportion of
correct detections of the signal and a proportion of in
correct detections or false alarms. If an observer adopts
a less cautious criterion level, he can improve his
correct detection score but will also make more false
alarm decisions. The criterion established by the
23
observer, as reflected in the proportion of correct
detections of signals relative to false alarms, provides
an indication of the observer's degree of caution with
regard to the amount of total neural excitation he re
quires before he will make the decision of "signal
present." An early use of the concepts of signal detec
tion theory and noise was in a study of absolute sensory
thresholds, which established support for the notion that
noise in the optic pathway is a limiting factor in the
sensitivity of the eye to light (Barlow, 1956). Noise in
this case was used to explain the variability of absorp
tion of quanta from a light flash in determining the
value of the absolute threshold. In a number of follow-
up experiments, Barlow (1957) and Barlow, Fitzhugh, and
Kuffer (1957a, 1957b) attempted to determine values of
retinal noise at visual threshold in single isolated
ganglion cells in the cat, using signal detection theory
concepts.
The first use of the concept of neurological noise as
an explanatory concept for some of the behavioral changes
associated with aging was by Crossman and Szafran (1956)
and by Gregory (1957). Support for the concept that neural
noise may increase with age is reported in a study of the
speed of discrimination by Crossman and Szafran (1956).
Their finding was that "easy discriminations, where the
24
ratio between the stimulus alternatives approached zero,
were more affected by age than more difficult discrimina
tions." According to Crossman and Szafran, "if...a random
disturbance is added to all signals before discrimination...
the signal-to-noise ratio would be little altered if nearly
unity, but much altered if near zero." They postulate that
an increasing level of "internal random noise in the brain
tends to obscure differences between signals" and results in
a lower signal-to-noise ratio with age. In a later study of
young and old pilots, Szafran (1965) used the concept of
neural noise to explain age differences in decision
processes in auditory perception under reduced signal-to-
noise ratio conditions. According to Gregory (1957) , age-
related slowing of behavior may be explained in terms of
increased noise in the form of random firing of neurons in
the central nervous system. Gregory postulates that the
absolute threshold describes the case where signals must be
i
detected against a background of residual neural noise,
whereas differential threshold is the case involving detec
tion of signals against neural noise plus other noise which
is added-’ externally. Gregory suggests that if the neural
noise level increases with age, sensory thresholds would
rise and acuity would deteriorate. A general slowing down
of central processes in the older nervous system might be an
25
adaptation which tends to increase the signal-to-noise
ratio in the nervous system which serves to improve detec
tion and sensitivity. Gregory attempted to test the
hypothesis that nerve deafness associated with aging may be
due to raised neural noise. He compared patients with nerve
deafness in whom higher noise levels might be postulated,
with the effect of experimentally raising the noise level
in young healthy subjects. This was accomplished by compar
ing differential intensity discrimination of short tone
pips against a continuous tone of the same frequency over a
wide range of frequencies and intensities of noise. He
found the curves relating intensity of the tone pip to the
continuous tone similar for the nerve deaf patients and the
subjects exposed to certain levels of added masking noise.
He concluded that the hypothesis that nerve deafness may be
caused by raised neurological noise was supported by these
data.
Neural noise has been used as an explanatory concept
to discuss the value of sensory thresholds, but with the
exception of Gregory's study of differential thresholds
(Gregory, 1947), attempts have not been made to manipulate
levels of neural noise experimentally. The validity of
using the concept of neural noise to explain slowing of
behavior with age also has not been tested experimentally at
26
the present time.
| The term activation can be historically traced to
Lindsley's use of the term in his "activation theory of
i
|emotion" (Lindsley, 1951). Although stated in the context
;of emotion, Lindsley intended the activation theory to be
much broader than an explanatory concept for emotional
behavior. The concept of activation was an outgrowth of
neurophysiology on one hand, and behavioral "energetics" on
;the other. Neurophysiological research with EEG has
I provided evidence that the primary role of the reticular
!formation is that of maintenance of wakefulness by means of
j activation of the cortex. Different degrees of excitation
are reflected in the spectrum of states of consciousness
from deep sleep to full alertness, as well as in the
accompanying changes in the activation pattern of the EEG.
Excitation of the reticular formation induces general
activity of the cortex, presumably by facilitating its
receptivity to sensory stimulation.
In an early study relating function of the reticular
I
j formation to activity in the cortex, direct electrical
j stimulation of the reticular formation in the cat produced
the activation pattern of the EEG in the cortex (Moruzzi
| and Magoun, 1949). Further evidence supporting the
I
j activating role of the reticular formation was presented
27
in a study in which destruction of the rostral end of the
reticular formation abolished activation of the EEG in the
cortex (Lindsley, Bowden, and Magoun, 1949) . It is of
interest to consider the relation between activation of the
cortex and behavior. Fuster (1958) found that concurrent
electrical stimulation of moderate intensity of the
reticular formation improved accuracy and speed of visual
discrimination response in monkeys. Fuster interpreted the
reticular facilitation to affect primarily "central inte
grative time" rather than the peripheral transmission time.
In addition, Fuster found that higher intensities of
stimulation had the opposite effect, creating fewer correct
responses and slower reaction times. These deleterious
effects from high intensity stimulation are of particular
interest because they might be predicted from the activa
tion theory. According to the activation theory, as
idiscussed and elaborated by Duffy (1957, 1962), Hebb
(1955), Lindsley (1951, 1957), and Malmo (1959), level of
activation extends along a continuum from deep sleep with
low activation to high excitement with high activation.
Activation is largely a function of cortical bombardment by
the reticular formation, and it is postulated that the
greater the cortical bombardment, the greater the activa
tion. In addition, the relation between activation and
28
behavior or performance is described as a U shaped function.
Performance, such as discrimination reaction time, is
facilitated more by a moderate level of activation than by
a lower or higher level of activation. For example, in the
-study of the relationship between learning of nonsense
syllables and muscular tension referred to earlier, Courts
(1939) found that learning increased with muscular tension
;up to 1/4 of maximum handgrip; greater tensions were
accompanied by reduced learning until at 3/4 of maximum
|handgrip, learning was below normal, and this level of ten
sion actually interfered with learning.
The concept of activation has not been applied as an
I
explanation of behavioral slowing with age in a systematic
manner. Various indices of central nervous system activa
tion such as EEG, however, have demonstrated lower levels
with increasing age. In a review of behavioral theories
of aging, Birren (1960) suggested that changes in the speed
of behavior of the older individual may be a consequence of
the lower level of activation of the nervous system.
|However, at the present time, the concepts of activation
,and changes in level of activation with age have not been
i
!tested' experimentally to establish possible relationships
;between age, level of activation, and performance.
29
CHAPTER III
EXPLORATORY INVESTIGATIONS OF ACTIVATION
PILOT STUDY A
In this, a preliminary study, auditory white noise and
induced muscular tension were selected as variables to be
investigated. The selection of an appropriate level of
induced muscular tension was based on an examination of the
available experimental literature. With no comparable
literature available to describe the effects of various
intensities of auditory white noise on the speed of re
sponding, this preliminary investigation was designed to
examine the variable of intensity of auditory white noise
and its effect on serial reaction time. In addition, three
levels of task complexity and three levels of irrelevant
information were investigated in relation to serial reac
tion times under conditions of auditory white noise,
induced muscular tension, and a control condition without
noise or tension.
METHOD
Apparatus
Subjects were tested on the PSYCHOMET apparatus
(Birren, Riegel, and Morrison, 1962; Rabbitt and Birren,
2 9
30
1967). The apparatus consisted of a control panel on a
table at which the subject sat. A horizontal row of ten
circular apertures, each one inch in diameter with their
perimeters one half inch apart, were inset across the upper
half of the panel. An identical row of ten circular
apertures was inset across the lower half of the panel. A
square contact response-grid, measuring one inch was mounted
six inches vertically beneath each upper circular stimulus
aperture. Contact-grid switches were closed by the sub
ject's skin conductance when any grid was lightly touched.
Stimulus lights located behind the top row of apertures
were switched on one at a time in sequences of any desired
order or length by programming a Rheem photoelectric tape
reader. Lights were also located behind the bottom row of
apertures, and these lights all remained on during a
stimulus-response sequence. Any stimulus (upper) light
could be paired with any response (lower) light and corre
sponding response-grid. The ten stimulus lights could be
presented with any predetermined order. Plexiglass panels
cr masks were mounted behind the apertures, in front of the
lights. Numerals placed on the plexiglass masks were visi
ble whenever the lights were on. The program tape specified
which contact grid must be touched in order to switch off
any stimulus light. On a correct response, the next
stimulus light designated in the program came on within 20
milliseconds. Incorrect responses were timed and recorded
31
but resulted in no change in the display. Intervals
between successive responses were timed to within 0.01
seconds by a version of SETAR (Welford, 1952), which
punched for each response a five-character code on paper
tape, giving the response time, identifying the grid
touched, and recording whether the response was correct or
incorrect.
A Smedly hand dynamometer calibrated in kilograms was
used to measure the relative strength of handgrip of the
subjects. Muscular tension was induced using weights sus
pended by rope over pulleys attached to the far side of the
table at which the subjects sat. The amount of tension
could be accurately varied by combining various weights
attached to the rope. The subject gripped a handle in his
non-preferred hand, lifted the weights off the floor, and
supported them by maintaining constant handgrip and arm
position. The subject's arm was then placed in an arm rest
attached to the table. The arm rest could be varied to
suit different size subjects. The handle was held so that
the pull of the weight was on the inside of the fingers
with the hand closed.
The noise generator used, a Western Electro-Acoustic,
produced a mixture of all frequencies in equal amounts from
100 to 7,000 cycles per second, up to a maximum level of 85
decibels. Auditory white noise was presented binaurally
through a set of Telephonic TDH-39-10z earphones with
32
MX41/AR cushions.
Subjects
Two women and four men were tested. Their age range
was from 22 to ->30 years. All subjects had completed four
years of college, and were free from known chronic illness. }
I
Experimental design j
i
The basic study was a three by three factorial design. I
One independent variable was task complexity using three
levels of complexity. The other independent variable was
the amount of irrelevant information, for which three levels
of irrelevant stimulus elements were used. A total of nine
stimulus-response sequences, factorally incorporating these |
I
three levels of complexity, and three levels of irrelevant j
information were presented under each of the three condi- |
tions of sensory stimulation. These three conditions of
sensory stimulation, included auditory white noise, induced
muscular tension, and a control condition without noise or
tension. The levels of white noise used for all subjects
were 50, 65, 75, and 85 decibels (sound pressure level) as
calibrated at the earphones using a Bruel Kjaer artificial
ear with a six cubic centimeter coupling cavity and a one
inch condenser microphone. The level of muscular tension
was determined individually for each subject, and was equal
to 20 per cent of maximum handgrip as registered on the hand
iynamometer at the end of a 30 second test.
33
The three levels of task complexity and three levels
of irrelevant stimulus information were a function of the
size, color, and arrangement of numerals on the plexiglass
masks. There were eleven plexiglass masks, one for each of
the nine stimulus-response sequences and two masks for the
training trials. The masks are described below.
Training mask, simple movement time
The subject touched the grid below the response lights
as quickly as he could from left to right, and then right to
i left, turning off the stimulus lights in numerical order,
one through zero, and then zero back to one. Buttons turn
ed off stimulus lights with corresponding numerals. Numbers;
below designate numerals on the plexiglass masks.
Stimulus lights 1 2 3 4 5 6 7 8 9 0
Response lights 1 2 3 4 5 6 7 8 9 0
Training mask, random movement time
Stimulus lights and response lights and buttons were
regularly paired top and bottom with numerical designations
as below, but presented in random order, i.e., stimulus
light behind numeral nine presented first, numeral three
Jsecond, four third, etc. Buttons turned off stimulus lights
!with corresponding numerals.
Stimulus lights 9 1 8 5 7 4 3 6 0 2
Response lights 9 1 8 5 7 4 3 6 0 2
The experimental masks factorially represented task
complexity and irrelevant information as indicated in the
design presented below.
COMPLEXITY
IRRELEVANT
INFORMATION
Level I Level II Level III
Level I Mask #1 Mask #2 Mask #3
Level II Mask #4 Mask #5 Mask #6
Level III Mask #7 Mask #8 Mask #9
As indicated above, masks three, six, and nine were
more complex than masks one, four, and seven, and masks
seven, eight, and nine contained more irrelevant information
than masks one, two, and three. Masks one, two, and three,
and masks four and seven will be described below. Masks
five, six, eight, and nine are factorially representative of
masks two, three, four, and seven, and will not be described
further.
Mask #1, task complexity level 1^
Stimulus numerals were in random order and stimulus
lights came on in random order, i.e., numeral six first,
three second, seven third, etc. Response numerals desig
nating response buttons were in regular order, one to zero,
from left to right. Irrelevant information level 1^. All
numerals were black arid .of equal size. Buttons turned off
lights with corresponding numerals above them.
35
Stimulus lights 6 1 8 3 9 4 0 5 7 2
Response lights 1 2 3 4 5 6 7 8 9 0
Mask #2, task complexity level II
Stimulus numerals were in regular order from left to
right, and stimulus lights came on in random order.
Response numerals designating response buttons were in
random order. Irrelevant information level ]E. All
numerals were black and of equal size. Buttons turned off
stimulus lights with corresponding numerals.
Stimulus lights 1 2 3 4 5 6 7 8 9 0
Response lights 5 7 0 6 3 9 1 4 2 8
Mask #3, task complexity level III
Stimulus numerals were in random order, and stimulus
lights came on in a different random order. Response
numerals designating response buttons were in random order.
Irrelevant information level I. All numerals were black
and of equal size. Buttons turned off stimulus lights with
corresponding numerals.
Stimulus lights 9 8 6 5 1 7 4 3 0 2
Response lights 8 5 2 7 3 9 0 6 1 4
Mask #4, task complexity level 1^
Stimulus numerals and lights came on in different
random orders as for Mask #1. Response numerals designat
ing response buttons were in regular order with numerals
one to zero, from left to right. Irrelevant information
36
level II. Stimulus and response numerals were black and
red and were of equal size. Buttons turned off stimulus
lights with corresponding numerals. Stimulus numerals
seven, six, nine, eight, and zero were red. Response nu
merals one, three, five, six, eight, and zero were red.
Stimulus lights 7 6 9 3 4 8 1 0 5 2
Response lights 1 2 3 4 5 6 7 8 9 0
Mask #7, task complexity level 1^
Stimulus numerals and lights came on in different
random orders as for Mask #1. Response numerals designat
ing response buttons were in regular order with numerals
one to zero, from left to right. Irrelevant information
level III. Stimulus and response numerals were black and
red, and large and small. Stimulus numerals :five, zero,
eight, nine and three were red. Response numerals two,
four,, five,: seven, and zero were red.
Stim ulus l i g h t s 2 5 0 | 8 4 9 3 6 7
Response l i g h t s 1 2 3 4 5 6 7 8 90
Procedure
Prior to experimental testing each subject's maximum
handgrip was measured in order that the amount of weight
used to induce muscular tension during the tension condi
tion should be proportional to the strength of the individ
ual. With his non-preferred arm on an arm rest, the sub
ject was instructed to exert maximum grip with the non
preferred hand on the hand dynamometer for 30 seconds. The
37
dial of the dynamometer was facing away from the subject
during this test, so he was not able to observe the reading.
The strength of grip at ther.end of 30 seconds was taken as
maximum grip. The amount of weight used to induce muscular
tension during the tension condition was 20 per cent of a
subject's maximum grip.
The training sequences were presented, and then each
subject was tested on all nine experimental stimulus
response sequences under each of the six conditions in two
one-hour experimental sessions. A stimulus-response se
quence consisted of 20 stimulus-response pairings. A five
minute rest period followed the presentation of each condi
tion. The three stimulus-response sequences programmed for
each condition were each constructed according to a balanced
schedule insuring that each of the possible ten pairings of
stimulus lights and response lights with corresponding
buttons was presented two times. There were three prear
ranged orders of presentation for the 20 stimulus-response
sequences, and each subject was exposed to each of the three
presentation orders. The three stimulus-response presenta
tion orders were used once within each condition.
Each stimulus-response sequence was begun by the sub
ject by pressing a start button located on the center of
the subject's console beneath the contact grids of the
PSYCHOMET. Pressing the start button caused the first
programmed stimulus light to come on, starting the
38
stimulus-response sequence. The same instructions were
given before all sequences; "press the button beneath the
numbered response light that corresponds to the numbered
stimulus light. Turn the lights off as quickly as you can.
When you are ready to begin, press the start button and turn
the lights off as quickly as you can."
During the auditory white noise condition, the noise
was turned off between stimulus-response sequences, although
the earphones remained on the subject. During the muscular
tension condition, the subjects were allowed to release the
weight between stimulus-response sequences. The noise was
turned on or the weight was lifted before the beginning of
the next stimulus-response sequence.
Results
All correct responses for each stimulus-response se
quence under each of the six conditions were used in the
statistical analysis. Means were computed for each
stimulus-response sequence, and comparisons were made using
a subject group mean of these means of the stimulus-response
sequences representing each of the nine combinations of task
complexity and irrelevant information.
To demonstrate the effects of the six conditions, all
of the means were combined to provide a representative
serial reaction time for each of the six sensory conditions.
The group mean in seconds for these conditions were:
39
Control Tension 50 db. 65 db. 75 db. 85 db.
1.12 sec. 1.05 sec. 1.10 sec. 0.98 sec. 0.99 sec. 0.98 sec.
As can be seen from the mean values, under all condi
tions of auditory white noise shorter group mean values
were observed than under the control condition, and the
muscular tension condition also yielded shorter mean values
than the control condition. On the basis of the mean
scores for the four levels of auditory white noise, 75
decibels was selected as the intensity level to be used in
subsequent studies. A higher level of intensity may appear
to be preferable from an examination of the means for the
four levels of auditory white noise. However, four of the
six subjects complained of the noise level at 85 decibels
as. being irritating, while no such comments were voiced at
the 75 decibel level. Since the differential effect of 75
decibels versus 85 decibels on mean values of serial reac
tion times appear to be minimal, and in order to avoid a
level of noise which might be irritating or stressful, 75
decibels was judged to be the best level of auditory white
loise to use in subsequent research.
To evaluate the effects of task complexity and irrele
vant information, the means of group stimulus-response se
quences were combined for an overall mean serial reaction
time for each of the nine sequences. These values were
then combined to provide a grand mean for each of the three
40
levels of task complexity, and for each of the three levels
of irrelevant information. The mean serial reaction times
computed in this way are presented below.
TASK COMPLEXITY
Level I Level II Level III Grand Mean
Level I 0.84 sec. 1.06 sec. 1.13 sec. 1.007 sec.
Level II 0.86 sec. 1.10 sec. 1.15 sec. 1.037 sec.
Level III 0.88 sec. 1.12 sec. 1.18 sec. 1.061 sec.
Grand
Mean
0.860 sec. 1.099 sec. 1.154 sec.
Serial reaction times increase with increasing task com
plexity and with increasing irrelevant information. The
grand means for the three levels of complexity and for the
three levels of irrelevant information also support this
finding.
The small relative differences between the overall
mean serial reaction times and between the grand mean serial
reaction times for the three levels of irrelevant informa
tion have indicated a small effect provided by this inde
pendent variable. For this reason, irrelevant information
was not included as an independent variable in the subse
quent dissertation study. In order to extend the range of
task complexity as an independent variable, some changes in
the composition of the stimulus-response sequences were made
for subsequent studies. Task complexity Level II was drop
ped, and Level I became complexity Level II. A less
41
complex stimulus-response sequence was introduced as task
complexity Level I. This change served to magnify the
differential effect of task complexity on serial reaction
times in the subsequent studies. These changes based on
analysis of pilot study A have been included in the
experimental design of the dissertation study.
PILOT STUDY B
The objective of this pilot investigation was to exam
ine the effects of auditory white noise and three levels of
muscular tension on serial reaction times of five elderly
subjects. Serial reaction time was measured for three task
complexities under the four conditions mentioned above, and
under a control condition of no noise or tension. Irrele
vant information was not included as a variable in this
pilot study.
METHOD
Apparatus
All apparatus was identical to that described for
pilot study A.
Subjects
Four women, ages 52, 55, 66, and 71, and one man, age
57, were tested.
42
Experimental design
The basic study was a five by three factoral design.
The first independent variable was sensory stimulation.
Three levels of induced muscular tension (lO^'per-rcentV- -20 0
per cent, and 30 per cent of maximum handgrip), one level
of auditory white noise (75 decibels), and a control condi
tion without noise or tension were presented. The second
independent variable was task complexity, and three levels
of task complexity were used. Task complexity levels I and
III from pilot study A were used as levels II and III, and
a more simple level of task complexity was included in this
study as level I. Task complexity level I was identical to
the Training-random movement time mask from pilot study A.
Task complexity level II was identical to task complexity I
(Mask #1) from pilot study A, and task complexity level III
was identical to task complexity III (Mask #3) from the
pilot study. Task complexities I, II, and III, in informa
tion theory terms, can be quantified as 3.3, 6.6, and 9,9
bits of information respectively. For full information con
cerning these values see Chapter V.
Procedure
The procedure used was identical to that described for
I
pilot study A, with the following exceptions. A stimulus-
response sequence was composed of 40 pairings instead of 20,
and five conditions were presented instead of six.
43
Results
All correct responses for each stimulus-response
sequence under each of the five conditions were used in the
statistical analysis. Subject's scores were grouped by
sensory condition and then by level of task complexity to
permit comparisons on the basis of these two experimental
variables.
To demonstrate the effects of the five sensory condi
tions, all of the group stimulus-response sequence means
were combined to provide an overall mean serial reaction
time for each of the five conditions. The group means were:
Control 75 db. Noise 10% Tension 20% Tension 30% Tension
1.07 sec. 0.94 sec. 1.01 sec. 0.97 sec. 0.99 sec.
The auditory white noise and tension conditions all resulted
in shorter times than the control conditions, and these
differences were significant at the .01 level of confidence.
In addition, the 20 per cent muscular tension mean was less
than either the LlOi per cent muscular tension or the 30 per
cent muscular tension condition mean. Because of differ
ences in the complexities of the tasks used, the data for
the two pilot studies are not directly comparable.
To evaluate the effects of task complexity, the group
stimulus-response sequence means were combined for all five
conditions to provide a mean serial reaction time for each
of the three levels of task complexity. The group means
44
for the three levels of task complexity were:
Level I Level II Level III
0.65 sec. 1.01 sec. 1.29 sec.
The main features here were that the means increased with
complexity level and that each mean was significantly
i
different from the others at the .01 level of confidence. j
Interpretation of findings from pilot studies A and B j
We are now in a position to take a broad view of the
serial reaction times collected from six young subjects, 22
to 30 years of age, and five elderly subjects, 52 to 71
years of age. It will be recalled that the young subjects
were tested under conditions of 50 decibels, 65 decibels, j
i
75 decibels, and 85 decibels of auditory white noise, 20 j
per cent of maximum handgrip muscular tension, and under a
control condition of no noise or tension. The elderly sub
jects were tested under conditions of 75 decibels auditory
white noise, ten per cent, 20 per cent, and 30 per cent of
maximum handgrip muscular tension, and under a control
condition of no noise or tension. However, because of
differences in the complexities of the tasks used for the
two pilot studies, the data for all tasks cannot be com
pared directly. Examination of mean serial reaction times
under the two different levels of task complexity which
are comparable for both pilot studies, indicates quite
clearly that rate of responding for young subjects under
45
all conditions is faster than the rate of responding for
the elderly subjects (Figure I.). This, of course, is to
be expected and simply supports other age-related studies
on the speed of behavior. Of particular interest, however,
are the effects of the conditions of noise and muscular
tension. For both age groups, mean values of serial reac
tion times are shorter under noise and under tension, than
under the control condition. The only exception to this
trend is at complexity level II for elderly subjects with
muscular tension at ten per cent of maximum handgrip, where
the mean serial reaction time is greater than for the con
trol condition.
The data provide tentative support for the activation
rather than for the neural noise hypothesis of the func
tioning of the aging central nervous system. According to
the activation hypothesis, it would be expected that addi
tional sensory stimulation, such as noise or tension, will
increase the level of activation of the central nervous
system, and result in shorter serial reaction times than
in the absence of noise or tension. On the other hand, the
neural noise hypothesis would predict that additional
sensory stimulation such as noise or tension would lower
the signal to noise ratio in the central nervous system
and serve to slow responding, relative to control condi
tions. Since the noise and tension conditions resulted
Mean
Serial
Reaction
Time
(Seconds)
1.50 • •
1.40 ■ ■
1.30 • •
1.20 -
1.10 ••
1.00
.90
.80
.70 -
/
J
.20 ■■
.10 -
.00
ooooooo
• • • •
+ ■ + ■ ■ + • ■+
Control
10% Tension
20% Tension
30% Tension
50 db. Noise
65 db. Noise
75 db. Noise
85 db. Noise
o Young
• Elderly
6.6 bits 9.9 bits
Task Complexity
Figure I. Mean serial reaction time for pilot studies A and B as a function of
of task complexity and sensory conditions for young and elderly subjects. Each point
was based on six subjects for the young group (pilot study A) and on five subjects
for the elderly group (pilot study B).
47
in serial reaction times which were shorter than for the
control conditions, the data can be regarded as supporting
the activation hypothesis. This by itself is, however,
inadequate for the interpretation of any change in the
function of the central nervous system related to age,
since the differences are in the same direction for both
the young and elderly subjects. To provide such an inter
pretation it must be shown that the improvement in serial
reaction time for the elderly is relatively greater than
for the young. To provide such a comparison, another anal
ysis was completed. Since comparable data for the two age
groups were available only for two levels of task complex
ity, analysis was carried out using data for task complex
ities II and III. The mean serial reaction times for task
complexity levels II and III were combined, giving a meas
ure of speed of responding for each condition for each age
group. Using the notation below, calculations were carried
out using combined means to evaluate the relative effects
of age on response times under the conditions of 75 deci
bels auditory white noise, 20 per cent handgrip muscular
tension, and the control condition without noise or tension,
j Where Y and E denote young and elderly subjects respec
tively, where MSRT denotes mean serial reaction times, and
where N, T, and C refer to conditions of noise, tension, and
control, the following statements must be true to support
the activation hypothesis as an explanatory concept
48
for behavioral slowing with age.
E MSRT C minus E MSRT T Y MSRT C minus Y MSRT T
> -------------------------
E MSRT C Y MSRT C
and
E MSRT C minus E MSRT N Y MSRT C minus Y MSRT N
> -------------------------
E MSRT C Y MSRT C
When appropriate figures were inserted to these
equations the following values were obtained:
Young subjects: control - tension = .069
control
control - noise = .085
control
Elderly subjects: control - tension = .096
control
control - noise = .120
control
! Comparison of the ratios for elderly and young •
I subjects reveals that the average effects of noise and
tension on mean serial reaction times are relatively greater
for the five elderly subjects than for the six young
subjects. Bearing in mind the small number of subjects, it
appears that the activation hypothesis may be used to
explain the differential effects of induced muscular
tension and auditory white noise on serial reaction times
of young and elderly subjects. However, further data are
necessary to provide substantial support of the activation
hypothesis of the aging central nervous system.
i CHAPTER IV
i
|
FORMULATION OF EXPERIMENTAL HYPOTHESES
The hypotheses to be considered in the following
research will concern age differences in the effects of
auditory white noise and induced muscular tension on
serial reaction time with three task complexities. The
i
|predictions and hypotheses to be stated below are based on
i
|data available in the literature, as well as the results
•of the two pilot studies, and are derived from the activa
tion and neural noise theoretical formulations.
The first prediction that can be made is simply that
elderly (E) subjects will have greater mean values of
iserial reaction time (MSRT), than young (Y) subjects.
This prediction follows from many studies of behavioral
'slowing with age, from Koga and Morant (1923) to current
research, and can be expressed by the following notation:
! MSRT E > MSRT Y
A second prediction that can be made is that on the
average, serial reaction times will be greater under the
control condition (C), than under the conditions of au
ditory white noise (N) or induced muscular tension (T) .
iThe prediction concerning the muscular tension condition
51
is based on studies of muscular tension and reaction time
by Davis (1940), Freeman (1933) , and Freeman and Kendall
(1940). The prediction concerned with the condition of
auditory white noise is based on the results of the two
pilot studies. These predictions can be expressed as
follows:
MSRT C > MSRT N and MSRT C > MSRT T
In addition, it can be predicted that the differences
between the noise and the control conditions, as well as
|
those between the muscular tension and the control will be
greater for the elderly subjects than for the young sub
jects. This prediction is derived from the theoretical
formulation of the activation hypothesis of the functioning
of the aging central nervous system. The prediction can
be stated as follows:
E C - E N > Y C-YN and E C - E T > Y C - Y T
The prior prediction states that the absolute differ
ences will be greater for the older than for the younger
subjects. In order to provide adequate evidence for an
activation hypothesis, however, these differences must also I
be expressed in relative terms. This amplified statement
becomes the first hypothesis: that the conditions of
auditory white noise or induced muscular tension will
52
produce serial reaction times which on the average are
relatively shorter by comparison with those in the absence
of auditory white noise or induced muscular tension: the
relative differences will be expected to be greater for
the older than for the young subjects. This hypothesis is
derived from the theoretical formulation of the activation
hypothesis and can be expressed as indicated below.
E C - E N ^ Y C - Y N and E C - E T \ Y C - Y T
EC ' Y C E C - ^ Y C
In addition to age and the conditions of noise, ten
sion, and control, complexity of the serial reaction time
task will be manipulated in the experimental design.
Formulation of the predictions concerning the effects of
task complexity can be stated most simply that mean serial
reaction time with the more complex tasks (Kg) will be
greater than that with the least complex task (K^). This
prediction can be expressed as follows:
MSRT K3 > MSRT ^
In addition, an age versus complexity interaction can
be anticipated, so that with increased age the differences
between the most complex and the least complex tasks is
likely to be increased. These predictions are based on
studies of the relationship between age and task complexity
by Goldfarb (1941) and by Birren, Reigel, and Morrison
53
(1962).. The interaction can be expressed as follows:
E K3 - E K1 > Y K3 - Y Kx
It can also be predicted that this difference will be
relatively greater for the elderly than the young. This
prediction is based on the formulations of the activation
hypothesis and can be expressed as below:
EK - E K YK - YK,
3_______1 ^ 3_______1
E K3 Y K3
An interaction between age, complexity, and sensory
conditions can also be predicted, so that induced muscular
tension and auditory white noise are likely to have a
greater effect at higher levels ofttask complexity for
elderly than young subjects in the.'direction of faster
relative serial reaction times. This statement, hypothesis
number two, is derived from the theoretical formulation of
the activation hypothesis and can be expressed as below:
(ECK - ENK3) - (ECK1 - ENK1) (YCI<3 - YNK-j) - (YCI^ - YNKj)
eck3 - enk3 ^ yck3 - ynk3
and
(ECK3. - ETK3) . - (ECI^ - ETK ) (YCK3 - YTK3) - (YCK1 - YTK^)
eck3 - etk3 yck3 - ytk3
54
The purpose of the main study of this dissertation
research was to gather data to test these predictions and
hypotheses.
CHAPTER V
METHOD |
| An experiment was designed for the purpose of testing
j the predictions and hypotheses presented in Chapter IV
| concerned with the functioning of the aging nervous system.
i ‘
I
|Independent variables of noise and tension, based on the
findings of the two pilot studies, were selected to be
:administered to elderly and young subjects. Level of task j
i I
!complexity was also selected as an independent variable to !
!be manipulated and thus this experiment was designed to i
evaluate the effects of age, sensory conditions, and task I
j
complexity on the speed of responding. I
i
l i
i |
!Apparatus
j All apparatus was as described for pilot study A.
I 1
|Subjects
I 1 " •" "
Thirty young subjects (ages £L8.jto 26 years) and 30
elderly subjects (ages 62 to 85 years) from the University
of Southern California were recruited as subjects for this
study.
j
The elderly sample was selected as follows. A copy j
of the Cornell Medical Index, a cover letter describing the
j research, and a stamped, self-addressed envelope were
; mailed to 203 University of Southern California faculty,
| retired faculty and spouses, and staff, over 62 years of
j
| age. Of the 101 Cornell Medical Indexes returned, those
i
indicating an absence of any chronic illnesses or dis
abilities were selected to be contacted. Those potential
I subjects not selected were those who reported visual or
i
i
auditory disabilities, heart trouble, previous occurrences
of coronary attacks or strokes, high blood pressure, or
| diabetes. This resulted in a subsample of 42 persons to
j be contacted. These 42 people were contacted by tele
phone and asked to participate in the study. Of the 42,
i six declined to participate, leaving 36 subject volunteers.
: Fifteen males and fifteen females were tested.
The fifteen male and fifteen female young subjects
, were recruited from summer classes in psychology at the
jUniversity. All potential subjects were given the Cornell
i
jMedical Index, and three subjects were rejected due to
Shearing defects or chronic illnesses. The same rejection
j criterion were used for both groups.
Experimental design
The basic study was a three by three factorial design.
One independent variable was stimulus complexity and three
i
j levels of complexity were used. The other independent
57
variable was conditions of sensory stimulation and three
different conditions of sensory stimulation were used:
auditory white noise, induced muscular tension,and a con
trol condition without noise or tension. A total of nine
serial reaction time sequences, factorially incorporating j
|
these three levels of complexity and three conditions of
sensory stimulation, were presented to all subjects in both j
I
groups. |
The level of auditory white noise used for all sub
jects was 75 decibels sound pressure level as calibrated
at the earphones, using a Brual Kjaer artificial ear with
a six cubic centimeter coupling cavity and a one inch
condenser microphone. The amount of weight used for in- j
l
duced muscular tension was determined individually for
each subject, and was 20 per cent of the maximum handgrip-.'
as registered on the hand dynamometer at the end of a 30
second test (see Appendix for tension data).
The three levels of task complexity were a function of
the arrangement of the ten numerals on the plexiglass
masks. The three levels of task complexity differed quan
titatively from the least complex to the most complex, and
can be stated in terms of the average amount of information
they presented. From least complex to the most complex,
the levels of task complexity were 3.3 bits, 6.6 bits, and
9.9 bits of information. The quantification of the level
58
of complexity is determined in the following manner:
Level I - by the position of the numeral on
the top row of ten lights, equivalent to log2 n =
3.3 bits.
! Level II - by the position of the numeral on
the top row of lights (3.3 bits), and the position
I on the bottom row of ten lights (3.3 bits), for a
| total equivalent of 6.6 bits.
Level III - by the position of the numeral on
the top row of ten lights (3.3 bits), the position
: of the numeral on the bottom row of lights (3.3 bits),
j as well as by misalignment of position between the
| two rows (3.3 bits), for a total equivalent of 9.9
bits.
It should be noted that this is but one of several
;ways in which the different levels of complexity here
employed could be expressed in information theory terms.
However, irrespective of the method of assessment that may
be preferred, the relative differences between the three
jlevels would remain substantially the same.
j In addition to these three quantifiable levels of task
!complexity, two simple tasks were included. The first of
i
!these, two-cycle-tapping was simply to tap as rapidly as
j
■possible between the adjacent apertures with numerals five
and six, turning off the lights, for a total of 40 res-
ppDnses. For the second of these simple tasks, simple move- ;
ment time, the task was to turn off the lights in order from
ione to zero, left to right, and then back from zero to one,
ifrom right to left, etc., for a total of 40 responses. For I
both tasks the Training - simple movement time mask from
pilot study A was used. These two simple tasks were includ-;
ed so as to have some base measures of response speed that j
would indicate maximum speed with a minimum of decision, j
;discrimination, or cognition time involved in the task. ;
These tasks were not specified in terms of units of infor-
jmation and were noto.directly comparable to the three
quantifiable tasks described earlier. These two simple
tasks were always presented first for any given condition,
two-cycle-tapping, and then simple movement time. Following;
these two tasks, the three stimulus-response sequences ;
incorporating the three quantifiable levels of complexity i
were presented.
: I
|
i :
i Procedure \
i :
The procedure was identical to that described for
jpilot study A, except for the following modifications. !
Three conditions were used instead of six, and a stimCiibus- ;
response sequence was composed of 40 stimulus-response
pairings instead of 20, in order to secure more reliable
estimates of the mean values. The order of presentation of
conditions was balanced so that each possible order of
presentation of conditions was presented an equal number
of times in each subject group (see Appendix B), and so
that all the data were collected in a single 30 minute
testing session. The three 40 stimulus-response paired
sequences for each condition were presented according to a
balanced and prearranged schedule, insuring that each of
the possible ten pairings of stimulus lights and response
lights with corresponding buttons were presented four
times. There were three prearranged orders of presentation
for the 40 stimulus-response sequences, and each subject
was exposed to each of the three presentation orders. The
three stimulus-response sequence presentation orders were
used once within each condition.
Prior to the testing session, each subject's hearing
was tested using a Beltone audiometer. This was to insure
that each subject met the Los Angeles County Medical
Association's criterion for normal hearing of less than 20
decibel loss in the better ear at 500, 1,000, and 2,000
cycles per second (1951 A.S.A. standard). In addition,
each subject's blood pressure was taken. Subjects whose
blood pressure was higher than 150/100 were not tested
further. After the subject had successfully passed the
hearing and blood pressure tests, his maximum handgrip was
61
determined, and he was then taken to the laboratory and
seated at the PSYCHOMET console. Instructions were read to
the subject (see Appendix C). Following the instructions,
the four training series were presented. Each training
series consisted of 20 stimulus-response pairings. The
first training series was two-cycle-tapping, the second was
simple movement, and the third training series was random
movement time using the Training-random movement time mask
from pilot study A. The last training series was similar
to task complexity III and used Mask #2 from pilot study A.
After the training series were presented, there was a five j
minute rest followed by the three experimental conditions,
each separated by a five minute rest break. During the
experimental conditions, the subjects wore the earphones
and placed the non-preferred arm on the armrest with the
weight handle held in the hand.
Data analysis
Detailed analysis of the data was carried out to an
swer the questions posed by the hypotheses. First the
effects of age, induced muscular tension, and auditory
white noise on serial reaction times were examined. Sec
ond, the interaction effects of task complexity, age, and
sensory conditions were examined.
The results of the experimental treatments were
analyzed using measures of serial reaction latencies. All
62
correct and incorrect responses for each stimulus-response
sequence were recorded. Because the first response in any
sequence was the time for the movement between the start
button and the appropriate response grid on the console,
and because its latency was typically much longer than that
of the later responses, the first response was eliminated
from data analysis. In addition, since the first two
responses following error responses have been shown to be
faster than the mean of a stimulus-response sequence
(Rabbitt, 1963), the two correct responses following an
error response were also eliminated. A mean was computed
for the series without these error-correction responses,
and this mean value was inserted in place of these two
response times so as to have 39 response times for each
stimulus-response sequence. Data analysis was then based
on these correct 39 response times. This practice of
insertation of means for removed data appears to be a
standard procedure for correcting for missing data, pro
vided the replaced values are not too numerous. This was
the case in the present study. The maximum number of
inserted values being six. It should also be noted that
subjects with more than three errors for any single
stimulus-response sequence were eliminated from the study.
There were one older and one younger subject in this
category.
Response latencies were submitted to a single
63
2x3x5 analysis of variance in which the main effects of
age, sensory conditions, and task complexity were analyzed,
as well as the interaction among these effects.
The ratio values as specified by the hypotheses were
not computed due to lack of differences for mean values for
the three sensory conditions and to the occurrence of a
practice effect, to be described later.
64
i CHAPTER VI
!
j RESULTS
!
j
i An examination of the mean values of serial reaction
I
|
times for the young and elderly groups (Table I.) reveals
that the two groups differed as expected in that the
elderly subjects responded more slowly than the young sub-
'jects. Analysis of variance of the serial reaction.time
. data indicates that these group differences were signifi
cant at the .01 level of confidence (F = 426.81, df = 1),
! and were consistent for each of the five different tasks
' (see analysis of variance table in Appendix D). The age
l
(differences for the various tasks varied from .12 seconds
I
'to .10 seconds for the two-cycle-tapping time and simple
movement time tasks, to a .32 second difference for the
!most complex task. For the two-cycle-tapping task, the
I young subjects as a group responded at 60 per cent of the
response time required by the elderly group. For the most
;complex of the three quantifiable tasks, complexity level
lIII, the young subjects as a group responded at a rate
equal to 80 per cent of the time required by the elderly
jsubjects (see Appendix E for individual stimulus-response
sequence means).
Individual distributions of time values are frequently
TABLE I
MEAN SERIAL REACTION TIME AS A FUNCTION OF TASK COMPLEXITY AND SENSORY
CONDITIONS FOR YOUNG AND ELDERLY SUBJECTS
Young Group Simple Tasks Task Complexity in Bits
Mean
N = 30
2-cy-T1 S.M.T.2 3.3 bits 6.6 bits 9.9 bits
Control
Tension
Noise
Grand Mean
.18 sec. .23 sec. .60 sec. .95 sec. 1.21 sec.
.19 sec. .23 sec. .58 sec. .95 sec. 1.20 sec.
.18 sec. .23 sec. .60 sec. .95 sec. 1.22 sec.
.186 sec.* .230 sec.* .59 2 sec.* . 953 sec.* 1.210 sec.*
Elderly Group Simple Tasks Task Complexity in Bits
Mean
N = 30
2-cy-T-'- S.M.T.2 3.3 bits 6.6 bits 9.9 bits
Control
Tension
Noise
Grand Mean
.31 sec. .33 sec. . 84 sec. 1.15 sec. 1.53 sec.
.31 sec. .33 sec. .85 sec. 1.15 sec. 1.52 sec.
.32 sec. .33 sec. .84 sec. 1.15 sec. 1.51 sec.
.314 sec.* .334 sec.* .842 sec.* 1.151 sec.*
!
1.520 sec.*
1 2-cy-T = two-cycle-tapping. 2 S.M.T. = simple movement time.
* These means were significantly different for the two age groups (£<^.01). ui
66
j skewed so an analysis, based on medians computed for each
| stimulus-response sequence, was conducted. As can be seen
1 from Table II., the median data support and are consistent
jwith the relationships found with the means (Table I.), so ,
I that all further data analysis can be based upon the means.!
i
! An examination of differences in the serial reaction
| times for the five levels of task complexity reveals that j
1 ~
i i
j mean serial reaction times increase significantly with
i :
i increasing task complexity (F = 269.84, df = 4,p < .01).
| A significant interaction between age and task complexity ]
' was also found (F = 20.90 , df = 4,£ <[ .01). This
j interaction seems to be made up of two components. For j
i I
i
the two very simple tasks, elderly subjects are probably
nearer their limit of speed of responding than young !
i subjects, and there may be less effect of differences '
I I
: between the tasks than is the case with the subjects who
i may be farther from their maximum speed or limits of per
formance. For example, if differences between two-cycle- j
tapping and simple movement!time are compared, simple
I movement time is slightly longer in all conditions for both;
I ;
age groups. However, the difference between two-cycle- I
i
tapping and simple movement time was greater for the young j
subjects (.04 seconds), than for the elderly group (.02) j
1 j
seconds. This suggests that elderly subjects reach sooner :
|
a limit of response speed for these very simple tasks, and j
I after this limit is reached, they cannot respond^ faster,___j
TABLE II
MEDIAN SERIAL REACTION TIME AS A FUNCTION OF TASK COMPLEXITY AND SENSORY
CONDITIONS FOR YOUNG AND ELDERLY SUBJECTS
Young Group Simple Tasks Task Complexity in Bits
Median
N = 30
2-cy-T1 S.M.T.2
3.3 bits 6.6 bits 9.9 bits
Control
Tension
Noise
Mean of
Medians
.18 sec. .22 sec. .59 sec. .92 sec. 1.16 sec.
.19 sec. .22 sec. .56. sec.. .92 sec. . 1.15 sec.. .
. 18 sec.. .23 sec. .59 sec. . .92 sec. 1.17 sec. . .
.183 sec. .223 sec. .580 sec. .920 sec. 1.160 sec.
Elderly Group Simple Tasks..... Task Complexity in Bits
Median
N = 30
. 2-cy-T1 S.M.T.2 3.3 bits. 6.6 bits 9.9 bits
Control
Tension
Noise
Mean of
Medians
.30 sec. .31 sec.. .78 sec. 1.10 sec. 1.46 sec.
.29 sec. .31 sec. .80 sec. 1.10 sec. 1.45 sec.
. 31 sec. . .31 sec. . 80. sec. 1.09 sec. 1.44 sec.
.300 sec. .310 sec. .793 sec. 1.097 sec. 1.450 sec.
1 2-cy-T = two-cycle-tapping. 2 S.M.T. = simple movement time.
68
even though the task is made more simple. The second
component of the interaction of age and task complexity
appears in the more complex tasks which are quantifiably
specified in terms of units of information. Elderly sub
jects slow to a greater extent than do the younger with
increasing complexity of the task. The data from Table I. ,
when plotted (Figure II.), indicate that for the three
quantifiable levels of complexity, as distinguished from
the two elementary tasks that have not been defined in
information terms, mean serial reaction times for both
groups increase in direct proportion to the increasing task
complexity. This relationship between task complexity and
mean serial reaction times was expected, and can be related
to the amount of information presented by each of the three
more complex tasks. However, the curves for reaction time
as a function of information, although roughly parallel
for young and elderly subjects, diverge somewhat at the
higher levels of complexity, with the elderly group appear
ing to slow to a greater extent with greater complexity
than the young. This spreading of the age curves with
increasing amount of information indicates that the older
individual in general tends to require more time to respond
than the younger subject as the task becomes more complex.
An examination of the mean values of response latency
in the three sensory conditions for both age groups
indicates no real differential effect created by auditory
Mean
Serial
Reaction
Time
(Seconds)
Control
Tension
Noise
2.00 o Young
• Elderly
1.80
1.60
1.40
1.20
1.00.
.80
.60
.40.
.20
.00
6.6 bits 3.3 bits 2-cy-T 9.9 bits S.M.T.
Task Complexity
Figure II. Mean serial reaction time as a function of task complexity and
sensory conditions for young and elderly subjects. Each point was based on the
mean of 30 subjects.
O ' *
VO
70
white noise, induced muscular tension, or the control
condition. This finding is inconsistent with results from
the two pilot studies, and in addition dispences with the
necessity of preferring either the neural noise hypothesis
or the activation hypothesis of the aging central nervous
system.
A practice or order effect was found as presented in
Figure III. The three sensory conditions were administered
according to a balanced design so that each possible order
of presentation of the conditions occurred an equal number
| of times in each age group. However, for both age groups,
|serial reaction times were faster as the subjects became
J more familiar with the instrumental setting. Both young
and elderly subjects were faster in every successive condi
tion following the first, regardless of what the conditions
might have been (F = 158.08, df = 2, d < .01). This was
true for 58 out of the 60 subjects tested. Serial reaction
times were shorter for the condition presented second,
than for the first and all subjects were fastest under
whatever condition was presented last. This order effect
appears to be differentially related to age in that the
l
jyoung group's mean reaction time decreased by .147 seconds
! from the first to the third condition, while the corren
sponding decrease for the elderly group was only .113 sec
onds. Similarly, as can be seen in Figure III, the young
group improved their speed of responding from the first
Mean
Serial
Reaction
Time
(Seconds)
00 Young
Elderly
.90
.80
.70
.60
.50
.40
.30
.20
.10
.00
1st Order 2nd Order 3rd Order
Presentation Order
Figure III. Mean serial reaction time as a function of presentation order
for young and elderly subjects. Each point was based on the mean of 30 subjects.
72
presentation to the second by .129 seconds while the
elderly group's improvement was .082 seconds. Further
examination of this practice effect in terms of sensory
conditions (Figure IV) reveals that during the first
exposure to the experimental task, both young and elderly
subjects were faster under the noise or tension conditions,
than under the control condition. The difference between
these two conditions and the control condition on first
testing was greater for the elderly group than for the
young subjects. Moreover, in the older group, with
repetition of the experimental tasks, for the control
condition when presented as second or third in the se
quence, the speed of responding tends to improve more than
for the noise or tension conditions. It should be recalled
that those subjects having the control condition for the
third exposure would be subjects who had noise and tension
for the first two runs. The data for the young group, do
not show this relationship, with the control condition
being shortest for the second run; for the third run the
mean values under conditions of muscular tension were
fastest, while the auditory noise condition yielded the
slowest serial reaction times.
As mentioned earlier, the sensory conditions were
presented according to a balanced design so that each of
the six possible orders of presentation was administered
to five subjects in each age group. Examination of the
1.00
.90
o
00
•
Mean .70
Serial .60
Reaction
o
in
•
Time .40
(Seconds) .30
.20
•
H
O
o
o
•
Control Young
Tension
Noise
• Elderly
8 =
1st Order 2nd Order
Presentation Order
3rd Order
Figure IV. Mean serial reaction time as a function of presentation order and
sensory conditions for young and elderly subjects. Each point was based on the mean
of ten subjects.
74
means for the five subjects under each of these orders of
presentation in the two age groups (Table III) indicates
that when noise and tension are presented before the
control condition, the reaction times are shorter. When
the control condition is presented first, followed by
muscular tension and auditory noise, or auditory noise and
muscular tension, mean serial reaction times are slower
than when the control condition is presented second or
third.
TABLE III
MEAN SERIAL REACTION TIME AS A FUNCTION OF AGE, TASK COMPLEXITY, AND PRESENTATION ORDER
YOUNG GROUE Simpie Tasks
?ask Complexity in Bits
Mean. Latency
N = 30
2-cy-T S.M.T. 3.3 bits 6.6 bits 9.9 bits
C T N*
Cl N T
.21 sec.
.20 sec.
.26 sec.
.23 sec.
.62 sec.
.61 sec.
1.01 sec.
.96 sec.
1.33 sec.
1.21 sec.
T C N
N C T
.19 sec.
.17 sec.
.22 sec.
.20 sec.
.56 sec.
.56 sec.
.94 sec.
.92 sec.
1.18 sec.
1.15 sec.
T N C
N T C
.18 sec.
.17 sec.
.24 sec.
.22 sec.
.60 sec.
.60 sec.
.95 sec.
.93 sec.
1.20 sec.
1.19 sec.
ELDERLY GROUP Simple Tasks
Task Complexi
ty
in Bits
Mean' Latency
N = 30 2-cy-T S.M.T. 3.3 bits 6.6 bits 9.9 bits
C T N*
C N T
.30 sec.
.39 sec.
.33 sec.
.40 sec.
.80 sec.
.98 sec.
1.13 sec.
1.30 sec.
1.50 sec.
1.79 sec.
T C N
N C T
.31 sec.
.32 sec.
.33 sec.
.31 sec.
.83 sec.
.83 sec.
1.12 sec.
1.08 sec.
1.46 sec.
1.47 sec.
T N C .29 sec. .33 sec. .82 sec. 1.15 sec. 1.48 sec.
N T C .28 sec. .30 sec. .79 sec. 1.11 sec. 1.4.4. sec. . .
* This designation indica
conditions of control (
:es that the
3) first, ten
presentation or
sion (T) second
der in this c
, and noise 1
ase was
N) last. ji
76
j CHAPTER VII j
i |
DISCUSSION
This study of 30 young and 30 elderly subjects
! |
I
explored some of the factors related to behavioral slowing i
with age. As expected, elderly subjects responded more
I ;
j slowely than young subjects on the serial reaction time
itask. This would reflect a general limitation with age on j
|the speed of behavior mediated by the central vervous !
i !
system as discussed in Chapter II. The speed of responding
was found to be directly related to the complexity of the
task in both groups so that serial reaction times became |
I
longer as the level of complexity was raised. These two '
:findings are consistant with the results of previous ;
:studies by Goldfarb (1941) and by Birren, Riegel, and
iMorrison (1962) as well as with those of many other inves- ;
i ' j
jtigators. The results of the present study also indicate j
j 1
ithat older subjects took proportionately longer than did thd
! I
i ;
lyoung to respond to the more complex tasks. This indicates j
t
| 1
Ithat whatever central nervous system processes are involved j
in behavioral slowing with age, they become more evident as I
|
the task becomes more complex. Examination of serial reac- j
jtion times for the two simplest tasks also proved interest- !
I ' :
ling. The relationship between the two-cycle-tapping task ,
77
and the simple movement task was different for the young
and the elderly. The young subjects responded faster in
the two-cycle-tapping task than for the simple movement
task. The elderly subjects, however, appeared to be at a
lower limit of responding in the simple movement task and
did not respond much faster in the simpler, two-cycle-
tapping task. This lack of real difference for the older
subjects between the two-cycle-tapping task in which the
requirement was to tap alternately between two contact
grids, and the movement time task which required the
touching of each of the contack grids, in order from left
to right and right to left was an unexpected ifihding. AtAt
first sight it might be thought that the simple movement
time task requires a greater degree of coordination and
dexterity for accurate and fast performance than the two-
cycle-tapping task. However, even though the initial
impression may be that the two-cycle-tapping task is a
simpler task which could be performed at a faster rate,
on further inspection it becomes clear that it does in fact
require coordination and dexterity. As stated earlier, the
largest component in any slowing of responses with age
appears to be related to the central processes of discrim
ination or coordination. It follows that these central
processes come to a progressively larger portion of the
response time as the task requirements approach the
78
subjects' lower limits.
The expectation of a reduced response latency under
the sensory conditions of induced muscular tension and
auditory white noise, as compared with the control condi-
i
tion of no noise or tension, was not supported by the data, j
j
Because the mean reaction times for the three conditions in j
I
both age groups did not differ significantly, the hypoth- l
esis that auditory white noise or induced muscular tension
would produce shorter response times than the control condi
tion, particularly in the older group, was not supported.
The activation hypothesis would predict that reaction times
under the conditions of noise or tension would be shorter
than in the absence of either type of added stimulation.
The rationale of this prediction has been presented in
Chapter II. The data reported in the two pilot studies
demonstrated a relationship as predicted by the activation
hypothesis. It will be recalled that in pilot study A,
six young subjects were tested, according to a balanced
design, under four levels of auditory white noise (50, 65,
75, and 85 decibels), one level of muscular tension (20
per cent of maximum handgrip), and a control condition. As
presented in Chapter III, the mean serial reaction times
for the six conditions were 1.12 seconds for the control
condition, 1.05 seconds for the muscular tension condition,
and 1.10, 0.98, 0.99, and 0.98 seconds for the 50, 65, 75,
j and 85 decibel noise conditions. In pilot study B, five
| elderly subjects were tested under three levels of muscular
; tension (10 per cent, 20 per cent, and 30 per cent of
i maximum handgrip) , one level of auditory white noise (75
j
| decibels), and a control condition. The mean serial reac-
I tion times were 1.07 seconds for the control condition,
j 0.94 seconds for the noise condition, and 1.01, 0.97, and
| 0.99 seconds for the 10 per cent, 20 per cent, and 30
i
Jper cent muscular tension conditions. Because of
differences in the complexities of the tasks employed in
; the two pilot studies, the mean values of reaction times
;for the young subjects cannot be compared directly with
those for the elderly subjects. In both pilot studies,
the auditory noise and muscular tension conditions yielded
I
j faster response times than the control condition. The
j data from the main experiment fail to confirm this rela
tionship, however, the mean serial reaction times for the
i
|three conditions do not differ significantly from each
i
i
j other.
! In addition to these predictions made by the activa-
t
Ition hypothesis, an age effect could also be predicted.
The theory would lead us to expect that older subjects
would benefit to a greater extent than younger subjects
from the influence of additional stimulation. On this
!
jview, the older central nervous system, being at a lower
80
level of activation than the young central nervous system
would be more affected by the activating influence of the
added stimulation. The young central nervous system, being j
j
at a higher level of activation than the older central j
nervous system would under conditions of auditory noise or
muscular tension enhance sooner the facilitating effects of
activation on performance. This prediction is not support-
I
ed by data from the main experiment although it is in line
with the results of the two pilot studies as well as those
6f other studies reported in the literature. Helson (.1964)
has shown that when two signals are presented within 25
I
to 35 milliseconds of each other the second signal serves
to speed up reaction times to the first signal. This is
true also when the two signals are in different modalities
such as a light signal and a tone. Helson1s results in
dicated that when a tone is the second signal and a light
is the first signal (or vice versa), and the two signals
appear either simultaneously or with an inter-signal-
interval of up to 25 to 35 milliseconds, the presence of
the second signal acts to facilitate or speed up the reac
tion to the first signal. It appears that the second
signal serves to amplify the effects of the first and to
increase the state of activation of the central nervous
system. Helson, did not touch upon the possible age
differences in this respect, but in a study of
i psychological refractory phase and aging by Brebner and
I Szafran (1961), similar results were reported for young
i j
I (20 to 39 years), middle age (40 to 59 years), and older |
j . i
I subjects (60 to 80 years). In this study subjects were i
i !
i
| required to respond to two streams of signals. One type j
of signal occurred at a regular rate of either one signal j
l
i
every two seconds or at a rate of one signal every four and;
i
I
;one half seconds, and the subjects respohded to these sig- i
j nals with their non-preferred hand. The high-information
signals were given at irregular intervals of 50, 150, 300, -
and 600 milliseconds following a response to the regular j
: signal. Responses to these irregular signals were made
;with the preferred hand. The data indicated an increased j
I I
|speed of responding to these high-information signals at j
: the faster rate of signal presentation in the regular sub- j
itask. The amount of this increased, speed correlated
significantly with age at all inter-signal intervals, e
;except the 300 millisecond intervals. Moreover, this |
increase in speed of responding was relatively greater at
the 50 millisecond interval than the 150 millisecond inter-
f ;
>val, and the difference between these two intervals were ;
marked for the older subjects. In other words, the elderly |
subjects reduced the latency of their responses to the j
i. !
;irregular signals proportionately more than the young sub- ■
\ . <
ijects at both the 50 and 150 millisecond intervals. The !
! !
| authors suggest that the regular signal and response serve
82
| to "improve the selective responsiveness of the cerebral
! cortex to a high-information signal." They speculate that
i
! this improved level of responsiveness can perhaps be inter-
i
I preted as corresponding to a higher level of activation in
|the cerebral cortex, affecting the older subjects rela-
| tively more than younger subjects.
| The neural noise hypothesis would predict that serial j
| I
;reaction times under the conditions of auditory noise or ;
i !
jmuscular tension would be longer than in the control condi-;
tion. As indicated in the preceding chapters, this predic-
;tion was not supported by either the pilot data or by the
data of the main experiment. It follows that neither
explanation of changes in functioning of the aging central j
• i
inervous system need be used to account for the data. |
However, it is interesting to note that the neural j
jnoise hypothesis would predict a different age effect
than the activation hypothesis with respect to one of the
conditions of added stimulation employed in this study.
|
jAccording to this theory, an externally introduced auditory
:noise would result in a lower signal-to-noise ratio. In
the case of auditory input, however, the effect of this
I lowered signal-to-noise ratio could be expected to be
less for the elderly than for the young. If the level
|of neural noise is higher in the auditory pathway of J
i !
|the older person, his brain would be affected to a
I i
I lesser degree by the added auditory stimulation j
than it would in a young nervous system. We should remind
ourselves at this point that a general lowering of signal-
to-noise ratio can be the result of either a lower signal
input or a greater amount of background noise. There are
reasons to believe that conditions such as muscular tension
and auditory white noise might serve to lower the signal-
to-noise ratio by different means. Feedback effects from
the contraction of skeletal muscles have long been recog
nized as a possible way of increasing the activity of the
central nervous system (Duffy, 1962). The sensory fibers
as they pass through the brain stem, give off collaterals
which, conduct impulses to the reticular formation. The
effect of muscular tension in terms of input along the
collateral fibers may serve to increase the level of
nervQus. discharge. Because of loss in the number of
functional neural cells with age, this augmented nervous
discharge may serve to increase the level of background
noise in the older nervous system and thus, by contrast
with the added auditory noise input, to alter the signal-
to-noise ratio unfavorably.
One of the well documented sensory changes accompany
ing increasing age is the loss of hearing acuity at high
frequencies, known as presbycusis. According to Szafran
(1968) , clinical evidence suggests there may be two types
of presbycusis. In one type, lesions are found in the
organ of Corti at the basal turn. The other type is
characterized by partial atrophy in the nerve supply at the
base of the cochlea (Crowe, Guild, and Polvogt, 1934). One
explanation for nerve deafness is that it may be caused by
a higher level of random neural activity related to damage
I
to the hair cells (Gregory, 1957). This increased level of j
i
neural noise in the older nervous system would tend to i
lower the signal-to-noise ratio with the paradoxical result
that high levels of externally introduced auditory noise
would have less effect for the elderly than for the young
individual (Szafran, 1968). In other words, according to
the neural noise hypothesis, the effects of added auditory
noise would tend to keep reaction times to non-auditory sig
nals unaffected, and this effect would be relatively more
pronounced for elderly than for the young individuals.
Several factors may contribute to the different find
ings of the pilot and subsequent dissertation studies.
First, the subjects participating in this study were
selected so as to be free from known chronic illness,
whereas the subjects tested in the two pilot studies were
not selected in this manner. It may be that the magnitudes
of auditory noise and muscular tension used in this study
affect healthy subjects differently from the way they
affect the less healthy subjects; it is,oof course, the
latter who may be more representative of the aging popula
tion. This kind of interpretation would be in line with
the findings of Birren, e-t.jd.. (1963), in a study of
85
healthy older men, as well as with those of Szafran with
extremely healthy pilots as subjects (Szafran, 1965).
These investigators have reported that healthy, older men
tend to respond significantly faster than men matched for j
age but not selected with respect to good health. In the
pilot study, where the relevant details concerning the
health status were unavailable, the young subjects respond
ed faster under the noise and tension conditions, than
under the control condition, while in the main study, where
the health information was obtained, the young subjects did
I
not respond differently to the three conditions. Although
one cannot be absolutely sure about this, the young sub
jects in the pilot study appeared to the author to be as
healthy and as free from chronic illness as the subjects
in the main experiment. To this extent, the differences
between the pilot study results and later results can
perhaps be regarded as not related in any important manner
to variations in health factors.
Other factors of importance in evaluating the effects
of the three conditions on performance, is the influence
of experience in the experimental situation as well as
that of the order of presentation. The three conditions
were presented according to a balanced design so that each.;
possible order of presentation would be used five times in
each age group. When the data were analyzed in terms of
responses in the first condition presented to the subjects,
86
regardless of what this condition was, these responses were
slower than the responses recorded in thevsecond and third
condition presented. It was found that subjects increased
their speed of responding with each exposure to the experi- j
mental task, regardless of what the nature of the condition
had been. Because of the effect of order of presentation,
the aim of any further research in this area should be to
try to control for this variable, in order to obtain a more
clear evaluation of the effects of the conditions of added
sensory stimulation, such as auditory noise or muscular
tension, on behavioral slowing with age. Two approaches !
could be used to control for the practice effects. Sep
arate groups of subjects could be tested under different
experimental conditions, so that the subjects were exposed
to only one condition and would not repeat the experimental
task under any other condition. Another approach might be
to expose subjects to'the experimental tasks a number of
times until no further practice effect could be observed.
Then subjects would be tested under each of the conditions
used according to a balanced design. The advantages of
this second design are that practice would have only a
minimal improvement effect for succeeding exposures to the
task and that data would be available for all subjects
under all conditions, thus allowing individual as well as
group comparisons of the effects of added sensory stimula
tion on the speed of responding.
87
A more detailed examination of the effects of experi
ence or practice in each of the three conditions revealed
an interaction effect between the order of presentation
and the three conditions. An analysis of the serial re
action times recorded during the first exposure to the j
I
experimental situation indicated that those subjects who j
i
were first tested in the auditory noise or the muscular |
tension condition had shorter reaction times, than those
subjects who were presented with the control condition ;
first. The differences between the response latencies in j
I
these two conditions and those in the control condition i
were greater for the elderly, than for the young subjects.
These effects would tend to support the activation
hypothesis. A>: further examination of the order effects
revealed that for the elderly group, during the second and
third exposure to the experimental situation, those sub
jects who were presented with the control condition first
had on the average shorter serial reaction times, than
those subjects who were tested in the noise or tension
conditions first. It should be noted that those subjects
who had the control condition second or third were the
very same subjects who had the auditory noise or muscular
tension conditions presented first. In other words, those
subjects who had the auditory noise or muscular tension
conditions first are not only faster than those subjects
who had the control condition first, but they continue to
88
respond faster under whatever conditions follow. It would
appear then that upon initial exposure, the auditory noise
and muscular tension conditions serve to activate the
elderly subjects and that the effects of this activation
carry over into subsequently presented conditions. By j
I
I
contrast, the data for the young subjects were not quite |
as clear in this respect. For the first presentation of I
the experimental conditions, those subjects who were j
exposed to the auditory noise and muscular tension condi- !
tions responded faster than those exposed to the control j
i
condition. For the second presentation, those subjects |
who had the control condition at this point in the sequence
were slightly faster than those subjects who had the aud
itory noise or muscular tension conditions. This pattern
is similar to that observed in the elderly group. However,
for the third order of presentation, those young subjects
who had the muscular tension condition at this point in
the sequence were faster than those who had the control
condition in this position. The subjects presented with
the control condition as third were faster than those
having the noise condition third.
In summary, we can say that the overall effects of
the order of the presentation in both age groups were to
balance any differences between conditions in such a way
as to leave the mean values of serial reaction times the
same. However, an examination of the interaction between
order and sensory conditions reveals that those elderly
subjects who were exposed to activating effects of auditory
noise or muscular tension during the first presentation of
the experimental task, not only responded faster in that
condition, but also continued to respond faster in the sub
sequent conditions. This interaction between order of
presentation and the sensory conditions can be interpreted
as supporting the activation hypothesis.
The reason for the interaction effects between the
order of presentation and the sensory conditions is not
clear. Nevertheless, tto the extent that it is possible to
generalize from these data, it is clear that if the meas
urement of serial reaction time is to be meaningful, the
role of practice and of the order of presentation of condi
tions must be specified.
The sensory conditions used to test the activation
hypothesis and the neural noise hypothesis in this study
have been postulated by a number of investigators to
create an effect on the level of activation or neural
noise in the central nervous system. Serial reaction time
has been used as a dependent variable in this study
because speed of responding has been demonstrated to be
highly related to cortical function as well as to levels
of activation of the central nervous system. The results
of this investigation indicate that either the added
sensory stimulation or the intensities in the modalities
90
selected, do not affect the level of activation or neural
noise in the nervous system, or that serial reaction time,
as measured in this study, does not reflect these changes.
IA third possibility to be considered is that the effects
of practice and order of presentation serve to counter
balance the effects of the added sensory stimulation.
A retrospective appraisal of the present study leads
to several suggestions concerning central nervous system
activation as a mechanism to be manipulated and investi
gated. The fact that the three conditions as employed
here did not differentially influence serial reaction times
indicates that possibly the effects of other levels and
types of added sensory stimulation should be considered.
Since the levels of auditory noise and muscular tension
used in the main study were chosen from the results of two
pilot investigations, with only six young and five elderly
subjects in each, similar studies should be conducted with
larger samples to determine more exactly the effects of
various levels of muscular tension and auditory white
noise. It is quite possible that the levels of auditory
noise and muscular tension chosen for the present study
were in fhct inappropriate in order to have maximum effect.
In addition, instead of using only serial reaction times
to reflect the functional state of the nervous system,
other indices should also be employed. It is suggested
that further research might include other indicators of
91
central nervous system function, such as EEG, heart rate,
respiratory rate, and galvanic skin response to determine
whether these indicies are correlated with serial reaction
times. The aim of future studies should be to consider to
I
what extent, if any, the central nervous system is affectedj
by the levels and types of added sensory stimulation and !
whether there are differential effects with age. In
addition to the effects created by added sensory stimula
tion, those of the order of presentation of experimental
conditions should also be considered. This variable of
experience must be evaluated more extensively before its
effect on central nervous system functioning can be
determined.
92
CHAPTER VIII
SUMMARY AND CONCLUSIONS
The objective of this study was to examine serial
reaction times in the light of the neural noise hypothesis
j and the activation hypothesis of the aging central nervous
system.
According to the neural noise hypothesis, slowing of
t
response speed with age is due to an increasing level of
internal random "noise" in the brain, which tends to reduce
the signal-to-noise ratio of the sensory input. This in
creased neural noise makes it more difficult for the brain
to distinguish a signal from the background and consequent
ly the older individual must accumulate data over a longer
period of time. This sampling time becomes longer as the
ratio of signal strength to background noise becomes small
er. It is postulated that conditions which would lower the
signal-to-noise ratio would tend to result in the slowing
of behavior. Exposure to added sensory stimulation, such
as auditory white noise or induced muscular tension, is
postulated to result in an increased level of background
noise, a lower signal-to-noise ratio, and slower response
speed.
93
The activation hypothesis would predict the opposite
result. The activation hypothesis assumes that because of
loss of functional neural cells with age or a reduction in
the overall level of stimulation, the central nervous
system in the older individual is at a lower level of
activation than in the younger individual. The older
nervous system, being less activated and less sensitive to
any input, must integrate data over longer intervals of
time and thus the older individual's responses become slow
er. It is postulated that the older, less activated,
central nervous system, when presented with added sensory
stimulation such as auditory white noise or induced mus
cular tension, would have the level of activation raised
and that consequently the response times would become
faster.
Serial reaction times were recorded using the
PSYCHOMET apparatus for five levels of task complexity
under sensory conditions of induced muscular tension (20
per cent of maximum handgrip), auditory white noise (75
decibels), and a control condition without noise or ten
sion. Thirty young (18 to 26 years) and 30 elderly (62 to
85 years) subjects, selected to be free from chronic
illness or sensory disabilities participated in the study.
All subjects were exposed to all conditions in a balanced
design. Mean serial reaction times for the two age groups
94
j differed as expected, in that the elderly subjects respond-
i
| ed more slowly than the young subjects at all levels of
!
| task complexity. Age differences ranged from .12 seconds
j for the simplest task to .32 seconds for the most complex
j task. Reaction time was found to increase linearly with
I task complexity, as measured. A significant interaction
|
| between age and task complexity was found. This consisted
: of two components. At the simplest level of complexity,
|elderly subjects appeared to be at a lower limit of re-
; sponding, and there was less of an effect of differences
| in complexity than in the case for the younger subjects.
In the more complex tasks, elderly subjects slowed to a
i
jgreater extent as the level of complexity was•increased
jthan did young subjects. Examination of serial reaction
!times in the three sensory conditions for both age groups.
:indicated no differential effect due to auditory white
noise or induced muscular tension, as compared with the
'control condition. However, an order of presentation
j
* effect was found indicating that the subjects tested first
l
!under the conditions of noise and tension had shorter
!response times under the first condition as well as under
jsubsequent conditions, than those subjects who were tested
!first under the control condition. These interaction ef-
i
]
;fects on reaction time were greater for the elderly than
I for the young subjects, and are interpreted as supporting
I
;the activation hypothesis. However, the overall mean re-
95
response latencies for the three sensory conditions were
not different and did not support the activation theory.
The lack of differences in reaction time under the
three conditions, and the interaction between conditions
and order of presentation appear to offer contradictory
evidence for the activation hypothesis. Suggestions are
given for further elucidation of the practice effect and
for the specification and measurement of activation
effects of added sensory stimulation on the latency of the
human response.
REFERENCES
REFERENCES
Barlow, H.B, Retinal noise and absolute threshold. J.
Opt. Soc. Am., 1956, 46> 634-639.
Barlow, H.B. Increment thresholds at low intensities
considered as signal/noise discriminations. J.
Physiol., 1957, 136, 469-488.
Barlow, H.B., Fitzhugh, H., and Kuffler, S.W. Change of
organization in the receptive fields of the cat's
retina during dark adaptation. J. Physiol., 1957a,
137, 338-354.
Barlow, H.B., Fitzhugh, H., and Kuffler, S.W. Dark adapta
tion, absolute threshold, and purkinje shift in single
units of the cat's retina. J. Physiol,, 1957b, 137,
327-337.
Beilis, C.J. Reaction time and chronological age. Proc.
Soc. Exper. Biol. Med., 1933, 30, 801.
Bills, A.G. The influence of muscular tension on the
efficiency of mental work. Amer. J. Psychol., 1927,
38, 227-251. “
Birren, J.E. Sensation, preception and modification of
behavior in relation to the process of aging. In:
Birren, J.E., Imus, H.A., and Windle, W.F. (Eds.):
The process of aging in the nervous system.
Springfield: Charles C. Thomas, 1959, 143-165.
Birren, J.E. Behavioral theories of aging. In: Shock, N.W.
(Ed.): Aging, some social and biological aspects.
Washington, D.C.: American Association for the
Advancement of Science, publication No. 65, 1960,
305-332.
Birren, J.E. Age changes in speed of behavior: its central
nature and physiological correlates. In: Welford, A.
T. and Birren, J.E. (Eds.): Behavior aging and the
nervous system. Springfield: Charles C. Thomas, 1965,
191-216.
97
98
Birren, J.E., Allen, W.R., and Landau, H.G. The relation
of problem length in simple addition to time required,
probability of success, and age. J. Gerontol., 1954,
9, 150-161.
Birren, J.E., and Botwinick, J. The relation of writing
speed to age in the senile psychoses. J. Consulting
Psychol., 1951, 15, 243-249.
Birren, J.E., and Botwinick, J. Speed of response as a
function of perceptual difficulty and age. J.
Gerontol., 1955, 1£, 433-436.
Birren, J.E., Butler, R.N., Greenhouse, S.W., Sokoloff, L.,
and Yarrow, Marian R. (Eds.): Human Aging.
Washington, D.C.: Public Health Service, publication
No. 986, 1963.
Birren, J.E., Riegel, K.F., and Morrison, D.F. Age differ
ences in response speed as a function of controlled
variation of stimulus conditions: Evidence of a
general speed factor. Gerontologia, 1962, £, 1-18.
Birren, J.E., and Wall, P.D. Age changes in conduction
velocity, refractory period, number of fibers,
connective tissue space and blood vessels in sciatic
nerve of rats. J_. Comp. Neurol., 1956, 104, 1-6.
Botwinick, J. Drives, expectancies, and emotions. In:
Birren, J.E. (Ed.): Handbook of aging and the
individual. Chicago: University of Chicago Press,
1959, 739-768.
Botwinick, J., and Birren, J.E. Mental abilities and
psychomotor responses in healthy aged men. In:
Birren, J.E. et. al. (Eds).: Human Aging. Washington,
D.C.: Public Health Service, publication No. 986, 1963.
Botwinick, J., Brinley, J.F., and Birren, J.E. Set in
relation to age. J. Gerontol. , 1957, _12, 3CL0J—305.,. . .
Botwinick, J., Brinley, J.F., and Robbin, J.S. The effect
of motivation by electric shock on reaction-time in
relation to age. Amer. J. Psychol., 1958a, 71, 408-
411.
Botwinick, J., Brinley, J.F., and Robbin, J.S. The inter
action effects of perceptual difficulty and stimulus
exposure time on age differences in speed and accuracy
of response. Gerontologia, 1958b, 2_, 1-10.
99
Botwinick, J. , Brinley, J.F., and Robbin, J.S. Further
results concerning the effect of motivation by
electric shock on reaction-time in relation to age.
Amer. J. Psychol. , 1959, 12_ , 140.
Botwinick, J., and Kornetsky, C. Age differences in the
acquisition and extinction of the G.S.R. J. Gerontol.,
1960, 15, 83-84.
Brebner, J., and Szafran, J. A study of the "psychological
refractory phase" in relation to ageing. Gerontologia,
1961, 5, 241-249.
Broadbent, D.E., and Gregory, R.L. Human responses to
classes of stimuli. Nature, Lond., 1962, 193, 1313.
Brody, H. Organization of the cerebral cortex III: A
study of aging in the human cerebral cortex. J. comp.
Neurol., 1955, 102, 511-556.
Cesa-Bianchi, M. Contribute alio studio delle medivicazioni
psichiche in rapporto con l1eta. Publicazioni dell'
Universita Cattelica del Sacro Cuere, Milan, 1955, 49,
1-61.
Courts, F.A. Relations between experimentally induced
muscular tension and memorization. J. Exp. Psychol.,
1939, 25, 235-256.
Critchley, M. Ageing of the nervous system. In: Cowdry,
E.V. (Ed.), Problems of ageing. Baltimore: Williams
and Wilkins, 1942, 518-534.
Crook, M.N., Alexander, Edith A., Anderson, Edythe M.A.,
Coules, J., Hanson, J.A., and Jeffries;, N.T. Age and
form perception. USAF, School of Aviation Medicine,
Randolph AFB, Texas, 1958, Report No. 57-124.
Crossman, E.R.F.W., and Szafran, J. Changes with age in
the speed of information intake and discrimination.
Experientia' Stipplementum, 1956, £, 128-135.
Crowe, S.J., Guild, S.R., and Polvogt, L.M. Observations
on the pathology of high tone deafness. Bull. Johns
Hopkins Hosp. , 1934, 5_5, 315.
Davis, R.C. Set and muscular tension. Indiana University
Publications, Science Series No. 10, 1940, 5-30.
100
Duffy, Elizabeth. The psychological significance of the
concept of "arousal" or "activation". Psychol. Rev. ,
1957, 64, 265-275.
Duffy, Elizabeth. Activation and behavior. New York: John
Wiley and Sons, Inc., 1962.
Fieandt, K. von, Huhtala, A., Kullberg, P., and Saarl, K.
Personal tempo and phenomenal time at different age-
levels. Reports from the Psychological Institute,
No. 2^, 1956, Helsinki: University of Helsinki.
Fitzhugh, R. The statistical detection of threshold
signals in the retina. J. gen. Physiol., 1957, 40,
925-948.
Freeman, G.L. The spread of neuromuscular tension during
mental work. J. gen. Psychol. , 1931, 5^, 479-494.
Freeman, G.L. Facilitative and inhibitory effects of
muscular tension on performance. Amer. J. Psychol.,
1933, 45, 17-52.
Freeman, G.L., and Kendall, W.W. The effect upon reaction
time of muscular tension induced at various prepara
tory intervals. J. exp. Psychol. , 1940, 2_7, 137-148.
Fuster, J.M. Effects of stimulation of brain stem on
tachistosEopic perception. Science, 1958, 127, 150.
Gerard, R.W. Aging and organization. In: Birren, J.E.
(Ed.): Handbook of aging and the individual. Chicago:
University of Chicago Press, 1959, 264-278.
Goldfarb, W. An investigation of reaction time in older
adults and its relationship to certain observed mental
test patterns. Teachers College Contribution to
Education, No. 831, New York: Columbia University,
1941.
Gregory, R.L. An experimental treatment of vision as an
information source and noisy channel. In: Cherry, C.
(Ed.): Information theory: Third London Symposium,
1955, London: Mefihuen and Co.
Gregory, R.L. "Neurological noise" as a factor in ageing.
Proc. Fourth Cong. Internat. A. Gerontology, 1957, 4^,
314-324.
101
Gregory, R.L., and Cane, Violet R. A statistical infor
mation theory of visual thresholds. Nature, Lond.,
1955, 176, 1273.
Hebb, D.O. Drives and the C.N.S. (conceptual nervous
system). Psychol. Rev. , 1955, 62^, 243-254.
Helmholtz, H. Uberdie Methoden, Kleinste Zeittheile zu
messen, und inre an wendung fur physilogiische aweck.
(Lecture) 1850. (Reported in Woodworth, R.S.,
Experimental psychology, 19 38, New York: Henry Holt
and Co., 299 .)
Helmholtz, H. NBast. Akad Wiss Berlin Monastsbr (1867),
1868, 228-234. (Reported in Woodworth, R.S.,
Experimental psychology, 19 38, New York: Henry Holt
and Co., 299.)
Helson, H. Current trends and issues in adaptation-level
theory. Amer. Psychologist, 1964, 19^, 26-38.
King, H.E. Psychomotor changes with age, psychopathology
and brain damage. In: Welford, A.T., and Birren, J.E.
(Ed.): Behavior, aging, and the nervous system.
Springfield: Charles C. Thomas, 1965, 476-525.
Koga, T., and Morant, G.M. On the degree of association
between reaction times in the case of different
senses. Biometrica, 1923, 1_4, 346-372.
Landahl, H.D. Biological periodicities, mathematical
biology, and aging. In: Birren, J.E. (Ed.):
Handbook of aging and the individual. Chicago:
University of Chicago Press, 1959, 81-118.
Leonard, J.A. Some experiments on the temporal relation
between information and action. Unpublished thesis,
Cambridge University, 1952.
Lindsley, D.B. Emotion. In: Stevens, S.S. (Ed.): Handbook
of experimental psychology. New York: Wiley, 1951,
473-516.
Lindsley, D.B. Psychophysiology and motivation. In: Jones,
M.R. (Ed.): Nebraska symposium on motivation, 1957.
Lincoln: University of Nebraska Press, 1957, 44-105.
Lindsley, D.B., Bowden, J., and Magoun, H.W. Effect upon
the EEG of acute injury to the brain stem activating
system. EEG clin. Neurophysiol. , 1949, JL, 475-486.
102
Magladery, J.W., Porter, W.E., Park, A.M., and Teasdall,
R.D. Electro-physiological studies of nerve and
reflex activity in normal man IV: The two-neurone
reflex and identification of certain action potentials
from spinal roots and cord. Bull. Johns Hopkins
Hosp., 1951, .88, 499-519.
Malmo, R.B. Activation: a neuropsychological dimension.
Psychol. Rev. , 1959, 6J5, 367-386.
Malmo, R.B., and Shagass, C. Physiologic study of symptom
mechanisms in psychiatric patients under stress.
Psychosom. Med. 1949, LI, 25-29.
McGrath, J.J. The effect of irrelevant environmental
stimulation on vigilance performance. Human factor
problems in ASW., 1960, Tech. Rep. No. 7, Los Angeles:
Human Factors Research, Inc.
McGrath, J.J., and Hatcher, J.E. Irrelevant stimulation
and vigilance under fast and slow stimulus rates.
Human factor problems in ASW. , 1961, Tech. Rep. No. f,,
Los Angeles: Human Factors Research, Inc.
Miles, W.R. Correlation of reaction and co-ordination
speed with age in adults. Amer. J. Psychol., 1931,
L3, 377-391.
Moruzzi, G., and Magoun, H.W. Brain stem reticular
formation and activation of the EEG. EEG. clin.
Neurophysiol., 1949, 1' 455-473.
Mundy-Castle, A.C., Hurst, L.A., Beerstecher, D.M., and
Prisloo, T. The electroencephalogram in the senile
psychoses. EEG. clin. Neurophysiol. , 1954, 6_, 245-
252.
Norris, A.H., Shock, N.W., and Wagman, I.H. Age changes in
the maximum conduction velocity of motor fibers in
human ulner nerves. J. Appl. Physiol. , 1953, _5, 589-
593.
Obrist, W.D. Simple auditory reaction time in aged adults.
J. Psychol., 1953, 35, 259-266.
Obrist, W.D. The electroencephalogram of normal and aged
adults. EEG clin. Neurophysiol. , 1954, 6_, 235-244.
103
Pieron, H. Nouvelles recherches sus l1analyse du temps de
latence seur sori 'elle at sur la loi aui relie le
temp l'intensitie d'ercitation. Annee. Psychol.,
1920, 22, 58-142.
I
Pinneo, L.R. The effects of induced muscle tension during j
tracking on level of activation and performance. J.
exp. Psychol., 1961, 62, 523-531.
Pollack, I. Speed of classification of words into super
ordinate categories. J. verb. Learn, and verb.
Behav., 1963, 2, 159-165.
Rabbitt, P.M.A. Effects of independent variations in
stimulus and response probability. Nature, Lond.,
1959, 183, 1212.
I
Rabbitt, P.M.A. Perceptual discrimination and the choice j
of responses. Unpublished Ph.D. Thesis. University j
of Cambridge, Cambridge, Eng., 1962, 260 pages. (Ref
erred to in Rabbitt, P.M.A., (1964).
Rabbitt, P.M.A. (Personal communication to J.E. Birren,
1963)
Rabbitt, P.M.A. Age and time for choice between stimuli
and between responses. J. Gerontol., 1964, 19, 307-
312.
Rabbitt, P., and Birren, J.E. Age and responses to
sequences of repetitive and interruptive signals.
J. Gerontol. , 1967, 22^, 143-150.
Shannon, C.E., and Weaver, B. The mathematical theory of
communication. Urbana, 111.: University of Illinois
Press, 1949.
Simonson, E. Performance as a function of age and cardio
vascular disease. In: Welford, A.T., and Birren, J.
E. (Eds.): Behavior, aging, and the nervous system.
Springfield: Charles C. Thomas, 1965. 401-434.
Sjoberg, H. Relations between different arousal levels
induced by graded physical work, and psychological
efficiency^ Reports from the P sy chologica1 Labora
tories. Stockholm: University of Stockholm, No. 251,
April, 1968.
Sommer, J. Synchronisierung motorischer impulse und ihra
bedeutung fur die neurophysiologische forschung.
Ztschr. ges. Neur. u. Psychiat., 1941, 172, 500-530.
104
Spieth, W. Slowness of task performance and cardiovascular
diseases. In: Welford, A.T., and Birren, J.E. (Eds.):
Behavior, aging, and the nervous system. Springfield:
Charles C. Thomas, 1965, 366-400.
Szafran, J. Age differences in sequential decisions and
cardiovascular status among pilots. Aerospace Med
icine , 1965, 36^ 303-310.
Szafran, J. Psychophysiological studies of aging in
pilots. In: Talland, G.A. (Ed.): Human aging and
behavior. New York: Academic Press, 1968, 37-74.
Talland, G.A. Initiation of response and reaction time in
aging, and with brain damage. In: Welford, A.T., and
Birren, J.E. (Eds).): Behavior, aging, and the nervous
sys'tem.. Springfield: Charles C. Thomas, 1965, 526-
561.
Tanner, W.P., and Swets, J.P. A decision-making theory of
visual detection. Psychol. Rev. , 1954, 61^, 401-409.
Venables, P.H. Slowness in schizophrenia. In: Welford, A.
T., and Birren, J.E. (Eds.): Behavior, aging, and the
nervous system. Springfield: Charles C. Thomas, 1965,
598-612.
Verzar, R. Lectures on experimental gerontology.
Springfield: Charles C. Thomas, 1963.
Wagman, I.H., and Lesse, H. Maximum conduction velocities
of motor fibers of ulner nerve in human subjects of
various ages and sizes. J. Neurophysiol., 1952, 15,
235-244.
Wallace, J.G. Some studies of perception in relation to
age. Brit. J. Psychol., 1956, £7, 283-297.
Weiss, A.D. Sensory functions. In: Birren, J.E. (Ed.):
Handbook of aging and the individual. Chicago:
University <5f Chicago Press, 1959, 503-542.
Welford, A.T. Age and learning: Theory and needed
research. Experientia Supplementum, 1956, £, 136-143.
Welford, A.T. Psychomotor performance. In: Birren, J.E.
(Ed.): Handbook of aging and the individual. Chicago:
University of Chicago Press, 1959, 562-613.
105
Welford, A.T. Arousal, channel-capacity, and decision.
Nature, Lond., 1962, 194, 365.
Welford, A.T., Norris, A., and Shock, N.W. Movement time
and age: A preliminary report (presented at the Sixth
International Congress of Gerontology. Copenhagen,
1963.)
Welford, A.T. Performance, biological mechanisms and age:
A theoretical sketch. In: Welford, A.T., and Birren,
J.E. (Eds.): Behavior, aging, and the nervous system.
Springfield: Charles C. Thomas, 1965, 3-20.
Woodworth, R.S. Experimental psychology. New York: Henry
Holt and Co., 1938.
APPENDICES
106
APPENDIX A
MUSCULAR TENSION DATA
108
APPENDIX A
SMEDLY HANDGRIP DYNAMOMETER DATA FOR YOUNG SUBJECTS
EIGHTEEN TO TWENTY-SIX YEARS: HIGHEST HANDGRIP RECORDED,
HANDGRIP READING AT FIFTEEN AND THIRTY SECONDS, AND AMOUNT
OF WEIGHT USED FOR INDUCED MUSCULAR TENSION CONDITION
Subject
Number*
Highest
Handgrip
Recorded
(kg.)
Handgrip
After 15
Seconds
(kg.)
Handgrip
After 30
Seconds
(kg.)
Weight Used
For Tension
Condition
(lb.)
1 39 31 20 8.4
2 45 28 18 7.9
3 40 38 36 15. 8
4 60 58 49 21.6
5 34 22 12 5.3
6 39 27 24 10.4
7 39 23 18 7.9
8 29 19 16 7.0
9 37 28 25 11.0
10 45 36 28 12.3
11 28 22 22 9.7
; 12 34 24 22 9.7
13 39 35 26 11.4
14 44 39 31 13.6
15 41 28 15 6.6
16 24 19 6 2.6
17 22 20 16 3.5
18 19 13 9 4.0
19 24 19 12 5.3
20 30 22 15 6.6
21 26 17 9 4.0
22 27 19 11 4.8
23 24 16 8 3.5
24 24 18 15 6.6
25 31 24 18 7.9
26 35 26 19 8.4
27 22 14 10 4.4
28 25 20 12 5.3
29 18 12 9 4.0
30 21 18 10 4.4 i
*Subjects 1-15 were male and subjects 16-30 were female.
109
APPENDIX A
SMEDLY HANDGRIP DYNAMOMETER DATA FOR ELDERLY SUBJECTS
SIXTY-TWO TO EIGHTY-FIVE YEARS: HIGHEST HANDGRIP RECORDED,
HANDGRIP READING AT FIFTEEN AND THIRTY SECONDS, AND AMOUNT
OF WEIGHT USED FOR INDUCED MUSCULAR TENSION CONDITION
Subject
Number*
Highest
Handgrip
Recorded
(kg.)
Handgrip
After 15
Seconds
(kg.)
Handgrip
After 30
Seconds
(kg.)
Weight Used
For Tension
Condition
(lb.)
31 24 18 12 5.7
32 48 40 35 15.4
33 35 35 27 11.9
34 27 25 20 8.8
35 42 39 34 15.0
36 31 20 14 6.2
37 47 41 35 15.4
; 38 40 30 25 11.0
39 30 28 25 11.0
40 22 12 10 4.4
' 41 29 28 25 11.0
42 28 26 21 9.2
43 36 34 25 11.0
44 26 20 17 7.5
45 24 23 17 7.5
i 46 23 16 12 5.3
i 47 14 12 11 4.8
48 17 15 12 5.3
49 11 10 8 3.5
50 24 20 14 6.2
51 28 26 20 8.8
52 30 27 20 8.8
53 20 17 13 5.7
54 25 23 14 6.2
55 21 21 15 6.6
56 22 18 15 6.6
57 20 17 13 5.7
58 24 22 17 7.5
59 18 17 16 7.0
60 30 25 12 5.3
.♦Subjects 31-45 were male and subjects 46-60 were female.
APPENDIX B
COUNTER BALANCE ORDER OF PRESENTATION OF CONDITIONS
Ill
APPENDIX B
COUNTER BALANCED ORDER OF PRESENTATION OF CONDITIONS
Subject Numbers*
Young Elderly
Order of.
1
Presentation
2
of Conditions
3
1 53 C* N*
T*
9 57 C T N
10 54 T C N
2 58 T N C
5 59 N T C
3 51 N C T
4 60 C N T
6 55 C T N
11 52 T C N
7 61 T N C
8 62 N T c
14 63 N C T
30 80 C N T
26 76 C T N
28 81 T C N
31 79 T N C
29 92 N T C
32 87 N C T
33 77 C N T
34 84 C T N
27 78 T C N
| 35 88 T N C
i 41 82 N T C
! 37 85 N C T
i 12 86 C N T
! 38 56 C T N
39 64 T C N
13 89 T N C
! 15 90 N T C
; 40
]
65 N C T
*Subject numbers 1-15 and 31-45 were male and subject
numbers 16-30 and 46-60 were female. C = Control, N =
Noise, and T = Tension conditions.
APPENDIX C
INSTRUCTION
113
APPENDIX C
INSTRUCTIONS
This apparatus is called a psychomet. With it we
will measure your speed of reacting to different kinds of
light signals. You will notice that there are lights that
come on and buttons to touch. The buttons turn the lights
off. Your job will always be to turn off the lights as
quickly as you can by pressing the correct button. If you
make a mistake the light will stay on. Please correct your
mistakes immediately by touching the correct button. When
a light goes off another light will come on. Keep turning
them off as fast as you can until all ofv the lights are out.
Try not to make any errors. Do not stop at any point once
the series has begun. Please ask any questions you want at
the beginning of each series. Please use only one finger
to touch the buttons. Your arm should be free, not resting
on the chair or table, so you can respond quickly.
Trial Serial, two-cycle-tapping
First we will do a trial run in order to give you a
chance to see how the lights and buttons can be paired.
Two lights will come on and you will turn off each light by
114
pressing the button directly beneath it. Turn them off as
quickly as you can. When you are ready, press the start
button and begin.
Trial Series, Simple movement time
- In this series the lights will come on in regular
order from left to right and back from right to left, start
ing with the first button on the left. Turn off the lights
as quickly as you can. When you are ready, press the start
button and begin.
Trial Series, random movement time
How we will do a different combination, using it again
as a trial run. Instead of the lights coming on in regular
order, this time they will come on in a somewhat random
order. However, no matter what order they come on, the
button beneath each light is the one which turns off the
light. Turn off the lights as quickly as you can. When
you are ready, press the start button and begin.
Now that you have seen three examples of the pairing
of lights and buttons we will now begin the main series of
the experiment. Please let me emphasize that I am inter
ested in seeing how fast you can turn off the lights with
as few errors as possible. Are there any questions?
Two-Cycle Tapping
Now we will begin the first series and measure your
115
speed of responding to the numbers and lights. In the
first series we will do the exact same thing as we did in
the first trial run, that is the lights behind the five and
six will come on, alternating back and forth, and you will
turn them off as rapidly as you can by pressing the buttons
beneath the light that is on. Turn off the lights as
quickly as you can. When you are ready, press the start
button and begin.
Simple movement time
This series will be like the second trial series and
the lights will come on in regular order from left to right
and then right to left. Turn off the lights as quickly as
you can. When you are ready, press the start button and
begin.
Complexity level !E, 3.3 bits
Complexity level II, 6.6 bits
Complexity level III, 9.9 bits
This time you will match up the numbered light on the
top row with the corresponding numbered light and button at
the bottom. Turn the lights off as quickly as you can.
When you are ready, press the start button and begin.
APPENDIX D
ANALYSIS OF VARIANCE
117
APPENDIX D
ANALYSIS OF VARIANCE OF SERIAL REACTION
TIMES AS A FUNCTION OF AGE, SENSORY
CONDITIONS, AND TASK COMPLEXITY
Source of
Variation
Sums of
Squares df
Mean
Squares
F
Ratios
Between-Ss
Age (A) 343.8824 1 343.8824 61.66*
S/A 323.4909 58 5.5775
Within-Ss
Conditions (C) .07
Complexity (L) 6585.2092 1646.3023 2118.79*
AC .1611 2 .0806 .11
AL
51.0148 4 12.7537 16.41*
CL .1864 8 .0233 .13
ACL
1.1143 8 .1393 .77
SC/A 82.0160 116 .7070
SSL/A
180.2533 232 .7700
SCL/A 83.9889 464 .1810
*£ . 01
APPENDIX E
STIMULUS RESPONSE SEQUENCE MEANS FOR INDIVIDUAL
YOUNG AND ELDERLY SUBJECTS
119
TABLE I
STIMULUS-RESPONSE SEQUENCE MEANS IN SECONDS
FOR YOUNG SUBJECTS
Task Complexity
Subjects 2-cy-T1 S.M.T.2 3.3 Bits 6.6 Bits 9.9 Bits
01 .23 .23 .62 1.04 1.37
i p 3
.22 .24 .56 .96 1.24
N3 .25 .24 .57 .96 1.25
02 C .17 .18 .70 .88 1.05
T .15 .19 .55 .81 .99
N .14 .18 .47 .78 1.03
03 C .16 .20 .53 .87 1.14
T .17 .20 .55 .93 1.25
N .15 .18 .48 .85 1.05
04 C .16 .20 .73 .95 1.06
T .21 .27 .75 1.20 .94
N .18 .21 .65 .94 1.10
05 C .15 .23 .58 .78 1.14
T .16 .21 .53 .81 1.18
N .16 .30 .64 .86 1.22
06 C .16 .18 .47 .85 1.25
T .15 .17 .46 .85 1.24
N .17 .19 .62 .94 1.32
07 C .15 .22 .65 .95 1.34
T .14 .18 .49 .82 1.03
N .15 .21 .53 .86 1.18
1 2-cy-T = two cycle tapping.
2 S.M.T. = simple movement time.
3 C = Control, T = Tension, N = Noise.
120
TABLE I (Continued)
I
! Task Complexity
Subjects 2-cy-T S.M.T. 3.3 Bits 6.6 Bits 9.9 Bits
I
08 C .19 .28 .77 1.50 1.74
T .18 .29 .68 1.35 1.64
N .15 .32 .67 1.26 1.60
09 C .20 .24 .55 1.10 1.43
T .21 .25 .58 1.12 1.55
N .22 .24 .59 1.08 1.36
10 C .16 .24 .52 .82 1.18
T .15 .23 .54 .93 1.32
N .14 .23 .63 .91 1.30
11 C .15 .20 .49 .80 .95
T .14 .20 .47 .83 .95
N .15 .21 .58 .84 1.04
12 C .14 .18 .51 .88 1.16
T .13 .18 .47 .97 1.06
N .18 .20 .53 .97 1.30
13 C .23 .26 .69 1.06 1.28
T .22 .23 .68 1.04 1.16
N .25 .27 .70 1.07 1.41
14
Q
.16 .25 .56 1.92 1.27
T .17 .22 .60 1.01 1.40
N .17 .23 .51 1.02 1.31
15 C .18 .20 .62 .99 1.16
T .17 .23 .66 1.05 1.36
N .19 .23 .72 1.21 1.65
16 C .20 .25 .78 1.07 1.30
T .18 .19 .61 .89 1.08
N .18 .22 .66 .99 1.13
TABLE I (Continued)
121
Task Complexity
Subjects 2-cy-T S.M.T. 3.3 Bits 6.6 Bits 9.9 Bits
I
17 C .22 .24 .59 .80 1.14
T .23 .23 .54 .84 1.16
N .20 .27 .62 .94 1.19
18 C .19 .27 .61 1.04 1.25
T .26 .27 .62 1.13 1.31
N .18 .30 .55 .96 1.28
19 C .19 .26 .50 .87 1.07
T .20 .26 .62 .98 1.22
N .19 .24 .54 .93 1.20
20 C .16 .22 .61 .90 1.02
T .17 .22 .58 .89 1.15
N .17 .22 .61 .95 1.29
21 C .20 .24 .63 1.06 1.15
T .19 .24 .67 .94 1.20
N .24 .26 .75 1.04 1.27
22 C .20 .24 .59 1.00 1.19
T .19 .21 .48 .86 1.10
N .20 .24 .47 .87 1.09
23 C .22 .26 .69 1.22 1.77
T .23 .29 .69 1.13 1.53
N .20 .26 .62 1.09 1.34
24 C .21 .21 .49 .84 .99
T .18 .21 .58 .86 1.07
N .20 .22 .52 .80 1.02
25 C .21 .25 .61 .93 1.17
T .29 .28 .65 1.03 1.29
N .21 .25 .63 .87 1.19
122
TABLE I (Continued)
Task Complexity
Subjects 2-cy-T S.M.T. 3.3 Bits 6.6 Bits 9.9 Bits
26 C .18 .22 .65 .95 1.16
T .18 .22 .59 1.07 1.26
N .18 .22 .70 1.01 1.26
27 C .15 .21 .49 .81 1.03
T .14 .20 .43 .74 .87
N .18 .22 .55 .88 1.02
28 C .24 .31 .71 .93 1.40
T .33 .36 .83 .83 1.16
N .23 .26 .46 .84 1.19
29 C .18 .19 .59 .84 .95
T .22 .20 .61 .88 1.12
N .18 .20 .58 .83 .99
30 C .16 .19 .57 .90 1.17
T .17 .19 .54 .93 1.11
N .16 .19 .71 1.07 1.16
TABLE II
STIMULUS-RESPONSE SEQUENCE MEANS IN SECONDS
FOR ELDERLY SUBJECTS
Task Complexity
Subjects 2-cy-T-!- S.M.T.2 3.3 Bits 6.6 Bits 9.9 Bits
31 c|
.41 .41 .80 1.24 1.73
T .25 .29 .80 1.14 1.39
N3 .42 .34 .81 1.12 1.45
32 C .26 .27 .86 1.11 1.38
T .23 .27 .71 1.07 1.24
N .20 .24 .71 1.01 1.18
33 C .41 .40 .74 .90 1.14
T .47 .39 .68 1.04 1.37
N .34 .33 .67 .90 1.14
34 C .23 .25 .85 1.15 1.40
T .31 .32 .98 1.23 1.60
N .23 .27 .73 1.24 1.56
35 C .21 .22 .57 , i 86 1.15
T .27 .26 .61 .89 1.23
N .33 .29 .60 .98 1.37
36 C .26 .32 1.02 1.36 1.52
T .26 .30 1.08 1.22 1.65
N .44 .42 1.49 1.30 1.67
37 C 1.00 .84 1.42 1.70 2.34
T .27 .37 1.24 1.54 1.98
N .64 .48 1.18 1.65 2.03
1 2-cy-T = two cycle tapping.
2 S.M.T. = simple movement time.
3 C = Control, T = Tension, N = Noise.
TABLE II (Continued)
124
I
| Task Complexity
I
s
I Subjects 2-cy-T S.M.T. 3.3 Bits 6.6 Bits 9.9 Bits
38 C .37 .42 .75 1.28 1.55
T .27 .38 1.03 1.10 1.70
N .28 .33 .77 1.32 1.52
39 C .30 .32 1.01 1.34 1.71
T .34 .30 1.35 1.52 1.59
N .21 .27 1.12 1.18 1.67
40 C .25 .27 .80 1.02 1.36
T .33 .36 .88 1.09 1.42
N .23 / .32 .87 1.01 1.34
41 C .23 .28 .71 1.17 1.63
T .33 .32 .87 1.17 1.23
N .47 .43 1.42 1.26 1.55
42 C .26 .27 .68 1.05 1.54
T .28 .27 .63 1.04 1.42
N .33 .35 .72 1.10 1.59
43 C .29 .37 .95 1.27 2.16
T .35 .37 .76 1.15 1.90
N .30 .33 .80 1.23 1.58
44 C .23 .27 .83 1.18 1.50
T .42 .43 .83 1.20 1.60
N .24 .27 .80 1.15 1.40
45 C .24 .36 .93 1.10 1.39
T .35 .35 .76 1.15 1.54
N .31 .40 .74 1.28 1.81
46 C .20 .32 1.28 1.19 1.85
T .20 .30 .83 1.17 1.57
N .19 .32 .94 1.18 1.64
*’ ~v - ' " ( S i V v ® V ^ ® ■ ■ ' S ’ * , £ * ? ? { ? g >
S '
125
TABLE II (Continued)
Task Complexity
Subjects 2-cy-T S.M.T.- 3.3 Bits 6.6 Bits 9.9 Bits
47 C .38 .42 .95 1.23 1.70
T .36 . .38 .76 1.15 1.61
. N .36 .45 .69 1.07 1.46
48 C .22 .33 .81 1.07 1.38
T .24 .30 .78 1.20 1.63
N .21 .28 .79 1.22 1.41
49 C .23 .29 .70 1.17 1.38
T .30 .68 .85 1.18 1.54
N .28 .38 .76 1.12 1.42
50 C .19 .22 .74 .98 1.29
T .21 .24 .74 .98 1.47
N .24 .22 .73 1.10 1.40
51 C .35 .30 .67 .97 1.27
T .28 .26 .68 .95 1.16
N .38 .31 .77 1.06 1.24
52 C .50 .45 1.07 1.35 1. 88
T .36 .37 1.04 1.31 1.69
N .45 .46 .86 1.25 1.63
53 C .38 .35 .86 1.28 1.85
T .36 .35 1.14 1.17 1.70
N .36 .36 .80 1.24 1.61
54 C .42 .46 .85 1.18 1.53
T .62 .52 1.03 1.35 1.92
N .37 .40 .84 1.17 1.68
55 C .27 .25 .79 1.15 1.44
T .38 .27 .74 1.27 1.68
N .46 .28 .78 1.20 1.52
TABLE II (Continued)
Task Complexity
Subjects 2-cy-T S.M.T. 3.3 Bits 6.6 Bits 9.9 Bits
56
57
58
59
60
C
T
N
C
T
N
C
T
N
C
T
N
C
T
N
.27
.27
.31
.22
.25
.33
.21
.19
.19
.16
.18
.18
.32
.36
.41
.29
.31
.36
.32
.25
.32
.23
.22
.23
.22
125
.22
. 33
. 35
. 35
.79
.78
.93
.81
.65
.81
.59
.69
.72
.63
.67
.58
.70
.90
.77
1.26
1.25
1.22
.996
.96
1.02
1.04
.992
.99
.94
.96
.92
1.02
1.08
1.16
1.48
1.46
1.52
1.48
1.44
1.62
1.31
1.21
1.21
1.14
1.29
1.23
1.44
1.46
1.61

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Simple Auditory Reaction Time Of Young Adult And Elderly Subjects In Relation To Perceptual Deprivation And Signal-On Versus Signal-Off Conditions

PDF

Biofeedback Control Of The Eeg Alpha Rhythm And Its Effect On Reaction Time In The Young And Old

PDF

Electrophysiological Properties Of Reaction Time And Aging

PDF

Changes In Memory As A Function Of Age

PDF

Signal Processing Time And Aging: Age Differences In Backward Dichoptic Masking

PDF

Differential effects of epinephrine and propranolol on shuttle box avoidance learning in rats of different ages

PDF

Age, Sex, And Task Difficulty As Predictors Of Social Conformity: A Search For General Tendencies Of Conformity Behavior

PDF

Age Differences In Primary And Secondary Memory Processes

PDF

Monocular Acquisition And Interocular Transfer Of Two Types Of Discriminations In Normal And Corpus Callosally-Sectioned Guinea Pigs

PDF

Audience Effect And Age Differences On Tasks Of Increasing Complexity

PDF

The Effects Of Reinforcement Pattern, Intertrial Interval Regularity, And adaptation On The Orienting Response And Classical Gsr Discrimination Conditioning

PDF

Mental Imagery As A Function Of Muscular Tension And Suggestion

PDF

The Effect Of Discriminability On The Partial Reinforcement Effect In Human Gsr Conditioning

PDF

Conflicting Motives In The Prisoner'S Dilemma Game

PDF

Aggression As A Function Of Arousal And Friendship Ties

PDF

Influences Of Interoperative Experience And Age On Recovery Of Visual Function Following Two-Stage Lesions Of The Striate Cortex

PDF

Measuring Thought Process As An Ego Function In Schizophrenic, Mentally Retarded And Normal Adolescents By Means Of The Rorschach

PDF

Psychomotor Performance And Change In Cardiac Rate In Subjects Behaviorally Predisposed To Coronary Heart Disease

PDF

Verbal And Motor Mediation In Normal Children Of Two Ages And Retarded Individuals

PDF

The Effect Of Anxiety And Frustration On Muscular Tension Related To The Temporomandibular-Joint Syndrome

Asset Metadata

Creator Jeffrey, Dwight Wendell (author)

Core Title Age Differences In Serial Reaction Time As A Function Of Stimulus Complexity Under Conditions Of Noise And Muscular Tension

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, experimental

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Birren, James E. (committee chair), Cliff, Norman (committee member), Garwood, Victor P. (committee member), Szafran, Jacek (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-388076

Unique identifier UC11361371

Identifier 7008527.pdf (filename),usctheses-c18-388076 (legacy record id)

Legacy Identifier 7008527.pdf

Dmrecord 388076

Document Type Dissertation

Rights Jeffrey, Dwight Wendell

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA