Sleep onset as a function of auditory stimulation rates, response requirements, and novelty of the environment

Material Information

Sleep onset as a function of auditory stimulation rates, response requirements, and novelty of the environment
Walker, Paul Mallory, 1944-
Publication Date:
Copyright Date:
Physical Description:
vi, 57 leaves. : illus. ; 28 cm.


Subjects / Keywords:
Acoustic stimulation ( jstor )
Analysis of variance ( jstor )
Breathing ( jstor )
Electroencephalography ( jstor )
Habituation ( jstor )
Mental stimulation ( jstor )
Philosophical psychology ( jstor )
Psychology ( jstor )
Respiration ( jstor )
Sleep ( jstor )
Dissertations, Academic -- Psychology -- UF ( lcsh )
Human beings -- Effect of environment on ( lcsh )
Psychology thesis Ph. D ( lcsh )
Sleep ( lcsh )
bibliography ( marcgt )
non-fiction ( marcgt )


Thesis -- University of Florida.
Bibliography: leaves 54-56.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
000577415 ( AlephBibNum )
13978065 ( OCLC )
ADA5110 ( NOTIS )


This item has the following downloads:

Full Text

Sleeo Onset as a Function of Auditory Stimulation
Rates, Response Requirements, and
Novelty of the Environment






The author expresses deep gratitude to the Chairman of

his Supervisory Committee, Dr. C. Michael Levy, for invaluable

advice throughout this research. He is also deeply indebted

to Dr. Wilse B. Webb for his direction in this project and

assistance in implementing it. Grateful acknowledgement is

also extended to Dr. William Mendenhall, III, Dr. Henry S.

Pennypacker, Dr. Madelaine M. Ramey, and Dr. Paul Satz for

their suggestions as members of his committee. Special

appreciation is due members of the NASA sleep lab for their

technical assistance in data collection and to Fred Coolidge

and Jim Kollan for their help in the data analysis. The

authcr's deepest gratitude is extended to his wife, Barbara,

for her constant assistance and encouragement.



ACKNOWLEDGEMENTS . . . . . . . . ... ii

LIST OF TABLES. . . . . . . . . .. iv

ABSTRACT ... . . . . . . . .. v

INTRODUCTION . . . . . . . .... . 1

METHOD. . . . . . . . . ... ... .16

RESULTS . . . . . . . . ... . . .20

DISCUSSION. . . . . . . . . . .. .32

APPENDICES . . . . . . . . . ... 39

REFERENCES . . . . . . . . .. .. 54

BIOGRAPHICAL SKETCH . . . . . . . ... 537



Table Page

1 Summary of Analysis of Variance of
Time to Onset of Brief Sleep. . . . ... 22

2 Mean Latency of Onset of Brief and
Extended Sleep for the C x R Inter-
action. . . . . . . . . ... .. 23

3 Mean Latency of Onset of Brief and
Extended Sleep for the C x D Inter-
action. . . . . . .. . . . 24

4 Summary of Analysis of Variance of
Time to Onset of Extended Sleep . . .. 26

5 Distribution of Depth of Sleep Attained
for the C x R Interaction . .. . . 27

6 Distribution of Depth of Sleep Attained
for the C x D Interaction . . . . .. 28

7 Summary of Analysis of Variance for
Breathing Rates . . . .. . . . 30

8 Mean Breathing Rates. . . . . . . 31

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Paul Mallory walker

June, 1972

Chairman: C. Michael Levy
Major Department: Psychology

T'l.c purpose of this study was to examine latency of

sleep onset as a function of rate of auditory stimulation,

response requirement, and responses competing with sleep.

All stimuli were 500 Hz tones of 1 s duration and 60 db

intensity presented at a rate of 30, 15, or 5 per minute.

The response requirements were: none, count the tone, or

breathe in rhythm with the tone. With each S participating

i. The experiment for two consecutive days, two levels of

res:_onses comr-eing with sleep were obtained. Fifty-four

m.:le students in an introductory psychology course partici-

pated as s:. with six randomly assigned to each Condition x onL.vnation. After each S was wired for an EEG record-

.inj he iwas instructed to rest on his back and go to sleep

in :.e dark, soundproofed testing room. Sessions were 45

minutes long and conducted between 2:30-4:30 p.m.

Results indicated that with no response demands and

when Ss counted the tone, the latency of sleep onset for the

15 per minute rate tone was significantly shorter than for

the 30 or 5 per minute rates. Sleep onset was faster on

the second day an S participated in the experiment, and

breathing rates approximated tone rates only when Ss were

instructed to breathe with the tones. The curvilinear re-

lationship between tone rate and latency of sleep onset

indicated sleep promoted by monotonous stimulation is not

solely the result of habituation to the stimulus, but also

due to cortical inhibitory processes.


Slee is a natural, complex, generalized response to

both :r.Ler-. and external environmental cues. Anecdotally,

fan irm-or-rant environmental clue for sleep is a boring)

situatic:. (Ez.:eri:nental attempts to establish that monoto-

nous sensocry stimulation induces sleep have been infrequent

.rc. the iesulcs inconsistent.' Perhaps the earliest proponent

of the hypothesis was !Sidis (_C1.908)_,who claimed sleep resulted

from a reduction of the variability of sensory impulses,

rather th-an a reduction in amount of stir.u]rla n-joj /He pro-

duced behavioral sleep in auinea pigs, cats, ciogs, and

ch.iLdren Dy limiting bodily movements and providing either a

rono-o.-.:'3US click of a metronome or singing some ditty. In

a opetjtion of Sidis' e::perimrents, Kleitman (192S) concluded

im.n.o;ilizaticn, not the monotonous stimulus, was crucial in

.rrom.oti :, sleep. i2r-.obilization led tc muscular relaxation

,.:",ih '.e ccnsidere3 the sine qua non for sleep induction

(leitm-an, 1963, p. 1.6) Kleitman's statement that no

m .ono.r,:ou~ stimulation: was necessary was justified, but his

data dc not ailow hLim to state that monotonous stimulation

could not induce sleep, since he did not provide a monotonous

stimulation-immobilization control.

Coriat (1912) also concluded that muscular relaxation

was the prerequisite for sleep and monotonous stimulation

was unnecessary. When relaxed, his Ss fell asleep in about

15 minutes whether a monotonous tone was sounded or not.

When Ss were made to maintain muscular tension, no sleep

ensued even with a monotonous tone. His procedures made it

impossible for the tone to produce relaxed muscles, and

therefore left tenable the hypothesis that monotonous

stimulation induces sleep by aiding muscular relaxation.

Addressing this problem, Lovell and Morgan (1942) con-

cluded that monotonous stimulation did produce muscular re-

laxation and in this way mediated sleep. The monotonous

stimulation consisted of a 60 Hz tone sounded both faster and

slower than an S's breathing rate, ranging from 10 to 25 times

a minute. A significant rise in palrar skin resistance beyond

a basal resting level was the criterion for relaxation. All

S: were assured of the harmlessness of the experiment and

then seated in a chair for 10 to 15 minutes to obtain a

basal resting level. Ten control Ss remained undisturbed

in the chair for 10 additional minutes. Experimental Ss re-

mained in the chair for 20 additional minutes, receiving

the stii.ulation faster than their respiration rate during

one 10 minute period, and slower in the other period. Order

of rate presentation was counterbalanced. In addition,

half of the experimental Ss were told to relax, and the

remainder was given no such instructions. The results were

that respiration rates of experimental Ss approximated the

stimulus rate. Significant relaxation was observed, but

it was independent of tone rate. Control Ss showed no

change in respiration nor increased relaxation. Since the

group instructed to relax did not show any differences in

relaxation from the group having no such directives, Lovell

and Morgan regarded suggestion as inoperative. Nine out of

their 24 Ss fell asleep during one of the 10 minute periods

and all but three Ss admitted feelings of drowsiness. They

concluded that responses to monotonous stimulation, such

as changes in respiration rate, aid muscular relaxation and

in this manner mediate sleep.

It would be surprising if the experimental Ss did not

relax, since they were in the chair for at least 30 minutes.

Therefore, Lovell and Morgan's conclusions were based pri-

marily on the control group, which participated for only 20

minutes, showing no increase in relaxation. The difference

in the ex-primental and control groups, moreover, was not

limited to the presence or absence of monotonous stimulation.

It included the presence or absence of knowing what to

expect. The control group was never told what would occur in

the 10 minute interval and possibly remained alert waiting

for something to happen, rendering relaxation impossible.

Even if the control Ss had concluded nothing was going to

happen, after the baseline recording they were in the chair

for only 1C minutes and could not be expected to achieve

the same level of relaxation as the Ss who were able to relax

for 20 minutes. Because of their design, Lovell and Morgan

did not indicate whether experimental Ss showed increased

relaxation after 10 minutes, but only after 20. It is

possible that no S showed increased relaxation after 10

minutes. Because of these problems, the hypothesis that

monotcnou.- stimulation can aid relaxation, thereby media-

ting sleep, remained untested.

Relaxation studies not concerned with monotonous stimu-

laticn provide some information. Sleep has been shown to

depend upon the attainment of a specific level of relaxation

(Jacobson, 1933, p. 38; Kahn, Baker, & Weiss, 1968; Miller,

19i25). Hodes (1961, 1962) and Zung (1970) have attributed

t;e sleep to decreased proprioceptive stimulation of the

cortex as a result of decreased muscle tension. The purpose

of rela::inc is to reduce not only proprioceptive stimulation

Lf che ascending reticular activating system (ARAS), but

also ccrtical stimulation of the ARAS from reflections

(Jacobson, 1938, p. 38; 1962, p. 87). This suggests how

monotonous stimulation might induce sleep. For Ss who have

difficulty shutting off cortical excitatory impulses to the

APAS, a monotonous stimulus becomes a focal point. Attending

to this stimulus effectively shunts other stimuli, peripheral

and cortical. Gradually the monotonous stimulus loses its

ability to maintain wakefulness because it becomes a non-

informational, non-excitatory stimulus, and sleep ensues.

Pavlov (1928, p. 311) attributed a more active role to

the stimulus. He asserted that monotonous stimulation acted

upon an organism's central nervous system to induce sleep.

He observed that a CS presented without the UCS inevitably

led to drowsiness and sleep. His explanation was that repeti-

tive stimulation of any point in the cortex gives rise to

inhibition which irradiated over the cortex. Thus, sleep

was inhibition that had spread over the cerebrum, the

entire hemispheres, and even into the lower midbrain.

In series of experiments, Oswald (1959) investigated

transmarginal inhibition, a mechanism to protect an S from

excessively strong stimuli. In one experiment Ss rested

cn a couch in a normally lighted room for three sessions,

each lasting 15-30 minutes. The Ss were required to click

.. switch when they heard a tone initially set at the level

or the absolute threshold for loudness. The tones were

presented at regular rates of 6, 12, or 20 per minute. The

rate was constant within a single session. Two tendencies

were observed: (a) sleep and alertness appeared to alternate

regularly with the rate of the signals, and (b)there was a

general drift toward sleep. The level of sleep for Ss was

identified by referring to sample EEG records. Respiration

rates tended to synchronize with the rhythm of stimulation.

In a third experiment the Ss lay on a couch and moved

their hands and feet in time with recorded jazz or blues.

EEG signs of sleep appeared for long periods, even though

movement continued. Again, alertness and sleep fluctuated

with the rate of the rhythm. Sleep appeared in some instances

during loud music, but more frequently when a smooth piece

followed a violent selection. There was a tendency for

respiration to become regular and related to limb movements.

To answer criticisms that sleep would not have been

possible if Ss had not closed their eyes, Oswald (1960)

reported further experiments in which Ss had their eyes

glued open. For each S the entire environment was dominated

by one rhythm of stimulation. A 60 minute tape of non-stcp

blues music was played with brief 800 Hz tones pulsating

with each beat of the musical rhythm. Visual stimulation

from four 60 watt bulbs also pulsated with the rhythm, as

well as electric shocks to the leg of the S. The first

three volunteers rested on a couch, making no movements.

The first S was sleep deprived, confounding the effects of

rhythm on sleep latency. The second S participated in the

experiment between 9:00 and 10:00 p.m. and showed low

-cltage slow. waves after 8 minutes. A sleep latency of 8

minutes was also obtained for the third S, with the experi-

ment conducted in the afternoon. Information on Ss' laten-

cies in a quiet environment was not obtained, which would

be prerequisite for an evaluation of the effects of the

rhythm on sleep latency. Two additional Ss sat in a chair,

received no shocks, and moved arms and feet in time with the

music and lights. The first sign of slow wave EEG activity

was after 10 minutes for one S and 15 for the other. Epi-

sodes of light sleep alternated with wakefulness for each

S. Oswald's experiments clearly demonstrated that people

can go to sleep when subjected to repetitive monotonous

stimulation, but do they sleep because of or in spite of

s-ch stimulation? A control condition of no stimulation

would be necessary to answer this question. Oswald (1966,

p. 46) claimed this control was studied in an experiment in

Marseilles. Scme of 150 Ss underwent a 4 minute period of

monotonously recurring lights or noises and over a third of

them went to sleep. By contrast, others whD experienced a

comparable period of silent darkness hardly ever went to

sleep. It is impossible to evaluate this study because

these were the only details presented.

Gastaut and Bert (1961) interpreted habituation to re-

peated stimuli as a form of protective inhibition, and

attempted to study this inhibition by examining habituation

of the blocking of the alpha rhythm with repeated stimuli.

Each S participated in five experimental sessions of approxi-

mately 4 minutes each, except the first which was a 7 minute

rest period. This was followed by a period of habituation

of the alpha rhythm by visual stimuli, given for 4 s every

20 s with a 75 watt lamp placed 1.5 m from the S's eyes.

The third session consisted of habituation of blocking of

the alpha rhythm by auditory stimuli presented at the same

rate as the visual stimulus. Next came a period of mental

addition and subtraction of increasing difficulty. A final

rest period concluded the experiment. Each of the 91 Ss

received the experimental treatments as outlined. In addi-

tion, 11 Ss repeated the procedure each week for five to

seven weeks. A third group of Ss with no or poor alpha was

considered a control group. Gastaut and Bert's results

indicated that. (a) EEG signs of sleep occur most frequently

during periods of habituation of blocking of the alpha

rhythm, (b) there is a positive correlation between rapidity

of habituation of blocking and appearance of signs of sleep,

and (c) there is a facilitation of the habituation when the

experiment is repeated.

It is necessary to evaluate Gastaut and Bert's methods.

rThey gave only one clue to the meaning of "signs of sleep";

reporting the frequency of the occurrence of the stage of

spindles and K complexes, they indicated that all of their

signs of sleep, except for 12 instances, must be stage 1.

No statistical tests were presented and there was no proper

control group. The rest periods might have been a control

condition, but the same sequencing of events was used for

all Ss, and differences attributable to length of time in

the experimental situation could not be evaluated. Prior

to each rest period, the S was engaged in active behavior,

but each habituation period was preceded by rest, with or

without stimulation.

Statements that monotonous stimulation caused sleep,

based on studies she considered uncontrolled, led Tizard

(1966) to investigate the amounts of sleep recorded during

periods of auditory stimulation compared to equal silent

periods, instructional effects on the amount of sleep,

and the result of varying stimulus intensity. Each of

12 Ss was seen for two sessions at 6 to 7 day intervals,

always at 2:00 p.m., for three 8.5 minute periods.

Two of the periods contained auditory stimulation, and one

was a control period. In the beginning, Ss were told there

would be periods of intermittent sounds and others when

nothing would happen. They were instructed to keep their

eyes closed throughout. At the start of the control period,

Ss were told there would be no sounds and to go to sleep

if they liked. Before one stimulation period they were

asked to listen very carefully for the sounds and press a

response bulb in their hand each time they heard a stimulus.

Before a second stimulation period they were instructed to

ignore sounds and sleep if they liked. The order of pre-

sentation for the three periods was randomized with a 3 x 3

Latin Square design. Half of the Ss were given unpleasantly

loud sounds in the first session and quiet sounds in the

second, with this order reversed for the other half. In

order to avoid habituation to one tone, four different fre-

quencies (800, 900, 1,000, 1,100 Hz) were used in the four

stimulation periods. The duration of a stimulus was 4 s and

the interval between stimuli was 20 s.

The EEG record of each 500 s experimental period was

diivided into 10 s epochs and evidence of sleep (Loomis,

Harvey, & Hobart, 1937, used as classification) assessed

in the first and every subsequent fifth epoch, resulting

in 10 measurements. During stimulation periods epochs were

always chosen before a stimulus. There was no significant

difference in amount of sleep between the control Ss and

those instructed to ignore the sound, but more sleep was

recorded in both of these Ss than for the Ss who attended

to the tones. The only other significant finding was that

amounts of sleep increased across weeks. In the control

and ignore sounds conditions, eight Ss reached stage C sleep

during the first session and all 12 in the second session.

Tizard concluded that the Pavlovian theory of the sleep-

promoting effects of rhythmic stimuli was not supported, but

a monotonous environment did favor sleep.

Tizard's conclusion that monotonous stimulation per

se does not promote sleep is consistent with the hypothesis

that the S's response to the stimuli is crucial. When a

motor response was required, S was asked to "listen very

carefully" for the sound. In effect, this condition became

a signal detection task and was not likely to produce sleep

in 8.5 minutes. Implicit in the instruction to listen care-

fully is the instruction to stay awake. Thus, while con-

cluding sleep produced in studies using monotonous stimula-

tion was due to instructions, Tizard demonstrated that sleep

could also be prevented by instructional manipulations. In

addition to the tone orienting instructions, the response

of bulb pressing was a possible source of stimulation

maintaining wakefulness.

More recent evidence indicates monotonous stimulation

promotes sleep even with no response requirement for the S.

Bohlin (1971) investigated the effects on arousal and sleep

onset of a 1,000 Hz tone of 4 s duration and 80 db intensity.

presented with a mean inter-stimulus interval of 30 s,

varied within a range of 20-40 s. Thirty Ss were exposed

to each condition of no stimulation and monotonous stimula-

tion, with the order of conditions balanced and a week inter-

val between sessions to minimize transfer effects. The

Ss were told the experiment was a study of physiological

response to auditory stimulation, and no overt responses

were to be given. The possibility of sleep was acknowledged,

but there were no direct instructions to sleep. Mean

latency of sleep onset was determined by EEG records and

was significantly shorter for the monotonous stimulation

than the control condition.

In a second experiment, in addition to no stimulation,

three monotonous stimulation conditions were obtained by

using the same stimulus but with three inter-stimulus inter-

vals: 5-15 s with a mean of 10 s, 20-40 s with a 30 s mean,

and 50-90 s '.-ith a 65 s mean. Forty-eight Ss were randomly

assigned to each condition, and given the instructions used

for the first experiment. The mean latency of sleep onset

for all. stimulation rates was shorter than that for the no

simulation condition. Compared to the longest inter-stimulus

interval, the shortest produced a significantly shorter mean

latency of sleep onset.

In a series of experiments designed to examine the in-

fluence of a repeated tone on latency of sleep onset, Webb

and Agnew (1971) studied the following conditions: silence,

monotonous sound, low level tones, low level tones plus

counting, low level tones plus eye opening, and silence with

sleep deprivation. The monotonous sound was a recording of

a fan motor, and low level tones were 800 Hz tones of 70 db

intensity and 2 s duration with a 2 s interval between tones.

All Ss were asked to sleep, and those asked to count were

required to do so at the beginning and end of each tone.

The Ss in the eye blink condition were required to close their

eyes when the tone began and open them at the end of each

cone. Eight different Ss served in each condition for 5

consecutive days, with each session lasting 30 minutes. For

the sleep deprivation condition, eight Ss arrived at the lab

a- 11:00 p.m. the night before the experiment and were kept

awake until the experimental session at 10:00 or 11:00 a.m.

the next morning, which were the times all Ss attempted

sleep. Results indicated neither the stimulation conditions

nor silence could produce sleep latencies as short as the

sleep deprivation condition. Excluding sleep deprivation,

the tone only condition induced sleep fastest, significantly

shorter than silence and the eye blink condition. Latency

of sleep onset was reduced across days for all conditions

except silence, with most of the reduction occurring from

Day 1 to Day 2.

Many studies have attempted to establish that monotonous

stimulation promotes sleep, and in the majority of them the

data were called into question due to errors in experimental

design. The more recent studies by Bohlin (1971) and Webb

and Agnew (1971) corrected for the previous errors and pro-

vide evidence that a repeated tone can facilitate sleep onset.

These studies suggest the following variables are important

determinants of the latency of sleep onset under auditory

stimulation: the S's response to the stimulus, rate of

stimulus presentation, and whether or not the S has tried

to sleep with auditory stimulation previously. Although the

effects of a response requirement and rate change have been

investigated in these experiments, it is possible that the

optimal rate depends on an S's response. The purpose of the

present experimentt was to examine this possibility and

include a response which approximates what an S usually does

naturally in going to sleep, i.e., breathe regularly. Any

response requirement should eliminate irrelevant cognitive

responses, force the monotony of the repetitive tone on the S,

and in this way mediate sleep. Therefore, when irrelevant

competing responses are at a maximum, such as the first time

anS serves in the experiment, a condition with a response re-

quirement should facilitate sleep more than one with tones

only. With reduced competing responses, in this case the

second time the S participates in the experiment, there

should be no differences between conditions with or without

a response requirement. The present experiment also tested

these hypotheses.


Overview. A 3 x 3 x 2 factorial experiment with re-

peated measures on the third factor was used to investigate

effects of the rate of auditory stimulation on latency of

sleep onset, the S's response to stimuli, and responses

competing with sleep. The stimuli were 500 Hz tones of 1 s

duration and 60 db intensity, presented at a rate of 30

(R-30), 15 (R-15), or 5 (R-5) per minute. The response

conditions studied were: tone only with no response (T),

ccunt the tone (T-C), and breathe in with the tone (T-B).

with each S participating in the experiment for 2 conse-

cutive days, two levels of responses competing with sleep

were obtained.

Subjects. Male students enrolled in an introductory

psychology course participated as Ss. The first 54 to

volunteer were selected, and six were randomly assigned to

each condition x rate combination. Although two Ss appeared

at each of the three times 2:30 p.m., 3:15 p.m., and 4:30 p.m.,

the data '.'ere considered as six replications of each con-

dition :, rate x day combination. This was dcne to simplify

the analyses, since initial computations, blocking for time

of day, indicated no time effects were significant. When

they volunteered for the experiment, all Ss were instructed

to refrain from alcoholic or caffeinated beverages on the

days of the experiment and to avoid altering their usual

sleeping schedule for the 2 days preceding the experiment.

In addition, they were asked to eat lunch at least 2 hrs

before the time they would be attempting to sleep. The Ss

ranged from 18 to 27 yrs (mean = 19.87, SD = 1.48).

Procedure. The Ss arrived at the lab 30 minutes prior

to an experimental session, and electrode application and

other preparations provided a common presleep experience.

Each S was wired so that an EEG recording could be obtained

between F -F7 and 03-0z z' as determined by the International

10-20 System for electrode application. The Grass Model VI

EEG was calibrated for 7.5 mm = 50 /,cv on all channels. The

recordings were made at a paper speed of 30 mm/s. A mercury

strain gauge was connected around the S's abdomen and a Parks

:Model 270 Flethysmrograph monitored respiration rates on the EEG

output. An S lay supine on a bed in a dark, soundproofed

roor and was instructed to sleep. Each S was told to make

no more body movements than absolutely necessary, since these

interfered with the recordings. Tape-recorded supplemental

instructions appropriate to the experimental condition (see

Appendix A) were then played. Tones were presented con-

tinuously at the preassigned rate throughout the 45 minute

session, and speakers placed above and on each side of the

S insured a constant auditory intensity level regardless

of the S's orientation. Each S was awakened at the end of

the experimental period and given a questionnaire designed

to determine if he followed instructions (see Appendix B).

Identical procedures were followed for the second experi-

mental session.

After each experimental period, the questionnaires

were examined to insure that each S had followed instructions.

Of primary interest were items concerning any difficulty

going to sleep at home under normal conditions and recent

happenings which might have affected sleep onset. Two Ss

were dropped from the experiment because of positive re-

sponses to the latter question. Even though every person

told to breathe with the tone reported difficulty doing so,

answers indicated that all tried to follow the instructions.

All of the Ss in the T-C Condition reported counting, and

half of them related problems with the requirement.

Scoring. The measures of dependent variables were de-

rived from the EEG records which were evaluated using a

blind scoring technique. EEG records were first scored in

1 minute epochs for onset of brief sleep, i.e., less than

30 s of 8-12 Hz occipital activity per epoch. The 11 Ss

who failed to sleep were assigned the maximum latency possi-

ble, which was 45 minutes. Scoring reliability was assessed

by correlating independent ratings of 30 randomly selected

records evaluated by a second scorer. The Pearson r was


In some instances an epoch was classified as sleep and

shortly afterwards an S woke up and remained awake for the

entire session. Since this sleep response is different from

one where an S remains asleep, EEG records were also scored

for onset of extended sleep. The criterion for sleep onset

was identical to that used for brief sleep, but it was

applied only to the period before which an S went to sleep

and remained asleep for the duration of the experiment, or

to the first epoch when S reached stage 2 (Dement & Kleitman,

1957). With these criteria, no extended sleep period was

iesc than 7 minutes long. The 27 non-sleepers were evenly

disLtr~ib>-ed among conditions, and each was assigned the

maximum latency of 45 minutes. Inter-scorer reliability

was again high (r = .99).

Sleep records were then scored for deepest stage of

sleep attained. Since breathing rates would change with sleep

onset, they were scored only for the first 8 minutes. The

average rate ner minute was calculated for four 2 minute Blocks.


General. For onset of brief and extended sleep,

planned comparisons were made with a t test for the Condi-

tion x Rate (22 df) and Condition x Day (34 df) interactions.

The t value is always given first in parentheses, followed

by the alpha level when less than .10. When interaction

comparisons were not significant, significant F ratios

for the appropriate main effects were reported. The bound

on error of estimation, i.e., 2 SD, was computed when means

were significantly different.

Since analysis of variance procedures were inappropriate

for the depth of sleep measure, a contingency table was set

up with depth of sleep for rows and the Condition x Rate

combinations for columns. A Chi-squared statistic could not

be computed since the expected cell frequencies were too

small (see I.:endenhall, 1967, p. 262). A descriptive analysis

was presented for this interaction. For the Condition x Day

interaction, differences between Conditions were analyzed

separately for Days 1 and 2 using the.normal approximation

to the binomial. The proportions considered were Ss re-

maining awake and those reaching stage 2 or deeper.


The primary purpose of examining breathing rates was to

ascertain that Ss followed instructions. The significant

condition x Rate x Block interaction was investigated with

Tukey's post hoc comparison suggested by Kirk (1969).

Brief sleep. The analysis of variance for onset of

brief sleep is summarized in Table 1. For Condition T,

R-15 produced shorter latencies than both R-30 (2.46, .015)

and R-5 (1.81, .05). In the T-C Condition, R-15 was more

efficient in inducing sleep than R-5 (1.61, .035), but not

R-30 (1.24). R-15 was not different from R-30 (.52) or R-5

(.71) for T-B. Table 2 contains the means for these com-

parisons, and the bound on error for each was 7.97. Planned

comparisons for Conditions at each rate indicated T-C produced

shorter latencies than T for R-30 (1.45, .085), with no other

differences. No differences existed for any of the Condition

x Day interaction planned comparisons (see Table 3 for the

appropriate means). A reliably shorter period was required

for sleep onset on Day 2 (15.18 minutes) than on Day 1 (18.99

minutes). For Day means the bound on error was 2.37.

Since Bohlin's (1971) results indicated a linear rela-

tionship between latency of brief sleep onset and tone rate,

sums of squares for Rates were parcelled into components for

linear (SS = 13.92) and quadratic trends (SS = 1457.78).

The F ,45 values for the linear and quadratic trends were

Table 1

Summary of Analysis of Variance of
Time to Onset of Brief Sleep

Source SS df MS F P

Conditions (C) 283.40 2 141.70 C1

Rates (R) 1471.70 2 735.85 3.86 <.04

C x R 407.27 4 101.82 <1

Ss w gps 8575.57 45 190.57

Days (D) 392.89 1 392.89 5.11 <.04

C x D .02 2 .01 <1

R x D 20.92 2 10.46 <1

C x R x D 500.16 4 125.04 1.63

D x Ss w gps 3458.10 45 76.85





Table 2 Latency of Onset of Brief and Extended
Sleep for the C x R Interaction


_S R-30 R-15 R-5

24.87* 11.01 21.18
29.19** 17.94 27.89

16.68 9.72 18.77
24.89 21.09 34.34

17.80 14.87 18.87
23.41 24.37 24.96

* Mean nu;i-ber of minutes to onset of

**lean nu:iber of minutes to onset of

brief sleep.

extended sleep.

Table 3

Mean Latency of Onset of Brief and
Extended Sleep for the C x D Interaction


Conditions 1 2

T 20.92* 17.11
30.40** 19.61

T-C 16.94 13.17
25.95 27.59

T-B 19.12 15.25
25.70 20.79

* Mean number of minutes to onset of brief sleep.

'**can number of minutes to onset of extended sleep.

1 and 7.65 (p <.01), respectively.

Extended sleep. The analysis of variance for extended

sleep latencies is summarized in Table 4. For the T Con-

dition, R-15 produced shorter sleep latencies than R-30

(1.59, .065) and R-5 (1.41, .09). In the T-C Condition,

R-15 induced sleep faster than R-5 (1.89, .04), but not

R-30 (.54). Neither R-30 (.14) nor R-5 (.08) was different

from R-15 in the T-B Condition. A study for Condition

effects at each rate indicated no significant differences.

Condition x Rate means for this variable are also presented

in Table 2, and the bound on error for each was 9.69.

Examination of the Condition x Day interaction indicated a

significant reduction in sleep latencies from the first to

the second Days for the T Condition (1.87, .035), with no

differences between Days for the T-C Condition (.28) or T-B

Condition (1.85). No differences existed between Conditions

on either day. Table 3 contains the means for this inter-

action, each having a bound on error of 5.36.

Depth of sleep. Table 5 contains the deepest stage of

sleep attained for the Condition x Rate interaction and

Table 6 for the Condition x Day treatments. Examination of

Table .5 implies that R-15 was more sleep inducing than R-30

and R-5 for the T Condition, and than R-5 for the T-C

Condition. There is no statistical evidence to support

Table 4

Summary of Analysis of Variance of
Time to Onset of Extended Sleep

Source SS df MS F P



C x R

Ss w gps



Rx D

C x Rx D

D x Ss w gps






















< .03






Table 5

Distribution of Depth of Sleep Attained
for the C x R Interaction


Deepest Sleep Rates Rates Rates
Stage Attained R-30 R-15 R-5 R-30 R-15 R-5 R-30 R-15 R-5

0 3* 0 0 0 0 2 1 0 2

1 4 2 4 3 2 6 3 5 1

2-4 5 10 5 9 10 4 8 7 9

*Entries are number of Ss.

Table 6

Distribution of Depth of Sleep Attained
for the C x D Interaction


1 2
Conditions Stage 0 Stages 204 Stage 0 Stages 2-4

T 3* 7 3 13

T-C 1 12 1 11

T-B 2 10 1 14

*Entries are number of Ss reaching the given stage, but no

this implication for this dependent variable, but it is in

agreement with the findings for brief and extended sleep.

For the Condition x Day interaction, on Day 1 more Ss in

the Conditions requiring a response (T-C, T-B) reached

stage 2 or deeper sleep than those in Condition T (z = 1.55,

p <.06). This comparison was not significant for Day 2

(z = .21), nor were similar comparisons for the percentages

of Ss remaining awake on Day 1 (z = .92) and Day 2 (z = 1.31).

Breathing rates. The analysis of variance for breathing

rates is summarized in Table 7. Tukey's post hoc comparison

of the significant Condition x Rate x Block interaction

indicated the following differences at the .05 level (see

Table 8 for appropriate means). The Ss in T-B with R-30

maintained breathing rates faster than both the groups told

to breathe at slower rates, and those receiving R-30 but

not in T-B. These differences were consistent for the four

Blocks. For R-5, with Condition T-B breathing rates were re-

duced compared to T and T-C for the first two Blocks only.

Breathing races were not different for the three instructional

Conditions for R-15. For Condition T-B with R-30, breathing

rates for Blocks 3 and 4 were significantly slower than for

Block 1. For every other Condition x Rate combination,

breathing rate changes across Blocks were not significant.

Table 7

Summary of Analysis of variance
for Breathing Rates

Source SS df Ms F

C 181.38 2 90.69 <1

R 1430.72 2 715.36 5.88

C x R 3685.88 4 921.47 7.57

Ss w gps 5478.48 45 121.74

D 44.40 1 44.40 2.76

C x D 8.23 2 4.12 <1

R x D 7.72 2 3.86 <1

C x R x D 27.06 4 6.77 <1

D x Ss w gps 724.33 45 16.10

Blocks (B) 4.71 3 1.57 <1

C x B 51.88 6 8.65 2.09

R x B 63.51 6 10.58 2.56

C x R x 5 143.59 12 11.97 2.89*

B x Ss w cps 559.28 135 4.14

D x B 4.33 3 1.44 <1

C x D x B 3.05 6 .51 <1

R x D x B 2.63 6 .44 <1

C x R x D x B 6.50 12 .54 <1

D x x Ss w qps 407.81 135 3.02
'Significant at .002 level.

Table 8

Mean Breathing Rates


Conditions Rates 1 2 3 4

















































Conditions x Rates. For the three sleep measures, the

optimal rate for sleep onset depended on the response re-

quirement, indicating a Condition x Rate interaction. This

was due to the fact that breathing with the tone was so

difficult that the Rate effect was not significant in this

Condition. For the T and T-C Conditions, R-15 was the

optimal rate for each measure of sleep. Although T-C tended

to be better than T for R-30, the Condition x Rate comparisons

indicated that it is more appropriate to discuss these main

effects than an interaction, since there is an optimal rate

to facilitate sleep, but its influence is minimal with a

difficult response requirement.

Rates. The relationship between latency of brief sleep

onset and tone rate was curvilinear. A medium rate tone facili-

tated sleep onset. A very rapidly presented tone produced ex-

cessive sensory and probably cortical stimulation, since Ss

reported the tone annoyed them. At a slow rate, silence was

approached and the tone ceased to serve as the same type of

monotonous stimulus. The absolute silence of a soundproofed

room produced a monotonous environment relative to the "silence"


an S experiences in his own bedroom and this abnormal lack of

sensory stimulation can impede sleep onset.

The results of the rate manipulation confirm Bohlin's

(1971) findings that this variable is an important determinant

of sleep onset but do not support his conclusion that sleep

onset and tone rate are linearly related. However, the

rates he investigated were all slower than R-5, the slowest

used in the present experiment; thus, the results of the

two studies are not entirely comparable.

Conditions x Days. This interaction tested the hypo-

thesis that requiring an S to respond to a repetitive tone

would reduce the effect of distracting environmental and

cortical stimuli which prevent sleep. On Day 1, the novelty

of the experimental situation provided maximum distractions

but the effects of these competing stimuli were reduced on

Day 2. The hypothesis tested had two parts. The first was

that on Day 1 T-C and T-B would produce shorter sleep

latencies than T. For the brief and extended sleep measures,

Day 1 latencies tended to be shorter for T-C and T-B than

for the T Condition, although these differences were not

statistically significant. For depth of sleep, the T Con-

dition tended to produce more non-sleepers than T-C and T-B,

but these differences were not reliable. However, a signifi-

cantly greater number of Ss in T-C and T-B reached deep sleep

than those in the T Condition. Therefore, the first time S

participated in the experiment, a response requirement reduced

the effects of distracting stimuli for only one sleep measure.

The second part of the hypothesis tested was that only

in Condition T would a significant reduction in sleep laten-

cies occur from Day 1 to Day 2. For brief sleep, shorter

latencies were observed on Day 2 than Day 1 for all con-

ditions, but for extended sleep, only in Condition T was

there a reduction in latency from Day 1 to Day 2. The ex-

tended sleep results indicated the T Condition was not

adequate to reduce the effects of distracting stimuli pre-

sent on Day 1, but the T-C and T-B Conditions minimized the

first Day effect so that no differences in sleep latencies

occurred between Day 1 and Day 2. Therefore evidence exists

for a Condition x Day interaction. The effects were weak

and probably would have been stronger if a situation were

created where distractions associated with a response re-

quirement were minimal compared to those of the experimental

situation. As it was, responses competing with sleep

generated by the novelty of the environment were minimal

compared to the distractions created by requiring a response.

Conditions. The weakness of the first Day effect

partially accounts for the failure to find Condition dif-

ferences. An additional explanation is with the T-C

Condition the S placed more emphasis on the instruction to

count than to sleep, and had difficulty releasing himself

from the response requirement. An additional study (see

Appendix C) indicated the T-C Condition could not produce

shorter sleep latencies when the S was experimentally re-

leased from the response requirement by terminating the tone

after the experiment was well under way. Therefore, the

release problem cannot be avoided.

The purpose of the T-B Condition was to facilitate sleep

onset through relaxation. Lovell and Morgan (1942) observed

increased relaxation when respiration rate increased or de-

creased. Since respiration rates become slower in sleep

(Oswald, 1962, p. 170), the slowest rate for T-B required S

to approximate one component of the complex sleep response.

The difficulties in maintaining extremely slow and fast rates

cancelled any advantage of the regulated breathing. Failure

for the T-B Condition to affect sleep onset was not due to

Ss' ignoring instructions. The significant Condition x Rate

x Block interaction for breathing rates indicated Ss main-

tained the required breathing rates until sleep onset ap-

proached. The possibility remains that regulated breathing

may facilitate sleep onset when the tone rate is more

compatible with this response. But this is unlikely since

R-15, a rate 1 or 2 units per minute slower than normal

breathing, tended to increase latency of sleep onset for

T-B compared to Conditions T and T-C.

Breathing rates. Although the primary purpose of

examining breathing rates was to insure an S followed in-

structions, a secondary reason was to determine if tone rate

influenced breathing independent of the instructions.

Rates for the T and T-C Conditions did not affect breath-

ing rates. The extremeness of R-30 and R-5 probably pro-

duced the failure to replicate Lovell and Morgan's (1942) -

findings that breathing rates approximate tone rate. In

no instance was the rate of their stimuli more than 10

units different from the .normal breathing rate. Both of

the extreme rates used in the present experiment were more

than 10 units different from normal respiration rates.

Supplementing the previous findings, the present results

on breathing indicate an upper and lower limit beyond which.

tone rate cannot be changed and still influence breathing.

Conclusions. There is evidence that a repetitive tone

promotes sleep (Bohlin, 1971; Webb & Agnew, 1971), but the

effects can be attenuated by instructions, since significantly

longer latencies for tones than for silence were recorded

when Ss were required to press a response bulb (Tizard, 1966)

or blink their eyes (Webb & Agnew, 1971) with the tones.

The present experiment and Webb and Agnew's (1971) indicate

a response more compatible with sleep, such as counting the

tone, can enhance the effects of the monotonous stimulus.

Bohlin (1971) concluded that tone rate was an important

determinant of sleep latency, but the present study revealed

rate control is weak, since its effects are diminished

with a difficult response requirement. The Day effect

observed in this and Webb and Agnew's (1971) study cannot

be considered strong either, since its control over sleep

onset is abated with a proper response requirement. When

compared to sleep need (Webb & Agnew, 1971), the three

variables manipulated in this experiment exercise minimal

control over sleep onset.

The curvilinear relationship between rates and latency

of sleep onset suggests a mechanism for sleep promoted by

repetitive sensory stimuli. Such sleep is not solely the

result of habituation to the stimulus, since the more rapid

the frequency of stimulation, the more rapid is habituation

(Thompson & Spencer, 1966). Habituation to the tone would

have been fastest for R-30, but latency of sleep onset was

shorter for R-15. This does not imply that habituation

is not important for sleep onset. It is possible that

habituation to repetitive stimuli triggers cortical inhibi-

tory processes. Thus, a time interval between habituation

to a stimulus and sleep onset exists and corresponds to

the time necessary for inhibitory processes to take over and

produce sleep. Any unfavorable attitude on the S's part would

interfere with this process. Concerning the localization of

inhibitory systems, Moruzzi (1960, 1963) concluded a

hypnogenic center for monotonous stimulation was located

in the area of the nucleus of the solitary tract. He sug-

gested these synchronizing structures could be driven by

repetitive sensory stimuli to reduce EEG arousal and lead to

drowsiness by blocking successively the phasic response and

tonic activity of reticular neurons. Although these

mechanisms are complex and the process can be interrupted

by an S's attitudes, there is evidence that monotonous

sensory stimuli do promote sleep.





It has been found that a repeated tone will help a per-

son get to sleep. We are trying to get an accurate measure

of the efficiency of this condition, and we are measuring

your sleep with the electrodes. Try to make no more body

movements than are absolutely necessary, since they will

interfere with the recordings. When you begin to fall

asleep, the volume of the tones will be gradually reduced.

Your primary task is to go to sleep. We will wake you at

the end of the experimental period.

Tone Count

It has been found that if a person counts a repeated

tone to himself for a little while, it will help him get to

sleep. We are trying to get an accurate measure of the

efficiency of this condition, and we are measuring your sleep

with the electrodes. Try to make no more body movements than

are absolutely necessary, since they will interfere with

the recordings. When you begin to fall asleep the volume

of the tones will be gradually reduced. You are to count

each tone to yourself until we reduce the volume of the

tones. If you should feel drowsy before we reduce the volume

of the tcnes, stop counting and go on to sleep. Your

primary task is to go to sleep. We will wake you at the

end of the experimental period.

Tone Breathe

It has been found that if a person breathes with a

repeated tone, it will help him get to sleep. We are trying

to get an accurate measure of the efficiency of this condition,

and we are measuring your sleep with the electrodes. Try

to make no more body movements than are absolutely necessary,

since they will interfere with the recordings. When you

begin to fall asleep, the volume of the tones will be

gradually reduced. You are to breathe in with the beginning

of each tone until we reduce the volume of the tones. If

-ou should feel drowsy before we reduce the volume of the

tcnes, stop thinking of breathing, and go on to sleep. Your

primary task is to go to sleep. We will wake you at the

end of the experimental period.



Day 1


Circle T for True or F for False.

I usually have little difficulty falling asleep. T F

I fell asleep in today's experiment. T F

I fell asleep faster than usual in today's
experiment. T F

I prefer a quiet room for sleeping. T F

I usually lie awake a long time before falling
asleep. T F

I prefer some background noise for sleeping.

If T for above, specify type of noise.

Something happened to me recently which affected
how fast I fell asleep today. T F

If T for above, explain.

Tone Count

The following questions were added to those asked for

the tone condition:

I had difficulty counting the tones. T F

The last number I counted was

I lost count and began counting again. T F

If T for above, number of times was

If T for above, last number counted each time

Tone Breathe

The following questions were added to those asked for

the tone condition:

I had difficulty breathing with the tones. T F

I quit trying to breathe with the tones before
the tones were faded out. T F

Day 2


Circle T for True or F for False.

I fell asleep in today's experiment. T F

I fell asleep faster than usual in today's
experiment. T F

Something happened to me recently which
affected how fast I fell asleep today. T F

If T for above, explain.

Tone Count

In addition to the questions asked for the tone condition

-n Day 2, those used on Day 1 for the tone count group were

also included.


Tone Breathe

In addition to The questions asked for the tone

condition on Day 2, those used on Day 1 for the tone breathe

group were also included.



The purpose of Study 2 was to determine if the T-C

Condition could be improved by experimentally releasing an

S from his response requirement. The Ss were 16 alpha

dominant male students in an introductory psychology course

between 20-27 yrs of age. They were divided equally into

No-Release and Release groups. The No-Release Ss were in-

structed to count the tone until they became drowsy, and

then stop counting and go on to sleep. Each Release S was

told the E would be observing his EEG ana would turn off the

tones when S became drowsy. He was also instructed to count

the tones, but only until they were silenced or he became

drowsy, whichever occurred first. General experimental pro-

cedures were identical to those used for T-C and R-15 on

Day 1 with the following exceptions: (a) respiration rates

were not recorded, and (b) for the No-Release Ss the tones

were turned off when S became drowsy, i.e., alpha dis-

appeared from his EEG for 15 s straight. Results indicated

no differences in sleep onset latencies (t14 = .24) between

No-Release Ss (mean = 20.50, SD = 11.40) and Release Ss

(mean = 21.79, SD = 9.54). Therefore, the T-C Condition

could not be improved by experimentally releasing an S

from the counting requirement.



Sleep Breathing Blocks
Conditions Rates Brief Extended Depth 1 2 3 4

R-30 16.50 22.16 4 17.5 17.5 19.0 19.5
34.00 36.33 1 12.5 14.5 16.0 16.0
16.66 34.33 1 11.5 09.5 10.0 12.0
23.00 23.00 4 17.5 14.0 10.0 14.0
45.00 45.00 0 15.5 16.0 15.6 13.0
45.00 45.00 0 16.0 16.5 16.0 18.0

T R-15 17.66 29.83 2 11.0 11.5 13.0 10.0
03.50 03.50 4 10.0 13.0 13.5 14.0
12.83 45.00 1 16.0 16.0 17.5 17.5
10.16 10.16 4 14.5 16.0 15.5 14.0
15.33 35.50 2 19.5 21.5 22.5 22.5
04.50 04.50 4 14.0 13.0 14.5 11.5

R-5 21.00 21.00 1 16.5 16.5 16.0 16.5
09.50 45.00 1 27.5 25.0 23.5 24.5
45.00 45.00 0 20.0 22.5 20.0 19.5
25.66 32.00 1 10.0 09.5 09.5 09.0
18.16 45.00 1 18.0 18.5 18.0 19.5
13.16 25.00 2 05.5 11.0 11.0 10.5

R-30 20.16 30.50 2 22.5 19.5 22.0 21.5
19.83 19.83 4 06.5 08.0 08.5 11.5
12.66 12.66 4 19.5 18.5 17.5 18.0
42.83 45.00 1 17.0 17.0 14.5 16.5
06.33 06.33 2 16.5 16.5 18.0 20.5
18.66 18.66 4 17.0 15.5 18.0 17.5

T-C R-15 13.00 15.50 2 13.5 15.5 14.0 13.5
08.00 08.00 2 17.0 15.5 16.5 15.0
12.00 21.83 4 21.5 19.0 19.0 16.5
12.83 45.00 1 20.0 18.5 19.0 17.0
15.66 15.66 4 15.5 18.5 19.0 19.5
14.16 14.16 4 11.5 10.5 10.0 09.0

R-5 19.66 19.66 2 14.5 13.5 12.0 14.0
22.16 36.50 1 12.0 09.5 09.5 13.0
06.50 32.33 2 23.5 23.5 23.0 23.0
05.33 35.50 1 20.0 21.5 19.0 20.0
10.16 45.00 1 17.5 17.5 20.5 19.5
45.00 45.00 0 11.0 11.5 10.5 10.5

Day 1 (Continued)

Conditions Rates

Brief Extended


Breathing Blocks
2 3 4

R-30 18.00

R-15 26.16



















Day 2

Sleep Breathing Blocks
Conditions Rates Brief Extended Depth 1 2 3 4

R-30 18.83 45.00 1 19.0 22.5 20.5 20.5
13.16 13.16 1 12.5 14.0 12.5 15.5
18.50 18.50 4 13.0 1.2.0 12.0 14.0
13.66 13.66 4 16.0 15.5 14.5 14.0
45.00 45.00 0 16.0 16.5 16.5 16.0
09.16 09.16 2 14.0 13.0 15.0 16.5

T R-15 26.33 45.00 1 11.0 13.5 13.5 11.5
05.33 05.33 4 13.5 13.0 13.0 14.0
06.33 06.33 2 17.0 18.0 19.0 18.5
10.83 10.83 4 12.5 18.0 20.0 19.5
13.50 13.50 2 20.5 18.5 21.0 22.0
05.83 05.83 2 13.5 13.5 12.5 12.5

R-5 05.16 05.16 2 17.5 19.0 19.5 19.0
07.50 07.50 2 26.0 22.0 22.5 22.0
45.00 45.00 0 20.5 20.5 18.5 18.5
45.00 45.00 0 07.0 09.5 08.5 12.5
11.00 11.00 2 16.0 19.5 18.5 19.5
08.00 08.00 2 10.0 12.0 14.5 14.5

R-30 09.33 22.66 2 23.5 23.5 23.0 21.0
22.16 45.00 1 10.0 13.0 15.0 17.0
08.83 08.83 4 20.0 18.0 17.0 16.0
20.16 20.16 2 14.5 17.0 15.0 17.0
10.50 24.00 4 15.5 15.0 19.0 22.5
08.66 45.00 1 16.0 14.5 14.5 14.0

T-C R-15 09.50 14.83 2 15.0 15.0 15.0 15.5
05.16 05.16 4 17.5 17.5 16.0 15.0
06.33 26.16 4 18.5 17.5 17.0 15.5
09.66 45.00 1 22.5 23.5 24.0 24.0
07.16 38.66 2 17.5 17.0 16.0 18.0
03.16 03.16 4 16.5 17.5 17.0 16.5

R-5 17.16 17.16 4 15.0 15.0 16.5 17.5
24.33 24.33 2 14.0 15.5 16.0 17.0
06.83 27.50 1 23.5 22.5 23.0 23.0
18.33 39.16 1 18.5 19.0 19.5 19.0
04.83 45.00 1 19.5 19.5 20.0 19.0
45.00 45.00 0 10.5 09.0 09.5 09.5

Day 2 (Continued)

Conditions Rates

Brief Extended

Breathing Blocks


1 2

3 4

R-30 05.33

R-15 14.33




















Bohlin, G. Monotonous stimulation, sleep onset and habitua-
tion of the orienting reaction. Electroencephalography
and Clinical Neurophysiology, 1971, 31, 593-601.

Coriat, I. H. The nature of sleep. Journal of Abnormal
Psychology, 1912, 6, 329-367.

Dement, W., & Kleitman, N. Cyclic variations in EEG during
sleep and their relation to eye movements, body motility,
and dreaming. Electroencephaloqraphy and Clinical Neuro-
physiology, 1957, 9, 673-690.

Gastaut, H., & Bert, J. Electroencephalographic detection of
sleep induced by repetitive sensory stimuli. In G. E. W.
Wolstenholnie, and M. O'Connor (Eds.), Ciba foundation
symposium on the nature of sleep. London: J. and A.
Churchill, 1961.

Hodes, R. Electrocortical synchronization (ECS) in cats from
reduction of proprioceptive drive caused by a muscle
relaxant (Flaxedil). Federation Proceedings, 1961, 20,

Hodes, P. Electrocortical synchronization resulting from
reduced proprioceptive drive caused by neuromuscular
blocking agents. Electroencephalography and Clinical
Neurophysiology, 1962, 14, 220-232.

Jacobson, E. You can sleeo well. New York:' Whittlesey
House, 1938.

Jacobson, E. You must relax. New York: McGraw-Hill, 1962.

Kahn, M., Baker, B. L., & Weiss, J. M. Treatment of insomnia
by relaxation training. Journal of Abnormal Psychology,
1968, 73, 556-558.

Kirk, R. Experimental design: Procedures for the behavioral
sciences. Belmont, California: Brooks/Cole Publishing
Company, 1969.

ileitman, N. Studies on the physiology of sleep. V. Some
experiments on puppies. American Journal of Physiology,
1928, 84, 386-395.

Kleitman, N. Sleep and wakefulness. Chicago: University of
Chicago Press, 1963.

Loomis, A. L., Harvey, E. N., & Hobart, G. A. Cerebral
states during sleep, as studied by human brain poten-
tials. Journal of Experimental Psychology, 1937, 21,

Lovell, G. D., & Morgan, J. J. B. Physiological and motor
responses to a regular recurring sound: A study in
monotony. Journal of Experimental Psychology, 1942,
30, 435-451.

Mendenhall, W. Introduction to probability and statistics.
(2nd ed.) Belmont, California: Wadsworth Publishing
Company, Inc., 1967.

Miller, M. Changes in the response to electric shock pro-
duced by varying muscular conditions. Journal of
Experimental Psychology, 1926, 9, 26-44.

Moruzzi, G. Synchronizing influences of the brain stem and
the inhibitory mechanisms underlying the production of
sleep by sensory stimulation. Electroencephalography
and Clinical Neurophysiolovg Supplement 13, 1960,

Moruzzi, G. Active processes in the brain stem during sleep.
Harvey Lecture Series, 1963, 58, 233-297.

Oswald, I. Experimental studies of rhythm, anxiety, and
cerebral vigilance. Journal of Mental Science, 1959,
105, 269-294.

Oswald, I. Falling asleep open-eyed during intense rhythmic
stimulation. British Medical Journal, 1960, 1, 1450-

Oswald, I. Sleeping and waking. Amsterdam: Elsevier, 1962.

Oswald, I. Sleep. Baltimore: Penguin Books, 1966.

Pavlov, I. P. Lectures on conditioned reflexes. New York:
International, 1928.

Sidis, B. An experimental study of sleep. Journal of Ab-
normal Psychology, 1908, 3, 1-32, 63-96, and 170-207.

Thompson, R. F., & Spencer, W. A. Habituation: A model
phenomenon for the study of neuronal substrates of be-
havior. Psychological Review, 1966, 73, 16-43.

Tizard, B. Repetitive auditory stimuli and the development
of sleep. Electroencephalography and Clinical Neuro-
physiology, 1966, 20, 112-121.

Webb, W. B., & Agnew, H. W. Stimulus control of sleep onset.
Technical Report No. 4, October 15, 1971, University of
Florida, Subcontract N0014-70-C-0350, Office of Naval

Zung, W. W. K. Insomnia and disordered sleep: A laboratory
approach. In E. Hartmann (Ed.), Sleep and dreaming.
Boston: Little, Brown, 1970.


Paul Mallory Walker was born in Pensacola, Florida,

on April 22, 1944. After graduation from Pensacola Catholic

High School in 1962, he entered the Society of Jesus. His

academic work for this period consisted of part-time studies

at Spring Hill College in Mobile, Alabama. In 1966, he

withdrew from the Society of Jesus, and at that time entered

the University of Florida, from which he obtained a Bachelor

of Arts degree in psychology in 1967. Since that time, he

has been enrolled in the Graduate School at the University

of Florida. He received the Master of Arts degree in

psychology in 1969.

He is married to the former Barbara A. Black of Columbia,

South Carolina. He is a second lieutenant in the army

reserves, a member of Psi Chi, and student affiliate of the

Am:erican Psychological Associaticn.

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

C. Michael Levy', Chairran
Associate Professor of Psychology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

Wilse B. Webb
Graduate Research Professor

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree doctor of Philosophy.

Henry S. Penny acker
Professor of Psychology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

Paul Satz
Professor of Psychology

I certify that I have read this study and that in my'
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

Madelaine I1. Rarmo v
Assistant Professor of Psychology

I certify that I have read this study. and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

William .me1 enhall, III
Professor and Chairman of Statistics

This dissertation was submitted to the Department of
Psychology in the College of Arts and Sciences and to the
Graduate Council, and '..'as accepted as partial fulfillment of
the requirements for the degree of Doctor of Philosophy.

June, 1972

Dean, Graduate School