Introduction
Interest In Sleep
The investigation of sleep has gained popularity in the area of psychology,
particularly the investigation of disruptive sleeping patterns and sleep
disorders. The increasing interest in sleep on behalf of psychology is
primarily focused upon to the significant correlations of sleep with mood,
cognition, and over-all functioning (Reilly & Piercy, 1994; Drummond,
et. al., 2000; Pilchler & Huffeutt, 1996). Furthermore, psychological
status has a renowned affect on sleep both qualitatively and quantitatively.
For example, anxiety can both reduce total amount of sleep as well as alter
the duration in each sleep stage (Savis, 1994). Sleep is also intriguing
because of the biological need people have to acquire some form of sleep.
Structure of Sleep
Stages of Sleep
Though sleep may seem to be a simple, unified behavior, it is shown to
be far more complex through empirically testing. The general construct
of sleep can be divided into two main categories: rapid eye movement (REM)
and non-rapid eye movement (N REM). These large categories can then
be subdivided into phases varying from the onset of sleep (sleep onset
latency) to the waking period after sleep has been completed (wakefulness
after sleep onset; Youngstedt, S., Patrick, J., Rod, K., 1997).
N REM sleep is divided into
four sleep stages that are labeled with numeric values ranging from one
to four. Stage 1 is known as transitional sleep characterized by
deepening respiration, less awareness of the surrounding environment, and
shifting thoughts. Although this stage's duration is quite short,
it bridges the gap between waking and sleeping states (Savis, 1994).
The state following Stage 1 sleep is referred to as Stage 2. This
phase of sleep accounts for roughly 50 % of total sleep time. For
the great length of time that is spent in this state, little evidence has
demonstrated that it provides the setting for any beneficial activities
for the individual (Savis, 1994). In fact, Stage 2 sleep is not even categorized
as restorative sleep, described later in Stages 3 and 4. In addition,
cognitive processes present during this form of sleep are abrupt and fragmented
(Hauri, 1982).
The combination of Stage 3 and Stage 4 sleep compose the sleeping behavior
known as slow wave sleep (SWS) or delta sleep (Savis, 1994; Taylor, Rogers,
Driver, 1997). The reason for the grouping of the two phases is due
to the similar nature of the physiological processes and electroencephalograph
(EEG) patterns present during these stages. Whereas metabolism, respiration,
heart rate, and core temperature are at their lowest during delta sleep,
growth hormone and cell division are peaked. Restorative sleep is
another title imposed on this form of sleep particularly because of its
responsibility to enhance recovery from expended energy and metabolic rate
(Savis, 1994).
In comparison to restorative sleep, REM sleep is characterized by an increase
in cerebral blood flow, brain temperature, brain protein synthesis, and
heart rate. REM sleep being associated with so many brain processes,
it is no surprise that researchers have shown a correlation between cognitive
functioning and REM duration. However, REM is not a constant state but
rather a fragmented stage whereby periods of REM sleep alternate with N
REM sleep. The first REM period lasts for only a short while.
The duration of each REM period increases as REM sleep occurs throughout
the night (Savis, 1994).
Healthy sleepers experience both N REM and REM sleep. These good
sleepers experience only a few minutes of Stage 1 sleep that is immediately
followed by Stage 2 sleep. Stage 2 sleep typically lasts up to 30
minutes initially and later alternates with REM periods. Stages 3
and 4 appear in the first half of the night and constitute only 15-20 %
of total sleep time. REM sleep follows, usually occurring 90 minutes
after sleep onset (Savis, 1994).
Circadian
Rhythms
Patterns of sleep throughout the day and night are also of interest to
the psychology field. These circadian rhythms refer to the occurrence,
duration, frequency, and onset of sleep, as well as the experience of fatigue
during daylight and nighttime hours. Most assume people awaken around
the same morning hour and retire for the day around the same time.
This, however, is a myth. Atkinson and Reilly (1995) note that depending
on different factors, sleep onset and waking occur at various points throughout
the day, leading to individual circadian rhythms.
The division of sleep into
stages gives sleep an appearance of being highly structured and consistent
across different individuals and populations. However, this is not
an accurate portrayal of sleep. Sleep patterns, as well as sleep
needs, have high variability when comparing groups or individuals (Savis,
Eliot, Bruce, Rotella, 1997). Just as people have individual differences
when referring to personality, likes, dislikes, and so on, so do people
have differences in sleep patterns.
Many factors can affect the sleep structure. For example, age has a significant
effect on circadian timing, quantity of slow wave sleep, and nocturnal
awakenings. As an individual proceeds through life, observable
changes occur in the person’s wakefulness in the morning hours. This
“morningness,” or increase wakefulness in the morning, is positively correlated
with age, as are day-time napping and nocturnal awakenings. Sleep
factors that are negatively correlated with age include variability in
acrophases (i. e., the specific time of day in which the person experiences
less fatigue), rhythm amplitudes ( i. e., the fluctuation of rhythms),
and the amount of slow wave sleep (Atkinson & Reilly 1995).
When an adolescent individual enters the college environment, the circadian
rhythm shifts to adapt to the new surroundings and stresses. Environment
factors, such as a loud dorm or a snoring roommate, have a significant
impact on the length of sleep. Also, emotional response to stress in this
particular population is a predictor of many sleep complaints (Verlander,
Benedict, Hanson, 1999). Large amounts of studying which create much
stress also alter the circadian rhythm by delaying bedtime and hastening
wake up time. However, there is a significant increase in the length
of sleep on the weekends (Machado, Varella, Andrade, 1998).
College Population and Sleep
The college-aged population has unique circadian rhythms and sleep patterns.
These patterns display high variability among sleep and rising times throughout
the week (Machado, et al., 1998). Along with this irregular sleep
schedule, college students also tend to experience less sleep than their
body physiologically requires. The amount of sleep in which they
partake averages 6.5-7 hours a night, with research showing that rate has
fallen steadily for the past two decades (Carskadon, 1990). These differences
when compared to other cohorts are related to high academic demands, increased
social activity, decreased parental supervision, and shifted caffeine and
alcohol consumption (Anch, Browman, Mitler, & Walsh, 1988). In
certain subgroups of the college population, such as athletes, additional
factors may further compound shifts in sleep patterns.
Athletes and Sleep
Amongst the college-age population, a sub-population exists with even more
factors that affect their circadian rhythms and quality of sleep.
College athletes experience much pressure academically, physically, and
psychologically. This group of individuals has a duty to fulfill
the standard workload demanded of all students while successfully performing
their positions as team members of a sport. This situation creates
much psychological and physiological stress (Savis, et al., 1997).
Much stress is present in this role as a team player, especially prior
to a competition (Winget, DeRoshia, Holley, 1985). Athletics also
taxes the body through weight loss, travel, and decreased immune system
functioning, as well as, disruption of sleep patterns (Shapiro, 1982; Worthen
& Wade, 1999).
Exercise and Sleep
Experiments pertaining to sleep patterns of athletes have produced findings
leading psychologists to believe this population are more “efficient” sleepers.
This “efficiency” lies in the ability to have quantitatively less total
sleep, however the duration of restorative sleep increases and Stage 1
and 2 sleep decreases in length. (Savis, et al. 1997; Taylor, et
al., 1997). The decrease in total sleep has been associated with
certain mood and physiological factors. One factor that is clearly
supported by the literature regarding disruption in sleep patterns among
the athletic population is competition anxiety. Athletes have reported
sleeping less than their average amount of sleep the night prior to the
competition. This factor does not explain the athlete’s loss of sleep as
a whole but does provide a possible contribution to the decrease in total
sleep time (Savis, et al., 1997).
Another noteworthy study concluded that the time of exercise in comparison
to the bedtime has an effect on the behavior of sleep. Exercise that
takes place fewer than four hours or more than eight hours before bedtime
will contribute to a considerably increased Sleep Onset Latency (SOL) which
is the amount of time it takes for a participant to engage in Stage 1 sleep.
In comparison, exercise scheduled four to eight hours before bedtime will
assist in the decrease of SOL (Youngstedt, et al., 1997). There still
appears to exist other yet unknown variables related to why athletes partake
in smaller amounts of sleep. At least one theory has been proposed
that may be used to examine this occurrence.
Exercise and Sleep Theory
Restorative Theory
The restorative theory seeks
to explain the relationship between physical activity and sleep behavior.
This theory states that sleep is a mechanism that follows exercise and
everyday activity. Increasing the amount of physical activity would,
according to this theory, increase the quantity of sleep or SWS.
This theory has much evidence to support its fundamental propositions such
as the findings produced by Taylor and colleagues (1997). Taylor and colleagues’
(1997) study showed that more rigorous training in elite female swimmers
significantly increased their amount of SWS. This within-subjects study
also showed that when the swimmers' activity decreased, amount of SWS also
declined.
Food and Sleep
Solid Versus Water-based Meals
Food is another factor that has a significant influence on sleep patterns
and fatigue. When comparing a water-based diet to a solid food meal,
the meal composed of solid food displays a negative correlation with SOL,
in that SOL decreases as solid-food intake increases. This reduction of
SOL and increase of fatigue after consumption of solid food is an enduring
effect persisting as long as three hours (Taylor, et al., 1997).
High- Carbohydrate Meals
The nutritional components that make up a meal are also related to sleep
behavior. Many have proposed that high carbohydrate diets will lead
to increased fatigue and decreased SOL (Spring, Chiodo, Bowen, 1987; Bell,
Rosekind, Hargrave, 1992; Blundell, 1992 ). Some speculate
this increase in sleepiness originates from the release of Cholecystokinin
(CCK) and insulin after consumption of carbohydrates. Both CCK and
insulin have been implicated in the regulation of sleep. (Kapas, et al.,
1991). Wells and her colleagues (1997) found evidence supporting
their hypothesis of a positive correlation between CCK and fatigue.
CCK peaked at roughly the same time period as fatigue and sleepiness hit
their maximum score. This research supports the hypothesis that carbohydrate
intake is positively correlated with feelings of fatigue.
In contrast, Orr and his colleagues (1996) have produced findings suggesting
no relationship between the composition of nutritional intake and sleep
patterns. They state the intake of carbohydrates and fat have no
effect on the athlete’s ability to and quality of sleep. Taylor and her
colleagues (1997) through the measuring of an unchanged diet and energy
levels replicated these findings. They found the athlete’s energy
level peaked during high competitive performance although no change had
occurred in their nutrition. Following this observation, they concluded
the composition of the diet had no effect on energy or sleep.
High- Fat Meals
High- fat meals have also been associated with an increased report of sleepiness.
This phenomenon may be a result of greater concentrations of CCK released
after a high- fat meal leading to increased fatigue. High fat lunches
have been noted to cause a great decline in feelings of alertness 2.5 hours
later (Lloyd, Green, Rogers, 1994). Also, subjects have reported
feelings of fatigue as well as dreaminess and feebleness after consuming
a high- fat breakfast (Wells, et al., 1995). In fact, a decrease
in SOL of newborn babies has been observed after ingestion of a high-fat
feed compared to a high-carbohydrate feed (Oberlander, Barr, Young, 1992).
Discrepancies Amongst Previous
Studies
Sample Selection and Size
Studies involving the relationships among sleep, exercise, and nutrition
have had several serious problems leading to ambiguous data and controversial
conclusions. The first problem when reviewing such sleep research
is that of sample selection and size. The samples studied have included
predominantly good sleepers involved in different types of sports that
reflect varying forms of exercise, including both anaerobic and aerobic.
When the subjects already have efficient sleeping patterns, there is minimal
room for improvement (Youngstedt, et al., 1997).
The variation among the different
forms of exercise in the athletes may be reflected through varying intensity,
duration, and frequency, making it difficult to compare across subjects.
In addition, most research that examines sleep behavior and athletic training
have recorded data from only a small number of participants. This
small sample size may be due to the reluctance of athletes to participate
in studies for fear of it having a detrimental result on their athletic
performance. The coaches may also discourage their athletes
from participation for the same reason, that it could decrease the level
of performance (Savis, 1994).
Methodology
Another problem pertaining to the study of sleep and athletics is the methodology
employed when performing the research. Subjective measures, such
as questionnaires, sleep diaries, and one-time self-report place too much
responsibility on the participant. A participant may disregard valuable
data when recording his sleep behavior due to lack of motivation or knowledge
about the study of sleep. With this missing data, the researcher
may be unaware of a key aspect of physical activity and sleep patterns.
Also, many questionnaires administered in the study of sleep regarding
physical activity have low reliability and validity (Savis, et al., 1997).
Yet, highly-controlled environments, such as sleep labs, can have an effect
on the outcome of the study, as well. Such effects include, for example,
the artificial nature of the sleep lab causing anxiety in the subject and
leading to an increase in sleep onset latency (Savis, 1994).
Experimenter Bias
Yet another factor that may contribute to biased data and conclusions is
due to the experimenter’s own expectancies. These expectancies may
be perceived by the participants as demand cues that elicit a certain reaction
from the individual to whom they are directed. The participants,
therefore, may respond to the questionnaire administered in correspondence
to the demand cues they perceive. Experimenters may also bias their
conclusions by selecting past research that represents their own hypothesis
to back up their findings while disregarding data that propose any contradiction
(Savis, 1994). Finally, assumptions proposed by the researcher and integrated
into the study may also present a misrepresentation of data. For
example, many assumptions are found in sleep exercise research regarding
gender and muscular fatigue. Gender has been assumed to play no significant
role in athletics and sleep, whereas muscle fatigue is often presumed to
be the sole variable inducing sleep due to exercise (Youngstedt, et al.,
1997).
Confounding Variables
Confounding variables are another source of inaccurate data. One
such confounding variable, the mood state of the athlete, can affect scores
on sleepiness. For example, if the athlete is in a depressed mental
state, his sleeping patterns may be disrupted due to his affect and not
his physical activity or diet. (Chelminshi, Ferraro, Petros, Plaud, 1999).
Alcohol and caffeine consumption are other factors that may influence the
data from one extreme to another (Jean-Louis, Von Gizycki, Zizi, Nunes,
1998). Level of college matriculation may also have an impact on
scores. An example of this would be that freshman entering a parent-free
environment will take advantage of their new-found freedom by staying out
to the early hours of the morning, whereas juniors and seniors would not
be as likely to engage in late night activities. Any drugs, such as Selective
Serotonin Reuptake Inhibitors (SSRI’s), can alter a person’s circadian
rhythm (Thase, 1998). Other drug forms that may create havoc in one’s circadian
rhythm patterns include anabolic and alternate forms of steroids (International
Society of Sport Psychology, 1993).
Current Study
This current study involves athletic participants voluntarily being connected
to an EEG machine where their brain-waves, heart rate, and muscle movement
was recorded. Also, their activity and food intake 24 hours prior
to the sleep lab was noted.
Hypotheses
This study focused on the relationships among perceived exercise intensity,
food consumption, sleep onset latency (SOL), and sleepiness. A positive
correlation between carbohydrate level and sleepiness was hypothesized.
A negative correlation was predicted between carbohydrate level and SOL.
Past studies have supported these predictions showing that high-carbohydrate
intake results in increased fatigue and decreased SOL (Blundell, 1992).
A positive correlation was
predicted among fat level and sleepiness. It was also hypothesized
there is negative correlation between fat level and SOL. Past literature
supports these hypotheses by stating that high-fat meals increase sleepiness
and decrease SOL (Wells, 1995).
It was also predicted that the athletic subjects would have a SOL of less
than ten minutes, implying they suffer from somnolence. The adult
population, ranging in age from 21 to 35 years, has an average SOL of ten
or more minutes. The standard for sleep pathology is 5.5 minutes
or less, which results in somnolence and severe impairment. The “diagnostic
grey area” has a SOL greater than 5.5 minutes but less than 10 minutes
(Richardson, et al., 1978; Carskadon, & Dement, 1987; Levine, et al.,
1988). Past research has supported the idea that athletes have a
decreased SOL, therefore falling in the pathological or diagnostic grey
area (Youngstedt, et al., 1997).
