Several important aquatics professionals and lifeguard training agencies have questioned
whether lifeguard surveillance techniques are based on science.
While many opinions abound about lifeguard scanning, few of them
are rooted in actual science. That’s too bad because science
has a lot to say about how to make guards more
One of the most reliable methods is a science called human information processing,
which deals with the way in which people process information,
especially during complex tasks. During scanning, lifeguards look
at, and think about, patrons in their area of responsibility.
Psychologists call scanning an example of human information
processing. You’re already familiar with this idea because
your five senses process information continuously. Reading this
article is human information processing.
Human information processing theories are based on decades of solid, peer-reviewed
research. The studies are just as valid today because though times
change, people, or at least their ability to process information,
That said, lifeguarding is a distinctive skill, and while human information
processing has much to teach us, more research is needed to figure
out exactly how it should be used. Evidence-based scientific models
are needed to examine the effectiveness, validity and reliability
of scanning practices.
Based on existing research, here are eight misconceptions about lifeguarding that
human information processing theories may help clarify:
1 Different settings call for different scanning methods.
Massive amounts of visual images from pools and beaches reach lifeguards’
retinas every second. They must limit their attention to patron and
facility surveillance images. Researcher Donald Eric Broadbent
described visual attention as a strainer that limits image
processing through visual pathways. The selection process occurs in
visual short-term sensory/visual short-term memory
When guards scan, they assign meaning to some visual images and dismiss mental
pictures of safe swimmers. My review of many human information
processing models has led me to conclude that the guard’s
eye/brain decision-making method for noticing the signs of drowning
and distress is the same for all settings
2 Thinking about patrons’ personal characteristics is the best way to help
lifeguards spot potential drowning persons or swimmers in distress.
Sometimes during scanning, lifeguards have doubts about a
person’s safety. The guard brings that visual image,
questions and educated guesses (hypotheses) about the bather into
working memory for further examination. In working memory, guards
look for more cues about whether the person is drowning, in
distress or swimming safely.
Unseasoned guards may briefly think about, but then get rid of, working memory images
without recognizing the person needs to be rescued. These guards
often think the drowning person is playing in the water, according
to my own research from 1959 to 1979. This research was supported
later by New York State Department of Health’s drowning
investigations from 1987 to 1990. More recently, it was discovered
that 89 percent of inexperienced lifeguards could not detect a
drowning person on a video clip in fewer than 30 seconds, according
to a 2002 study by researcher Billy Doyle. He found that 70 percent
of the participants would “wait more than two minutes to see
what would happen,” and 30 percent indicated they would take
Many studies have shown our working memory is limited to approximately seven bits or
chunks of information and three to four hypotheses. Therefore, when
teaching guards to scan, it may be wise to avoid surveillance
practices that fill up their working memory with too many images
Cognitive psychologists agree that working memory is a major human
information processing bottleneck. When working memory is at or
near capacity, guards divide attention between looking at visual
images and making educated guesses about a person’s safety.
Based on these precepts, lifeguard training agencies and
instructors should avoid surveillance policies and scanning rules
that exceed lifeguards’ working memory
Learning, however is a different matter. Learning allows the guard to store new
information and reactivate past information from the two parts of
long-term memory. Declarative long-term memory covers knowledge
about ideas, definitions of terms, and visual images of drowning
people and distressed swimmers. Procedural long-term memory
consists of skill execution, such as rescue tube use and in-water
spinal injury mobilization.
3 Scanning patterns are the best way for lifeguards to spot drowning persons
and swimmers in distress.
With training, lifeguards can learn to automatically recognize
drowning or distress images. When they use automatic processing,
drowning people or distressed swimmers will “jump out”
at the guard. These guards can avoid mulling over an image because
training allows them to immediately match a visual image to a
picture in their long-term memory. Automatic processing reduces
working memory dependence, and it can be 25 times faster than
controlled processing of visual images, according to researchers
Walter Schneider and Richard Shiffrin.
A scientific study was conducted to teach lifeguards to use automatic processing to
detect the signs of an active drowning person. Beginning in 1959, a
21-year observational study of thousands of drowning people at
Orchard Beach, Bronx, N.Y., was carried out. This research led to
my drowning vs. distress water crisis classification, and was
contained in the 16 mm film “On Drowning.” This study
documented the difference between a drowning person and a swimmer
In 1995, the United States Lifesaving Association’s manual adopted the drowning
person and distress swimmer classification. The USLA built on the
original 1970 water crisis classification and added 13 distressed
swimmer descriptions to its training manuals.
4 All scanning methods work the same.
When scanning, lifeguards’ eyes move rapidly from one point
in the visual field to another. The American Red Cross advises
guards to “scan from point to point, rapidly glancing at all
the movements of people in your area.” Scientists call a
single rapid eye movement within the visual field a
Within each saccade,the guard’s eyes and brain work together to examine all the
bathers. Images of people bathing safely are filtered out and need
not remain in working memory. The guard’s brain selects some
people for closer examination. Selecting, focusing and discarding
images are continued as the guard examines people in each
During saccade-based sweep scanning, guards do not neglect examining people in any part
of their zone. During consecutive saccades, they glance at and
evaluate every person in each saccade.
This process assigns an equal chance that every bather might be a drowning person or
distressed swimmer. Saccadic sweep scanning comfortably fits into
guards’ eye and brain system, by using familiar human
information processing and natural eye movements.
5 Lifeguards should look out for “high-risk” patrons.
The instruction to lifeguards to “check high-risk
patrons” suggests that a different observation and evaluation
method is used for “nonhigh-risk” bathers. Asking
guards to mentally rehearse images of “high-risk
guests” who are neither drowning nor in distress after a
guard completes a zone scan is unnecessary. This practice increases
working memory load and can hasten the onset of sensitivity
Further, this practice does not increase detection of drowning persons, according
to my own research. Other research by Raja Parasuraman in 1979
found that any vigilance task requiring a continuous load on
working memory leads to a signal sensitivity loss.
6 Lifeguards don’t need frequent breaks.
Studies suggest that the best way to increase lifeguard
attentiveness is giving breaks. “You should take a break at
least once an hour,” state the American Red Cross’
lifeguard manuals. The USLA encourages facilities to provide guards
on continuous surveillance duty with a 15-minute break every hour.
Likewise, the YMCA recommends sufficient guard surveillance
Researcher N.H. Mackworth developed the visual sensitivity loss model. Using
classic clock task experiments, signal detection performance often
declined during the first half hour of the watch. Later experiments
found five- to 10-minute breaks reset the vigilance level to its
Unless lifeguards are given regular surveillance breaks, the chances increase that as
their shift progresses, they will be less likely to detect images
of drowning people or distressed swimmers. Broadbent pointed out
that the attention necessary to fixate continuously on images
causes visual fatigue. Another reason for surveillance breaks:
There are considerable environmental and psycho-physiological
factors that cause guard fatigue.
The low arousal theory usually applies to lifeguards with long watches who do not
have regularly scheduled surveillance breaks.
Facilities marked by few preventive actions and rare rescues can cultivate within the
guard decreased central nervous system arousal.
Preservice and in-service discussions of the causes and solutions of low arousal
are important. Individuals could be trained with biofeedback
techniques to suppress brain theta waves indicative of low arousal,
according to researchers Jackson Beatty, Arana Greenberg, W. Philip
Deibler & James I. O’Hanlon. The National Sleep
Foundation in Washington, D.C., offers tips for recognizing low
arousal signs, and recommends that lifeguards take a break at least
once an hour.
Some pool managers in the state of New York clear their single lifeguard pools every
hour for 10- to 15-minute guard surveillance breaks. The manager
remains on deck during the guard’s break.
Every lifeguard training agency should discuss the necessity of providing regularly
scheduled surveillance breaks from the many stressors that fatigue
guards. Pool management companies should list regularly scheduled
break periods in their lifeguard manuals and safety plans. Aquatics
professionals can no longer encourage guards to “stay
alert” without advocating hourly surveillance breaks to
guards assigned continuous surveillance duties.
Multiple lifeguard facilities also should list surveillance break times within their
rotation system. Because rotation only changes guards’
positions, alternating their positions is not equivalent to a
Arousal theory accounts for some aspects of vigilance performance. If
inexperienced guards think there’s a low possibility for a
drowning or near-drowning, they lower their vigilance.
7 The fewer thepeople, the less likely a drowning will occur.
Novice lifeguards at facilities with few drownings or
near-drownings decrease their vigilance levels by silently noting
to themselves that a water crisis is unlikely. This faulty
conclusion allows them to lower their vigilance level.
This means lifeguards might look only at certain areas in their zone,
according to a study by Neville Moray.
Researcher C.H. Baker linked expectancy vigilance decrements (for example
“nobody drowns at this facility”) to decreases in the
perceived likelihood of target events. A target event would be a
drowning person, swimmer in distress or person engaging in a
Douglas C. Sackett,assistant director for the New York State Health Department’s
Bureau of Environmental Health, is developing a matrix for studying
causes contributing to drownings during an 18-year span at
lifeguarded facilities in New York. This matrix is based on data
from 1987 to 2005. It provides valuable data for preservice and
in-service discussions about the ways low arousal levels and
decreased drowning person expectancies contribute to guards missing
the signals of drowning or distress. For example, because 33
percent of pool drownings occurred with one to five bathers
present, managers can prove to guards they must remain vigilant
even when a single bather is using the pool.
8 Lifeguarding isinherently boring and tasks must be assigned to fight it.
Patron surveillance is a repetitive task. However, repetitious
activities do not invariably lead to boredom. A 1985 study looked
at the cognitive and affective features of boredom and noted
sensory monotony alone does not induce boredom, according to
research by R.E. Perkins and A. B. Hill. Using the research of
other scientists, Perkins and Hill suggested boredom results when
stimuli lack meaning to the person. In-service training and
lifeguard supervision practices need to continually remind guards
that watching swimmers is a meaningful activity.
Describing rectangular pools as “boring” predisposes inexperienced
lifeguards to accept this biased evaluation. Portraying scanning as
a dull, undemanding activity similar to filling up your bathtub
with water, pulling up a chair and watching the water for half an
hour creates a self-fulfilling prophesy for unseasoned
Opinions that patron surveillance is boring and monotonous have led to pattern scanning
involving alphabet, geometric and multistage strategies.
Pattern scanning superimposes letters or geometric figures over the guard’s
visual field. While the guards’ eyes follow the pattern, they
are looking only at people in the areas they guess will contain
drowning people or distressed swimmers. Testing whether one pattern
results in faster water crisis detection than another pattern has
not been conducted.
Multistage scanning contains quadrant grouping, focal person identification, pattern
scans, postural changes, and rescue visualization.
Several lifeguard training agencies now recommend guards combat boredom by imagining
(visualizing) a rescue if they are bored during patron
surveillance. Visualization is forming a mental image. While
composing mental images, guards cannot give full attention to
people in their areas of responsibility. My research suggests that
visualization while scanning, creates self-induced inattentional blindness.
Individuals can miss obvious signals during attentionally demanding working memory
tasks, according to studies by Arien Mack and Irwin Rock, as well
as another by Daniel J. Simons and Christopher F. Chabris. Many
individuals in the Simons study failed to see a person dressed in a
gorilla suit walk through the middle of a 62-second basketball
Multistage scanning uses a series of mental manipulations and postural changes over a
consecutive five-minute interval. Research questions and hypotheses
supporting the theoretical basis for quadrant grouping and focal
person selection within each quadrant have not been identified.
Furthermore, the consecutive scans in the multistage model have
research design weaknesses.