The Briefing Room - Learning from Experience

EFFECTIVE COMMUNICATION
IN THE AVIATION ENVIRONMENT:
WORK IN PROGRESS
By Anne Isaac, Ph.D.
Annes early experience in ATM and airline operation was followed by six years with the
Human Factors team at EUROCONTROL, where she was associated with the development
of tools and techniques to help identify human error and risky performance in the ATM
environment, as well as developing the Team Resource Management (TRM) concept for
European ATM. Anne now heads a team in Human Factors integration within the Division
of Safety in NATS, UK.
Effective communication is a basic
human requirement and in the aviation
environment an essential pre-requisite
to safety. So why do we continue to get
it so wrong? - and we do get it wrong
about 30% of the time. In a recent radio
telephony survey it was found that
80% of RTF transmissions by pilots
were incorrect in some way. However
pilots are not the only ones in the communication process, and there are
some startling statistics from the air
traffic controllers as well:

30% of all incident events have
communication errors, rising to
50% in airport environments.

23% of all level-bust events involve
communication errors.

40% of all runway incursions also
involve communication problems.
None of these statistics are surprising
when we realise the demand we place
on the verbal communication process,
and most of us know some of the obvious traps: call sign confusion, the problems with native language, the use of
standard phraseology and the increasing traffic and complexity leading to
frequency congestion and overload, as
well as a high percentage of technical
failure of the communication system
itself. However, what might not be so
obvious is the complexity of effective
communication and the aviation cul-

HINDSIGHT N5

ture which reinforces operational staffs

trust in other colleagues.
The following graph indicates the most
numerous problems, however this only
illustrates half the story.
Perhaps more importantly we should
ascertain the most serious issues
caused by these activities and the context in which they are likely to increase
the risk to the system.
The leading events, which encompass
some of the above issues are: mis-hearing information over the RTF, often
caused by incorrect pilot read-back of
information (but by the correct pilot)
and transmission and/or recording of
incorrect information by either the
pilot or controller. In all cases the problems are embedded in the complexity

Page 31

of the communication process itself. In

order to transfer information, both the
person sending and receiving the
information must be able to formulate,
listen, hear and interpret the message
correctly as well as verify the information for completeness, and at any of
theses stages things could go wrong.
The most risky situation is when one of
the parties does not identify or recognise an error, since then they are
unable to recover from the situation
themselves. Some of these risks are
embedded in the way we ascertain

July 2007
January

The Briefing Room - Learning from Experience

information from equally qualified colleagues.

We tend to ask confirmatory questions
to solve a problem when we are unsure
in these situations. The example below
is taken from the Danair 1008 air accident at Tenerife:
Co-pilot : gosh, this is a strange hold,
isnt it?
Captain : yes, it doesnt ............................,
it doesnt parallel the runway or anything.
Co to Engineer : its that way isnt it?
Engineer : that is a 3 isnt it?
Co : yes, well, the hold is going to be
there, isnt it?
Captain to Co : did he say it was 150
inbound?
Co : inbound, yeah
Captain : well, thats....................................,
I dont like that
Co : they want us to keep going all
around, dont they?
Another very risky situation, in terms of
the above issues, are conditional clearances. Conditional clearances are used
on the understanding that both parties
are assured of the message they hear.
Since most of the information which is
found in the conditional clearance
information is standard and known by
both parties, it is very rare for one of
the parties to question part of this
communication. Usually you will hear
the person receiving the message say,
Oh he must have said that, or she
must mean this. This situation is made
more risky when the actual communication is correct but incomplete.
Almost all runway incursion incidents
which involve conditional clearances
are also the result of incomplete com-

July 20072007
January

munication strings. This is particularly

risky for both parties since an incomplete transmission is not so easy to pick
up as an incorrect transmission.
Another example regarding communication and feedback to colleagues
within the aviation industry is the issue
of seniority and expertise. Air traffic
control assistants as well as cabin crew
believe that it is not their place to
question or challenge a colleague who
is more qualified or in a position of seniority. The following example illustrates
this and had fatal consequences.
On March 9th 1989, an Air Ontario
Fokker F-27 was getting ready to take-off
from a small airport in Northern Ontario.
Take-off was delayed as the tower waited
for a small private aircraft to land. It had
been lost in a spring snow storm. Whilst

Page 32

the aircraft waited for take-off clearance,

several passengers took note of the accumulation of snow on the wings. One of
them brought it to the attention of the
flight attendant, who assured him that
there was nothing to worry about. Many
of the aircrafts occupants were concerned about the snow, but no one,
including the flight attendants, thought
it appropriate to say anything to the
flight crew. When asked about this during the course of the investigation, the
one surviving crew member, a flight
attendant, stated that she did not feel it
was her job to inform the pilots of potential problems. She had never been trained
to question an area that in her mind was
clearly a pilot responsibility.
Moshansky, 1992.
Since then both the development of
Crew and Team Resource Management

HINDSIGHT N5

The Briefing Room - Learning from Experience

activities have enabled clarification

and challenge to be an acceptable part
of this working environment.
One of the most prevalent errors in all
aviation communication is information
which is mis-heard or not heard at all.
The reasons for this are again many
and varied, which is why ICAO and
National Air Navigation Service
Providers train their operational staff to
use standard radio telephony. So why
dont we stick to these rules? Research
would indicate that there are several
human traits which make following
rules more problematic. Firstly people,
even controllers, assistants, pilots and
aerodrome drivers never believe they
could be involved in a serious incident
or accident. The fact that these events,
compared to the number of aircraft
movements, are relatively rare, helps to
perpetuate this belief. This trait is not
exclusive to aviation professionals, we
all believe the best when we step outside into the hazardous world, not
appreciating we could be the victim of
many and varied serious incidents.
Secondly, having developed standard
phraseologies, individuals as well as
Centres, Units and even National
Providers and Airlines believe, because
they are different, they need to apply
for an exemption or change to the rule.
These changes are rarely associated
with a study to establish the reason for
the changes and the best consequent
solutions. Again it is rare that procedure specialists would ask the advice of
the human performance specialists
about how humans process both written and spoken information. This often
leads to the use of incorrect phraseologies being delivered in the wrong
order. Some of these risky words and

HINDSIGHT N5

phrases have been identified as follows:

In turn - intended sequence is
unclear;

Next exit - whos next are you referring to;

Pull forward - clearance is not clear;

One hundred and eleven hundred
- as in flight level;

Three digit numbers ending in zero
- heading often confused with
flight level;

Similar sounding letters and numbers - B,G,C, D and 3;

Made a ... interpreted as Mayday;

Holding position interpreted as
hold in position;

Climb to, two thousand - action, followed by qualifier.
Many other errors are made because of
the problems of expectancy. Because
we use standard phraseology, we often
expect to hear a particular request or
reply in a familiar situation. If the message we receive is distorted in some
way, such as due to other noise or cut
off, it is easy to assume we heard what
we expected to hear instead of confirming the message. Hearing what we
want to hear, guessing at an insignificant part of the spoken message, and
filling in after the fact, are commonplace. We also reconstruct parts of messages unintentionally - and we do so
with the utmost confidence that we
hear what we actually reconstructed,
not what was said.
Another reason for the prevalence of
information which is mis-heard or not
heard is associated with interruption
and distraction. Usually a verbal message or phone call will interrupt almost
any activity, and by the time we realise
that this interrupting message is of lit-

Page 33

tle importance, it is too late to retrieve

the activity we were engaged in when
the message or phone call started. This
results in the two tasks, whether they
were verbal (receipt of a message) or
another action (scanning, writing)
being incomplete. When two activities
compete for our limited working
capacity we usually end up losing all
the communication channels, and have
to start again.
This problem is particularly obvious
when working under a high task load.
Task load is dependent on work load
(the sheer volume and complexity of
traffic) and contextual conditions such
as:

Weather;

Experience;

Fitness;

Time on position;

Stress.
Task load is a personal experience, different for everybody and depending
on many things. The limitations of the
human information processing system
are first observed in our ability to communicate. Overloading this system
inevitably leads to less effective communication due to tunnel vision (and
tunnel hearing), reduction of scanning
cycles, less investment in time to execute feedback and a rising temptation
to fall for the trap of expectation bias.
This results in more incorrect information which leads to further incorrect
communication, and finally decisions
and actions which are error-prone. We
all have a tendency to dismiss the need
to invest time in effective communication when it is most needed; under
high task load.
The main issues which have been iden-

July 2007
January

The Briefing Room - Learning from Experience

tified during incident investigation and

safety trend analysis are the following:

Pilot reads back incorrectly and the
controller does not recognise and
correct the error, often since it is
from the correct pilot;

Pilot reads back correctly, however
this is followed by an incorrect
action on the flight-deck;

Pilot reads back correctly however
the controller records the information incorrectly, resulting in a subsequent error.
Statistics would also suggest that controllers can often pick up errors in communication more quickly than pilots.
Cardosi, in her 1997 study, recorded the
fact that controllers correct 50% of
pilot read-back errors on ground control frequencies and 89% on en-route
frequencies. The reason for this is possibly because not only do controllers
have more and varied R/T communica-

July 20072007
January

tion to deal with, but also because they

are constantly tested for their proficiency in these skills.
Well, having explored some of the traps
that cause humans to make errors,
what are the solutions? These, like the
traps themselves, are not easy to manage and implement since the communication process itself is highly complex. However, here are some tips for
both pilots and controllers which may
help:

Use clear and unambiguous
phraseology at all times; challenge
poor RTF;

Try to avoid issuing more than two
instructions in one transmission;

Be aware that you tend to be less
vigilant when speaking in your
native language;

Always insist on complete and
accurate read-backs from pilots;

Set the clearance given, not the

Page 34

clearance expected;

Both pilots should monitor the frequency whenever possible;

On frequency change, wait and listen before transmitting;

ATC instructions should be
recorded where possible;

Use standard phraseology in faceto-face telephone coordination;

Monitor all read-backs, try to avoid
distractions - especially the telephone;

When monitoring messages - write
as you listen and read as you speak;

If you are unsure, always check!
The European Action Plan for AirGround Communication Safety contains more information and advice on
effective communication. Copies may
be obtained by completing the form
on the EUROCONTROL web-site at
http://www.eurocontrol.int/safety/publ
ic/standard_page/documentation_distrib.html

HINDSIGHT N5