http://eos.kub.nl:2080/w3thesis/Groupwork/communication_channels.html (Einblicke ins Internet, 10/1995)
Media for communication
Often, communication is thought of in a rather traditional way: it succeeds
whenever the message is received by the receiver. According to Weigand [1994], this interpretation
of communication is too limited; it passes over the influential character of
communication as well as over the shared object of communication.
Communication is defined by Short
et. al. [1976] as
the physical signals whereby one individual can influence the behaviour of
another.
The object of intergroup communication is denoted as the
common ground by Clark and
Brennan [1990]. All actions are built on common ground and its accumulation.
The common ground provides a shared context for action and at all time reflects the
progress that is made in accomplishing the group's task. As I will point out,
establishing a mutual feeling of common ground is essentially the goal
of the human-computer interface in a computer-mediated cooperative working
environment. However, I leave the computer as a means for communication aside for
now.
Short et. al. [1976] define the
medium of communication as
the system of constraints on the physical signals available in any particular
situation.
According to McGrath and Hollingshead
[1994], human conversation can be regarded as a series of interlocked
communication cycles. In turn, each cycle involves
a series of operations on a given message: composition, editing, transmission, reception,
feedback (acknowledgement of receipt), and reply. The time to complete such a
communication cycle differs for different communication media.
Daft [1992] organises different
communication channels into four categories ranging from highest to lowest in
information richness, which he defines to be the information carrying capacity
of data communicated through that particular medium or channel. For each
lower medium, the communication cycles become longer.
Under conditions of uncertainty (that is, in complete absence of information)
when data must be gathered that will reduce uncertainty, this may not be a problem.
Technology-based information systems, referred to as high tech means of
interaction by Daft can provide this data
effectively and efficiently. Such systems suffice when a firm
common ground is already established.
However, longer communication cycles effectively hinder high touch
interaction, which is particularly necessary for dealing with ambiguous situations
in which issues cannot be objectively analysed and understood and when
additional data that will resolve an issue cannot be gathered through high tech
means of communication. Higher ambiguity requires creative solutions, which
are typically not communicated through high tech systems, instead these need
to be communicated through high touch interaction.
When equivocality is high, individuals are likely to have different interpretations
of problems. As a result, they may disagree as to what information is needed to come
to a solution. Under these conditions, individuals must first create a shared sense
(a common ground) of the situation before,
through negotiation and feedback, a common response can be formulated. This
requires a rich communication medium, one that provides interactivity and expressiveness
(Kraut et. al. [1992]),
in short, a high touch medium.
Daft distinguishes the following
communication channels:
- Face-to-face communication. Face-to-face communication is the
richest medium. Communicating this way keeps communication cycles extremely
short. Besides, it provides many cues in addition to verbal modalities, such as
facial expression, direction of gaze, and body language or posture. Even the way
interactors dress as well as the physical distance between them seems to matter.
Short et. al. [1976] list six
functions, identified by Argyle,
for non-verbal cues during face-to-face interaction. In summary, this list looks
as follows:
- Mutual attention and responsiveness. In order to provide continuous
evidence that the other is paying attention, he will respond by nodding his head or
by making gestures. Utterances such as 'yes', 'ah ha', which are often used in
telephone conversations, are likely to replace these non-verbal signals.
- Channel control. Head nods and eye movements are used in
determining who shall speak and for how long. In face-to-face conversation, the
interactors do not appear to be aware of making or receiving this signals. However,
this function is likely to be more seriously affected by the removal of the visual
cues.
- Feedback. An important property of non-verbal signalling is that it
can go on simultaneously with verbal communication without interrupting it. This is
especially important for providing feedback: in order to plan his utterances, the
speaker needs to know how the other is reacting to what has gone before. If the
visual channel is locked out, the speaker must wait for a verbal reply from the
listener before he receives any feedback on his remarks.
- Illustrations. Often, speech is accompanied by hand movements that
are used to illustrate an object or action, for emphasis, or to point to objects (e.g.,
in the case of giving driving directions). Apparently, illustrations introduce some
informational redundancy. However, the clarity of the message will be improved
by gesturing.
- Interpersonal attitudes.Non-verbal cues may be used by the listener
as a vital source of information about the speaker's attitude to him. Moreover, he
may draw conclusions regarding the speaker's intentions of, for example, threat.
Non-verbal cues can indicate sudden changes in a relationship, for instance the
change from a relaxed to a more formal posture. Also, non-verbal cues can express
the speaker's affective reactions or true emotions (e.g., embarrassment or
uncertainty which will make the speaker blush; known to happen during
graduation speeches :-) in what may be verbally a completely emotion-free
statement.
- Emblems. This term is used to refer to gestures being used instead
of a word. For instance, a head-shake instead of speaking out the word 'no'. By
definition, emblems are replaceable by words in the absence of the visual channel.
Therefore this category can be considered as less important than the others.
- Telephony. The telephone and related personal electronic media
such as voice mail are next in richness. Though the telephone lacks a visual
channel, it is a relatively rich medium for communication because feedback is fast
and messages are personally focused (the only exception being group telephone
meetings).
- Written, addressed documents. Addressed documents such as letters,
memos, notes, and faxes are lower still in richness. All operations on a message
(composition, editing, transmission, reception, feedback, and reply) take more time
compared to the richer media. Moreover, the number of visual cues is reduced to a
minimum. Despite these obvious disadvantages (especially in a cooperative
working environment) some people prefer written interaction. For example, while
writing a letter, they can carefully weigh all the words. Whereas in a face-to-face
conversation they may need to respond more rapidly and as a consequence allow
themselves less time to think before they speak.
- Written, impersonally addressed documents. Impersonally addressed
documents include bulletins, standard computer reports, and printouts. These are
the leanest channels of communication. Such documents are not amenable to
feedback and are often quantitative in nature. This channel is best for conveying a
large amount of data to numerous people.
Different media impose different constraints on intergroup communication.
For example, for synchronous communication to take place, all group members
must be available for interaction at the same time. Clark and Brennan [1990] identify eight
constraining factors:
- Copresence: the interactors share the same physical environment.
- Visibility: the interactors are visible to each other.
- Audibility: the interactors communicate by speaking.
- Cotemporality: the receiver receives at roughly the same time as
the sender produces.
- Simultaneity: the interactors can send and receive at once and
simultaneously.
- Sequentiality: the interactors' turns cannot get out of sequence.
- Reviewability: the receiver can review the sender's messages.
- Revisability: the sender can revise messages before he actually sends
them.
When a medium lacks one of these characteristics, group members are forced
to use alternative techniques for establishing the common ground. These grounding techniques change with the
medium. For example, consider a face-to-face conversation in which the
interactors are able to time their utterances almost intuitively and with great
precision. Grounding techniques that rely on the same precision of timing but use
an asynchronous communication channel instead, will inevitably increase the
efforts or costs of communication.
For the previously distinguished media, these are the associated
constraints:
- Face-to-face: copresence, visibility, audibility, cotemporality,
simultaneity, and sequentiality
- Telephone: audibility, cotemporality, simultaneity, and sequentiality
- Written documents: reviewability, revisability
Sjoerd Michels, Tilburg, The Netherlands