LIVING ALOFT: Human Requirements for Extended Spaceflight






[194] Many communications that are essential for the space traveler will not occur under face to face conditions. Interorganizational communications between physically remote units, such as a satellite and a surface explorer or a space vehicle and the Earth, must be transmitted electronically. Although there are numerous similarities between direct and mediated communication, these forms of communication differ in many obvious and not so obvious ways. In this section we will consider the special characteristics of mediated communication.


Planning Considerations


We are in a period of rapid growth in computer applications. At the same time, we are witnessing the long predicted revolution in the field of communication. But more important than the growth of either of these industries is the fact that these two giants are now merging to provide capabilities far in excess of the added possibilities of both. Distinctions that formerly existed between communication and computers have blurred as computers control and direct communication operations while voices and dynamic video images are digitized, compressed, stored, and even sent over telephone lines. New and exciting products that promise to change the way we conduct our businesses and run our homes and our lives appear with a frequency that bedazzles even the most committed electronic communications buff. It is in this rapidly changing milieu that planners must attempt to address the communication needs and possibilities of extended spaceflight. In technological development, production trails knowledge, leaving innovations outmoded even as they are disseminated; when the vehicle using the technology is in space, the planning problem is truly challenging.

[195] From the design perspective, one of the most pressing requirements for communication systems for space is that they be capable of adapting to industry changes. In the early evolution of computer technology, emphasis was placed almost entirely on hardware design. A shift in attitude occurred with the realization that software could introduce enormous flexibility to a system. A third phase evolved when it became obvious that the software portion of the system, being skilled labor intensive, was the most costly component of the system. The present and presumably final stage of the process acknowledges the high costs of software by placing even greater emphasis on this component. The emerging view is that hardware may last only a short period, but that software must be preserved. With this view, the emphasis shifts to the coordination and integration of programming packages and interfaces that will allow developed techniques to be used with each other and with any equipment. Although the hardware to be used in space cannot follow the throwaway approach, the current direction of systems' development toward greater flexibility and compatibility should benefit the space application by providing a wide range of opportunities to update the system without changing the equipment.


Systems Requirements


A distinction that is often drawn in the communication literature is that between "data" (emphasizing the capacity of the carrier or the rate of the transmission) and "information" (emphasizing the intelligence in the message).1 Great quantities of data can contain little information, whereas important information can be conveyed in a single (yes/no) bit.

Following STS-1, astronauts commented on the excessive amount of useless data they had received during the flight. Similarly, principle investigators reported receiving so much data that were irrelevant to them during the Spacelab simulation (Helmreich, et al., 1979a) that important information may have been overlooked. A significant consideration for any system, but especially for one in space, is that the communication contain a high percentage of useful information. Although the goal of providing useful information is easy to enunciate, it can be difficult to put into operation.

[196] What is "useful" information depends both on the particular circumstances surrounding the event and on other considerations. Dervin (1981) draws an interesting distinction between two contrasting assumptions concerning the utility of information. The first view of information is that it has some intrinsic or objective value which the user must be persuaded to adopt. The second view is that information gains its value only through interpretation and use by the individual. The importance of Dervin's distinction lies primarily in the standards it suggests for communication systems. For instance, the objective view could be satisfied by a system that is predominantly one dimensional; the subjective or interpretive view would require a highly interactive system. As Dervin points out, the objective or persuasional approach is the dominant model of most delivery systems in use today. It is perhaps overly ambitious to expect that spacecrew systems will be among the first to deviate from this pattern. Yet neither should we feel bound by limiting or inappropriate models. In designing systems for future spacecrews, we must attempt to eliminate unnecessary data and to strike some balance between providing necessary information (as judged by others) and meeting the perceived needs of involved individuals.

Even the best designed communication systems have finite information processing capabilities. Information overload occurs when the system is confronted with more information than it can effectively or efficiently process. Miller (1960, 1978) has identified several possible responses to input overload which are common to all information systems. These include queueing, selective processing, omission, and error. Two general procedures are available to ensure that important information is not missed. One is a system of prioritization such that critical information is given preferential treatment in the course of queueing and filtering. The other involves the establishment of reserve or emergency communication channels. Both of these procedures are probably needed to ensure that the communications systems used in each mission will make provision for identifying and promptly processing crucial information, even under overload conditions.


Systems Effects


During the last two decades there has been increased interest in understanding the effects resulting from the use of mediated communication systems, such as those which will provide the link between the spacecraft and the ground. This area of investigation is [197] directly applicable to the requirements of spacecrews and is presented here in some detail.

An operating communication system involves a complex mix of hardware and people variables. Several authors have developed descriptive taxonomies in attempting to put into order the many variables which influence systems use and impact (Bailey, Nordic, and Sistrunk, 1963; Casey Stahmer and Havron, 1973; Johansen, Miller, and Vallee, 1975). Broadly represented, these taxonomies divide the various influencing factors into antecedent variables (conditions which precede the interaction, such as the physical environment and the hardware/software system), and intervening variables (conditions which operate during a particular interaction, such as "gatekeeping" activities or the availability of feedback). Criteria variables are the measures by which the system is assessed. Criteria variables include both effectiveness and satisfaction measures.

Mediated systems influence the dynamics of the communication process in two related but distinct ways. First, there are the effects common to all mediated modes. Second, there are the effects specific to particular classes of systems.

Overview of mediated systems effects- Numerous studies (see Chapanis, Ochsman, Parrish, and Weeks, 1972; Westrum, unpublished Ph.D. Thesis, Purdue University, 1972) have shown that mediated communication differs most from direct face to face communication when the activities engaged in require a high level of interaction among participants, and when the expression of emotion and its accurate perception are related to the outcome. Compared with direct communication, mediated communication contains reduced socioemotional content. As a result, mediated communication tends to be less effective for tasks such as getting acquainted with another or tasks involving bargaining or negotiation, than for tasks such as giving and receiving information, asking questions, and exchanging opinions (see reviews by Short, Williams, and Christie, 1976; Hough and Panko, 1977; Johansen, Vallee, and Spangler, 1979). Studies by the Communications Studies Group (Champness, 1971) also indicate that people are more willing to compromise or "go along" in a direct exchange and are less easily moved from their established positions in a mediated exchange. If follows from the various findings that mediated communication is better equipped to handle routine, businesslike, and formal exchanges than significant, social, and informal exchanges.

[198] Various hypotheses have been offered to explain the differences observed between face to face communication and mediated communication. These hypotheses can be subsumed under three general categories. The social presence hypothesis (see Short et al., 1976) emphasizes how real or immediate other individuals seem to the communicating individual. The channel hypothesis assumes that changes in the nature of the exchange are a result of changes in the amount of information received. Direct contact brings into play numerous communication channels such as visual, auditory, olfactory, and tactile. With information flowing over so many pathways, the system is more than adequate, it is naturally redundant. As the number of channels is reduced, redundancy is lost; as the number of channels is further reduced, information is lost. The task/ socioemotional ratio hypothesis states that, as the medium of communication changes, the relative importance of interpersonal information and task related information changes. Here the emphasis is not on the net gain or loss of total information, but rather on the relative emphasis given to different kinds of information.

Some data are better explained by one hypothesis than another. For instance, it has been demonstrated that channel impairment can adversely affect the assessment of another's opinion and the reaching of an agreement in bargaining situations (Communications Studies Group, 1972). In support of the task/socioemotional ratio hypothesis, it has been found that real opinion change (as opposed to willingness to compromise) is greater after an audio only conversation than after a face to face meeting (Communications Studies Group, 1972; Short, 1972a, 1972b, 1973). However, all three explanations describe many of the differences observed between direct and mediated communications. They are distinguished primarily by the value they assign to particular modes of communication. The social presence and channel hypotheses suggest an overall superiority of direct communication; the task/socioemotional ratio hypothesis does not suggest a superiority of one system over the other, except in relationship to a particular task or goal.

Differences among media- As discussed above, face to face communication can be contrasted with mediated communication. It can also be conceptualized as one point on a communication continuum. Other major points are full motion video, audio, and computer or telegraphic communication. Full motion video is often referred to as broadband communication because its transmission requires approximately one thousand times the bandwidth of an audio link. Audio, computer, or slow scan video can be transmitted over [199] physical links which have only limited capacity, and thus are called narrowband communication.

Differences among media parallel those of face to face versus mediated communication. Interactions involving tasks with little affective component, such as exchanging information, are not as influenced by the particular medium of communication; interactions with high affect, such as bargaining sessions, are more influenced by the choice of medium. Knowing the other person well can offset some of the problems of, for instance, remote negotiations; however, tasks with an emotional component are generally handled better with direct communication or via a video link than with audio only or computer communication. Some interactive tasks such as problem solving may or may not be influenced by the medium, depending on the particular exchange.

Differences among media can also be explained in terms of social presence, channels, and the task/socioemotional ratio. For instance, extending the channel hypothesis, we can consider the various media as differing in the number of modes of verbal and nonverbal channels that they employ. As shown in table 1, when compared with direct communication, video, audio, and computer or telegraphic systems employ successively fewer modes.

Video systems- As noted above, direct communication, and to some extent systems with a video component, convey affective information. The visual dimension also serves another important function - it helps regulate the conduct of an interactive meeting. Eye movement and body position announce who wishes to speak and with what urgency, aiding the smooth flow of an interaction (Argyle, 1969) In a group situation, visual information also provides a continuous status update, keeping the remote participants aware of who has joined and who has left. Visual information also serves to "refresh" impressions of participants.

When more than two people are communicating, the presence of nonverbal cues makes a direct or a video supported exchange far easier to control than an audio or a computer/telegraphic exchange. However, there is no required relationship between the presence of these cues and the management of the interaction. Management without visual cues can be accomplished by providing a system that can identify speakers and give participants a way of indicating their wish to talk. (For a description of such a system, see Stapley, 1973.)


Face-to-face Video Audio Computer/teletype

Proxemic (distancing or placement)


Kinesic (facial expression and gestures)



Paralinguistic (amplitude, rate, and tenor of speech)




Linguistic (written or spoken word)






Among the media, video most closely approximates the effects of direct communication (Goldmark, 1973); however, significant differences remain For instance, in a video exchange someone must decide who or what to focus on, whether to show close ups or distant views, etc. This preplanned selection and focusing requirement considerably limits the choice and ease of viewing that is available in a face to face interaction. Also, the direction of gaze in a video exchange can introduce anomalies. When meeting face to face, people alternately establish and break eye contact. When meeting over video, one tends to fix on the monitor; this can result in the perception of being ignored (if the camera is displaced from the monitor) or of being stared down. One highly advanced (and costly) video design attempts to deal with some of the problems of direction of gaze. Decisions and Designs Inc., under contract to the Defense Advanced Research Projects Agency, have developed a concept which they call "virtual space." Through the use of multiple cameras and monitors, participants at each location appear to turn toward the particular individual they are addressing. Evaluators report that the system has been well received (Sticha, Hunter, and Randall, 1981), suggesting that the verisimilitude of direction of gaze may be a significant factor in the smooth flow of an exchange.

Audio systems- Audio systems can provide a great deal of information, not only of the subject being discussed, but also of the [201] attitudes of the participants. However, achieving high technical quality in an audio exchange involving groups of individuals is a challenging task. Free arrangements (i.e., where individuals are not attached to microphones and earphones) are routinely beset by problems of feedback, loss of signal, etc. A fully duplex system where both locations can speak at will further compounds these problems. Yet any compromise with this open approach to communication usually results in reduced satisfaction and sometimes with reduced use of the system. If the technical barriers can be overcome, audio exchanges can satisfy the requirements for a high percentage of exchanges. It has been concluded that about 40% of the interactions which now take place face to face could readily be handled by an audio only medium (Cornell, 1974; Short et al., 1976). The audio only medium may have the advantage of perceived privacy. Goldmark (1973) found that participants judge a telephone conversation to be more private than even a face to face interaction.

Audio transactions can be augmented as necessary by the transmission of limited motion visual materials. Pictures, graphs, textual materials, or slow scan video can be transmitted over the same narrowband linkages which serve the audio connection. These visuals are limited in providing the socioemotional information that makes participants aware of each other as individuals (Morley and Stephenson, 1969) and in providing the important procedural/ control information available with full motion video. However, they can accommodate substantial informational exchange in support of a given task.

Computer/telegraphic systems- Although capable of transmitting only textual or graphic materials, computer/telegraphic systems are emerging as a flexible and powerful means of communication through work processing, text construction and editing, document manipulation, store and forward messaging, and computing capabilities. An advantage of computer based communication systems is that they do not require simultaneous interactions. Senders can input messages at their leisure, and receivers can acquire and respond to these messages as they choose. Alternately, all participants can input at once without conflict. Received messages can be ordered and prioritized. Such systems have a high capacity for storing and retrieving messages and are useful for transmissions that may require later documentation.

Written communication, as represented by computer/telegraphic systems, tends to be the most formal mode of exchange. [202] As mentioned previously, face to face interactions result in a greater willingness to compromise than does mediated communication. When compared with other forms of mediated communication, written communication tends to result in the most intransigent positions (Sinaiko, 1963; Smith, 1969). Of course, the willingness to compromise speaks only to the likelihood of consensus, not to a change in opinion or to the quality of the decision.

Media preference- Communication systems begin by emulating and extending the functions of the systems they replace. In the mass media, television began its existence by imitating radio, as radio had imitated the printed media which preceded it. Exchanges between separated individuals or groups of individuals are now paralleling this development, with face to face communication often the assumed standard with which other systems are compared (Hough and Panko, 1977).

Face to face or video interactions are judged more positively than either audio or written interactions, and people encountered over video are perceived more positively than those encountered over other mediated systems (Weston and Kristen,1973; Ryan and Craig, 1975). For complex tasks and for tasks involving considerable social interaction, there is often a preference expressed for a video channel, and, as we have seen, video aids these interactions.

However, the preference for video is far from universal, especially among experienced users. Broadband systems can have disadvantages for certain types of businesslike communication and for negatively toned communication involving such issues as criticism and censure (Hough and Panko, 1977) in which one would choose to deemphasize the interpersonal aspects of the exchange. Narrow band communication emphasizes the argument rather than the presenter, and individuals tend to prefer it if their arguments are strong and to avoid it in favor of face to face or video communication if their arguments are weak (Morley and Stephenson, 1969, 1970; Short, 1971, 1974). Because of the tendency of narrowband media to emphasize rational rather than social considerations, some users express a general preference for it. One user (in Johansen, Vallee, and Collins, 1977) expressed this view poetically in describing computer communications as acting (p. 383): a filter, filtering out irrelevant and irrational interpersonal "noise" and enhanc(ing) the communication of highly informed "pure reason"-a quest of philosophers since ancient times . .

[203] It can be concluded, then, that although there is a basic preference for direct communication, mediated communication is not a poor substitute for face to face exchanges. Rather, it is another form (or several other forms) of communication. Although all forms of mediated communication seem to be aided by the previous acquaintance of the participants, preference for one system over another is highly dependent on the specific conditions.

System acceptance- Any communication system, whatever its capability, will fail if it is not used by its intended beneficiaries. Many systems have been resisted because (1) the designers did not understand fully the users' needs, (2) the potential benefits were never made clear to users, or (3) the users were never properly trained in the system (Elton and Carey, 1980). Systems have also failed because of unanticipated user fears or because users felt that the new medium would somehow disrupt the normal pattern of their lives. Elton and Carey have offered a number of useful suggestions on planning and implementing communication systems.

First, potential users should be included in the planning and implementation process as early as possible. There are three reasons for this: User involvement helps designers understand user attitudes and needs, users become committed to choices to which they themselves have contributed, and there develops a cadre of users who have sufficient familiarity with the system to be able to explain it to their peers.

Second, planning should take the psychological aspects of the physical arrangement into account. The physical arrangement refers to such variables as equipment location, physical setup, and operator machine interface. A location should be sought which is at once accessible and private. Although it may be necessary to locate the equipment in a multipurpose area, steps should be taken to minimize the extent to which the system's use and other area activities interfere with one another. The equipment should be as simple and gimmick free as possible in order to provide minimal distraction from the messages themselves. In general, users resist systems that are remote or inaccessible, that are difficult to use, or that raise issues of territoriality or privacy.

Third, system planning should take the social context into account. The social context includes such variables as group norms, expected degree of formality during the communication, user social status, and the like.

[204] Fourth, hardware development should be accompanied by the development of a protocol that specifies who uses the system, how it is used, and under what conditions it is used. The intent of such protocols is to facilitate interaction, reduce the potential for embarrassment, and eliminate uncertainties which can contribute to negative attitudes. For example, an appropriate protocol would overcome such problems as the uncertainties generated by not knowing that others have entered or left the assemblage.

Fifth, once systems have been designed and installed, efforts must be taken to ensure that all of the system's potential users understand its strengths, limitations, and procedures for operation. The expectations of users have been found to play a significant role in system acceptance or rejection. Training should precede the system's first use, as users' initial experiences are likely to have a disproportionate impact on their attitudes about the system. Users will often balk if a system fails to perform to their expectations on demand. However, these same users will frequently persevere with a lesser system, provided its limitations are understood in advance. Similarly, the perceived reliability of a system may be at least as important in user acceptance as its actual reliability, with operational/environmental factors playing a significant role in the shaping of perceived reliability (Bair, 1978).

Finally, ongoing evaluations should continue throughout the life of the system. Such assessment should address the system's performance, its impact on relevant aspects of the users' lives, and users' attitudes. Maintaining a high level of rapport with users is essential to effective systems evaluation.

In summary, the abruptness with which communication systems can be dismissed or devalued suggests that the planning and implementation of communication systems is a major task and requires a high degree of effort and attention.


Application to Space


Careful planning can help avoid some of the problems associated with matching the requirements of users and the offerings of communication systems for space. However, the findings reported above suggest that some difficulties could persist, whatever the system or implementation plan. Mediated communication systems handle simple tasks better than complicated tasks, information exchange better than persuasion or influence, and work related [205] activity better than social activity. Yet it is anticipated that extended missions will increase the need to deal with complex tasks, to reach agreement on issues, and to engage in affective exchanges with those at home. Among mediated systems, broadband systems are more effective than narrowband systems in handling tasks that are complicated or that contain high emotional content. Yet it is anticipated that for extended missions, narrowband computer based systems will have certain technical advantages that may counter other considerations. For missions involving very large distances, transmission delays may render simultaneous or quasi simultaneous communication impossible. Here, computer communication would have an advantage, since it is easily used in a sequential mode.

Although suggestive, studies reported to date on the impact of mediated communication systems are not definitive for space application. First, the subjects of these studies have been neither isolated nor confined. Second, these studies have cast electronic exchanges as an option rather than a necessity for interaction, and users have not been required to rely on these systems exclusively for prolonged periods of time. Third, studies to date have focused on goal oriented groups functioning specifically in their work roles, rather than on socioemotional groups functioning in their family or friendship roles - one of the requirements for space. Many of the questions concerning the relationship of media variables and social interaction will have to be reexamined to take into consideration the unusual conditions of space.


1 Information Theory (see Shannon, C. E.; and W. Weaver: The Mathematical Theory of Communication. U. of Illinois Press, Urbana, III., 1963) is a significant exception to this lexical distinction.