Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 329
SOCIAL STRESS, ~R-HEDLAl~= FORNICATION SYSTEMS,
AND IN ~OWC~lIVIrY :~ SPACE STATIONS: A RESEARCH AGE
Karen S. Cook
INT~DUCT1ON
We sheer complexity of the space station program is enough to boggle
the mire of any academic trained in a single discipline. Certainly,
space station design requires the ultimate In interdisciplinary
teamwork and denigration of basic arx] applied pa Trams of research. In
this sense, the project demands knowledge an] insights not easily
produced An an isolated discipline, be it en~in==rinq, aeronautics or
sociology. '~ as a cnalleng mg task and one that shculu call forth the
best efforts of those touched by the allure of extending the boundaries
of human knowledge.
For a sociologist there are a myriad of research problems which come
to mind On even a cursory glance into the window of the future as
envisioned by those currently Dlanm no the space station program.
,
_, e ~ _ _ e , ~ _
_ _ ~
~ _ ~ _ ~ ~ J
clearly, a wide range of processes and factors must be taken into
account when considering the more social aspects of this enterprise.
These include technologist constraints, pressures,
-physiologic limits, psychological processes (including cognitive
capacities and mativationa~ factors), and the many interfaces between
tenant' and machine required b~v the intense interdePendenci~= of human
and technoloci~1 forces in snare
_ _ -_ ~ As__ Such intense interdependencies in
this extreme are much less often observed on earth (with the possible
exception of certain medical contexts in which life is tenuously
maintained by sheer technological support).
Given this reality, one cannot extrapolate easily frog what is known
about society as we experience it on earth and "life aloft." It has
even been said that humans may become a very different species while in
space. Similarly, social systems which emerge to support and maintain
life in this context may deviate along many dimensions from those
social structures an] processes that are a part of our daily existence
and often so "routine" that they are taken for granted. ~
Nothing must
He considered as Routine An a novel environment. It must be said at
the outset that what we transport from earth in the way of social,
norms,
rules, shared expectations, roles, etc.) may prove much less functional
than we envisioned given a ccm-plet~ly altered social and technological
environment. Because we have virbnally no scientific evidence
psychological and organizational adaptive mechanisms (e.q.
329
OCR for page 330
330
concerning the parameters of life after eighty-four days in space (that
is, there is no U.S. experience to rely ony, one is forced to engage
An speculation an] extrapolation despite the potential pitfalls.
My reading of the documents we have been supplied with concerning
the space station program in the l990s and beyond and my very limited
exposure to ROSA through a two-day symposium, lead me to several
tentative conclusions regarding the most critical social contingencies
(besides the issue of conflict addressed by Michener) confronting NASA
as it plans for the extended duration existence of groups of
individuals in space with limited opportunity for replacement or exit.
These critical contingencies include the social and psychological
management of stress (regardless of the nature of the stressors) and
determination of the most efficient and socially productive mechanisms
for handling interpersonal communications (e.g. within the crew,
betwe ~ crews of different modules, and between the crew and the
"ground," including family melters and friends). The successful
management of both stress and interpersonal communications is critical
to individual and y~vup-level performance, productivity and ultimately,
'mission success." While there are many other issues which could be
investigated profitably from a sociological Perspective, time and space
limit the scope of this first foray into life as currently envisioned
on space stations.
STRESS, INDIVI1~L ~FORMPNCE AND UP PE%ODUCTIV~
Stress has been identified as a contributing factor in the etiology of
certain acute and Manic illnesses (e.g. ulcers, high blood pressure,
heart attacks, nervous dis ~ ances, etc.~. It has been demonstrated
to have consequent== not only for the hearth status of individuals, but
also for individual performance, decision-ma-ding and productivity.
With respect to spa~-relat=1 research Foushee (1986) shade= that an
important goal is "to under stand and munlmize the effects of acute and
long_duration stresses on group functioning." Although there is
enormous literature on the effects of stress on individuals,
researchers have been slow to address the impact of stress on groups.
Furthermore, the bulk of the existing research examines the physiologic
and psychological consequences of stress. There is much less work on
the antcr=~ents of stress, in particular the stresses creaked by social
factors (PearlLn, 1982~. Another limitation to existing research is
the tendency for investigators especially in experimental work to focus
on single, isolated stressors. This work is extremely important, but
it does not inform us about the interactive anchor cumulative effects
of ~rn~tiple stresso~.
Deficit and Measuring Stress
The most commonly city definition of sass is Hans Selye's, "the
nonspecific, that is, canon, result of any Planar upon the body, be
the effect any or somatic." In the tradition of research initial
OCR for page 331
331
by Selye (e.g. 1936, 1956, 1974) this "result" or reaction of the body
to stress is referred to as the "general adaptation syndromes' LEAS) or
"biologic stress syndrome." It consists of an alarm reaction
biologically detectable in such organs as the adrenal glands, thymes,
lymph nodes and stomach, followed by the stage of resistance
accompanied also by marked physiologic responses, then the stage of
exhaustion at which point Selye argues the acquired second-stage
adaptation is lost.
Other researche ~ emphasize the significance of the "cognitive
appraisal" of stressors (see Brezn~tz and Goidberger, 1982, etc.),
noting the importance of the "subjective, phencmenologi~a~ experience
of stress" which lies between the stressor and its effects. Some
definitions of stress include reference to cognitive appraisal, others,
like Selye's, do not. Currently, there is no agreed upon definition of
the term and existing differences reflect major unresolved theoretic=]
t~c in the field. Though they disagree on the significance of
cognitive appraised, researchers do agree on the common goal of
understanding adaptations to stress or the nature of coping
mechanisms. Mach of the current research focuses upon specifying the
nature of these mechanisms. Before discussing adaptations however, let
tic examine the problems associated with the measurement of stress.
Varicus approaches have been adopted to the problem of measuring
stress; none of them completely satisfactory. One of the most common
approaches to Rena DcmcOt, popular over the past two decades because it
can be applied outside experim~ta1 settings, is he "life~rents"
scale (e.g. Holes and Rahe, 1967) or the modified life events scale
(l~hr~ ark Did, 1974a, 1974b). Life~vents typically Ian
"bbje~ive events that disrupt or threaten to disrupt the individual's
usual activities" (see D ~ end and ~ reran end, 1974b:133, 1984).
Events listed on such scales include both he=~th-related (onset of
chronic illness, major illness or accident, etc.) and non
health-related events such as divorce, separation, increase ~ family
income, retirement, death of a spouse, pregnancy or remarriage, etc.
(see Thoits, 1981, for a cogent critique of the life-events approach).
m e main debate in this research tradition has been over whether or
net only undesirable events contribute to stress or whether events that
require change either desirable or undesirable produce stress. The
latter has been referred to as the "toter change" approach to measuring
stress, the former, the "undesirability" approach (Thoits, 1981~.
Ihoits (1981) identifies several studies suggesting that only the
undesirable changes significantly affect stress levels, although she
goes an to critique these studies as well as many of the total change
sues for fat ing to include independent indicators of their
independent and dependent variables. Her findings also suggest that
'Cohen he~lth-related events are controlled, other undesirable events
have small and nonsignificant effects upon psychophysiological
distress" (as Measured by reports of psychosomatic symptoms using the
MacMillan Health Opinion Survey Index). the main conclusion she draws
relevant to current research is that previously well-established
correlations between undesirable events and distress may have been
inflated *ue to the Operational confounding of hen1th-related items on
OCR for page 332
332
the indeperxlent and dependent variable scales." A major contribution of
new research on sass wed be to refine existing usurp of stress
and to develop more sensitive anal reliable ~asur~nt techniques.
Iaboratory research employs quite different methodologies than
surrey research, however, as Hoir~ and Tea s (1982:26) poirlt OUt,
"lab paradigms An biological science have rennet to isolate stress
Ins fee he psy~ologi~8~1 ~ 8~ it con ~ ·~' ~ ~ ~
measurement problems are reduced in this way, little knowledge is
~ inert ~ n ~ ~ e ~ ~ lay of physiologic, ~~ologi~s=8
8 _ ~ ~ _ _ _ '8 ~ * -- ~ 8888_8 ·8
8~ ial ~ at · Hol ~yd a ~ ~ z ~ s (1982:30) ~8aJ1 for "field
research that exam mes stress in the psychosocial context" and more
descriptive work on the sources of stress "that operate in naturalistic
settings." m e space station environment is a "natural" laboratory for
this type of research.
Multiple Stressors in Space
m e reality of space station existence includes the potential for
canto and intermittent exposure to multiple stressors. In this
regard it is not at all clear that much of the existing research,
except that done ~ analogous environments, can be extrapolated to
apply to the space station. Both the number and the magnitude of
stressors in the space environment is likely to be at the high end of
existing scales, and quite possibly off the scale. Only research in
rare, high stress situations will contain insights of direct relevance.
Sources of potential stress in space stations include sensory
deprivation, environmental factors like noise level, crawling, spatial
arrangements, and invasion of privacy, as well as isolation,
confinement, and the possibility of life-threatening dangers or crises
situations. Nickerson in his chapter for this volume includes in the
category of potential stressors: weightlessness, unfamiliar motion,
motion restriction, sensory and perceptual restriction as well as sleep
interference and acute medical problems. Work-related factors like
variety and intensity of assigned tasks, and workload, etc. may peso
be stressors in the space station environment. hoper (1983) indicate
that in many work environments work or job overload is a major
stressor. There is sane irrigation that working intensity arc time
pressure were factors that contributed to the problems experienced by
~ w FINS aboard the Skylab 4 Mission. Accordi~ to Holrgyd and
Lazarus (1982:24), "the individual who is constantly challenged by even
relatively innocuous occupational and social demands and who is, as a
result, repeatedly mobilized for struggle may be particularly
vulnerable to certain disorders (Glass, 1977~." Given the duration of
planned space station missions, the cumulative physiological,
psychological and social impact of intermittent and continue exposure
to multiple stressors must be investigated.
Another significant factor in space stations related to multiple
stressors is the recognition that the stressors will be produced by
quite different types of events and forces. Stressor~ may be produced
both by the astronaut's home environment, to the extent that s/he has
· ~
OCR for page 333
333
information about significant events occurring on earth (e.g. ~ the
lives of his/her close relatives and friends, etc.), and by life
aloft. Within the apse capsule, factors contributing to stress are
environmentally indho d resulting in both p ysiological and/or
psychological distress ~e well as socially induced, created by factors
associated with the ~ter~nal ernrir~nt, especially the intense
interred of the crew fez. Since both Eibysiologi~al and
psychological factors have been given more consideration in the
existing literature, ~ will emphasize the social forces likely to
indu<::e stress.
Identifying Socially Pod - stressors
Outside of the life-events tradition and research focusing upon
national stress (e.g. Cooper and Payne, 1978), there have been few
investigations of stress produced by interpersonal factors in small
group sett logs (Levine and Scotch, 1970). Potential causes of stress
settings requiring intense interdependence among group members
include hectic personality conflicts, incompatibilities in interpersonal
orientation an] style, an inefficient or inequitable division of labor,
a lack of perceived 1agit;macy cancer m ng the allocation of leadership
responsible ities or aubbority, the inequitable allocation of individual
or collective rewards, lack of a clear definition of role or task
responsible ities, uncertainty regarding the timing, coordination or
sequencing of related tasks especially when synchronization is a
critical factor, and the arbitrary or inappropriate exercise of
authority or influence (i.~. violating role prescriptions or norms
conocrning the use of private timed. Many of these factors have been
demonstrated to have significant impacts upon group functioning
non~ssful situation ~ may or may not be exacerbates ~
si - cations of high Ire. Pearl or ~unta~n~li~bing teams
i~icat-= that deer periods of high stress many of then= problems
became earthly salient ad ~ some coca= result in aborts attends
to reach We summit. Int ~ r ~ 1 conflicts appear to be a major
problem for many ~ itions especially when the goal of reaching the
summit is highly values by all and where there is a great d=~1 of
uncertainty about achieving the goa1. Connors (3985:147) also notes
that in s;~3ation roe arch, 'members of isolated and confined groups
who were incompatible showoff increased stress, withdrawal, and
territorial behavior. ~
Many of these potential stressors have not ~ ermine in the
context of group functions primarily Ruse the predominant Ides in
this area of ir~ry has In one of ir~ivi~1 functior~ir~. ~ will
Dent Are upon the limitations of such a Entice in a s~ibs~erlt
section of the paper.
OCR for page 334
334
Mongering Stress
Relate to the problem of measuring stress are identifying the
ante—ents of stress is the prdblen of monitoring sages. Unc~n~sive
Romania for anchoring Swiss at bath ~ individual (physiologic
and psyc~hologi~1) petrel and the group leered need devel ~ t, given
the potential dele~=rious consequences of high levels of stress for
individual and group functioning. An important byproduct of such
monitoring is that it will give us same insight into the interactive
and Negative impact of various stressors. Furthermore, it will
enable us to ads issues still urKler Agate r~ardir~ He extent to
dish the effects are linear, cUzVil~=r, or approximate a
ste~function (or three hold function). It may also be the case that
the effects of Canaan stressors are Story given that not all
the effects are potentially negative. me positive impact of stress
has been given little attention In the literature.
Personal ~acter~stics, Crew CX—osition and Stress
As several authors have suggest, the "right stuff" may be the 'dry
stuff" den it Ad; to the selection of crew ~ who will not only
have the nectary Animal and professional skills, but w;11 also
have the psychological arm social cx~etencies Reid for the
creation of effective Ant ~ rsonal relations and relatively s ~ h
grasp functioning on space station 'fissions." According to Biersner
and Hogan (3984:495), veterans believe that "social compatibility is as
important as technical skills for overall Antarctic adjustment" to
isolation. Social competence will become even more critical as a basis
for selection and training in the future as NOVA envisions shorter
training periods for some astronauts (e.g. teacher and congress=embers
~ space programs). me potential for commercial joint ventures with
NINA not only increases crew he~enei=, but also beans that same
apace station Seers ~ the U.S. Mule will In all lik~ihoa] rx~
have Me benefit of intense N~;A training (and selection).
Intriguing research by He~rei~h and his colleagues (e.g. He~ich
et al., 1980) on this basic topic suggests that at least one
aracteristic typically associated with the "right stuff"
constellation of traits, interpersonal competitiveness, may be
dysfunctional for producing smooth group functioning depending upon the
mix of personnel and the'' traits in any particular crew. As Connors
(1985:155) notes, Helmreich et al. (1980) "hypothesize that the
combined interests of task accomplishment and social compatibility will
be best served if crew Seers straw a strong work and mastery
orientation, but relatively little campetitiveness."
-
Group Decision~ g Ur~=r Stress
Relearn of particular interest to 2~;A is Me research on the
relationship between stress and decision-makir~ which Eric that
OCR for page 335
335
the experience of stress generally interferes with psychological
processes relater to effective decision-m2Xing. Janis (1982), for
example, reports the following reactions associated with stress during
decision-making:
(~) narrowing of attention span and range of perceived
alternatives,
(2) reduction in problem-solving capabilities,
(3) oversight of long-term consequences,
(4) inefficiency In information sear h,
t5) premature closure, and
(6) with intense fear, there is also temporary loss of perceptual
acuity and perceptual-motor coordination (Daffy, 1962).
Evidence further suggests that accelerating time pressure greases
the probability of these reactions, although clearly more r ~ is
needed on the temporal aspects of stress reactions as well as sibilation
specific/individual difference interaction effects. (Individuals in
certain situations are likely to respond differently both to stress and
to the demands of the decision-making tack.)
Janis (3982) also specifies five basic patterns of dec~sion-making
under stress. The first four patterns in the list represent
''5efective" patterns of response, the fifty is the term Janis uses for
the most adaptive respo ~ e pattern. Observed patterns of response
under stress delude:
(1) unconflicbeJ inertia
(2) unconflicted change
(3) defensive avoidance
(4) hypervigilance, and
(5) vigilance.
Of the four defective response patterns, hypervigilance is found to
be the dc~unant reaction under conditions of high stress or
n~ar-panic. As Janis (1982:77) note=, "Ex~=c~ive alertness to all
signs of potential brat rats ~ diffusion of attention...one of
the main so of cognitive inefficiency whenever scone fines
hypervigilant, and it prc}'ably a ~ 3unts for sam of be failures to
meet the criteria for effective dec~sion-malci~." R~U1LS also sew est
that other problems emerge in high stress situations. "slang with
cognitive constriction there is a marked tendency toward stereotyped
thinking in terms of oversimplified categories and reliance on
Oversimplified decision rules" (Ja m s, 1982:781. Two conditions appear
to enhance the probability of hypervigilanoe occurring in stressful
situations: unconflicted inertia (or the failure to react to early
warrungs) and defensive avoidance (e.g. procrastination). Additional
factors identified by Janis as associated with the antecedents of
hypervigilan~= are the lack of contact with family members or other
support persons, lack of perceived control and lack of preparatory
information alcut potential stressful events.
OCR for page 336
336
The prevention of "defective" patterns of response in threaten mg
situations has focused An recent years upon several strategies
including "benign preexposure to the threatening situation, stress
inoculation via preparatory communications" and various types of
relaxation techniques designed to mitigate physiologic reactions
(Janis, 1982:82; see also, Janis et al., 1982~. Research on these
techniques and the extent to which they are successful under specific
circumstance== continua. Extrapolation to situations likely to be
encountered in space stations must be done m~reful~y. Some techniques
may be effective for single stressors, but less effective in the face
of multiple stressors. Aga m, further research is needed. Certainly,
however, this research gives us some clues as to problems associated
with decision-making in highly stressful contexts.
A Comment on the Limits of Medic and Psycho~ogi~1 Models of
Stress: The underlying framework a researcher adopts to the analysis
of a problem often circumscribes both the nature of scientific inquiry
as well as conceptions of potential solutions. Thus it is not
surprising that medical research on stress tends to exam me primarily
physiologic response patterns and the impact of drugs on the
functioning of the individual undergoing stress. Psychologists
similarly focus on cognitive and emotional factors, examining
individual differences associated with cognitive appraisals of stress
and reactions. The solutions they consider include biofeedback, stress
"inoculation", and various types of individual training and therapeutic
techniques. All of this research is necessary since the problem
entails both physiologic and psychologist dimensions. What is
missing, however, flus much of the current work is the investigation of
the system properties of stress and exam mation of solutions to the
problems created by multiple stressors at the group or collective level
lasso sometimes called the system level. Inquiry of this type would
examine the interpersonal Dynamics related to stress responses and
adaptive strategies rather than treating the problem purely from an
~ntraindividual perspective. Adoption of an Interpersonal or system
level perspective would lead to quite different conceptions of adaptive
mechanisms. In Connors (1985:146) words, "Given that future missions
will require increase levels of cooperative functionir~, selection and
trainer'; procures ~st not only yield effective individuals, they
st yield effective g~vups."
The dominant Characteristic of Efface station missions in the near
future involving 6-8 crew m ~ cars marooned in space for approximately
ninety day intervals of the high degree of interdependence among the
group members (and possibly between groups in different Accuses at some
point). Stressors which significantly impact any single group member
will, of necessity, influence group function~ng--even if it simply
entails the reassignment of duties or tasks for brief periods of time
or temporary isolation of a group member. IT addition, group members
may be impacted similarly by stressors and thus collective solutions
should be explored. Strategies might be developed, treating the Group
~~ a social system (as Michener does) of interdependent parts and group
members sight be trained in specific response patterns through a
division of labor. For example, roles could be assigned such that each
OCR for page 337
337
attends to a specific problem associated with inefficient
decision-ma-ding under high stress. One crew member might be assigned
the task of vigilance with respect to only alternatives, another to
long-term conseguen~=, etc. and coordination might be achieved either
by an assigned group leader or some sort of computerized decision-aide.
Ccmputer-a'~ed systems could be developed which help to meliorate
common deficiencies observed in cognitive processing during peak
periods of stress. Coping strategies of this type are more like Janis'
suggestion that an appo~nt=d "devil's advocate" be used to mitigate the
negative consequences of "groupthink." They have the possible advantage
that "failure" is not localized in a single individual (typically, the
"leader") who ~ st assume full responsibility for group decisions in
"crisis" or intensely stressful situations. Furthermore, a clean
division of labor also reduces the workload on any single individual
under stress.The work on distributed decision-m2Xing by Fischhoff and
others may well provide models for this ~ of coping mechanism.
Relevant work on computer-aided decision-making should alto be
explored.
Mediators of Stress and Adaptation
In the words of Ho~royd and Lazarus (1982:25), "It has been
increasingly acknowledged that health Outcomes are a product of
effective coping rather than simply a consequence of the presence or
absence of stress." Identifying factors that result in effective
coping is an important research agenda item, however current
investigations focus more on drug therapy, biofeedback and
"cognitive-behavioral" interventions to codify responses to stress and
facilitate coping. The social and organizational management of stress,
as noted above, has not been examined. Psychological approaches take
us one step beyond physiologically focused management strategies, but
even they have not been evaluated extensively.
Coping mechanisms and adaptation responses form one axis of current
research, the second axis is extensive work on factors that ' ~ diate"
the stress response. Such factors include individual difference= which
relate not only to susceptibility, but also to cognitive appraised and
effective coping. Variables incorporated into these investigations are
ethnicity, age, gender, occupation, income, level of education, marital
state=, health status and ares to social support (i.e. personal
resources and network supplied resources), among others. Access to
social support, for example, has been demonstrated to mitigate some of
the effects of stressful entreats (e.g. Caplan arm Killilea, 1976~.
Much of this work is useful for general medical and scier~tific
pi, kilt caution must be exercised when attesting to generalize
these findings to astronauts and the Apace station ~nriror~nent. me
range of variation on same of these variables is quite restric ~ In
the astronaut population, although increasing heterogeneity must be
assumed along many of these dimensions (i.e. gender, age and
ethnicity) in the future.
OCR for page 338
338
Rewarm linking gamer to stems, for example,
varied of sties Off wan are more susceptible
indicate= ~ a
to stress (e.g.
Kessler arid MeRae, 1981~; given certain levels of stress they report
higher levels of distress as reflect Epically In Apology
(prim rily self-reports). Pa ear ~ discussed by Kessler art MbI ~ d
(1984) documents that waken ten] to be more affected by undesirable
life events than men even though they do not report significantly more
_
~ = ~ To___ ~ a IT ~~= ~ ~ ^^ A ~ ___— ~ ~ TV_— _-D ' _~.L~
::~U~1 Novell Hi- ~~1~:L Olaf l~l~;L~U \1:70~) ~~W~.~.t,. .L=~ r!3::i ==,~ lot t~
that women are more vulnerable to network' events, events that happen
to significant others in their networks' than men, and it is this
difference that accounts at least An part for previously observed sex
differences in responses to stress. Thus, they argue that waken are
not "pervasively more vulnerable than men to stress," but vulnerable
specifically to stress linked to the important people in their lives as
a result of their "greater emotional involvement in the lives of those
around them." P=1le (1983) refers to this fact as the "stress of
r~r~g~t
There are mans unanswered congestions ~n~rnirm the link between
_ _ a, , _ _ ~ _
_~_ _~ ~~_~ ret ~_' - ~ .: ~` p~1 ~ ^~' ___ ~ ^_ it__
qua = 1~1 ~ ~~ ~ 1-11~: AL ~Q WIll=1 L - All ~~VI1=U~ ~ lllL' ~=
vulnerable to secrets than Pro e astronauts is an open question. FEW ot
the existing stud)'== include in their samples waken On such high stress
occupations and it may well be that women with high capacities for
coping with stress self-select into these occupations (e.g. as is
likely the case for women mountain climbers). It should also be noted
that many of these studies reporting sex-related stress differences are
based on sample day obtained in the 19SOs, 1960s and early 1970s;
little evidence exists hack on more recent data including samples of
women In more varied occupations contexts and roles.
Impact on Productivity: Individual and Grcup-Level Effects
The link between stress and productivity has been demonstrated to be
somewhat complex. Mangler (1982:94) argues that "the problem of stress
is twofold; both the initial autonomic signals and the conditions that
generate these signals require some conscious capaci~y...an] therefore
interfere with the performance of targeted tasks." What is not clear is
specifically how and under what conditions performance is impaired. An
fact, as Mangler (1982:96) indicates, like noise, stress reduces
"attentional capacity and narrows it to central t~Sks,t' thus if the
target task is central , "then autonomous arousal may improve
performance." mis depends upon both the centrality of the target bask
and specific characteristics of the task, or bask sequence which
requires performance. Early research on this topic seemed to suggest
that there is a curvilinear relationship between arc usal and
performance such that performance is enhanced by moderate levels of
arousal, but impaired significantly at both very low and very high
levels of arousal. The generality of this effect is still under
debate. Mangler (1982:95) concludes that "understanding the relation
between efficiency and stress requires an analysis of specific
stressors, an approach to arousal that assigns it definable
OCR for page 339
339
properties..., and knowledge about the requirements of the task."
Research by Baddeley (1972) and others indicates that stress associated
with dangerous environments "affects performance through its influence
on the subject's breadth of attention...but we still do not know what
mechanisms mediate the effect of arc usal" on attention span or even
what is entailed in the adaptation to fear.
Evidence suggests that problem~solving abilities are affected by
stress in much the way Janis indicates that decision-making is
impacted. In particular, "if much of problem-solving involves the
manipulation in consciousness of alternatives, choices, probable and
possible outcomes and consequences, and alternative goals," then stress
interferes with efficient problem-solving. Few alternatives are
actually considered and the thought process is guided more by
habituation and stereotyping than by the conscious weighing of
alternative strategies. What is needed, he argues, is "f~ne-gra~ned"
analyses of these processes. It; t.?;lh The ',~ ;
~ . . . ~ . . _
--~wu~a~ ~ =11 W401 =~ - ~~ it
has restricted experlmenb=1 work on these problems (candler
1982:101). A related shortcoming is the failure to consider the social
context of problem-solving behavior. The balk of the research days
with individual tasks, not collective or highly interdependent tasks.
A Peso arch Agenda: System-Level Responses to Stress
In the previous era when highly trained male pal ots were selected as
astronauts on the basis of p ysical stem m a, high tolerance for stress,
psychological stability and technical competence for space missions
involving relatively short-term exposure to multiple stressors in
dangerous environments, l ~= attention was paid to research on stress.
In fact, Candler (1967) noted in his early sees of highly tra wed
astronauts a lack of anticipated stress responses; these men had been
"trained to have available response sequences, plans and
problem-solving strategies for all imaginable emergencies" thus
emergencies were transformed into llrcutine sip mbions't and therefore
not experienced as stressful. At this stage in the space program
harm was the primary focus of Bach selection and trainer. Even
E pace capsule design divisions were not frequency made in order to
runi~ze ernrir~rent~lly ironed stress or to impose Inhabitability"
(Clearwater, 19851.
me future holds form a different scenario. First, astronaut
selection procure have changed to include Ron white ones and
scientific personnel as well as pilcdcs. m ere ~ greater dived ity
among potential astronauts in optional training, gender, age,
ethnicity, and personality traits. Given this heterogeneity and the
increased oa~lexi~y and duration of space station missions, emphasis
most now be placed (as Helmreich, 1983; Foushee, 1984; and other social
scientists have argued) on the selection and training of highly
compatible crews especially as group size increases to eight or more in
relatively small modules. In addition, only recently has habitability
become an integrated aspect of the space station design process.
Alterations in selection processes to maximize crew compatibility and
OCR for page 345
345
decision-making contrasting computPr-mediated communication with
face-to-face groups. However, Herr and Hiltz (1982:121-122) identify a
wide range of other group and organizational level impacts, some of
which correspond to Kiesler's concerns. the group-level hypothesized
cognitive impacts include: (~) the creation of "on-1ine" groups or
'communities of int~rmSttt, (2) improved group decisions, and (3) an
increase in "knowledge-based authority," etc. With respect to group
decisions, the findings cited are mixed. On the positive side results
suggest that the capabilities of data base searches, increased ads
to information and access to decision-aides enhance group
problem-solving and decision-making. As Turoff and Hiltz (1980:123)
indicate "the cc mputer can aid in gathering subjective estimates within
a group" and facilitate the resolution of disagreements.
While Kerr end Hiltz (1982) indicate some empirical support for "at
least the same qualify of solution" den ~arir~g cc~uter-mediated to
face-t~face grumps (Turnoff, 1980; Hiltz et al. 1981~; Kiesler et al.
(1984) and Siegel et al. (1986) report a decrement at least with
rasped to time to solution for the ~uter-mediat~ grips. Others,
Barr and Hiltz (1982) note, (SAC J6hans~ et al., 1979) argue that more
conflict may result face the increased access to alternative vim and
that a "false sense of Have consensus" mav arise (Barr arm Hiltz
1982:125).
.,— — —a. . __ _ ~ ~ ,
On group prciblem-solving Burr arm Hiltz (1982:124) cite the work of
Lipinski et al. (1980:158-159) which suggests that when considering,
the "~k-focused c=~cations ~i~ by groups involved in joint
prdblen solving, ~uter-bas~ Fornication systems are appropriate
in the structuring, evaluating, and locating phases of problem
solving, since time delays are acceptable, written rinses are
appropriate, ark face-t~face contact is not essential." However, they
go on to state thErt He "implementing, s~r~, and conceptualizing
stages of problem solving are less amenable to this technology." More
resort Is never concerning the phases of prdblern solving and the
effects of cc~nputer mediation.
The following list includes same of the hypothesized behavioral
ilr pacts on grad identified by Burr ark Hiltz (1982:132-133). Many
have not been sufficiently investigated to praise definitive
evidence. (Adapted fan Burr and Hiltz:)
1.
2.
3.
4.
Cauterized Fornication Uses cross~group c~nication.
It incises lateral network linkages armors organizations.
It incenses Bateau network Mirages within organizations.
Catheterized Fornication may Charge social struck from
pyramid or hierarchical to net~rk-shap~l.
5. It Ranges He c~tr~ity of fibers within grips.
6.
7.
8.
It increases the possible Span of control.
It can increase tile effective limits on the size of working grips.
It Onuses the Persia of social networks, increasing
connect~ness.
9. It increases Opportunities for decentralized c~nication.
10. Winterized Fornication may increase informal cc~nications.
11. It Farces who tows to Awn.
OCR for page 346
346
12. Groups take longer to reach agreement and consensus is less likely.
13. Computerized communication scmetim£s mats it difficult to focus
. . .
c 1scusslons.
14. Regularity of indivi*u21 participation is sometimes difficult to
enforce.
15. There is greater equality of participation than An conventional
media.
Communication Network Structure, Centrality and Power
Prior research on communication networks In the social sciences
provided evidence that the specific configuration or structure of the
network affected the efficiency of problem solving groups. But more
recent research tents to indicate that these results may not be valid
for mandated communication systems. Subjects in various fcur-person
network structures, given telephone contact capabilities, were able to
come to consensus on group decision problems without much variation in
degree of consensus or t;~ to achieve consensus across structures (see
Frie~kin ark Cook, 1987). Results freon the co~put~r-mediated version
of this e~per;,nent are not yet complete.
Centrality has been Tire to ocher in various studio= of
communication at in networks in which r~r~ other than information
are unharmed (see Furman, 1979; Cook et al., 1983~. ~
c ~ puter-m ~ lath corns nication networks centralizer is lir ~ to ache ss
to information an] control over the flow of information. To the extent
that computer-mediation alters these parameters decentralization of
power may I. Peso arch is needed which examines the relationship
between the structure of the communication network an] control over
information channels. Certainly as Kerr and Hiltz (1982:150) indicate
"opportunities for decentralized communication are increased" in
computer-mf~;ated networks, "becalm=- it is easier to keep all those
concerned with issues informed and up to date." Thus the efficient
flow of information is enhanced. But efficient decision-ma-ding in
grc ups in which commune cation is computer-m£diated may require
structured access to information rather than open access during the
final stages of decision-making. Levels of access to information
rather than the availability of communication channels becomes the
critical determinant of positional centrality and thus power in this
circumstance. Further research on these topics is needed.
Communication Networks, Authority and Control
Kerr and Hiltz (1982:125), among others, predict that computerized
communication increases the "appreciation of knowledge- rather
than hierarchical authority." If this result is general, it will be
Important to study the conditions under which conflict can arise
between knowledge-based and hierarchical authority structures.
Efficient group functioning and problem solv m g is likely to be
enhanced when there is minimal conflict between these sources of
OCR for page 347
OCR for page 349
OCR for page 350
OCR for page 351
OCR for page 352
OCR for page 353
OCR for page 354
OCR for page 355
Representative terms from entire chapter:
multiple stressors
347
authority. Furthermore, hierarchical authority and command systems
must be designed in such a way that information flow is not tightly
hierar~hic,~ly stn~cb~.
As noted above, in particular, in systems involving highly tra wed
professionals the upward flow of critical information must not be
circumvented by bureaucratic praceduras or restricted communication
channels. Maximization of group productivity and problem solving
efficiency is likely to occur under conditions of open access to
communication channels rather than strict hierarchical access under
conditions of complex tasks, high uncerta sty and a highly
professionalized staff. ~ '~ -
b~l~lC ~ on opts =~tive
authority structures urKler varying animation network stricture are]
task conditions is r~i~.
With respect to authority anal control in systems using
~uter-mediated communication networks, two additional impacts city
by Kay and Hiltz (1982:150-151) are relevant. May argue (p. 150)
that "greater delegation of authority is possible winch the capacitor for
account~hili~r and reprising decisions ~ a timely and orderly
mariner." Secorx], they argue (p. 151) that it "in~ase5 the possible
span of control" ark "allayers more centralized control oven
geographically disper ~ units." Computerized d~
348
Existing relearn is focused on earth bash Fornication networks
primarily a~r~g colleagues or remote xnb~s of interest gray where
the exercise of authority is rarely an issue. Information expunge is
frequently the primary or sole goal of the interaction. Thus
e~crapolation foam the r ~ ts of s ~ ies on these networks must be
treated as highly speculative. New research must be designed around
the specific problems and parameters facing crews in space.
Simulations could be designed which would mirror some of the most
critical circumstances and used to evaluate alternative network
structures, systemic of controlled versus c pen access to information,
seven different tulles and levels of complex tasks. Problem solving
Decency and group productivity woula He a primary focus of the
Usury, although other issues sup as increase social c~runi~tion
between crew melters an] ground personnel would also need to be
addressed ~ terms of the impact on mission success, broadly defined.
Priority should be place] on the development an] evaluation of on-~ne
data collection systems for post-IOC space station missions and other
long-~uration, 'reannex" missions concern mg the multiple impacts of
computer-mediated communication systems.
_ - .
am_ e 8 _ ~ , 8 8,
.
Summary Statement Concerning Research Needs
The 1986 Challenger disaster was as much a failure in organizational
decision-m2Xing as a technical failure in the right rocket booster on
ache shuttle. This fact atoms to the envy in organizational
condoms for scientists and managers to focus attention primarily on
the technological aspens of systems rancher then the social aspens of
system design. Historically, ~ Me social scions=, as well as the
physical sciences, productivity has been viewed furrily as a
problem of ~hnim~1 system or organizational design and innovation.
those Ho design and evaluate complex systems which rewire human
pa~cicipation, hammerer, ~ st eventually recognize the significant role
of psychological and social factors in productivity. Human factors are
now incorporated in NASA's research program, but this is a recent and
fairly ~m~11 beginn mg given the time frame within which research
commitments are necessarily made.
My recommendations assume that technical and social systems can not
be designed in isolation of one another arm that interdisciplinary
research High cusses the invisible bouncy between the physical and
social sciences is rewired. `=ignir~ apace stations which are
maximally habitable and which optimize human comfort, satisfaction and
pa ~ ctivity and m ~ Size the sense of isolation ark ache stresses
associated with risk and uncerta sty, as well as the potential for
~ntra-grcup and ~nter-group conflict is as critical a goal as the
flawless design of structures which will provide the technical support
for "life aloft".
Research on many critical ~ of social system design is simolv
- ~- In Dart this is because the technologies under
, _ _ _ , ~ , _ ~ ~ __ _ Be, _~
nck available. _ , ~ -_ _ ~ _
consideration are new (e.g. computer-m£diated networks to facilitate
interpersonal ccmmunication are relatively recent); but also in part,
349
this state of the art is a function of national priorities arm
budgetary constraints. Hopefully, this situation will change. The
quality of life in space in the t~r~ty-first century will hinge upon
decisions we may during this decade as to wit reseat is necessary
to T~mize not only productivity, the bott~n line for many, but also
less tangible qualities such as habitability, sociability and
livability. The space station is, after all, a place to be inhabited,
a mini-society which at some not too distant time in the future must
begin to cope with not only the technologist requirements of its
environment, but also the psychological and social newt of its
i ~ itants and the social constraints and requirements of an emerging
s ~ icily. Recruitment, selection, training, sustenance an] replacement
of persons will be as critical as the maintenance and replacement of
Pal.
The following is an Abbreviated list of research newt (see Table 1)
which I have suggested in the text of this report related to social
factors involved An space station design during the post-IOC phase.
The emphasis in this report has been placed on issues related to
stress, its causes and consequence=, and the impacts of
c~uter-m~iated c~nication systems (since that is currently the
primary Modality envisioned. ) I heave only scratched the surface.
In conclusion, it is important to note that as with many of the
r~ear~hpr~r~; of NINA ar~Un~versi=-based scientists, the benefits
to be derived freon the pry research extend far beyond the limits
. . · . . .
-
purposes of future spa ~ station missions. Improved methods for coping
with multiple stressors in hostile environments and a better
understanding of the social and psychological effects of
computer-m-~;ated communication systems have great potential
applicability in a wide range of human social contexts. m e payoffs
for society as we know it on earth are potentially even greater than
the payoffs for life as we envision it on space stations in the next
century.
350
ABLE 1 Self Research New: Social Factors ~ ivity
I~-l~ration Space Station Missions
Social stress, Eon Activity am Group E~tionir~:
(1) Develop more oc~z~sive arm praise neasu~ of stress levels for
situations i~lvir~ multiple stressorse
(2) Pesear~ arm develop stress Ditorir~ A, on-lir~ data acll~ion
pros, arxt more Abusive measures of stress.
(3) At r~ ~ ;~rsanal characteristics= (~.g. personality divisions,
Gerber, e=.) awl -Pacific responses to str~;s and adaptations to stress on
lo duration Span station m, cci=S.
(7)
(4) Amine gr=2p composition factors Rich maximize efficient group functioning
~ multiple stressors.
(5) Prearm ~ Pacific i—acts of t ins of multiple str~sors ~ irxtivi~1
arm grow deci_icn-0ci~ pa. Assess the effectiveness of different
ocp~ strategies arxt decision aides urger v~yir~ levels of stress arm
c~inatia~ of stepsons.
(6) E~1 r~ an He camps of stress to in`:lu~ as well as Ecological
arx! p~siol~ical faiths social factors such as gram size, gram
cc~ositian, division of 1~, workload, Empties of equity in the
~ccignme~ of tasks arm ~;ibilities, styles of leadership, type arm
Dane of Stack with significant others, etc. on lam - uratian missions.
Begin to develop process wars which relate stress to i*xtivi~
performance and griever h~tionir~ arm specify the asr~itims Sexier
whim the ~ of ir~ivi~1 ~rforma~e seriously Apses= group
Tinier.
iated ~icati~ As, Eon ~ctivit,, arm Alp
tianir~:
(1) ~ em research cn me social Pacts of a~t~iated
= cation gist—E; on irxtivi~al decisi~-naki~ and group pmblem
solvers.
(2) Investigate the effects of cxn~r~iatioa~ in relation to Of phases of
group problem solvers, oapplexity of the tasks am variations in the levels
of envi~ta1 stress am A.
(3) C~ct demean can canp~ter~;~ted Fornication Stems am ~
distribution of poser arm authority. lr~restigate in particular the potential
for conflict between knthrledge~ and hieratical authority structure
and Off link between centrality arm the exen:ise of pager arm influence.
(4) Tr~restigate the E=ter~tial a= of Alter radiated canmu~cation
between crew Airs am significant athers ~ earth atter~ir~ to issues of
privacy, social support arm the effects on responses to isolation,
nfi~rit arm ~ r stressors on apace station missions.
In the future, research the differential impacts on individual performance
and group functioning of various types of mediated ~`u.uDnication systems
(including audio and video channels).
(6) Examine factors related to communication Orbits and aortas to
cereal mcation channels which inhibit the upward flow of critical information
(especially negative information) and mechanisms which circumvent this
E,~1em.
(7) Consider the effects of computer-mediated communication on the relations
between crew members and ground personnel and between crews of different
modules With respect to the potential for intergroup conflict and develop
mechanisms to mitigate conflict.
351
F~RENGF~
Baddeley, A. D.
1972 Selective attention and performance ~ dangerous
environments. British Journal of Psychology 63:537-546.
Belle, D.
1982 m e stress of caring: women as providers of social support.
Pp. 496-505 in L. Goldberger and S. Breznitz, ens., Handbook
of Stress: Theoretical and C1 mimal Aspects. New York: The
Free Press.
Biersner, R. J., and Hogan, R.
1984 Personality correlates of adjustment in isolated work groups.
Journal of Research in Personality 18:491-496.
Breznitz, S., and Goldberger, L.
1982 Stress research at a crossroads. Pp. 3-6 On L. Goldberger
and S. Brezn~tz, eds., Handbook of Stress: Theoretical and
Cl~nir=1 Aspects. New York: The Free Press.
Broadbent, D. E.
1971 Decision and Stress.
CaplanJ G. J and Killilea M., -do.
1976 Support Systems and Putual Help:
Explorations. New York: Grune and Stratton.
New York: Academic.
Multidisciplinary
Champness, B. G.
1971 Bargaining at Bell Laboratories.
Paper E/71270/CH,
Communication Sues Group, London, England.
Clearwat~r, Y.
1985 A human place ~ cuter space.
34-43.
Psychology Today. July,
Cook, K. S., Emerson, R. M., Gillmore, M. R., and Yamagishi, T.
1983 The distribution of power in exchange networks: theory and
~ ~ American Journal of Sociology
expert 1 results.
89:275-305.
Cooper, C. L., ed.
1983 Stress Research: Issues for the Eighties. Chicest~r:
John Wiley & Sons.
352
Cooper, C. L., and Payne,
1978 Stress at Work.
R.
Chicester: J6hn Wiley & Sons.
Connors;, M. M., Harrison, A. A., arm Atkins, F. R., - a.
1985 Living Aloft: Iran Requirements for Extended Spaceflight.
W~ton D.C., VIA.
Danford, G. S., Ericson, J. E., Freeran, H.
Lewis Jr., W. C., Sc~hoor~oven, C. K., Sims,
Wellens, A. R., and Wise J. A.
1983 humane space stations. Pp. 63-95 in Autonc~7 and the Ivan
Elenent In Suace. Final Resort of the ~SA/ASEE: Sammy
Faculty Workshop, NASA.
Dohre~rend, B. S. and Ddhrer~r~, B. P.
J ., Hays , D. G.,
H. P., Walsh, P. J.,
1974a Overview and prompts for r~r~ on stressful life events.
Pp. 313-331 ~ B. S. Dohrerw~ and B. P. Dcibr~r~, eds.,
Stressful Life Events: Their Nature and Effects. Near York:
Wiley.
Dohrenwend, B. S. and Bohr ~ end, B. P., eds.
1974b Stressful life Events: Their Nature and Effects. New York:
Wiley.
1984 Stressful rife Events and Their Contexts. New Jersey:
Rutgers University Press.
Duffy, E.
1962 Activation and Behavior. New York: M~Graw-Hill.
Foush^=, C.
1982 the role of communications, socio-psychologi~1 and
personality factors in the maintenance of crew coordination.
Aviation, Soace, and Environmental Medicine 53:1062-1066.
1984 Dyads and triads at 35,000 feet: factors affecting group
process arxl aircrew perfor~nc-~. American Psychologist
39:885-893
Foushee, C. and Helmreich, R. L.
1987 Group interaction and flightcrew performance. 1b appear in
E. L. Wiener and D. C. Nagel, Is., Iran Factors in Modern
Aviation. New York: Academic P=ss (forthcoming).
Freeman, L. C.
1979 Centrality in social networks: conceptual clarification.
Social Networks 1:215-239.
353
F~i~ki=, N., and Cook, K. S.
1987 fornication Network Stances and Social Influence
Excess== In Small ~oblen-Solving Groups. To be presented
at the annual meetir~s of the American Sociological
Association, Chicago, August, 1987.
Glass, D. C.
1977 Behavior Patterns, Stress end Coronary Disease. Hillsdale:
Erlbaum
Hel~eic h, R. L.
1983 Applying psychology ~ outer Apace: w fulfilled promises
revisited. Pp. 445-450 On American Psychologist. April.
Helmreich, R. L., Wilhelm, J. A., and Range, T. E.
1980 Psychological considerations ~ future space missions. Pp.
1-23 in T. S. Cheston and D. L. Winters, eds., Human Factors
of Out ~ Space Production. Boulder' Colorado: Wes~view
Press.
Hiltz, S. R., Johnson, R., and Turoff, M.
981 The quality of group decision-making in face-to-face versus
computerized conferences. Presented at the annual meetings
of the American Sociological Association, Toronto, Canada.
Holmes, T. H., and Rahe, R. H.
1967 m e social readjustment rating scale. Journal of
Psychosomatic Research 11:213-218.
Holroyd, K. A. and Lazarus, R. S.
1982 Stress, Coping, and Somatic Adaptation. Pp. 21-35 ~
- L. Goldberger and S. Breznitz, eds., Handbook of Stress:
Theoretical and Cl Apical Aspects. New York: The Free Press.
Janis, I. L.
1972 Victims of Groupthink. New York: E arc curt, Brace.
1982 Decisionmaking under stress. Pp. 69-87 ~ L. Goldberger and
S. Brezn~tz, ens., Handbook of Stress: Theoretical and
Clinical Aspects. New York: m e Free Press.
Janis, I. L., Defares, P. B., and Grossman, P.
1982 Hypervigilant reactions to threat. In H. Selye, ea., Selye's
Guide to Stress Research. Vol. 3. New York: Van Nostrand.
Johansen, R., Vallee J., and Spangler, K.
1979 Electronic Maatings: Technological Alternatives and Social
Choices. Addison-Wesley: Reading, Mess.
3S4
Kerr, E. B., and Hiltz, S. R.
1982 Pp. 114-151 ~ ~ter-Miadiated Cation Systems:
Sta—~-= and Evaluation. New York: Academic Ens.
Kessler, R. C. ~ ark Mood, J. D.
1984 Sex differences in vulnerability to undesirable life events.
American Sociologic Reviser 49: 620-631.
Kessler, R. C., ark florae Jr., J. A.
1981 Trots in the relationship between sex and pi hological
distress: 1957-1976. American Sociological Reviser
46: 443-452 .
Kiesler, S., Siegel J., and M~Guire, T. W.
1984 Social psychological aspects of ccmput~r-mediated
communication. American Psychologist 39:1123-1134.
Levme, S., and Scotch N. A., eds.
1970 Social Stress. Chicago: Aldine Publishing Co.
Lipinski, H., Spang S., and Tydeman, J.
1980 Supporting task-focused communication. Pp. 158-160 in
A. R. Benenfeld and E. J. KazlausXas, ens., Communication
Information: Proceedings of the 43rd ASIS Annual Meeting.
Knowledge Industry Publication: White Plains, New York.
Meddler, G.
1967 Invited commentary. In M. H. Appley and R. Trumbull, eds.,
Psychological Stress: Issues In Research. New York:
Appleton Century Crofts,
1982 Stress and thought processes. Pp. 88-104 in L. Goldberger
and S. Erezn~tz, eds., Handbook of Stress: m eoreti~1 and
Clinical ~ s. New York: m e Free Press.
Pearl m, L.
1982 The social context of stress. Pp. 367-379 in L. Goldberger
and S. Brezn~tz, eds., Handbook of Stress: Theoretical and
Clinical Aspects. New York: me Free Press.
Selye, H.
1936 A syn~rc=e produced by diverse nocuous agents. Nature
138:32.
1956 The Stress of Life. New York: M~Graw-Hill.
1974 Stress Without Distress. Philadelphia: Lipp~ncott.
Siegel, J., Dubr~vsky, V., Kiesler, S., and Require, T. W.
1986 Group processes in computer-meSiated communication.
Organizational Behavior and Human Performance 37:157-187.
355
Simes, D. K., and Sirsky, P. A.
1985 Human factors: an exploration of the psychology of
human-cc mputer dialogues. On H. R. Hartson, ea., Advances
Human-Computer Interaction 1:49-103. Norwood, N.J.: Ablex
Publishing Corp.
Thoits, P. A.
1981 Undesirable life events and psychophysiological distress: a
problem of operational confounding.
Review 46:97-109.
American Sociological
Turoff, M.
1980 Management Issues in Human Communication Via Computer.
Presented at the Stanford Conference on Office Automation.
Stanford, California.
Turoff, M., and Hiltz, S. R.
1980 Structuring ~x~munications for the Office of the Future.
Presented at the National Computer Office Automation
Conference, Atlanta, Georgia.
Valice, J., Jchansen, R., Iipinski, H., Spangler, K., and Wilson, T.
1978 Group Communication Through Computers 4:Rep. R-40. Institute
for the Future, Halo Park, California.
