The restorative benefits of nature: Toward an integrative framework

STEPHEN KAPLAN
Department of Psychology, University of Michigan, Ann Arbor-, MI 48109-1109, U.S.A.
Abstract
Directed attention plays an important role in human information processing; its fatigue, in turn, has farreaching
consequences. Attention Restoration Theory provides an analysis of the kinds of experiences that
lead to recovery from such fatigue. Natural environments turn out to be particularly rich in the characteristics
necessary for restorative experiences. An integrative framework is proposed that places both directed
attention and stress in the larger context of human-environment relationships.
0 1995 Academic Press Limited
Introduction
Evidence pointing to the psychological benefits of
nature has accumulated at a remarkable rate in a
relatively short period of time. Whether a theoretical
understanding of these restorative influences
has kept pace with the empirical work is, however,
less clear. As Hartig and Evans (1993) have pointed
out, theory in this area has been dominated by conflicting
positions, one emphasizing stress reduction
(Ulrich, 1983) and the other concerned with recovery
of the capacity to focus attention (Kaplan &
Talbot, 1983; Kaplan & Kaplan, 1989). While it
might be argued that these positions are hopelessly
far apart, Hartig and Evans hold out hope for a synthesis.
The purpose of this paper is to propose a way
in which such an integration might be achieved.
A synthesis requires first that there be something
to synthesize. There must be entities or ideas distinct
enough and useful enough to warrant synthesis.
One piece of this is simple to achieve; there
is no disagreement over the point that stress is a
meaningful concept and that stress reduction is
aided by natural environment experience. Ulrich et
al. (19911, however, have questioned the usefulness
of the attentional concept in this context, and have
suggested that the performance deficits found in
research on attentional fatigue can be understood
simply as effects of stress.
Given these questions and alternative interpretation,
it seems appropriate to begin by focusing on
‘directed attention’ and its role in the Attention Restoration
Theory. Several studies that examine the
relationship between the natural environment and
restoration are then presented. This lays the
groundwork for proposing an integrative framework
which focuses on the causal matrix that connects
stress and attention. Finally, the paper offers some
directions for future research and concludes with a
brief analysis of the merits of distinguishing
between stress and attentional factors.
Directed attention
An important source for the attention constructs
central to Attention Restoration Theory is the work
of William James (1892). His ‘voluntary attention’
concept concerned the kind of attention that went
‘against the grain’, as it were. It was to be employed
when something did not of itself attract attention,
but when it was important to attend nonetheless.
Thus James emphasized the centrality of effort in
the employment of this kind of attention. Certainly
the themes of ‘voluntary’ and ‘effort’ suggest the
functioning of the will, a topic of considerable interest
to James. And in fact, by turning to his discussion
of will, it is possible to &in still further
insight into his thinking on this issue. In exploring
169
170 S. Kaplan
how one can support a weak intention, such as a
thought of something one ought to do but finds difficult,
James indicated that the only hope was to
inhibit all distractions. There was no way, he felt, to
strengthen the weak intention itself. The only support
one could provide required protecting it from
competing thoughts.
While James did not explicitly relate his discussions
of the will and of voluntary attention, it is
evident that in both cases the central construct is
that of focus, of supporting difficult mental activity
in the face of potential distraction. Bringing these
two Jamesean themes together yields a more general
mechanism than either would be separately. It
also suggests an interesting and far-reaching
hypothesis about the operation of this mechanism,
namely that it is inhibitory in nature. Although
James did not use the term ‘inhibition’, a mechanism
whose operation depends on the suppression of
competing activity must exert an inhibitory
influence.
An attentional mechanism that requires effort,
that can be brought under voluntary control, and
that depends upon inhibition for its operation offers
substantial explanatory promise. James’ writings
suggested this juxtaposition about a century ago; it
is remarkable that this potentially powerful theoretical
tool has been ignored for so long.
Although James emphasized the role of effort in
the voluntary attention concept, he did not address
the possibility that this mechanism was susceptible
to fatigue. While not having as well developed a concept
of attention, the eminent landscape architect
Frederick Law Olmsted not only understood the
possibility that the capacity to focus might be
fatigued, he also recognized the need for urban
dwellers to recover this capacity in the context of
nature. This was evident both in his approach to the
design of parks, and, quite explicitly, in his writing
as well (Olmsted, 1865).
More recently, clinical neurologists, working with
brain-damaged patients, have identified a remarkably
similar mechanism that they refer to as
‘directed attention’ (Mesulam, 1985). Interestingly,
they relate deficits in directed attention to damage
to the prefrontal cortex, a portion of the brain that
has long been associated with an inhibitory role in
mental functioning (Rothbart & Posner, 1985).
Since the concern of clinical neurologists is with
rehabilitation, their emphasis is less on specific
information-processing capabilities than it is on the
broader issue of getting along in the world. Thus
they have been particularly concerned with the role
of directed attention in ‘executive functioning’, the
capability necessary to lead an organized and purposeful
life (Lezak, 1982; Stuss & Benson, 1986).
While inspired by James’ notion of voluntary
attention for some time (e.g. Kaplan, 19731, we have
shifted to calling this concept ‘directed attention’ as
it avoids confusions others have had with James’
terminology (Kaplan & Kaplan, 19891.
Directed attention fatigue and basic processes.
Drawing together these various themes yields a
mechanism with the following properties: it
requires effort, plays a central role in achieving
focus, is under voluntary control (at least some of
the time), is susceptible to fatigue, and controls distraction
through the use of inhibition. While the
coherence and usefulness of this concept may not be
immediately obvious, in fact it is quite familiar.
What is familiar, however, is not the mechanism
itself, but the state of mind that accompanies its
fatigue. Any time one has worked intensely on a
project and subsequently finds oneself mentally
exhausted, one has experienced this unwelcome
state. The typical state of mind of students at the
end of a semester is a familiar example. In fact,
even a thoroughly enjoyable project, if sufficiently
intense and sufficiently prolonged, is likely to lead
to this same outcome.
More formally, any prolonged mental effort leads
to directed attention fatigue. It might seem peculiar
that a mechanism so intimately involved with
human effectiveness would be so susceptible to
fatigue. Yet, in evolutionary perspective, this apparent
limitation might have been quite reasonable. To
be able to pay attention by choice to one particular
thing for a long period of time would make one vulnerable
to surprises. Being vigilant, being alert to
one’s surroundings may have been far more important
than the capacity for long and intense concentration.
Further, much of what was important to the
evolving human-wild animals, danger, caves,
blood, to name a few examples-was (and still is)
innately fascinating and thus does not require
directed attention. It is only in the modern world
that the split between the important and the interesting
has become extreme. All too oRen the modern
human must exert effort to do the important while
resisting distraction from the interesting. Thus the
problem of fatigue of directed attention may well be
of comparatively recent vintage.
The importance of directed attention. One might
wonder just how important directed attention
fatigue is. Granted that scholars who need to concentrate
for long periods of time might be disadvan-
The Restorative Benefits of Nature 171
taged by this limitation. But there remains the
question of how significant the fatigue of directed
attention would be for people in general. It may
seem unlikely that a deficiency one has never even
heard of before could play a major role in human
thought and human effectiveness. Nonetheless
there are theoretical grounds for suspecting that
directed attention fatigue can, and often does, have
devastating impacts.
Selection. The capacity to solve problems is often
viewed as the crowning achievement of the human
mind. Even a cursory look at problem-solving
reveals the central role of attention. Consider a
mature individual approaching a problem. Such an
individual has available a remarkable array of
capabilities; there is considerable stored knowledge,
along with a multiplicity of perceptual possibilities
and a vast repertoire of possible actions. Much as
these are powerful resources, most of them are irrelevant
to the solution of any particular problem. The
very richness of possibilities presents a daunting
challenge. Solving a problem requires a focus on the
tiny portion of one’s repertoire that is pertinent to
the problem at hand. It is essential to select appropriately
from among the knowledge, the possible
percepts, and the potential actions. Ironically, the
larger the store of possibilities, the more essential is
the capacity for selection. In his chapter on Reasoning
(where he treated material closely akin to what
we would now call problem-solving), James (1892)
both emphasized the significance of selection and
illustrated it in the same passage: ‘To me now, writing
these words, emphasis and selection seem to be
the essence of the human mind. In other chapters
other qualities have seemed, and will again seem,
more important parts of psychology’ (p. 223).
In routine behavior much of the selectivity comes
from associative connections; one looks for the crucial
stimulus or performs the suitable action
because that solution has become habitual. But in
problem-solving, where well-learned connections
are often not available, some other means of selectivity
is essential. This is precisely the role that has
traditionally been assigned to attention (Moray,
1987). In problem-solving, where routine, associatively-based
attention can not be counted on, an
attentional capacity under voluntary control is particularly
important.
Inhibition and afrcect. Just as following one’s welllearned
patterns is often inappropriate to the solving
of problems, following the uncensored dictates of
one’s affective system can be equally maladaptive.
As Pennebaker (1991) has pointed out, there are
many times when inhibiting one’s impulses and
inclinations is essential. There are times one must
inhibit the inclination to flee, or to act without
thinking, or to act in socially unacceptable ways.
Carrying out actions that are unpleasant but necessary
also requires considerable inhibition. Indeed,
this function is so basic that Pennebaker has
characterized inhibition as ‘the linchpin of health’,
because, like the linchpin of a wagon, it is what
holds everything together. An inhibitory capacity
that is under voluntary control is thus an indispensable
mechanism for behaving appropriately.
Fragility. Directed attention is not, in itself, more
important to problem-solving than knowledge or
perception or action. Likewise, it is not necessarily
the most important component of the system necessary
to generate appropriate behavior. But unlike
these components, it is fragile. It is susceptible to
fatigue, and as such is more likely to be deficient
than are the other components. It is, in other words,
often the weak link in the chain. And for this reason
it may well be a critical resource in problem-solving
and in human effectiveness in general.
As one might expect of a mechanism with such
pivotal roles, it is central to the smooth and effective
operation of basic information processing functions.
And as is so often the case with psychological
mechanisms, its importance is most readily
observed by the consequences of its absence:
Perception. Lacking effective directed attention,
an individual becomes highly distractible, resulting
in impaired perception of material that is not
inherently engrossing.
Thought. Directed attention is necessary for
stepping back from the situation one is facing, for
pausing to get a larger picture of what is going on.
Thus without the aid of directed attention, it is difficult
to deal with situations in which the appropriate
action is not immediately obvious. It is also hard to
plan and to follow a plan. This leaves the individual
caught up in the demands of the immediate situation,
unable to transcend momentary pressures
and temptations.
Action. Inhibition is essential to delay and
reflection. Lacking this capability an individual
behaves in a less adaptive and appropriate fashion.
Without the patience and endurance necessary to
carry out difficult or unpleasant tasks, behavior
becomes more oriented to the short term. Social
172 S. Kaplan
behavior, which also depends upon inhibition,
becomes less appropriate. There is also a greater
inclination to be impulsive, to take unnecessary
risks, and to act in an impatient and hasty manner.
Feeling. Irritability is a hallmark of a person
who cannot draw on directed attention. There is an
interesting contrast here to stress, which is characterized
by anxiety. Anxiety often leads to seeking to
be with others; irritability tends to have the
opposite effect. Thus it is not surprising that there
is evidence suggesting that under these conditions
people are far less likely to be willing to help
one another (Sherrod & Downs, 1974; Cohen &
Spacapan, 1978).
Directed attention is, thus, a key ingredient in
human effectiveness. The fatigue of directed attention
is similarly a key ingredient in ineffectiveness
and human error. Unfortunately, even momentary
lapses in directed attention, at critical times, can
have dire consequences. Airplane pilots, ship captains
and operators of nuclear or chemical plants
provide vivid examples, since for each of these roles,
at least one major accident has occurred under conditions
when directed attention would be predicted
to be at a low ebb (Moore-Ede, 1993). In fact, a
study of airline crashes when equipment was not at
fault found in every instance that there were disruptions
of sleep schedules for key personnel (Wolfe,
1992). As the work of Broadbent et al. (1982) suggests,
a state of mental fatigue may well be at fault
in a large percent of those cases in which accidents
are attributed to ‘human error’.
The Restorative Experience
The restoration of effectiveness is at the mercy of
recovery from directed attention fatigue. Sleep provides
one approach to recovery. While useful, it is
insufficient. Certainly for serious cases of directed
attention fatigue, insomnia is likely to set in long
before full recovery has taken place. In order to rest
directed attention, it is necessary to find some other
basis for maintaining one’s focus. What is needed
is an alternative mode of attending that would
render the use of directed attention temporarily
unnecessary.
Fortunately there is a way to meet these requirements
and, also fortunately, it is widely available.
This notion too, like the directed attention concept,
derives from James’ perceptive analysis of attentional
processes. In his discussion, James contrasted
voluntary and involuntary attention. The
latter is a form of attention that requires no effort.
Although James does not deal with this explicitly, it
seems reasonable to assume that both types of
attention are similar in being inhibitory and in
having their effect through suppression of competition.
And to venture clearly beyond James’ analysis,
involuntary attention, requiring no effort, is
likely to be resistant to fatigue. Further, while the
individual is in involuntary mode, directed attention
should be able to rest. Since James’ terminology
has been confusing for many people and the
circumstances that call on the effortless attention
are intrinsically compelling, we have substituted
the term ‘fascination’ for ‘involuntary attention’
(Kaplan, 1995).
There are many sources and types of fascination.
Some of these derive from process. For instance,
otherwise normal individuals have been reported to
rouse themselves out of bed at an early hour in
hopes of catching a glimpse of a small, feathered
animal whose identity is uncertain. Likewise many
are addicted to books in which the identification of
the guilty party is difficult but not impossible to
predict, and generally is not resolved until the end,
even though far more efficient ways to transmit the
same information are surely available. Predicting
despite uncertainty as practised by gamblers provides
another example of process fascination.
Fascination can also come from content. As previously
noted, wild animals and caves are among
the many contents that do not require directed
attention. In some cases extremes of size lend to the
fascination of objects or settings. Fascination can
also derive from extremes along a ‘soft-hard’ dimension.
Thus, there is the ‘hard’ fascination of watching
auto racing and ‘soft’ fascination of walking in a
natural setting. Soft fascination-characteristic of
certain natural settings-has a special advantage in
terms of providing an opportunity for reflection,
which can further enhance the benefits of
recovering from directed attention fatigue (Kaplan,
1993). We have used the concept of ‘Restorative
experiences’ or ‘Restorative environments’ to refer
to such opportunities for reducing the fatigue of
directed attention (Kaplan & Kaplan, 1989).
Fascination is thus a central component of a restorative
experience. That is not to say, however,
that the presence of fascination guarantees that
directed attention can rest. Fascination is a necessary,
but not sufficient basis for recovering directed
attention. It is evident that some researchers have
failed to understand that fascination, although
important, is but one component of the model.’ We
have, in fact, proposed three additional components
The Restorative Benefits of Nature 173
that are integral to our analysis of what makes an
environment restorative (Kaplan & Talbot, 1983).
(1) Being away, at least in principle, frees one
from mental activity that requires directed attention
support to keep going. In fact, people often use
‘getting away’ as a shorthand for going to a restorative
place. But continuing to struggle with the old
thoughts in a new setting is unlikely to be restorative.
Clearly being away involves a conceptual
rather than a physical transformation. A new or
different environment, while potentially helpful, is
not essential. A change in the direction of one’s
gaze, or even an old environment viewed in a new
way can provide the necessary conceptual shift.
(2) The environment must have extent. It must,
in other words, be rich enough and coherent enough
so that it constitutes a whole other world. An endless
stream of stimuli both fascinating and different
from the usual would not qualify as a restorative
environment for two reasons. First, lacking extent,
it does not qualify as an environment, but merely an
unrelated collection of impressions. And second, a
restorative environment must be of sufficient scope
to engage the mind. It must provide enough to see,
experience, and think about so that it takes up a
substantial portion of the available room in one’s
head.
(3) There should be compatibility between the
environment and one’s purposes and inclinations.
In other words, the setting must fit what one is trying
to do and what one would like to do.
Compatibility is a two-way street. On the one
hand, a compatible environment is one where one’s
purposes fit what the environment demands. At the
same time the environment must provide the information
needed to meet one’s purposes. Thus in a
compatible environment one carries out one’s activities
smoothly and without struggle. There is no
need to second guess or to keep a close eye on one’s
own behavior. What one does comfortably and naturally
is what is appropriate to the setting (Kaplan,
1983).
The relationship of compatibility to one’s purposes
has several interesting implications. First,
one’s purposes generally are more readily achieved
when one has prompt and useful feedback from the
environment. An environment that is compatible
will thus be a responsive environment. Second, different
people’s purposes vary widely. If one’s purpose
is to be frightened by a horror movie, a snake might
be a compatible as well as a fascinating stimulus.
Likewise, members of a ‘snake-oriented religious
sect might find snakes both fascinating and compatible.
For a great many humans, of course, a snake
would fail the compatibility criterion. Third, carrying
out purposes often involves solving the problems
one meets along the way. As we have seen, solving
problems involves exercising selectivity, a key fimction
of directed attention. An ambiguous or distracting
environment raises many irrelevant possibilities,
placing more demand on directed attention.
A compatible environment requires less selectivity
and hence less directed attention.
A note on cognition
It may be appropriate at this point to address a central
misunderstanding of the role of information
processing in restoration. Ulrich et al. (1991) argue
that the compatibility concept is based on cognition,
and that cognition is too slow a process to play a role
in restoration. There is no doubt that if one defines
cognition as conscious, language-based, and dependent
upon reasoning, then it will in fact be a relatively
slow process. Ulrich et al. seem to adhere to
such a limited view of cognition.
By contrast, those who consider perception to be a
cognitive process-among them James (18921,
Bruner (1957), Attneave (19621, Shepard (19751,
Hebb (19801, and Margolis (1987)-make none of
these assumptions. And perception is, of course, a
very rapid process. Another example of rapid cognition
is the implicit prediction that people continuously
make about what will happen next in the
environment (Macphail, 1987). This process is generally
unconscious and typically surfaces only when
it is falsified by subsequent events. Thus, in an
example Hebb has used, one is not aware of making
a prediction when picking up a coffee cup at the
breakfast table. Should it, however, turn out to
contain cold beer, one is suddenly vividly aware of
having made a contrary prediction.
Another example is the inference of three dimensions,
made from a two dimensional array on the
retina. This process is likewise fast and unconscious.
There is also no doubt, terminology aside,
that rapid, unconscious human information processing
occurs in many forms and at many levels of
the system.
Nature and the Restorative Environment
An important theme of this paper, namely, the role
of the natural environment in human effectiveness,
has not been forgotten. We have talked about the
significant role that directed attention plays as a
component of effectiveness and about the need to
174 S. Kaplan
reduce the fatigue of directed attention in order to
restore effectiveness. We can turn now to the question
of how nature relates to restoration. One
approach to this issue is to consider the ways in
which natural settings are particularly likely to
meet each of the four requirements for a restorative
environment. The second approach will be to look to
the empirical literature to see what evidence is
available.
Revisiting the components of restorative
environments
Being away. Natural settings are often the
preferred destinations for extended restorative
opportunities. The seaside, the mountains, lakes,
streams, forests, and meadows are all idyllic places
for ‘getting away’. Yet for many people in the urban
context, the opportunity for getting away to such
destinations is not an option. However, the sense of
being away does not require that the setting be distant.
Natural environments that are easily accessible
thus offer an important resource for resting
one’s directed attention.
Fascination. Nature is certainly well-endowed
with fascinating objects, as well as offering many
processes that people find engrossing. Many of the
fascinations afforded by the natural setting qualify
as ‘soft’ fascinations: clouds, sunsets, snow patterns,
the motion of the leaves in the breeze-these readily
hold the attention, but in an undramatic fashion.
Attending to these patterns is effortless, and they
leave ample opportunity for thinking about other
things.
Extent. In the distant wilderness, extent comes
easily. But extent need not entail large tracts of
land. Even a relatively small area can provide a
sense of extent. Trails and paths can be designed so
that small areas seem much larger. Miniaturization
provides another device for providing a feeling of
being in a whole different world, though the area is
in itself not extensive. Japanese gardens sometimes
combine both of these devices in giving the sense of
scope as well as connectedness. Extent also functions
at a more conceptual level. For example, settings
that include historic artifacts can promote a
sense of being connected to past eras and past
environments and thus to a larger world.
Compatibility. The natural environment is experienced
as particularly high in compatibility. It is as
if there were a special resonance between the natural
setting and human inclinations. For many
people, functioning in the natural setting seems to
require less effort than functioning in more ‘civilized’
settings, even though they have much greater
familiarity with the latter (Cawte, 1967; Sacks,
1987).
It is interesting to consider the many patterns of
relating to the natural setting. There is the predator
role (such as hunting and fishing), the locomotion
role (hiking, boating), the domestication of the wild
role (gardening, caring for pets), the observation of
other animals (bird watching, visiting zoos), survival
skills (fire building, constructing shelter), and
so on. People often approach natural areas with the
purposes that these patterns readily fulfill already
in mind, thus increasing compatibility.
A nearby, highly accessible natural environment
cannot provide the context for all of these goals and
purposes. Yet even such a setting is likely to be supportive
of the inclinations of those who seek a respite
there. Consider the factory worker, racing off
during the lunch period, fighting traffic and distractions,
in search of a spot in the shade of a tree for a
peaceful break. If the peaceful effects were to be
worn off totally by the time the return trip is made
at the end of the hour, would this ritual be repeated
again the next day?
Attention restoration theory and natural
environments: empirical findings
Olmsted (1865) was particularly sensitive to the
role of ‘natural scenery’ in restoration: it ‘employs
the mind without fatigue and yet exercises it; tranquilizes
it and yet enlivens it; and thus, through the
influence of the mind over the body, gives the effect
of refreshing rest and reinvigoration to the whole
system’ (p. 22). His inspiration, as well as the
insights of others with intimate knowledge of the
natural setting, were most influential in the development
of national parks and in numerous other
conservation efforts (Nash, 1968). These early writings
relied on personal experience and literary talent.
Thoreau’s perceptiveness and foresight are perhaps
even more appreciated today than in his own
time (Anderson, 1968; Stern, 1970).
While such writings have great power and provide
deep inspiration for some, scientific evidence is
more compelling for others. The empirical literature
on nature benefits has been growing steadily
(Altman & Wohlwill, 1983; Kaplan & Kaplan, 1989;
Francis & Hester, 1990; Relf, 1992). Several studies
have addressed the role of nature scenes in holding
The Restorative Benefits of Nature 175
attention and interest (e.g. Ulrich, 1979, 1981; as
cited by Ulrich & Parsons, 1992).
There are four studies that speak directly to the
relation between restorative experiences and information-processing
effectiveness, a crucial issue in
establishing the distinct role of directed attention.
What is particularly striking about these studies is
how parallel the results are, despite wide variation
in setting and procedure. Several of these studies
are clinical- or field-oriented, an important type of
research for a theory that is intended to make a difference
beyond the laboratory. Such studies bring
with them their own set of challenges and limitations
of experimental control. The consistency of
findings both with respect to each other and with
respect to laboratory research is thus particularly
useful in establishing the construct validation of the
concept.
In the first study, Hartig et al. (1991) compared
wilderness vacationers with urban vacationers and
a non-vacationing control group. Only individuals
with backpacking experience and who were engaged
in regular physical fitness regimens were included
in the study. Following their trip, the wilderness
group showed a significant improvement in proofreading
performance, a task that is highly
demanding of directed attention. By contrast, the
other two groups showed a pre-test-to-post-test
decline. Interestingly, the wilderness groups had
the lowest overall happiness score at the post-test.
At a 3-week follow-up, however, they showed the
highest levels of overall happiness.
In the second study, also reported by Hartig et al.
(1991), participants were randomly assigned to one
of three conditions. Those in the ‘natural environment’
and ‘urban environment’ conditions drove to
field sites where they completed attentionally
fatiguing tasks before walking for 40 minutes in the
respective setting. Those in the passive relaxation
condition completed the same set of tasks before
spending a comparable period listening to soft
music and reading magazines. The pre-treatment
manipulation was intended to ensure attentional
fatigue; a check on this manipulation was also
included. Hartig et al. reported that in this more
controlled study, as in the initial quasi-experimental
one, those in the nature-setting condition performed
better on the proof-reading task. The second
study also included self-report evaluations of the
40-minute experience in terms of being away, fascination,
coherence (an aspect of extent), and compatibility.
These were summarized in a ‘perceived
restorativeness’ score which was, on average, highest
for the natural environment group and was positively
correlated b-=0*22) with the proof-reading
score.
The other two studies used a new measure that
was theoretically derived specifically to test the
effect of restoration on directed attention. This measure
is based on the Necker cube, a wire frame
drawing whose perceived orientation in depth
appears to reverse over time. Necker cube fluctuation
is an individual difference measure of long
standing. A common interpretation of the apparent
reversals is the satiation of the representation of a
given perspective, thus allowing the alternative perspective
to dominate (Orbach et al., 1963; Cornwell,
1976). This interpretation suggests a direct parallel
to situations requiring directed attention, namely
the potential cessation of a given mental activity
due to competition from some other mental activity.
If this interpretation is correct, then by intentionally
focusing on the cube as it appears at a given
moment, its apparent change to the other possible
orientation should be delayed. In other words, it
should be possible to slow the rate of reversals
through exertion of directed attention. Further, the
more fatigued the directed attention, the less effective
such an effort should be. Thus a high ‘Necker
Cube Pattern Control’ score (the degree to which
the rate is slowed relative to the individual’s base
line) should provide an indication of the strength of
this inhibitory capacity.
Cimprich’s (1992, 1993) clinical study of
recovering cancer patients also supported the link
between the restorative experience and enhanced
effectiveness. Cancer patients are generally
instructed in necessary self-care following discharge
from the hospital. They frequently have difficulty in
remembering such information, thus seriously jeopardizing
attaining optimal treatment outcomes and
quality of life. It has also been observed that following
cancer treatment patients with a clean bill of
health from a medical perspective nonetheless
experience persistent and diverse coping problems,
including difficulties in interpersonal relationships
and limitations in returning to former activities.
Feeling that these clinical observations suggested
serious problems of directed attention fatigue,
Cimprich studied recovering breast cancer patients
at four time points during the 3 months after surgery,
using a wide range of attentional and other
measures. Participants were randomly assigned to
either the experimental intervention or usual care
control group. The former involved having each person
sign a contract agreeing to participate in three
restorative activities (of at least 20 minutes each)
per week. While the notion of restorative activities
176 S. Kaplan
was explained in broad terms with numerous
examples, participants in the experimental group
generally selected nature-based activities such as
walking in nature and gardening to fullil their contracted
time. The control group received no information
about the proposed attention-restoring
activities until after the study was completed; however,
to ensure that they received equal attention,
time was spent in discussing the importance of
usual self-care activities, such as frequent rest
periods and monitoring of untoward symptoms.
Cimprich reported that the participants in both
groups showed severe attentional deficits after surgery
before the intervention was initiated. The
experimental (restorative) group showed significant
improvement in attentional performance over the
four times they were measured; the control group
did not. The Necker Cube Pattern Control measure
was particularly sensitive to attentional changes
such that the restorative group showed improvement
in the capacity to limit pattern reversals,
while the control group showed a significant decline
in this capacity by the end of the 3-month study
period. Mood scores, however, did not show a significant
relationship with the measures of attentional
capacity.
The intervention also appeared to have an impact
on life patterns. In the restorative group, participants
went back to work and were more likely to
return full time. Another striking difference was the
inclination of members of the restorative group to
start new projects (such as losing weight, music
lessons, and volunteer work). No new projects were
reported by the control group participants. And
finally, experimental group members showed significantly
greater gains on quality of life ratings by
the end of the study period.
What is particularly remarkable about this study
is the effect of a very modest intervention (three
activities of at least 20 minutes a week) on a problem
that, according to the literature in the area, can
undermine the capacity to deal with and to adjust to
the effects of a serious illness such as cancer and its
treatment.
Finally, Tennessen, C. & Cimprich, G. (in press)
studied the possible restorative benefits of a natural
view from a college dormitory window using a battery
of attentional and other measures. Dormitory
views ranged from all natural to all built views.
Controlling for the geographic location of the buildings,
undergraduates with more nature in their dormitory
view scored significantly higher on the
Necker Cube Pattern Control measure and the
Symbol Digit Modalities Test. Those with more
natural views also tended to rate themselves as
functioning more effectively in daily life activities
requiring directed attention than those with more
built views. The observed differences in attentional
performance were not related to age, gender, or year
in school. Finally, there were no differences in mood
state based on type of dormitory view.
Taken together these studies make it reasonable
to suspect that there is a link between the restorative
experience and directed attention. Further, the
majority of these findings were obtained without
corresponding influences on mood. Since mood
enhancement is a frequent goal of stress reduction
procedures (presumably to reduce the aversiveness
of the stress inducing situation), this finding is
suggestive of a distinctive effect of restorative
experiences on directed attention.
Toward an Integration
It is time to turn to the synthesis of the stress-oriented
and the attention-oriented theories of restorative
experiences that Hartig and Evans (1993)
encouraged. Since the stress-oriented position put
forward by Ulrich et al. (1991) does not permit a significant
role for attention, the challenge can be
defined as developing an integrated theory of stress
that permits such a role. Ulrich and his colleagues
emphasize attentional decline (and performance
decline in general) as a consequence of stress. The
integration proposed here, which focuses on the
various factors that lead to stress, creates a quite
different perspective on the role of attentional
decline.
The causes of stress
A distinction is often made between physiological
and psychological theories of stress (Evans, 1982).
The former concerns the autonomic nervous system
reaction to harm or to a threat of harm. The latter,
following Lazarus (19661, tends to focus on a cognitive
appraisal of whether the individual has the
resources necessary to deal with a given challenge.
Fisher et al. (1984, p. 771, however, make the useful
point that ‘physiological and psychological stress
reactions are interrelated, and do not occur alone’.
Furthermore, they distinguish between stimuli
‘aversive enough in themselves’ to evoke a stress
response and those requiring more information processing
before a stress response occurs. The integration
I propose follows the lead of Fisher et al.
with respect to both these points. It differs in plac-
The Restorative Benefits of Nature 177
ing greater emphasis on the factors leading to a
stress response, and particularly in the central role
of resource inadequacy as a causal factor.
Let us begin with the generally accepted notion
that the stress response is an organism’s adaptive
mobilization to deal with a potentially negative
situation. The schematic organization of the factors
that I propose as leading to this adaptive mobilization,
depicted in Table 1, is based on two major categories,
harm and resource inadequacy.
Harm can be direct, as when one is injured (either
physically or psychologically). Alternatively, it can
be signalled by a perceptual pattern. The latter case
is in the spirit of Zajonc’s (1980) ‘preferences need
no inferences’ theme. In other words, certain patterns,
like something suddenly looming in one’s
face, automatically indicate the threat of impending
harm.
‘Resource inadequacy’, focuses on whether one
has the resources necessary to deal with the situation
one is facing (or anticipates facing at some
point in the future). It is divided into three subcategories.
The first, following Lazarus (19661,
emphasizes the appraisal by which the individual
determines that the available resources are insufficient.
This implies a deliberative process by which
a conclusion is reached.
The second subcategory also involves information
processing but of a variety so much faster and so
unlikely to be conscious that the term ‘appraisal’
would not be an appropriate description. Rather
‘intuition’ or even ‘pre-attentive process’, following
Neisser (1967) would be more appropriate.
An example is provided by the mechanism by
which an individual begins the process of making
sense out of a scene or visual array. A crucial first
step is the segmentation of a scene into 5+2 objects
or regions (Lesperance, 1990). This simple but
powerful process, which is believed to depend on a
primitive brainstem structure (Bruce et al., 19861, is
essential to the process of understanding and dealing
with the visual environment. As a highly
adapted mechanism, it is usually successful. This is
not, however, always the case. While some scenes
parse readily, others resist attempts to organize
them. It is hypothesized that in such cases an
TABLE 1
Factors leading to stress
Harm Resource inadequacy
Direct Determine’d via appraisal
Perceptual pattern Determined via intuition
or signal Occurring throqh gradual depletion
immediate sense of discomfort will occur. One is
suddenly faced with a situation of potential confusion,
a situation in which one cannot be sure one
has what it takes to resolve it.
These two types of resource inadequacy mark
ends of a continuum. While the degree of consciousness
and the speed of processing might seem to be
parametric issues not worthy of major theoretical
distinctions, it is, as we have seen, at the heart of an
unfortunate misunderstanding of the role of information
processing in stress and restoration. Ulrich
et al. (1991) appear to avoid allowing cognition to
play a role in the stress process, due to the same
misconception that led them to deny the relevance
of the (presumably cognitive) compatibility concept
in restoration. Once again, definitional issues aside,
there is overwhelming evidence that information
processing can occur rapidly and without consciousness.

The third type of resource inadequacy involves
neither assessment nor appraisal. Rather, it concerns
circumstances in which a task, either neutral
or even pleasant, gradually draws down some basic
resource, leading to a stress reaction. Unlike the
other members of this category, where the resource
inadequacy is either predicted or anticipated, in
this case there is a reaction to the actual depletion
of the resource. This type of resource inadequacy is
closely akin to that described by Hancock and
Warm (19891, which is a central part of their model
relating vigilance or sustained attention to stress.
Regardless of the type one is considering, the
resource inadequacy category inevitably raises
the question of what sort of resources might be
involved. Sometimes resources external to the
organism such as money or friends in high places
are undoubtedly crucial. At other times the physical
strength of the individual may be the key resource
at issue. But much of the time it is likely that
psychological resources are the crucial limiting factor.
How to conceptualize such resources is, then,
our next task.
A possible alternative to consider would be that
there are a large number of resources that play this
role. One might wonder, for example, if one had the
needed empathy to handle a situation, or the
needed insight, or the needed patience. The problem
with this alternative is that such a list would be
endless, and the time required to come to a conclusion
would be prohibitive. Thus, on adaptive
grounds, it seems more reasonable that there would
be some basic underlying resource. This is consistent
with people’s behavior; when someone states
that they just don’t have what it takes to deal with
178 S. Kaplan
some forthcoming challenge, they speak as if referring
to a broad, global concept.
What would the requirements be of such a
resource? It would have to be important to the individual’s
functioning, and pervasive in its influence.
It would also have to function like a resource; in
other words, it would have to be subject to depletion
and to subsequent inadequacy. It is perhaps hardly
surprising at this point in the discussion to discover
that directed attention fits these requirements
remarkably well. Directed attention is important
because of the central role of selectivity in human
information processing, and because of the significance
of inhibition in managing behavior. It is also
important for the very reason that it is fragile, that
it is susceptible to fatigue. As the weak link in the
chain, it is a highly likely cause of incompetent or
inappropriate behavior.
A potential drawback in this argument is that
attention has only rarely been treated as a resource
(Simon, 1978). Furthermore inhibitory attention is
rarely discussed, and the fatigue of this mechanism
is essentially not considered. Perhaps it is for these
reasons that Ulrich et al. (1991) favor the ‘more
mainstream terms’ relating to stress. In fact, there
is more support than this would imply, although the
conflicting uses of the term ‘stress’ are part of the
difficulty. Hancock and Warms (1989) provide an
interesting example of this. Attention, which they
consider equivalent to ‘psychological adaptability’,
is in fact the key psychological resource in their
model. They characterize this attentional resource
as being reduced by task demand (which they refer
to, following their distinctive terminology, as ‘input
stress’). The loss of attentional resources in turn
elicits a physiological response, which other investigators
refer to as stress.*
(As Lazarus (1966) pointed out long ago, the original
meaning of stress in the engineering context
was the force or pressure directed at an object. Selye
(1956) reversed the meaning of the term, using it to
refer to a reaction to a pressure rather than to the
pressure itself. Others, however, including Hancock
and Warm, retain the original engineering usage,
referring to environmental pressure as ‘input
stress’, or, at times, simply as ‘stress’.)
Thus the integration proposed here and the model
proposed by Hancock and Warm have several
remarkable similarities, despite the fact that the
conclusions have been reached by quite different
routes. In the model proposed here, as in theirs,
insufficient attentional resources will often be an
antecedent of stress. The proposed model differs,
however, in incorporating anticipated resource
insufficiencies and non-resource issues into a
framework that attempts to address a broader
range of factors that lead to stress.
Implications for research, past and future
While the proposed integration is consistent with
much of the theorizing in the stress domain, it is in
direct conflict with a particular class of theories. It
is important to recognize that theories of stress vary
widely, ranging from the modest to the expansive.
The latter can become so broad and diffuse as to
cover everything and explain nothing.
One of the claims characteristic of the very broad
theories of stress is that stress not only has physiological
and experiential consequences, but that it
has extensive impacts on performance as well. A
number of studies have been cited as supporting one
aspect or another of this very broad approach. However,
one of the contributions of the integration proposed
here is to indicate how challenging it is to
carry out research that speaks to these issues in an
unambiguous fashion. The central difficulty is that
under a wide range of circumstances one would
expect resource deficiencies and stress responses to
occur together. Thus, although the contributions of
directed attention and stress are distinct, the frequent
co-occurrence could readily lead to the
assumption that one concept subsumes the other.
Figure 1 illustrates three patterns that would
tend to lead to the joint presence of resource
deficiencies and stress responses. In Pattern A
resource deficiency is a precursor of stress. In Pattern
B, by contrast, stress is not resource-based (e.g.
stress associated with pain and injury), but rather
leads to resource deficiency. Pattern C includes
those many circumstances (including manipulations
popular in stress research) that simultaneously
cause both stress and resource depletion. An interesting
example of this pattern is provided by the
work of Pennebaker (1991). He has shown that
listening to a description of a traumatic experience
is itself stressful. He also points out that people
tend to avoid hearing about the traumas of others
when they can avoid it. It follows that behaving in a
socially (and humanely) appropriate way in such a
situation (as opposed to running away) requires
inhibitory control. Being stuck in a situation where
one is exposed to a depiction of trauma (as occurs in
studies such as that of Ulrich et al.) would thus be
expected to be both fatiguing of the basic attentional
resource and stressful at the same time. Comparably,
the exhausting task used in the second
study reported by Hartig et al. (1991) was very prob-
The Restorative Benefits of Nature 179
& Stress response
(al Task demand – Resource decline
– Impaired performance
(bl Stress A Severe – Resource – Impaired
response distraction decline performance
– Resource decline – Impaired performance
CC) Aversive stimulus
– Stress response
FIGURE 1. Causal linkages associated with impaired performance, illustrating the potential impacts of both the harm and resource
inadequacy components of the proposed integration.
ably stressful as well as fatiguing. Clearly the confounding
of these variables is a common property of
much research in this area which has tended to
stand in the way of an understanding of the causal
relationships involved.
Given this confounding of stress and attention in
so much research, the co-occurrence of stress and
performance decline needs to be interpreted with
caution. It is inappropriate to jump to the conclusion
that stress is the cause of impairment in
such cases. Fatigue of attention may well be the
causal factor, for the very reason that selectivity
and inhibitory control are so central to effective
information processing.
minal as an independent variable (Lundberg et al.,
1993). Several physiological measures of stress were
elevated following termination of the task. This
study has been cited by proponents of broad stress
theory as an example of stress occurring without an
aversive stimulus. It would seem, however, that the
resource depletion concept allows a more adequate
explanation. Even tasks perceived as pleasant by
participants can lead in time to resource depletion
and ultimately to a stress response.
The challenge facing research attempting to
obtain unambiguous evidence in this area is illustrated
by two studies which, although cited as supporting
a broad stress theory, are entirely consistent
with the expectations of the integrative
framework proposed here.
The fact that resource depletion takes time is part
of a larger set of temporal issues that cry out for
exploration. Attentional fatigue is slower to develop
than is stress. It is also slower to recover. Hartig
(1993) found that while stress recovery occurs more
rapidly, it also dissipates more rapidly. Recovery
from mental fatigue, while a slow process, may also
be more durable. Clearly more information on these
temporal parameters is urgently needed.
Bohnen et al. (1990) studied the impact of continuous
mental tasks (accompanied by noise) on
stress and performance. Although performance was
not impaired for the group as a whole, the participants
with higher cortisol reactions (the stress measure
used in this study) showed a greater attentional
deficit. While the authors prefer the ‘stress
leads to impaired performance’ interpretation, they
openly acknowledge that the stress response may be
‘a marker for ineffective coping with the demands of
continuous tasks’, or in other words, an outcome of
resource depletion.
A second set of temporal issues has implications
for the design of future studies in this area. As Fig.
1 suggests, a resource decline can lead to stress and
a stress response can lead to a resource decline.
Thus, even if care is taken to select manipulations
that are relatively pure with respect to their impact
on either directed attention fatigue or stress, the
duration of the manipulation must be carefully
chosen to keep the attentional fatigue from becoming
stressful or vice versa.
Conclusion
Another study suggestive of the effect of resource
depletion utilized a pleasant (as rated by The concept of stress is invoked in common usage
participants) learning task at a video display ter- under a wide assortment of circumstances. One is
180 S. Kaplan
‘stressed out’ when tired, pressured, anxious, exasperated.
Feelings of stress can certainly lead to a
sense of ineptitude and to being distraught. There
are, however, both practical and theoretical reasons
for being more analytic about the diverse situations
that are so casually cast as ‘stress’. The purpose of
this paper has been to propose a framework that
distinguishes between the stress-related and the
attentional components that lead people to seek and
benefit from restorative experiences. To earn its
keep, the proposed framework should provide
insight into matters not illuminated by previous
theory. Here are four examples where this is the
case:
(11 It explains how information-processing
demands are related to attentional fatigue, and,
under certain circumstances, to stress.
(2) It explains how one can enjoy what one is
doing, be good at it, and confident of a positive outcome
and still be exhausted from it.
(3) It explains why the two classes of experiences,
one involving fatigue and the other, stress, can feel
so different. Any theory that ignores these profound
phenomenological differences is throwing away
some important information.
(4) It explains how the same task can be stressful
at one time and not at another. An individual
already fatigued (experiencing resource inadequacy)
could find a challenge overwhelming even
though the same individual in a rested state might
consider the same challenge to be minor. Individuals
coming back from vacation sometimes find that
a matter long avoided is experienced as relatively
trivial given their rested state of mind.
In summary the proposed integration points to
the existence of two distinct, albeit interacting,
benefits of restorative experiences. Both of these
benefits have an important role in an individual’s
life. Quite obviously anything that aids in the management
of stress is desirable. At the same time,
however, directed attention also plays a significant
role. It is essential to a coherent life and to the
identification and carrying out of worthwhile purposes.
Looking back on a life of purpose and productivity,
even if one experienced some stress along
the way, might well be more satisfying than looking
back on a stress-free life in which little was
accomplished.
The proposed integration also makes a contribution
to a larger theory of how humans relate to
their environment. It points to the significant role
that directed attention, a key psychological
resource, plays in coping with challenges. In this
perspective the role that natural environments play
is a powerful one. Experience in natural environments
can not only help mitigate stress; it can also
prevent it through aiding in the recovery of this
essential resource.
Acknowledgements
Many individuals helped in the formation and writing
of this article. I would like to thank Lisa
Bardwell, Lisa Canin, Bernadine Cimprich, Gary
Evans, Carrie Gilker, Terry Hartig, Tom Herzog
and Rachel Kaplan. The members of SESAME
(Seminar on Environmentally Sensitive Adaptive
Mechanisms) were, as always, a great help. And
special thanks to David Schwartz for invaluable
assistance in the process of revising the manuscript.
References suggested by an anonymous reviewer
also played a central role in this final version. Work
on this article was supported in part by the U.S.
Forest Service, North Central Forest Experimental
Station, Urban Forestry Project through several
cooperative agreements.
Notes
(1) Ulrich et al. (1991, p. 206) state, ‘However an assessment
of the viability of a restoration explanation that
emphasizes fascination should also take into account the
fact that several scientific studies have shown that settings
containing certain types of natural stimuli, such as
snakes and spiders, do elicit strong “involuntary” attention
or fascination, yet the effects are anything but restorative’.
Although previous papers have discussed the
four proposed components of a restorative experience,
Ulrich and his colleagues have apparently misinterpreted
the other three components as alternative sources of fascination.

(2) The Hancock and Warm (1989) model is in fact more
complicated than this, depending upon an optimal task
demand that can vary ‘. . . between extreme values of
underload and overload. A zone of comfort is located at
the central position of this continuum’ (p. 527). Despite
this complication, however, their basic model focuses on
resource depletion. Task demands tax the individual’s
attentional resources, ultimately resulting in a stress
response.
References
Altman, I. & Wohlwill, J. F. (Eds) (1983). Behavior and
the Natural Environment. New York: Plenum.
Anderson, C. R. (19681. The Magic Circle of Walden. New
York: Holt, Rinehart & Winston.
Attneave, F. (1962). Perception and related areas. In S.
The Restorative Benefits of Nature 181
Koch, Ed., Psychology: A study of a science, vol. 4. New
York: McGraw-Hill, pp. 633-659.
Bohen, N., Houx, P., Nicolson, N. & Jolles, J. (1990). Cortisol
reactivity and cognitive performance in a continuous
mental task paradigm. Biological Psychology,
31,107-116.
Broadbent, D. F., Cooper, P. F., FitzGerald, P. & Parkes,
K. R. (1982). The Cognitive Failures Questionnaire
(CFQ) and its correlates. British Journal of Clinical
Psychology, 21, 1-16.
Bruce, C. J., Desfmone, R. & Gross, C. G. (1986). Both
striate cortex and superior colliculus contribute to visual
properties of neurons in superior temporal polysensory
area of Macaque monkey. Journal of Neurophysiology,
55, 1057-1075.
Bruner, J. S. (1957). On perceptual readiness. Psychological
Review, 64, 123-152.
Cawte, J. E. (1967). Flight into the wilderness as a psychiatric
syndrome. Psychiatry, 30, 149-161.
Cimprich, B. (1992). Attentional fatigue following breast
cancer surgery. Research in Nursing and Health, 15,
199-207.
Cimprich, B. (1993). Development of an intervention to
restore attention in cancer patients. Cancer Nursing,
16,83-92.
Cohen, S. & Spacapan, S. (1977). The aftereffects of
stress: an attentional interpretation. Environmental
Psychology and Nonverbal Behavior, 3,43-57.
Cornwell, H. G. (1976). Necker cube reversal: sensory or
psychological satiation? Perceptual and Motor Skills,
43,2-10.
Evans, G. W. (Ed.) (1982). Environmental Stress. New
York: Cambridge University Press.
Fisher, J. D., Bell, P. A. & Baum, A. (1984). Environmental
Psychology, 2nd Edn. New York: Holt, Rinehart &
Winston.
Francis, M. & Hester, R. T. (1990). The Meaning of
Gardens. Cambridge, MA: MIT Press.
Hancock, P. A. & Warm, J. S. (1989). A dynamic model of
stress and sustained attention. Human Factors, 31,
519-537.
Hartig, T. A. (1993). Testing restorative environments
theory. Doctoral dissertation. University of California-Irvine.

Hartig, T. & Evans, G. W. (1993). Psychological foundations
of nature experience. In T. Garling & R. G.
Golledge, Eds., Behavior and Environment: Psychological
and geographical approaches. Amsterdam:
Elsevier/North Holland, pp. 427-457.
Hartig, T., Mang, M. & Evans, G. W. (1991). Restorative
effects of natural environment experience. Environment
and Behavior, 23,3-26.
Hebb, D. 0. (1980). Essay on Mind. Hillsdale, NJ:
Lawrence Erlbanm.
James, W. (1892). Psychology: The briefer course. New
York: Holt.
Kaplan, R. (1973). Some psychological benefits of gardening.
Environment and Behavior, 5,145-152.
Kaplan, R. & Kaplan, S. (1989). The Experience of Nature:
A psychological perspective. New York: Cambridge
University Press.
Kaplan, S. (1983). A model of person-environment compatibility.
Environment and Behavior, l&311-332.
Kaplan, S. (1993). The role of natural environment aesthetics
in the restorative experience. In P. H. Gobster,
Ed., Managing Urban and High-use Recreation Settings.
St. Paul, MN: Forest Service, USDA. General
Technical Report NC-163, pp. 4649.
Kaplan, S. (1995). The urban forest as a source of psychological
well-being. In G. A. Bradley, Ed., Urban Forest
Landscapes: Integrating multidisciplinary perspectives.
Seattle: University of Washington Press.
Kaplan, S. & Talbot, J. F. (1983). Psychological benefits of
a wilderness experience. In I. Altman & J. F.
Wohlwill, Eds., Behavior and the Natural Environment.
New York: Plenum, pp. 163-203.
Lazarus, R. (1966). Psychological Stress and the Coping
Process. New York McGraw-Hill.
Lesperance, R. M. (1990). The location system: using
approximate location and size information for scene
segmentation. Doctoral dissertation. University of
Michigan.
Lezak, M. D. (1982). The problem of assessing executive
functions. International Journal of Psychology, 17,
281-297.
Lundberg, U., Melin, B., Evans, G. W. & Holmberg, L.
(1993). Physiological deactivation after two contrasting
tasks at a video display terminal: learning vs
repetitive data entry. Ergonomics, 36,601-611.
Macphail, E. M. (1987). The comparative psychology of
intelligence. Behavioral and Brain Sciences, 10,
645-695.
Margolis, H. (1987). Patterns, Thinking, and Cognition.
Chicago: University of Chicago Press.
Mesulam, M. (1985). Principles of Behavioral Neurology.
Philadelphia: F. A.. Davis.
Moore-Ede, M. (1993). The Twenty-four Hour Society.
Reading, MA: Addison-Wesley.
Moray, N. (1987). Attention. In R. L. Gregory, Ed., The
Oxford Companion to the Mind. New York: Oxford
University Press, pp. 59-61.
Nash, R. (Ed.) (1968). The American Environment: Readings
in the history of conservation. Reading, MA:
Addision-Wesley.
Neisser, U. (1967). Cognitive Psychology. Englewood
Cliffs, NJ: Prentice-Hall.
Olmsted, F. L. (1865). The value and care of parks.
Reprinted in Nash, R. (Ed.) (1968), The American
Environment: Readings in the history of conservation.
Reading, MA: Addison-Wesley, pp. 18-24.
Orbach, J., Ehrlich, D. & Heath, H. A. (1963). Reversibility
of the Necker cube: I. An examination of the
concept of ‘satiation orientation’. Perceptual and
Motor Skills, 17,439458.
Pennebaker, J. W. (1991). Inhibition as the linchpin of
health. In H. S. Friedman, Ed., Hostility, Coping and
Health. Washington, DC: American Psychological
Association, pp. 127-139.
Relf, D. (Ed.) (1992). The Role of Horticulture in Human
Well-being and Social Development. Portland, OR:
Timber Press.
Rothbart, M. K. & Posner, M. I. (1985). Temperament and
the development of self-regulation. In L. C. Hartlage
& C. F. Telzrow, Us., The Neuropsychology of
Individual Differences. New York: Plenum Press,
pp. 93-123.
Sacks, 0. (1987). The Man who Mistook his Wife for a Hat
and Other Clinical Tales. New York: Harper.
182 S. Kaplan
Selye, H. (1956). The Stress of Life. New York: McGrawHill.

Shepard, R. N. (19751. Form, formation, and transformation
of representations. In R. L. Solso, Ed., Znformation
Processing and Cognition. Hillsdale, NJ:
Erlbaum, pp. 103-117.
Sherrod, D. R. & Downs, R. (19741. Environmental determinants
of altruism: the effects of stimulus overload
and perceived control on helping. Journal of Experimental
Social Psychology, 10,468-479.
Simon, H. A. (1978). Rationality as process and as product
of thought. American Economic Revielo, 68, 1-16.
Stern, P. V. D. (19701. The Annotated Walden. New York:
Clarkson N. Potter, Inc.
Stuss, D. T. & Benson, D. F. (19861. The Frontal Lobes.
New York: Raven Press.
Tennessen, C. & Cimprich, G. (1995). Views to nature:
effects on attention. Journal of Environmental Psycholgy,
15,77-85.
Ulrich, R. S. (1983). Aesthetic and affective response to
natural environment. In I. Altman & J. F. Wohlwill,
Eds., Human Behavior and Environment: Advances in
theory and research (vol. 6). New York: Plenum,
pp. 85-125.
Ulrich, R. S. & Parsons, R. (1992). Influences of passive
experiences with plants on individual well-being and
health. In D. Relf, Ed., The Role of Horticulture in
Human Well-being and Social Development. Portland,
OR: Timber Press.
Ulrich, R. S., Simons, R. F., Losito, B. D., Fiorito, E.,
Miles, M. A. & Zelson, M. (19911. Stress recovery
during exposure to natural and urban environments.
Journal of Environmental Psychology, 11,201-230.
Wolfe, S. M. (1992). Pilot fatigue could kill you. Health
Letter, 8, l-3.
Zajonc, R. B. (1980). Feeling and thinking: preferences
need no