Content uploaded by Rachel Severson
Author content
All content in this area was uploaded by Rachel Severson on Feb 03, 2021
Content may be subject to copyright.
Journal of Environmental Psychology 28 (2008) 192–199
A plasma display window?—The shifting baseline problem
in a technologically mediated natural world
Peter H. Kahn Jr.
a,
, Batya Friedman
b
, Brian Gill
c
, Jennifer Hagman
b
, Rachel L. Severson
a
,
Nathan G. Freier
b
, Erika N. Feldman
a
, Sybil Carre
`re
d
, Anna Stolyar
e
a
Department of Psychology, University of Washington, Seattle, WA 98195-1525, USA
b
The Information School, University of Washington, Seattle, WA 98195-2840, USA
c
Department of Mathematics, Seattle Pacific University, Seattle, WA 98119-1957, USA
d
Department of Psychology, California State University San Bernardino, San Bernardino, CA 92407-2397, USA
e
Department of Medical Education and Biomedical Informatics, University of Washington, Seattle, WA 98195-7240, USA
Available online 8 May 2008
Abstract
Humans will continue to adapt to an increasingly technological world. But are there costs to such adaptations in terms of human well
being? Toward broaching this question, we investigated physiological effects of experiencing a HDTV quality real-time view of nature
through a plasma display ‘‘window.’’ In an office setting, 90 participants (30 per group) were exposed either to (a) a glass window that
afforded a view of a nature scene, (b) a plasma window that afforded a real-time HDTV view of essentially the same scene, or (c) a blank
wall. Results showed that in terms of heart rate recovery from low-level stress the glass window was more restorative than a blank wall; in
turn, a plasma window was no more restorative than a blank wall. Moreover, when participants spent more time looking at the glass
window, their heart rate tended to decrease more rapidly; that was not the case with the plasma window. Discussion focuses on how the
purported benefits of viewing nature may be attenuated by a digital medium.
r2007 Elsevier Ltd. All rights reserved.
Keywords: Nature; Views; Technology; Plasma display; Physiology
1. Introduction
Digital technologies will increasingly mediate the human
experience of the natural world. We know, for example, that
childrencomeofagemoreandmorewithnaturerepresented
through television, video, and the Web rather than experienced
directly (Kahn & Kellert, 2002). Other technologies may well
take hold. For example, there is some evidence that robotic
pets (e.g., Sony’s AIBO) will offer partially compelling
substitutes to live animals for children (Kahn, Friedman,
Perez-Granados, & Freier, 2006;Melson et al., 2005)andthe
elderly (Beck, Edwards, Kahn, & Friedman, 2004). Or
consider a ‘‘Telegarden’’: a community garden that allows
users to plant and tend seeds in a remote garden by controlling
a robotic arm through a web-based interface (Goldberg, 2000;
Kahn, Friedman, Alexander, Freier, & Collett, 2005). And
there has also been ‘‘Telehunting’’: one goes online and can kill
a real animal remotely by controlling a rifle by means of a web-
based telerobotic installation (Root, 2005). If such technolo-
gical trends continue, which seems likely, then an important
question needs to be answered: What are the physical and
psychological effects of experiencing technologically mediated
nature? This paper reports a study that investigated this
question.
An empirical starting point for this study builds on the
literature that supports the proposition that people benefit
by experiencing many aspects of a diverse natural world
(Beck & Katcher, 1996;Frumkin, 2001;Kahn, 1999;
Wilson, 1984). In one canonical study, for example, Ulrich
(1984) examined the potential differences in the recovery of
patients after gall bladder surgery depending on whether
the patients were assigned to a room with a view of a
natural setting (a small stand of deciduous trees) or a view
ARTICLE IN PRESS
www.elsevier.com/locate/jep
0272-4944/$ - see front matter r2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jenvp.2007.10.008
Corresponding author. Tel.: +1 206 616 9395; fax: +1 206 616 5149.
E-mail address: pkahn@u.washington.edu (P.H. Kahn Jr.).
of a brown brick wall. Patients were paired on relevant
variables that might affect recovery (e.g., age, sex, weight,
tobacco use, and previous hospitalization). Results showed
that ‘‘patients with the natural window view had shorter
postoperative hospital stays, had far fewer negative
comments in nurses’ notes (‘patient is upset,’ ‘needs much
encouragement’) and tended to have lower scores for minor
postsurgical complications such as persistent headache or
nausea requiring medication. Moreover, the wall-view
patients required many more injections of potent painkil-
lers, whereas the tree-view patients more frequently
received weak oral analgesics such as acetaminophen’’
(Ulrich, 1993, p. 107). Other studies have shown that
interactions with animals (such as a dog, cat, bird, dolphin,
or even small turtle) increase the physiological health,
social competence, and learning opportunities of children,
the elderly, and the general population (Beck & Katcher,
1996;Myers, 1998). Contact with nature can also lead to
‘‘enjoyment, relaxation, and yincreased levels of satisfac-
tion with one’s home, one’s job, and with life in general’’
(Kaplan & Kaplan, 1989, p. 173).
If we accept that interacting with many aspects of nature
can benefit people, then a question arises. There are over
six billion of us on this planet, and through human activity
we pollute the air (and most scientists now recognize a
‘‘greenhouse effect’’), pollute the waters, deplete soil,
deforest, create toxic wastes, and extinguish over 27,000
species each year (100–1000 times the background extinc-
tion rate). If contact with the diversity of nature is so
good for us, then why are we destroying it? One possible
answer is that the baseline for comparing what is normal
continues to shift downward as environmental conditions
degrade.
As a case in point, consider a study of the environmental
views and values of African American children in the inner-
city of Houston, Texas (Kahn & Friedman, 1995). Results
showed that a significant number of the young children
interviewed in that study understood about the idea of air
pollution; but they did not believe that Houston had such a
problem even though Houston was then (and still remains)
one of the most polluted cities in the US. In interpreting
these results, Kahn and Friedman (1995) suggested that
these children may have lacked a comparative experiential
baseline from cities with less pollution by which to
recognize that Houston was itself a polluted city. Building
on these results, Kahn and Friedman also proposed that
people across generations experience psychologically some-
thing quite similar to the children in Houston, that people
construct a conception of what is environmentally normal
based on the natural world encountered in childhood. The
crux is that with each ensuing generation, the amount of
environmental degradation can increase, but each genera-
tion tends to take that degraded condition as the non-
degraded condition, as the normal experience: a condition
that Kahn (1999, 2002) has termed environmental genera-
tional amnesia, and which more broadly we refer to as ‘‘the
shifting baseline problem.’’
This problem has been recognized in other fields, as well.
Pauly (1995), for example, has written of what he calls the
‘‘shifting baseline syndrome’’ of fisheries:
Essentially, this syndrome has arisen because each
generation of fisheries scientists accepts as a baseline
the stock size and species composition that occurred at
the beginning of their careers, and uses this to evaluate
changes. When the next generation starts its career, the
stocks have further declined, but it is the stocks at that
time that serve as a new baseline. The result obviously is
a gradual shift of the baseline, a gradual accommoda-
tion of the creeping disappearance of resource speciesy
(p. 430).
Along similar lines, Evans, Jacobs, and Frager (1982a,
1982b) showed that long-term residents in Los Angeles,
compared to recent arrivals, showed a greater desensitiza-
tion in their judgments of the severity of the smog problem
to their health. Dubos (1980), too, has argued: ‘‘Any
disease, or any kind of deficiency, that is very widespread in
a given social group comes to be considered as the ‘normal’
state and consequently is accepted as a matter of course
within that group’’ (pp. 250–251).
While the research literature has shown physical and
psychological benefits of experiencing many aspects of
nature, there are also indications that some of these
benefits carry over when experiencing visual representa-
tions of nature. For example, static photographs of nature
have been shown to confer health benefits in a hospital
setting (Ulrich, 1993). People also tend to appreciate and
feel psychologically restored by photographs of nature
(Heerwagen & Orians, 1993;Kaplan & Kaplan, 1989;
Orians & Heerwagen, 1992). There is also some evidence
that videotapes of nature-dominated scenes (compared to
artifact-dominated scenes) can confer quicker recovery
from stress and greater immunization to subsequent stress
(Parsons, Tassinary, Ulrich, Hebl, & Grossman-Alexander,
1998).
Building on these findings, Friedman, Freier, and Kahn
(2004) created a technological installation—a ‘‘plasma
window’’—where they installed an HDTV camera on top
of an university building on their campus, and displayed, as
the default image, a real-time local nature view on 50-in
plasma screens. The rationale behind this study was that if
static photographs of nature were beneficial in people’s
lives, then why not capitalize on more cutting-edge
technologies that convey more realistic portrayals of nature
and investigate the effects of a real-time ‘‘streaming’’
nature view for people who otherwise lacked access to the
outside environment. The screens were mounted on the
interior walls (as ‘‘windows’’ to the outside) in windowless
offices of seven faculty and staff within the same building.
Results showed that participants reported an increase in
their psychological well being, cognitive functioning,
connection to the wider social community, and connection
to the natural world. Thus there is some initial evidence
that a plasma window with a compelling nature view can
ARTICLE IN PRESS
P.H. Kahn Jr. et al. / Journal of Environmental Psychology 28 (2008) 192–199 193
confer benefits to people who would otherwise be working
in a windowless environment.
A critical question, however, is not just whether there are
benefits of the technologically mediated form, but if there
are, then how those benefits compare to the direct natural
form. For if the technologically mediated form comes up
short compared to the direct form but still confers a
benefit, then in the years ahead we may unknowingly allow
the technologically mediated form to substitute for our
direct experience of nature, shifting the baseline lower for
what can be considered optimal human functioning.
Thus, in the current study, we investigated the physio-
logical effects of experiencing (a) an HDTV quality real-
time view of nature (a similar installation as used by
Friedman et al., 2004), (b) essentially the same view
through a glass window, and (c) no window at all.
Participants came into the office setting under the guise
of a task performance study. While they worked on four
tasks and had two specified resting periods, participants’
heart rate was assessed. Prior to the start of each of the six
activities (the four tasks and the two waiting periods), a
researcher gave each participant instructions for the new
activity. This form of social interaction typically elevated
participants’ heart rate, and thus functioned as a low-level
stressor. A camera (time-synchronized with the physiolo-
gical recording equipment) focused on participants’ faces
to allow coding of the frequency and duration of looking
behavior out the glass window and plasma window, and at
the blank wall. We sought this behavior data so as to allow
us to examine whether physiological recovery might
depend not only on participants being in one of the three
office conditions but on their actually looking out the
window.
We had two hypotheses and two open questions. First, we
hypothesized that in terms of the rate of heart rate recovery
from the low-level stressor the glass window would be more
restorative than the blank wall; it was an open question how
the plasma window would compare to the blank wall.
Second, we hypothesized that when participants spent more
time looking at the glass window, their heart rate would
decrease more rapidly; it was an open question in terms of
rate of recovery when participants spent more time looking
at the plasma window.
2. Methods
2.1. Participants
Ninety participants (all undergraduate students; age
18–34; M¼20.8; SD ¼2.53) were recruited through flyers
posted on a university campus. All participants were
recruited during the summer, prior to the return of students
for the fall term, which started at the end of September.
Thirty participants were assigned to each of the three office
conditions, balanced by gender within condition. Each
participant was compensated with $20.00. The experiment
lasted approximately 1 h.
2.2. The three conditions
Each of the three conditions employed the same office on
the university campus. In the glass-window condition, the
south-facing view through a glass window overlooked a
nature scene that included water in the foreground, as part
of a public fountain area, and then extended to include
stands of deciduous trees on one side, and a grassy expanse
that allowed a visual ‘‘exit’’ on the other. This office view
was chosen to include features that people usually find
esthetically pleasing and restorative in nature (Kaplan &
Kaplan, 1989;Orians & Heerwagen, 1992). The office size
was approximately 13 ft by 81
2ft, with off-white colored
walls, matte finished, 101
2foot ceilings with fluorescent
ceiling lights. An office desk, 71
2ft by 2 ft (32 in high), was
placed in front of the window. A swivel chair was locked
into position on the floor so as to keep constant the
distance to the window.
In a second condition, the plasma window condition, a
50-in plasma screen was inserted into the office window,
entirely covering it (Fig. 1). We then mounted an HDTV
camera approximately 15 ft higher on top of the building
and, through hard cabling, displayed on the plasma screen
essentially the same glass-window view one would see from
inside the office itself. The size of the glass window and
plasma window were virtually identical. The desk was
moved approximately 4 ft from the plasma window so as to
mimic as close as possible from a viewer’s perspective the
experience of the same view through the glass window.
In the third condition, the blank wall condition, we first
sealed off the original glass window with light-blocking
material, and then covered the sealed window with drapes,
ARTICLE IN PRESS
Fig. 1. Demonstrator in the plasma window condition. The plasma
window covered up the same-sized window used in the glass-window
condition. The camera that recorded looking behavior can be seen poking
out from the drapes to the left of the plasma window. The drapes were
pulled across the entire wall for the blank wall condition.
P.H. Kahn Jr. et al. / Journal of Environmental Psychology 28 (2008) 192–199194
in effect turning the space into a windowless office. The
desk was moved back to the position as established in the
glass-window condition.
Installing the plasma window required many hours, and
uninstalling it even longer, since the surrounding window
area required re-plastering and repainting after removing
the hardware. As a result, it was not feasible to repeatedly
switch data collection efforts back and forth between the
glass window and plasma window conditions, and thus it
was not possible to randomly assign participants to the
three conditions. We collected all of the plasma window
data prior to collecting data for the glass window. That
said, all of the participants came from a summer-school
population. The plasma window condition was run from
early July to early August, and the glass-window condition
was run from mid-August to mid-September. The blank
wall condition was interspersed throughout the other two
conditions from early July to mid-September. Moreover,
participants during the first 5 weeks of data collection were
assigned to either the plasma window condition (30
participants) or the blank wall condition (15 participants),
while participants during the remaining 5 weeks of data
collection were assigned to either the glass-window
condition (30 participants) or the blank wall condition
(15 participants).
The lighting was kept congruent with the anticipated
contexts of use: the glass-window condition had both
natural light and fluorescent light; the plasma window
condition had light from the display and fluorescent light;
and the blank wall condition had fluorescent light. Because
both light intensity and outside weather had the potential
to differ across conditions, data on these variables were
collected, and subsequent analyses were conducted to
examine their potential role in heart rate recovery.
2.3. Procedure
At the start of the experiment, participants had a 5-min
‘‘waiting period’’ during which they could (depending on
their condition) look out the glass window or plasma
window, or at the blank wall, if they so chose. Then
participants completed a series of four tasks: a 10-min
proofreading task, a 3-min ‘‘name-a-Droodle’’ task that
asked for clever labels for ambiguous drawings, a 7-min
‘‘invent-a-Droodle’’ task that asked for the creation of
one’s own Droodle, and a 10-min ‘‘tin can unusual uses’’
task that asked for different uses for a tin can. The tasks
were chosen to allow for different forms of mental
engagement. Following the tasks, participants had another
5-min waiting period.
2.4. Assessments
To assess heart rate, we used a Biopac MP 100
physiological system with a 2-lead configuration to collect
electrocardiogram (ECG) waveform data at a rate of 200
samples per second. Cardiac interbeat interval (IBI) was
determined from the ECG waveform based on the interval
between R-waves, and heart rate was computed as the
reciprocal of IBI.
To assess looking behavior, we recorded the face of
participants during the experiment by means of a camera
(visible in Fig. 1) that was time-synchronized with the
physiological recording equipment. We subsequently coded
on a second-by-second basis the frequency and duration of
looking behavior out the glass window or plasma window,
or at the blank wall.
Outside weather conditions and light intensity inside the
office were recorded for each experimental session. Cloud-
cover conditions for the start and end of each session were
obtained from surface weather observations at the Seattle
NOAA/NWS Lake Washington weather station,
1
located
approximately 3 miles from the office where the experiment
was conducted. These weather observations were then
collapsed into one of three categories (sunny/mostly sunny,
cloudy/mostly cloudy, or mixed) as a single evaluation of
the general weather conditions during the session. Also, a
Minolta TL-1 Illuminance meter was used to measure the
light intensity on the work surface of the desk at the start of
each session. When the desk surface was in direct sunlight
in the glass-window condition, light intensity sometimes
exceeded the meter’s light measurement range of .11—
21,517 lx (.01–1999 ft-c) and hence was recorded as over
21,517 lx (2000+ft-c).
Prior to the start of each of the six activities (the two
waiting periods and four tasks), a researcher gave each
participant instructions for the new activity. This form
of interaction typically elevated participants’ heart rate,
and thus functioned as a low-level stressor. To assess
the rate of heart rate recovery from the resulting low-
level stress, the slope of the least-squares regression line
for each participant’s heart rate was computed as a
function of time during the first 60 s of the activity,
beginning from the moment the researcher left the side of
the participant (to sit behind a partition in a different part
of the experimental room). These slopes were computed
separately for each participant and each activity, yielding
six measures of short-term heart rate recovery for each
participant.
2.5. Intercoder reliability
Two individuals independently coded the looking
behavior for 10 participants. Intercoder reliability, using
Cohen’s kappa, showed that k¼.79, which is considered
excellent agreement (Fleiss, Levin, & Paik, 2003).
3. Results
Table 1 provides summary statistics for the slopes of the
least-squares regression lines for heart rate as a function of
ARTICLE IN PRESS
1
Archived weather data available at http://www.atmos.washington.edu/
data/old_obs.cgi.
P.H. Kahn Jr. et al. / Journal of Environmental Psychology 28 (2008) 192–199 195
time during the first 60 s of each activity.
2
Visually, these
data are summarized in Fig. 2. The mean slope was
negative across all combinations of condition and activity,
indicating that heart rate typically declined during the first
minute of each activity regardless of condition.
To address our first hypothesis about heart rate recovery
by condition, general linear models (GLM) were used,
treating the six linear regression slopes for each participant
as repeated measures of the rate of heart rate recovery. As
hypothesized, there was more rapid heart rate recovery in
the glass-window condition compared to the blank wall
condition (F(1,58) ¼4.204, p¼.045, Cohen’s d¼.538). In
turn, there was no difference in the heart rate recovery
between the plasma window condition and the blank wall
condition: (F(1,58) ¼.003, p¼.955, d¼.015). Thus, in
terms of this measure of heart rate recovery from low-level
stress, the glass window provided a significant physiologi-
cal benefit over a windowless office (the blank wall
condition), while there was no evidence of a similar benefit
from a plasma window when compared to a windowless
office.
Another way of comparing the functionality of a plasma
window compared to a glass window is in terms of what
participants did with their eyes. How often did they look
out each window, and for how long? Our results showed
that participants looked just as often at the plasma window
(median 58 occurrences per participant) as at the glass
window (median 52 looks per participant) (Mann–Whitney
test, U¼370, p¼.324, ^
y¼.425
3
). However, the total
duration of looking time was significantly greater in the
glass-window condition (median ¼622.0 s) than in the
plasma window condition (median ¼491.5 s) (Mann–Whit-
ney test, U¼299, p¼.039, ^
y¼.656). Participants spent
much less time looking at the blank wall (median ¼55.5 s)
than either the real window (U¼26, po.0005, ^
y¼.029) or
the plasma window (U¼21, po.0005, ^
y¼.024). In other
words, both windows just as frequently garnered partici-
pants’ attention, and on this measure our results showed
equivalent functionality between the two windows. But the
glass-window view held participants’ attention longer than
the plasma window view.
To address our second hypothesis about heart rate
recovery and looking behavior, we conducted the following
analysis. For each participant in the glass window and
plasma window conditions, the Pearson correlation coeffi-
cient was computed within subject between the duration of
looking during the first 60 s of each activity and the heart
rate slope during the same 60 s. Negative correlations
indicate a tendency for more rapid heart rate recovery with
more looking, while positive correlations indicate a
tendency for slower recovery with more looking. Results
showed the mean correlation for participants in the glass-
window condition (M¼.218, SD ¼.445) was signifi-
cantly less than 0 (t(29) ¼2.683, p¼.012, Cohen’s
d¼.498), while the mean correlation in the plasma
window condition (M¼.144, SD ¼.535) was not signifi-
cantly different from 0 (t(29) ¼1.478, p¼.150, d¼.274).
Thus, on tasks where participants spent more time looking
at the glass window, their heart rate tended to decrease
more rapidly than on tasks where they spent less time
looking at the glass window. This was not the case with the
plasma window, where no relationship was found between
duration of looking at the plasma window and rate of heart
rate recovery.
The experimental design did not directly control for light
intensity and outside weather conditions. Thus we con-
ducted additional analyses to explore differences in these
variables across the three experimental conditions, and the
possible impact of these variables on heart rate recovery.
In terms of outside weather conditions across the three
experimental conditions, the weather was sunny or mostly
sunny for 47% of the sessions with the glass window, 47%
ARTICLE IN PRESS
Table 1
Summary statistics for linear regression slopes of heart rate recovery by
activity and condition (in beats per minute per minute (bpm/min))
Activity Glass
window
Plasma
window
Blank
wall
All
conditions
Waiting period 1
Mslope 5.10 1.44 3.12 3.24
SD 13.32 15.30 9.12 12.78
Name-a-Droodle
Mslope 5.16 4.74 3.30 4.38
SD 9.42 6.60 6.60 7.62
Invent-a-Droodle
Mslope 7.80 4.80 4.98 5.88
SD 8.22 7.62 6.24 7.44
Tin can
Mslope 7.02 6.54 5.52 6.36
SD 15.06 11.28 8.22 11.76
Proofreading
Mslope 8.88 7.38 6.48 7.56
SD 9.66 8.28 8.58 8.82
Waiting period 2
Mslope 15.18 6.30 8.22 9.90
SD 11.76 17.82 11.58 14.40
Average across all activities
Mslope 8.19 5.19 5.26 6.21
SD 6.91 5.30 3.69 5.58
2
There were no statistically significant differences between the two
groups of 15 participants in the blank wall condition, either on
demographic variables (gender, age) or on our outcome measures (heart
rate slopes and duration and frequency of looking at the wall). As a result,
the two groups were collapsed into a single group of 30 participants for all
analyses.
3
^
yis a probabilistic measure of effect-size obtained from the
Mann–Whitney U-statistic by ^
y¼U=mn, where m,nare the group sizes.
It provides an estimate of the probability that the value of the variable for
an individual in the first condition will exceed the value for a randomly
selected individual from the second condition (Acion, Peterson, Temple, &
Arndt, 2006).
P.H. Kahn Jr. et al. / Journal of Environmental Psychology 28 (2008) 192–199196
plasma window, and 50% blank wall. The weather was
cloudy or mostly cloudy for 20% of the sessions with glass
window, 33% plasma window, and 30% blank wall. For
the remaining sessions, the weather was a mix of sun and
clouds for 33% of the sessions with the glass window, 20%
plasma window, and 30% blank wall. Pearson’s w
2
-test
(w
2
¼2.541, 4 d.f., p¼.637, f¼.166) showed no signifi-
cant difference in weather conditions across the three
experimental conditions. Furthermore, there was no
evidence that the average of the heart rate slopes differed
significantly by outside weather condition, either with all
three experimental conditions combined (ANOVA,
F(2,87) ¼1.469, p¼.236, Z
2
¼.033) or within each con-
dition separately (glass window: F(2,27) ¼1.939, p¼.163,
Z
2
¼.125; plasma window: F(2,27) ¼.032, p¼.969,
Z
2
¼.036; blank wall: F(2,27) ¼.469, p¼.631, Z
2
¼.125).
Next we compared light intensity on the work surface of
the desk across the three experimental conditions. Results
showed that light intensity was quite similar in the blank
wall condition (median ¼456.4 lx, range 418.7–480.1 lx)
and plasma window condition (median ¼462.8 lx, range
338.0–477.9 lx) (Mann–Whitney test, U¼357.5, p¼.240,
^
y¼.411). This finding was as expected since these two
conditions occurred in the same office with the same
fluorescent lighting as the primary light source. However,
with the addition of natural light in the glass-window
condition, light intensity was much greater and more
variable (median ¼5010.6 lx, range 277.7 to over 21,517 lx)
than with either the plasma window (U¼29, po.0005,
^
y¼.967) or the blank wall (U¼30, po.0005, ^
y¼.966).
Given the large amount of variability in light intensity in
the glass-window condition, we then examined the
association between light intensity and heart rate recovery
within this condition. For each participant in the glass-
window condition, we computed an average rate of
recovery by finding the mean of the six linear regression
slopes for the first 60 s of each activity. Using the non-
parametric Kendall’s tau-b, the correlation between light
intensity and average rate of recovery was .140, which is
not significantly different from zero (p¼.287). (Descrip-
tively, the positive correlation indicates a slight, though not
statistically significant, tendency toward slower heart rate
recovery with higher light intensity.) In the blank wall and
plasma window conditions there was also no significant
correlation between light intensity and average rate of
recovery, but this is not surprising given the very limited
variation in light intensity in these conditions.
4. Discussion
This study established three key findings. First, in terms
of heart rate recovery from low-level stress, working in the
office environment with a glass window that looked out on
a nature scene was more restorative than working in the
same office without the outside view (the blank wall
condition). Second, in terms of this same physiological
measure, the plasma window was no different from the
blank wall. Third, when participants looked longer out the
glass window, they had greater physiological recovery; but
that was not the case with the plasma window, where
increased looking time yielded no greater physiological
recovery. Thus the results from this study provide a check
on what might otherwise be an unbridled positive judgment
about plasma windows (cf. Friedman et al., 2004),
particularly if one has the option of building or inhabiting
spaces with glass windows offering natural views.
ARTICLE IN PRESS
Fig. 2. Heart rate recovery from low-level stress. Values are the mean slope of heart rate (in beats per minute per minute (bpm/min)) during the first 60 s of
each activity. Negative values indicate decreasing heart rate, and points lower on the graph represent more rapid decreases in heart rate. (The activities are
ordered by the overall average slope across all six activities.)
P.H. Kahn Jr. et al. / Journal of Environmental Psychology 28 (2008) 192–199 197
Other research has suggested that people may accrue
physiological and psychological benefits simply by experi-
encing daylight in otherwise inside spaces (Ku
¨ller &
Lindsten, 1992;Leather, Pyrgas, Beale, & Lawrence,
1998; cf. Ku
¨ller & Wetterberg, 1993). Thus it could be
argued that the effects of our current study could be
completely explained on the basis of daylight: that the
glass-window condition was the only condition of the three
that had actual daylight (as opposed to digitally repre-
sented daylight through the plasma window). To partly
address this potential confound, we examined the associa-
tion between light intensity and heart rate recovery within
the glass-window condition. We did not find a significant
correlation. In other words, our data indicate that in the
glass-window condition there was no tendency for heart
rate recovery to be more rapid when the light intensity was
greater. In contrast, when we examined the relationship
between time spent looking at the window and heart rate
recovery in the glass-window condition, we did find a
significant association. While this result does not rule out
the possibility that part of the observed difference in heart
rate recovery between the blank wall condition and the
glass-window condition is due to natural daylight vs.
artificial light, our data do provide evidence that actually
looking out the window plays a significant role in heart rate
recovery.
Various theories have been advanced for why nature
views may be physiologically and psychologically restora-
tive. For example, according to attention restoration
theory (Kaplan & Kaplan, 1989), nature views have
properties that engage involuntary yet undemanding
attention, and thus promote recovery from mental fatigue.
Alternatively, according to one version of psycho-evolu-
tionary theory (Ulrich et al., 1991), many aspects of nature
accord in humans a quick positive affective reaction which
subsequently benefits physiological and psychological
processes. What is striking about our findings is that the
physiological and psychological experience of nature would
appear to differ depending on the medium (transparent
glass or digital display) through which one views nature.
Granted, the difference may be due to the lack of full
fidelity in the digitized real-time display; for example, the
plasma window did not afford parallax (the apparent
shifting of objects when viewed at different angles), a
difficult but tractable technical problem (Radikovic,
Leggett, Keyser, & Ulrich, 2005). But we suspect—and it
awaits further study—that the difference is due to more
complex reasons, involving not only technical issues of
parallax, pixilation, and 2-D as opposed to 3-D depth
perception, but judgments by viewers about what it means
for a view to be ‘‘real’’ as opposed to ‘‘represented,’’ and
how such judgments feed back into the physiological and
psychological system.
This study also speaks to the problem of the shifting
baseline. The problem is characterized well in the context
of the human–nature relationship by Fredston (2001) who,
over several decades, rowed more than 20,000 miles of
some of the wildest coastlines in the arctic waters. During
one of her later expeditions, she and her husband were
rowing along portions of Norway. She notes that much of
Norway’s built environment has an esthetic that most
towns in Alaska (where she lives) lack. But then she adds:
Still, even the undeniably beautiful portions of the
Norwegian coast that send visitors from more devel-
oped, congested parts of Europe into raptures seemed
sterile to usyThat experience frightened us to the
marrow. It made us realize that, like the perpetually
grazing sheep [in Norway], centuries of human habita-
tion have nibbled away not only at the earth but at our
perception of what constitutes nature. When we do not
miss what is absent because we have never known it to
be there, we will have lost our baseline for recognizing
what is truly wild. In its domestication, nature will have
become just another human fabrication (p. 217).
This problem of the shifting baseline takes on greater
import when one recognizes that not only are we quickly
degrading the natural world (and thus limiting our
opportunities to interact with healthy and diverse ecosys-
tems), but more and more the human experience of nature
will be mediated by technological systems. Of course,
humans will continue to adapt to such technologies. But it
is important to address the issue of whether such
adaptations are not just different but impoverished from
the standpoint of human functioning and flourishing, and
whether such technological systems and resulting interac-
tions are shifting the very baseline of what we can recognize
as impoverishment. The current study can be understood
as an initial foray into this largely uncharted territory; and
our results, even in this early stage, provide some
cautionary thoughts.
Acknowledgements
This material is based upon work supported by the US
National Science Foundation under Grant nos. IIS-
0102558 and IIS-0325035. Any opinions, findings, and
conclusions or recommendations expressed in this material
are those of the authors and do not necessarily reflect the
views of the National Science Foundation. We thank
Judith H. Heerwagen, Gordon H. Orians, and James
A. Wise for early discussions about this research.
References
Acion, L., Peterson, J. J., Temple, S., & Arndt, S. (2006). Probabilistic
index: an intuitive non-parametric approach to measuring the size of
treatment effects. Statistics in Medicine,25, 591–602.
Beck, A. M., Edwards, N. E., Kahn, P. H., Jr., & Friedman, B. (2004).
Robotic pets as perceived companions for older adults. Paper
presented at the tenth international conference on human–animal
interactions, Glasgow, Scotland.
Beck, A., & Katcher, A. (1996). Between pets and people. West Lafayette,
IN: Purdue University Press.
ARTICLE IN PRESS
P.H. Kahn Jr. et al. / Journal of Environmental Psychology 28 (2008) 192–199198
Dubos, R. (1980). Man adapting. New Haven: Yale University Press
(Original work published 1965).
Evans, G. W., Jacobs, S. V., & Frager, N. B. (1982a). Behavioral
responses to air pollution. In A. Baum, & J. E. Singer (Eds.), Advances
in environmental psychology, Vol. 4 (pp. 237–269). Hillsdale, NJ:
Erlbaum.
Evans, G. W., Jacobs, S. V., & Frager, N. B. (1982b). Human adaptation
to smog. Journal of the Air Pollution Control Association,32,
1054–1057.
Fleiss, J. L., Levin, B., & Paik, M. C. (2003). Statistical methods for rates
and proportions (3rd ed.). New York: Wiley.
Fredston, J. (2001). Rowing to latitude: Journeys along the Arctic’s edge.
New York: North Point Press.
Friedman, B., Freier, N. G., & Kahn, P. H., Jr. (2004). Office window of
the future?—Two case studies of an augmented window. In Extended
abstracts of SIGCHI conference on human factors in computing systems
(p. 1559). New York: ACM Press.
Frumkin, H. (2001). Beyond toxicity: Human health and the natural
environment. American Journal of Preventive medicine,20, 234–240.
Goldberg, K. (Ed.). (2000). The robot in the garden: Telerobotics and
telepistemology on the Internet. Cambridge, MA: MIT Press.
Heerwagen, J. H., & Orians, G. H. (1993). Humans, habitats and
aesthetics. In S. R. Kellert, & E. O. Wilson (Eds.), The biophilia
hypothesis (pp. 138–172). Washington, DC: Island Press.
Kahn, P. H., Jr. (1999). The human relationship with nature: Development
and culture. Cambridge, MA: MIT Press.
Kahn, P. H., Jr., & Friedman, B. (1995). Environmental views and values
of children in an inner-city black community. Child Development,66,
1403–1417.
Kahn, P. H., Jr., Friedman, B., Alexander, I. S., Freier, N. G., & Collett,
S. L. (2005). The distant gardener: What conversations in a telegarden
reveal about the user experience of telepresence. In Proceedings of the
14th international workshop on robot and human interactive commu-
nication (pp. 13–18). Piscataway, NJ: Institute of Electrical and
Electronics Engineers (IEEE).
Kahn, P. H., Jr., Friedman, B., Perez-Granados, D. R., & Freier, N. G.
(2006). Robotic pets in the lives of preschool children. Interaction
Studies: Social Behavior and Communication in Biological and Artificial
Systems,7, 405–436.
Kahn, P. H., Jr., & Kellert, S. R. (Eds.). (2002). Children and nature:
Psychological, sociocultural, and evolutionary investigations. Cam-
bridge, MA: MIT Press.
Kaplan, R., & Kaplan, S. (1989). The experience of nature: A psychological
perspective. Cambridge, MA: Cambridge University Press.
Ku
¨ller, R., & Lindsten, C. (1992). Health and behavior of children in
classrooms with and without windows. Journal of Environmental
Psychology,12, 305–317.
Ku
¨ller, R., & Wetterberg, L. (1993). Melatonin, cortisol, EEG, ECG and
subjective comfort in healthy humans: Impact of two fluorescent lamp
types at two light intensities. Lighting Research and Technology,25,
71–81.
Leather, P., Pyrgas, M., Beale, D., & Lawrence, C. (1998). Windows in the
workplace: Sunlight, view, and occupational stress. Environment and
Behavior,30, 739–762.
Melson, G. F., Kahn, P. H., Jr., Beck, A. M., Friedman, B., Roberts, T.,
& Garrett, E. (2005). Robots as dogs?—Children’s interactions with
the robotic dog AIBO and a live Australian Shepherd. In Extended
abstracts of the conference on human factors in computing systems
(pp. 1649–1652). New York: ACM Press.
Myers, G. (1998). Children and animals: Social development and our
connections to other species. Boulder, CO: Westview Press.
Orians, G. H., & Heerwagen, J. H. (1992). Evolved responses to
landscapes. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The
adapted mind: Evolutionary psychology and the generation of culture
(pp. 555–579). New York: Oxford University Press.
Parsons, R., Tassinary, L. G., Ulrich, R. S., Hebl, M. R., & Grossman-
Alexander, M. (1998). The view from the road: Implications for stress
recovery and immunization. Journal of Environmental Psychology,18,
113–139.
Pauly, D. (1995). Anecdotes and the shifting baseline syndrome of
fisheries. Trends in Ecology and Evolution,10, 430.
Radikovic, A. S., Leggett, J. J., Keyser, J., & Ulrich, R. S. (2005).
Artificial window view of nature. In Extended abstracts of the SIGCHI
conference on human factors in computing systems (pp. 1993–1996).
New York: ACM Press.
Root, J. (2005). Hunt game in the wild from the comfort of home. Fort
Worth Star-Telegram.
Ulrich, R. S. (1984). View through a window may influence recovery from
surgery. Science,224, 420–421.
Ulrich, R. S. (1993). Biophilia, biophobia and natural landscapes. In S. R.
Kellert, & E. O. Wilson (Eds.), The biophilia hypothesis (pp. 73–137).
Washington, DC: Island Press.
Ulrich, R. S., Simons, R. F., Losito, B. D., Fiorito, E., Miles, M. A., &
Zelson, M. (1991). Stress recovery during exposure to natural and
urban environments. Journal of Environmental Psychology,11,
201–230.
Wilson, E. O. (1984). Biophilia. Cambridge, MA: Harvard University
Press.
ARTICLE IN PRESS
P.H. Kahn Jr. et al. / Journal of Environmental Psychology 28 (2008) 192–199 199
