ENVIRONMENT AND BEHAVIOR / July 2001Kuo, Sullivan / AGGRESSION AND VIOLENCE
AGGRESSION AND VIOLENCE
IN THE INNER CITY
Effects of Environment via
FRANCES E. KUO is an assistant professor at the University of Illinois, Urbana-
Champaign. Her research examines effects of the environment on healthy human
functioning in individuals, families, and communities.
WILLIAM C. SULLIVAN is an associate professor at the University of Illinois,
Urbana-Champaign. His research focuses on the psychological and social benefits of
urban nature and citizen participation in environmental decision making.
ABSTRACT: S. Kaplan suggested that one outcome of mental fatigue may be an
increased propensity for outbursts of anger and even violence. If so, contact with
nature, which appears to mitigate mental fatigue, may reduce aggression and vio-
lence. This study investigated that possibility in a setting and population with rela-
tively high rates of aggression: inner-city urban public housing residents. Levels of
aggression were compared for 145 urban public housing residents randomly assigned
to buildings with varying levels of nearby nature (trees and grass). Attentional func-
tioning was assessed as an index of mental fatigue. Residents living in relatively bar
ren buildings reported more aggression and violence than did their counterparts in
greener buildings. Moreover, levels of mental fatigue were higher in barren buildings,
and aggression accompanied mental fatigue. Tests for the proposed mechanism and
for alternative mechanisms indicated that the relationship between nearby nature and
aggression was fully mediated through attentional functioning.
The power of the physical environment to influence human aggression is
well established. Crowding, high temperatures, and noise have all been
linked to aggression and violence (Baker, 1984; Baum & Koman, 1976;
Donnerstein & Wilson, 1976; Rule, Taylor, & Dobbs, 1987). Each of these
features of the physical environment has been associated with heightened
levels of aggression; are there features of the physical environment that work
to diminish levels of aggression and violence? This study examines whether
ENVIRONMENT AND BEHAVIOR, Vol. 33 No. 4, July 2001 543-571
© 2001 Sage Publications
natural elements such as trees and grass can decrease aggression. In addition,
it tests a potential mechanism by which natural features—and by extension
other environmental features—may affect aggression. In doing so, it sug
gests a new role for environment and behavior research in an important public
policy domain—addressing aggression and violence in inner cities—and
contributes possible new insight into the psychological factors underlying
There are hints in the literature that exposure to nearby nature, for
instance, a garden or a grassy area with trees, may reduce aggression. For
instance, violent assaults by Alzheimer patients were compared during two
consecutive summers in five long-term care facilities, two in which exterior
gardens were installed and three without gardens (Mooney & Nicell, 1992).
In Alzheimer patients, increases in the number of aggressive assaults each
year are typical as a consequence of the progressive deterioration of cognitive
processes; and indeed, in the facilities without gardens, the incidence of vio
lent assaults increased dramatically. By contrast, in the other facilities, the
incidence of violent assaults stayed the same or decreased slightly after gar-
dens were installed. More recently in another study, some subsets of prison
inmates reported less hostility after participating in a gardening project than
before, although these findings were not consistent across different analyses
(Rice & Remy, 1998).
Why might we expect the findings from these two studies to reflect a more
general, systematic phenomenon? By what mechanism might exposure to
nearby nature leave individuals in a less aggressive state? Here, we review
544 ENVIRONMENT AND BEHAVIOR / July 2001
AUTHORS’ NOTE: A portion of these findings was presented in invited testimony to
the National Urban and Community Forestry Advisory Council (NUCFAC) and at the
27th International Conference of the Environmental Design Research Association,
Salt Lake City, Utah, May 1996. The data for this study were drawn from the Coping
With Poverty archive, a multistudy research effort supported by a grant from NUCFAC
(F. E. Kuo and W. C. Sullivan, principal investigators). This work was also supported
by the Cooperative State Research, Education and Extension Service, U.S. Depart
ment of Agriculture, under Project No. ILLU-65-0387. Thanks go to Rebekah Coley
for training and supervising resident interviewers, Esther Davis and Doris Gayles for
recruiting and interviewing participants, the Robert Taylor Homes residents for their
participation, and the management of the Chicago Housing Authority for their as
sistance throughout. Ann Schlosser suggested the Conflict Tactics Scale and made
many contributions during the early stages of this project. Correspondence concern
ing this article should be addressed to Frances E. Kuo, Human-Environment Re
search Laboratory, University of Illinois, 1103 S. Dorner, Urbana, IL 61801; e-mail:
theory and evidence suggesting first that natural settings assist in recovery
from mental fatigue and second that aggression may increase with mental
fatigue and decrease with its recovery. We then present an analysis suggest
ing that residents of disadvantaged inner-city neighborhoods may be subject
to chronic mental fatigue. Finally, we test the possibility that, in an urban
public housing community, the presence of trees and grass lowers the inci
dence of aggressive and violent behavior among residents living nearby.
NATURE AND MENTAL FATIGUE
Attention restoration theory (S. Kaplan, 1995) proposes that exposure to
nature reduces mental fatigue, or more precisely, directed attention fatigue.
S. Kaplan (1995) noted that many settings, stimuli, and tasks in modern life
draw on the capacity to deliberately direct attention or pay attention. The
information-processing demands of everyday life—traffic, phones, conver
sations, problems at work, and complex decisions—all take their toll, result-
ing in mental fatigue, a state characterized by inattentiveness, irritability, and
impulsivity. In contrast, natural settings and stimuli such as landscapes and
animals seem to effortlessly engage our attention, allowing us to attend with-
out paying attention. For this and a number of other reasons, S. Kaplan sug-
gested, contact with nature provides a respite from deliberately directing
Indeed, there is growing empirical evidence of the attentionally restor-
ative effects of natural settings. Evidence of cognitively rejuvenating effects
comes from a variety of “natural” settings, including wilderness areas
(Hartig, Mang, & Evans, 1991; R. Kaplan, 1984), prairies (Miles, Sullivan, &
Kuo, 1998), community parks (Canin, 1991; Cimprich, 1993), views of
nature through windows (Ovitt, 1996; Tennessen & Cimprich, 1995), and
even rooms with interior plants (Lohr, Pearson-Mims, & Goodwin, 1996).
Moreover, these studies have demonstrated links between contact with nature
and more effective attentional functioning in a variety of populations—AIDS
caregivers, cancer patients, college students, prairie restoration volunteers,
participants in a wilderness program, and employees of large organizations.
MENTAL FATIGUE AND AGGRESSION
If contact with nature is attentionally restorative, how then might
attentional restoration mitigate aggression? S. Kaplan (1987) suggested that
one of the costs of mental fatigue might be a heightened propensity for “out
bursts of anger and potentially . . . violence” (p. 57). The following analysis
Kuo, Sullivan / AGGRESSION AND VIOLENCE 545
shows how each of three symptoms of mental fatigue might contribute to
First, mental fatigue may contribute to aggression because of its effects on
cognitive processing. A common theme in some recent theories of aggression
is that information processing plays a central role in managing social situa
tions, especially potential conflicts (e.g., Crick & Dodge, 1994; Dodge &
Crick, 1990; Dodge & Schwartz, 1997; Martinko & Zellars, 1998). For
example, Dodge and Crick (1990) proposed that a child’s behavioral
response to a social stimulus is a function of the following five steps of infor
mation processing: encoding of social cues, interpretation of social cues,
response search, response evaluation, and enactment. The proposal here is
that in problematic social situations, relatively automatic, effortless cognitive
processing (e.g., “Bob took my computer station!”) is more likely to generate
conflict-escalating behavior than is more reasoned, effortful, reflective pro
cessing (e.g., “Hmm...didIleaveanyclues that I was working there?”). As
the individual’s willingness and ability to engage in more reflective, effortful
processing decreases with mental fatigue, social behavior is likely to become
increasingly thoughtless, tactless, and unstrategic, allowing conflicts to spi-
ral out of control (see Rubin, Bream, & Rose-Krasnor, 1991, for a similar
proposal with respect to children’s social problem solving).
There is some evidence to suggest that deficits in effortful processing are
indeed associated with aggression. In school settings, deficits in effortful
processing are likely to manifest in inattentiveness, and inattentiveness has
been closely tied to aggression in both children (Stewart, 1985) and adoles-
cents (Scholte, van Aken, & van Lieshout, 1997). Indeed, the tie between
attention deficits and hyperactivity on one hand and conduct problems and
aggression on the other has been so strong that there has been some debate as
to whether these disorders are distinct (see Hinshaw, 1987, for a meta-analysis
indicating that these disorders are distinct although correlated). Conversely,
Rabiner, Lenhart, and Lochman (1990) found that when aggressive children
were encouraged to be more reflective in their responses to problematic
social situations, their generation of conflict-escalating responses dropped to
the same levels as their nonaggressive, nonrejected peers. Thus, it seems
plausible that the deficits in effortful processing that are symptomatic of
mental fatigue may contribute to aggression.
Mental fatigue may also contribute to aggression because of its effects on
emotion—specifically, heightened irritability. Irritability appears to be a fre
quent side effect of mentally fatiguing tasks, such as the vigilance tasks
involved in air traffic control (Thackray, Bailey, & Touchstone, 1979; Warm &
Dember, 1986). Not surprisingly, irritability is linked with aggression in
numerous studies (e.g., Caprara & Renzi, 1981; Coccaro, Bergeman,
546 ENVIRONMENT AND BEHAVIOR / July 2001
Kavoussi, & Seroczynski, 1997; Kant, Smith-Seemiller, & Zeiler, 1998;
Kavoussi & Coccaro, 1998; Stanford, Greve, & Dickens, 1995). Irritable
individuals are prone to aggression when faced with frustration (Caprara &
Renzi, 1981), and pharmacological treatments that reduce aggression also
reduce irritability (Kant et al., 1998; Kavoussi & Coccaro, 1998). Thus, it
seems plausible that the irritability symptomatic of mental fatigue might con
tribute to aggression.
Finally, mental fatigue may also contribute to aggression because of its
effects on behavior—specifically, decreased control over impulses. S. Kap
lan (1987) noted that one of the hallmarks of mental fatigue is a difficulty
inhibiting behavioral impulses. Impulsivity in turn is associated with aggres
sion and violence in a variety of populations (for reviews, see Brady, Myrick, &
McElroy, 1998; Markovitz, 1995; Tuinier, Verhoeven, & Van Praag, 1996).
Violent parolees are more impulsive than nonviolent parolees (Cherek,
Moeller, Dougherty, & Rhoades, 1997), maritally violent men are more
impulsive than maritally nonviolent men (Barnett & Hamberger, 1992), and
among depressed males, impulsive individuals are more likely to be aggres-
sive than nonimpulsive individuals (Hynan & Grush, 1986). Not surprisingly,
then, Luengo and colleagues (Luengo, Carrillo-de-la-Pena, Otero, &
Romero, 1994) found in their 1-year longitudinal study that present
impulsivity ratings predict future antisocial behavior, including aggression.
In sum, each of these factors—impairments in effortful cognitive process-
ing, irritability, and impulsivity—has been independently implicated in
aggression. To the extent that mental fatigue combines these three factors,
mental fatigue seems likely to contribute substantially to aggression.
INNER CITIES AND CHRONIC MENTAL FATIGUE
Poor, inner-city neighborhoods may be an especially promising context in
which to study the effects of nature and attentional restoration on aggression.
As the following analysis, drawn from Kuo (1992), suggests, the attentional
demands associated with poverty and the inner-city environment are likely to
place this population at special risk for chronic mental fatigue and
fatigue-related aggression. As a consequence, residents of poor, inner-city
neighborhoods may have a special need for the mental respite provided by
The attentional demands of poverty are many and unremitting. For the
poor, even basic concerns such as rent, utilities, and food are ongoing chal
lenges that require effortful problem solving and reasoning. Added to these
are the attentional challenges posed by major life events. Poverty brings
with it a greater susceptibility and vulnerability to drastic life changes.
Kuo, Sullivan / AGGRESSION AND VIOLENCE 547
Underinsured and having no financial cushion against setbacks, even a minor
temporary trauma such as a child’s illness can have far-reaching effects,
eventually necessitating major readjustments in life, family, and work
domains. Making these adjustments requires sustained, high levels of mental
Moreover, the environmental characteristics of inner-city neighborhoods
place additional demands on attention. First and foremost, the ever-present
possibility of crime or violence places high demands on attention (see
Cohen & Spacapan, 1978, for an analysis of the attentional demands imposed
by unpredictable stressors). Danger requires individuals to be vigilant for
signs of impending trouble, to continuously consider possible responses to
new situations, and to consider the ramifications of those responses. Second,
the home environment may place further demands on attention; lack of ade
quate space and facilities makes purposive functioning more effortful as
more problem solving is required to accomplish goals in unsupportive or
inadequate settings. Problem solving may be made all the more fatiguing by
the lack of quiet, safe settings in which to think. And finally, for the many
inner-city residents who lack natural settings in their everyday environment
(nearby parks, views to green spaces, and gardens), recovery from mental
fatigue may be especially rare.
Over time, the ongoing and acute attentional demands of poverty, in com-
bination with the mentally fatiguing characteristics of the inner-city environ-
ment, seem likely to yield chronic high levels of mental fatigue. Thus, among
inner-city inhabitants lacking ready access to attentionally restorative set-
tings, we might expect chronic high levels of mental fatigue and a heightened
propensity for aggressive behavior. Conversely, among residents with ready
access to nature, we might expect comparatively low levels of mental fatigue
Two questions are central to this study. First, does nearby nature reduce
aggression and violence? And second, if so, is this effect mediated via
attentional restoration? To examine these questions, structured interviews
and attentional tests were conducted with urban public housing residents.
Because official adult residents are predominately single mothers, the struc
tured interviews focused on intrafamily aggression and violence rather than
other forms of violence. Attentional performance and self-reports of aggres
sion were then compared for residents living in buildings with relatively high
versus relatively low levels of nearby nature, and mediation tests were used to
examine whether attentional restoration might account for a relationship
between nature and aggression.
To explore possible alternative accounts for a nature-aggression relation
ship, a number of additional tests were conducted. A test for spuriousness
548 ENVIRONMENT AND BEHAVIOR / July 2001
(Evans & Lepore, 1997) was conducted to guard against alternative accounts
in general. In addition, the following three particular alternative accounts
were given specific attention: (a) Positive mood, (b) stress recovery, and
(c) social support were each identified as theoretically plausible explanations
for a link between nature and reduced aggression. Positive mood has been
linked directly with contact with nature (Hull & Michael, 1995), and it seems
plausible that positive moods could reduce the propensity for aggression
(Pihl & Zacchia, 1986, tested this notion but found no evidence for it). Simi
larly, stress (or more precisely, recovery from stress) has been linked directly
with contact with nature (Ulrich et al., 1991), and stress also appears to con
tribute to aggression (Bolger, Thomas, & Eckenrode, 1997; Chang, 1994).
And finally, there is some indication that neighborhood social ties and sup
port networks are stronger around greener neighborhood spaces (Kuo,
Sullivan, Coley, & Brunson, 1998; Kweon, Sullivan, & Wiley, 1998); in turn,
child abuse is less prevalent among parents who have social support
(Garbarino & Sherman, 1980; Roth, 1986).
THE SITE: A NATURAL EXPERIMENT
ON THE EFFECTS OF NEARBY NATURE
A number of methodological criteria were employed in the selection of a
site for this research. Robert Taylor Homes (RTH) in Chicago was rare in that
it simultaneously met each of these criteria.
First, although the amount of vegetation outside the buildings at RTH var
ies considerably from building to building, other environmental features are
held remarkably constant from one building to another. Because the build
ings are architecturally identical, at RTH, building size, building layout,
building facilities, architectural detail, and the number of residential units per
building are held constant (see Figure 1). Moreover, because the buildings are
placed in single file along a 3-mile corridor, the features of the surrounding
landscape are similar from one building to another. Each building is bordered
on the west by an interstate highway and railroad tracks and bordered on the
east by a six-lane municipal thoroughfare and wide sidewalk.
Second, public housing policies result in de facto random assignment of
residents with respect to levels of nearby nature at RTH. Although housing
applicants to the Chicago Housing Authority can specify their choice of
development (e.g., RTH vs. some other development), they have little choice
Kuo, Sullivan / AGGRESSION AND VIOLENCE 549
550 ENVIRONMENT AND BEHAVIOR / July 2001
Figure 1: Attrition Has Left Some Buildings Surrounded by Only Concrete and
Asphalt and Others With Pockets of Green
of where they will be assigned within a development (i.e., this apartment vs.
another apartment within RTH).
Moreover, the scale of the Chicago Housing
Authority precludes the placement of “better” (e.g., more responsible and
less aggressive) residents in “better” (e.g., greener) locations. Clerks in a cen
tral office handle all assignments of residents to apartments for 40,700 units
in 1,479 buildings across 17 developments throughout the city. They gener
ally have never met or seen the applicants for housing and most likely have
never set foot in most of the Housing Authority’s developments. It is im
plausible that anyone could remember the characteristics of so many build
ings, let alone take them into account in assigning apartments.
Third, residents at RTH have little role in the landscaping outside their
building. When RTH was originally built in the 1960s, trees and grass were
planted around each of the 28 high-rise buildings. Over time, the majority of
these green spaces have been paved in an effort to keep dust down and main
tenance costs low; this paving has killed many of the original trees, leaving
some buildings with completely barren common spaces, others with a few
scattered trees, and still others with leftover pockets of green. Ongoing land-
scape maintenance at RTH is handled entirely by a small landscaping crew
serving all of the developments managed by the Chicago Housing Authority;
residents are not involved in maintenance, and funds are inadequate to fulfill
special requests from residents. Thus, a relationship between greenness of
common spaces and aggression in this setting cannot be explained by a pro-
cess in which especially effective or cooperative residents have made their
In sum, RTH constitutes a naturally occurring experiment on the effects of
residential vegetation, with random assignment of residents to vegetation
conditions, no control of residents over levels of vegetation, and a host of
environmental variables held constant. An additional methodologically
desirable feature of RTH for this study is that the residents are strikingly
homogeneous with respect to many of the individual characteristics that
might be expected to affect aggression—income, education, life circum
stances, and perhaps most important, economic opportunities.
PROCEDURE, PARTICIPANTS, AND DESIGN
To maximize participants’ ease in responding, interviewers were selected
to be as similar to interviewees as possible. Three African American female
residents of RTH were hired and trained to conduct the recruitment, inter
viewing, and testing for this research. All three were longtime residents of
RTH (19 years or more) residing in buildings outside the study sample. Thus,
Kuo, Sullivan / AGGRESSION AND VIOLENCE 551
interviewers were matched to interviewees not only in major demographic
characteristics such as race, gender, and socioeconomic status but also in life
circumstances, background, and more subtle social cues such as patterns of
speech and dress.
In preparation for interviewing and testing, interviewers completed exten
sive training (50 hours of general training in interview methods, 12 hours
learning the specific interview measures used, and 14 hours of supervised
and unsupervised practice in performing practice interviews). In addition, an
on-site research supervisor met regularly with the interviewers to review pro
cedures and address any difficulties or questions. Interviewers did not inter
view individuals with whom they were previously familiar, and interviewers
were counterbalanced for nature condition.
Recruitment was conducted door to door in buildings spanning the range
of vegetation of RTH. Sampling was restricted to 18 buildings—buildings
adjacent to parks, police stations, and other relatively unique features were
excluded to minimize effects of extraneous factors on residents’ access to
nearby nature. Within buildings, sampling was restricted to apartments on
Floors 2 through 4, where residents had maximal physical and visual access
to the trees outside their building (there are no residences on the first floor).
Recruitment criteria included not only environmental factors but also resi-
dent characteristics. Women heads of household younger than the age of 65
were invited to participate in a University of Illinois study about life at RTH.
Recruitment focused on women because official adult residents in urban pub-
lic housing are overwhelmingly female—80% in RTH (Chicago Housing
Participants were told that they could refuse to answer any
question and could stop the interview at any time and that they would receive
$10 on completion of the interview.
Of the 158 qualified residents invited to participate, 92% chose to partici
pate, yielding a final sample of 145 residents, 69 with relatively low levels of
nearby nature and 76 with relatively high levels of nearby nature. The com
posite participant profile is that of a 34-year-old African American single
woman with a high school or high school equivalency diploma raising three
children on an annual household income of less than $10,000.
Individual interviews were conducted during summer and fall months in
participants’ apartments. Residents’ attentional capacity, aggression, and a
number of control variables likely to be associated with aggression were
assessed as part of a 45-minute structured interview.
552 ENVIRONMENT AND BEHAVIOR / July 2001
Nearby nature. Levels of nearby nature were assessed using standardized
sets of photographs and multiple independent raters. For each of the 18 build-
ings to be sampled, a standardized set of photographs was taken from eye
level of the area immediately surrounding the building. As Figure 2 shows,
each standardized set comprises 16 photographs taken from specified van
tage points; most showed views looking out from the building, and the
remaining showed views looking across the building.
To obtain ratings of the nearby nature for each building, the photographs
were arranged at 18 stations (drawing tables in a design studio), with each
station showing all 16 photographs for a given building. Undergraduate and
graduate students in horticulture then independently rated levels of nearby
nature for each of the 18 buildings. First, raters visited each of the stations to
familiarize themselves with the range of vegetation in the 18 buildings. Sec
ond, they visited each station again in turn and provided a single greenness
rating for each building based on the 16 photographs. Raters were encour
aged to use the entire response scale from 0 to 4 (0 = not at all green ,1=a
Kuo, Sullivan / AGGRESSION AND VIOLENCE 553
Figure 2: Plan View of an Apartment Building at Robert Taylor Homes With
NOTE: The numbers within the building indicate apartments. The arrows indicate the position from
which photographs were taken (for each building) that were then rated by 22 independent raters.
Note that despite the presence of trees outside a building, residents in particular apartments may
have little or no visual access to trees.
little green,2=somewhat green,3=quite green, and 4 = very green). For
each building, greenness ratings from the 22 raters were averaged to produce
a summary greenness rating. These averaged greenness ratings ranged from
0.8 for the building with the least nearby nature to 3.6 for the greenest
With these data, agreement between raters is analogous to the reliability of
items in a scale; the hope is that different raters will respond to a particular
building in a similar fashion. Thus, to assess interrater agreement, a
Cronbach’s alpha was calculated, with individual raters treated as individual
items in a scale and individual buildings treated as individual respondents.
This procedure yielded an alpha of .97, indicating a high level of agreement
between raters with regard to building greenness.
Greenness ratings were used as the basis for assignment to conditions.
Buildings whose ratings fell below the midpoint of the range were designated
barren; buildings whose ratings were at or above the midpoint were desig
nated green. Greenness ratings for the 7 buildings in the barren condition
ranged from 0.8 to 1.7, with a mean of 1.2. Greenness ratings for the 11 build-
ings in the green condition ranged from 2.0 to 3.6, with a mean of 2.6. In inter-
preting these ratings, it should be noted that because raters were encouraged
to use the entire response scale, even a high greenness rating of 3.6, or very
green, is relative to the range of vegetation at RTH; as Figure 1 shows, even
the greenest pockets at RTH are neither especially large nor especially lush in
There were no systematic differences between barren and green buildings
in environmental factors such as pedestrian or automobile traffic, nearness to
parking, or nearness to parks, schools, or other facilities. There was no sys
tematic pattern in the sequence of green and barren buildings along the 3-mile
corridor; green and barren buildings were not clustered but rather haphaz
ardly interspersed. Some barren buildings were oriented north-south, others
east-west; similarly for green buildings. For barren buildings, in the places
where trees or grass might have been, there was only bare dirt or asphalt, and
even the green buildings were surrounded by large areas of bare dirt or
To check for possible condition differences in participant characteristics
for barren versus green buildings, a series of t tests was conducted. As would
be expected given random assignment of residents to nature conditions, no
significant condition differences were found in demographic characteristics,
household characteristics, or other variables potentially related to aggres
sion. Specifically, green and barren participants did not differ in age, educa
tion, employment, income, size of household, marital status, number of
554 ENVIRONMENT AND BEHAVIOR / July 2001
children, years in apartment, years in public housing, health ratings, health
symptoms, alcohol use, prescription drug use, or other drug use.
Attentional functioning. The capacity for directed attention was assessed
with the Digit Span Backwards (DSB) test. Digit Span Backwards is a stan
dardized neurocognitive measure and is used in the measurement of
attentional fatigue (Cimprich, 1993; Schwartz, 1994; Tennessen &
Cimprich, 1995) and in the clinical measurement of attention (Lezak, 1983;
Mesulam, 1985). DSB is particularly useful for field settings because it is
easy to administer: The administrator reads aloud a series of digits (e.g., “2, 5,
1”), and participants are asked to repeat back the series in reverse order (e.g.,
“1, 5, 2”). Series are administered in increasing length; if a participant fails a
series of a given length, a second series of equal length is administered.
Scoring was based on the longest series performed correctly within two
Aggression. The Conflict Tactics Scale (CTS) (Straus, 1979) is a widely
used self-report measure designed to assess levels of intrafamily aggression
and violence. It has been used in more than 100 studies (see bibliography in
Straus, 1995). The CTS has a test-retest reliability of .97 (parent-to-child
aggression) (DuRant, Pendergrast, & Cadenhead, 1994), an internal consis-
tency of .88 (wife-to-husband aggression) (Straus, Gelles, & Steinmetz,
1980), and good concurrent validity with other measures of parental
psychosocial distress (Wissow, Wilson, Roter, Larson, & Hope, 1992).
To elicit reports of aggressive behavior, participants are asked to think of
situations in which they had a disagreement or were angry with a specified
family member and to indicate how often they used each of 18 conflict tac
tics, beginning with socially acceptable tactics (e.g., reasoning) and ending
with violent tactics. Table 1 shows the 14 aggressive conflict tactics from the
CTS. The responses to these 14 items provide an index of overall aggression.
The first 6 items index psychological aggression: verbal and symbolic acts
intended to cause psychological pain or fear. The remaining 8 items index
violence: the use of physical force or violence. These 8 items comprise both
mild violence (3 behaviors unlikely to cause injury) and severe violence (5
behaviors likely to cause injury).
For each of the specific conflict tactics, participants are asked to indicate
how often they have used it in the past 12 months on a 7-point scale (0 = never
to6=more than 20 times, with X = don’t know). If participants reply never or
don’t know, they are then asked, “Did you ever...?”(yes/no). These ques
tions are asked in reference to two specified family members—first the
Kuo, Sullivan / AGGRESSION AND VIOLENCE 555
respondent’s partner (or the adult with whom they are closest) and then their
child (or if they have more than one, the child with whom they have the most
This procedure yields estimates of both the frequency of specific aggres-
sive behaviors used in the past year and the range of aggressive behaviors
employed over the respondent’s lifetime. In general, more aggressive per-
sons employ a wider range of aggressive behaviors; for instance, Person A
might only use verbally aggressive tactics, whereas Person B might use not
only verbally aggressive tactics but also physically aggressive tactics. For
each of the different categories of aggression (overall aggression, psycholog
ical aggression, and so forth), the range of tactics a respondent has used in
that category is calculated by taking the proportion of the number of different
tactics employed out of the total number of different tactics. For instance, an
individual who has employed each of the 14 different overall aggression tac
tics would have an overall aggression range of 1.0; an individual who has
employed only 7 of those tactics (typically, the less serious ones) would have
an overall aggression range of 0.5. Similarly, the range of violent tactics
would refer to the proportion of the 8 violent tactics that a respondent had
employed. It should be noted that the term range is not used in the statistical
sense here; the different conflict tactics are not assumed to be on an interval
556 ENVIRONMENT AND BEHAVIOR / July 2001
Overall Aggression Items From the Conflict Tactics Scale
Insulted or swore at the other
Sulked or refused to talk
Stomped out of the room or house
Did or said something to spite the other
Threatened to hit or throw something
Threw or smashed or hit or kicked something
Pushed, grabbed, or shoved the other one
Slapped the other one
Threw something at the other one
Kicked, bit, or hit with a fist
Hit or tried to hit with something
Beat up the other one
Threatened with a knife or gun
Used a knife or gun
Other factors likely to affect aggression. Three additional variables that
seemed likely to be related to nearby nature, attention, or aggression were
assessed through self-report using a 5-point scale (0 = not at all,1=a little,2=
a medium amount,3=quite a lot, and 4 = very much). Positive mood was
assessed with the Positive Mood subscale of the Profile of Mood States
(POMS). Participants rated themselves on six adjectives (lively, active, ener
getic, cheerful, full of pep, and vigorous). The POMS has been shown to be a
valid and reliable instrument for the measurement of mood (McNair, Lorr, &
Droppleman, 1981). Stress was assessed with the question “How stressful is
this period in your life?” And social integration was assessed with an 8-item
scale, alpha = .80, that included items such as “How well do you know the
people next door?”; “Are people here concerned with helping and supporting
one another?”; and “Is there a strong feeling of belonging here?” (see Kuo
et al., 1998, for details).
Results are presented in three subsections. First, descriptive statistics on
intrafamily aggression and participants’ attentional resources are presented.
Second, the central hypothesis is tested. And third, potential mechanisms
underlying the relationship between nearby nature and aggression are
AGGRESSION AND MENTAL
FATIGUE AT ROBERT TAYLOR HOMES
Consistent with previous research, levels of aggression in this population
were much higher than in national samples. A majority of participants in this
sample (61%) reported having engaged in a violent act against their partner at
least once in their lives, a rate approximately 4 times that reported in two
national probability samples of couples in the United States (Straus, 1979;
Straus & Gelles, 1988) but consistent with rates from a sample of formerly
married African American women (57%) (Neff, Holamon, & Schluter,
1995). Aggression against children showed much the same pattern. A major
ity of the participants in this sample reported hitting their child with some
thing at least once in their lives (56%), approximately 4 times the rate
reported in a national sample of two-caretaker households with at least one
child (Straus & Gelles, 1986).
Kuo, Sullivan / AGGRESSION AND VIOLENCE 557
Participants’ DSB scores provide an index of attentional resources in this
inner-city population. According to Lezak (1983), scores of 4 or 5 on this ver
sion of the DSB test are within normal limits, depending on the individual’s
educational level. In this sample, the mean DSB score was 4.8, with a stan
dard deviation of 1.1, indicating substantial variation in attentional
TESTING THE CENTRAL HYPOTHESES
If the availability of nearby nature reduces the propensity for aggression,
then residents living in green conditions should report less aggression than
their counterparts living in barren conditions: less frequent aggression in the
past year and a narrower range of aggressive tactics used over the course of
their lifetime. A series of planned, one-tailed t tests were conducted to exam
ine condition differences in frequency and range of aggression, first against
the respondent’s partner and then against their child.
Aggression against partner. Tables 2 and 3 show the findings with respect
to the frequency of aggression against partner in the past year and the range of
aggressive tactics used against partner over the lifetime.
As the first row of Table 2 shows, there was a significant condition differ-
ence in the frequency of overall aggression against partner during the past
year. That is, residents living in green conditions reported significantly less
overall aggression against their partners than did their counterparts living in
barren conditions. The following rows in Table 2 show the findings for spe-
cific forms of aggression. The second row indicates that there was a signifi
cant condition difference in psychological aggression against partner;
residents living in green conditions were significantly less likely to have
engaged in psychological aggression against their partners than were resi
dents living in barren conditions. Does this effect extend to more violent
forms of aggression? Because the violence indices produce extremely
skewed distributions, Straus (1979) recommended dichotomizing these indi
ces into violent and nonviolent categories. If a participant had engaged in at
least one of the eight violent conflict tactics during the past year, they were
designated violent; otherwise, they were designated nonviolent. As rows 3
through 5 of Table 2 show, nearby nature is related to the use of violence
against partner during the past year. Violence scores were significantly lower
for residents living in green conditions than those living in barren conditions.
Furthermore, this pattern held for both the more mild forms of violence and
the more severe forms of violence. Both mild violence rates and severe
558 ENVIRONMENT AND BEHAVIOR / July 2001
violence rates were significantly lower in the green condition than in the bar
ren condition. Mean differences between the green and barren conditions for
the various aggression subscales ranged from one third to one half of a stan
Table 3 shows the findings with respect to the range of aggressive conflict
tactics used against partner over the participant’s lifetime. As the first row
shows, there was a significant condition difference in the range of overall
aggression tactics used. That is, residents living in green conditions report
Kuo, Sullivan / AGGRESSION AND VIOLENCE 559
Mean Rates of Aggression Against Partner
During Past Year in Green Versus Barren Condition
MSD MSD t p
Overall aggression 1.04 0.88 0.76 1.07 1.68 < .05
Psychological aggression 1.47 1.20 1.00 1.26 2.24 .01
Violence 0.73 0.45 0.49 0.50 2.99 < .01
Mild violence 0.73 0.45 0.49 0.50 3.06 .001
Severe violence 0.48 0.50 0.31 0.47 2.10 < .05
NOTE: The response scale for the original items in this table was from 0 (
more than 20
). Because the violence, mild violence, and severe violence indices were skewed, we followed
Straus’s (1979, p. 80) recommendation that these scales be dichotomized into violent and nonvio
lent categories. Thus, for these three indices, 0 indicates having engaged in none of the specific tac
tics during the past year, and 1 indicates having engaged in at least one of the tactics during the past
year. Degrees of freedom ranged from 136 to 140.
Range of Aggression Tactics Used Against Partner
in Lifetime in Green Versus Barren Conditions
MSD MSD t p
Overall aggression .44 .28 .32 .30 2.39 < .01
Psychological aggression .58 .36 .44 .35 2.46 < .01
Violence .32 .27 .24 .32 1.54 .06
Mild violence .52 .39 .35 .40 2.50 < .01
Severe violence .19 .24 .16 .29 0.80 .22
NOTE: In response to questions about having ever used specific aggressive conflict tactics, partici
pants responded never (0) or yes (1). Standard deviations are in parentheses; degrees of freedom
ranged from 136 to 140.
using a narrower set of aggressive conflict tactics against their partners over
their lifetime than did their counterparts living in barren conditions. The
following rows in Table 3 show the findings for specific forms of aggres
sion. The second row indicates that there was a significant condition differ
ence in psychological aggression against partner; residents living in green
conditions used a significantly narrower set of psychologically aggressive
conflict tactics than did residents living in barren conditions. The third row of
the table suggests that nearby nature may be related to the range of violent
conflict tactics used against partners. For residents living in green conditions,
the set of violent tactics used was 25% smaller than for those living in barren
conditions, a marginally significant difference (p = .06). Although there was
no difference in the range of severe forms of violence used against partners,
there was a significant condition difference in more mild forms: Residents
living in green conditions report using a smaller set of mildly violent conflict
tactics against their partners over their lifetime than did their counterparts liv
ing in barren conditions. Mean differences between the green and barren con-
ditions ranged from more than one quarter to one half of a standard deviation.
Aggression against a child. The conditions leading to aggression against
an adult family member may be quite different from those leading to aggres-
sion against one’s child. Conflicts with children are likely to be more fre-
quent than those with adult family members and often center around
disciplinary issues. At the same time, some forms of aggression may be less
socially acceptable against children than against adults. Does the relationship
between nearby nature and aggression found for women and their partners
exist for women and their children? Condition differences were examined for
aggressive conflict tactics used with children, specifically, the child with
whom the participant had the most conflicts. A t test showed that greenness
was related to the range of psychologically aggressive tactics used against
children: Lifetime scores for proportion of psychologically aggressive tactics
used were significantly lower for participants living in the green condition
than for their counterparts living in the barren condition (.54 vs. .62), t(140) =
1.83, p < .05. But the effect did not hold for the frequency of psychological
aggression during the past year or for the frequency or range of more violent
forms of aggression against children.
In sum, there were a number of indications that nearby nature has a miti
gating effect on aggression and violence: Nearby nature was systematically
related to lower scores on multiple indices of aggression against partners and
one index of aggression against children.
560 ENVIRONMENT AND BEHAVIOR / July 2001
TESTING FOR UNDERLYING MECHANISMS
What mechanism or mechanisms might underlie the association between
nearby nature and aggression? The following analyses tested for each of the
following: the proposed mechanism, the possibility that some unspecified
mechanism might be at work, and three specific alternative mechanisms.
If effects of nearby nature on aggression operate through attentional resto
ration, a number of predictions follow. First, residents living in green condi
tions should show higher levels of attentional functioning than their
counterparts living in barren conditions. A planned student t test showed that
indeed, mean DSB scores were significantly higher in the green condition (M =
5.0, SD = 1.0) than in the barren condition (M = 4.6, SD = 1.2), t(138) = 1.74,
p < .05, differing by more than one third of a standard deviation.
Second, if effects of nearby nature on aggression operate through
attentional restoration, then attentional functioning should be systematically
related to aggression. Using the lifetime measure of overall aggression
against partner as a summary index of aggression, an ordinary least squares
regression was conducted using DSB scores to predict levels of aggression.
As predicted, there was a significant negative relationship between DSB per-
formance and overall aggression (β = –.26, R
= .07, F = 9.9, p < .0025). By
this summary measure of aggression, the better a participant’s attentional
functioning, the less aggression she had engaged in.
And finally, if effects of nearby nature on aggression operate through
attentional restoration, the relationship between nature and aggression
should statistically depend on the relationship between attention and aggres-
sion. These interdependencies are important to examine when hypothesizing
mediation because significant associations among three variables are possi
ble without there being a mediation relationship between them. For example,
in this case, nearby nature might enhance attention and reduce aggression but
influence aggression through some other mechanism than attention. In that
case, the nature-aggression relationship would most likely be statistically
independent of the nature-attention relationship. If, on the other hand, con
tact with nearby nature reduces aggression via the restoration of attentional
resources, we would expect the nature-aggression relationship to diminish or
disappear when attention is statistically controlled.
Accordingly, a multiple regression was used to test for the joint relation
ships among nearby nature, attentional performance, and levels of aggres
sion. When DSB (the proposed mediator) was controlled in a regression
between greenness and overall aggression, greenness was no longer a signifi
cant predictor (β = –.13, p = .11, R
= .09, F = 6.5, p < .0025). Complete, or
Kuo, Sullivan / AGGRESSION AND VIOLENCE 561
“perfect,” mediation requires that the independent variable has no additional
predictive power when the mediator is controlled (Baron & Kenny, 1986);
thus, these findings indicate that attentional restoration could be the sole
mechanism underlying the nature-aggression relationship found here.
Could the links among nature, attention, and aggression be explained by
some unspecified confounding variable or some alternative mechanism?
Evans and Lepore (1997) suggested addressing what they referred to as “the
spuriousness problem” by conducting an analysis in which the relationship
between the hypothesized mediator and the outcome variable is examined
while the independent variable is controlled. By their reasoning, if there is
some unspecified confounding variable responsible for the relationships
among nature, attention, and aggression, then attention will not be signifi
cantly related to aggression when nature is controlled. In fact, the multiple
regression described earlier addresses this possibility: DSB was a significant
predictor of overall aggression (β = –.24, p < .01) when greenness was con
trolled. This finding indicates that some unspecified mechanism cannot
account for the relationships among DSB, greenness, and overall aggression.
These findings were echoed in follow-up analyses examining the follow-
ing three specific, theoretically plausible, alternative mechanisms: positive
mood, stress, and social integration. Planned student t tests showed that
greenness was unrelated to positive mood, t(142) = –.04, p = .48, and stress,
t(140) = .17, p = .43, but was related to social integration, t(140) = 2.7, p < .01.
Correlational analyses showed that overall aggression was related to neither
mood, r(141) = –.07, p = .48, nor stress, r(139) = .135, p = .11, nor social inte-
gration, r(142) = –.06, p = .48. Together, these results indicate that neither posi-
tive mood, nor stress, nor social integration mediate the nature-aggression
relationship found here. Moreover, these analyses reinforce the aforemen
tioned mediation and spuriousness findings, indicating that the effect of
nature on aggression found here may be wholly mediated through attentional
In 145 adult women randomly assigned to a series of architecturally iden
tical apartment buildings, levels of aggression and violence were signifi
cantly lower among individuals who had some nearby nature outside their
apartments than among their counterparts who lived in barren conditions.
Furthermore, as would be predicted if this relationship were mediated by
mental (attentional) fatigue, (a) residents living in greener settings
562 ENVIRONMENT AND BEHAVIOR / July 2001
demonstrated reliably better performance on measures of attentional func
tioning, (b) attentional performance predicted scores on a summary index of
aggression, and (c) the relationship between nearby nature and aggression
scores became nonsignificant when attention was controlled. Finally,
follow-up analyses examining potential alternative mediators revealed no
significant relationships between aggression and mood, stress, or social inte
gration, and a test for unspecified mediators similarly ruled out alternative
It should be noted that the predicted relationship between nearby nature
and aggression was not consistently found for more violent forms of aggres
sion or for aggression against children. Of the various forms of aggression
examined in this study, these may be the most susceptible to social desirabil
ity effects. Future research might use other strategies to examine the
nature-aggression relationship for forms of aggression that are most difficult
to assess through self-report.
To what extent can the nature-aggression relationship found here be inter-
preted as an effect of nearby nature on aggression? The following consider-
ations lend confidence in a causal interpretation of these data: the random
assignment of residents to nature condition; the consistently negative find-
ings across numerous checks for condition differences in participant, house-
hold, and interviewer characteristics; the consistency of architectural and
other environmental features over the two conditions; the use of multiple
buildings per condition; and the use of double-blind measures for both
nearby nature and aggression. Numerous tests were conducted to identify the
particular causal pathway between nature and aggression. Results from all of
these tests were of one accord: The mediation tests indicated a pathway
through attention, and the spuriousness test and direct tests of alternative
mediators all worked to rule out other possible pathways. Although other
possibilities cannot be ruled out entirely, the only interpretation consistent
with the complete set of findings here is that nearby nature reduces aggres
sion by supporting attentional functioning. At this juncture, attention restora
tion theory (S. Kaplan, 1995) provides the best explanation for the link
between nature and aggression.
Having addressed the question of internal validity, we turn now to external
validity. To what extent do the relationships found in this study generalize to
the real world? External validity depends in large part on how the constructs
in a study are operationalized. In this study, the constructs were
operationalized as directly as possible; to the extent we could, we avoided
using surrogates or proxies. For example, measuring the vegetation around
participants’ homes was a more direct way to assess the effects of residential
Kuo, Sullivan / AGGRESSION AND VIOLENCE 563
nature than, say, showing slides of nature in a classroom. Similarly, using a
performance measure of attention provided a more direct measure of
attentional functioning than asking participants to rate how attentive they
feel. And, asking participants to estimate the actual frequency of specific
aggressive behaviors in the past year provided a more direct measure of in
situ aggression than obtaining ratings of feelings of aggression in a labora
tory setting or eliciting hostile attributions in hypothetical contexts. Relying
on relatively direct measures of nature, attention, and aggression lends
greater confidence that the relationships found here are true outside of this
study. The large sample size employed (145 participants) further strengthens
the case for external validity.
At the same time, there is reason for caution in assuming that these effects
generalize to forms of aggression not studied here or to aggression in other
populations and settings. Although the mental fatigue hypothesis should
apply to many forms of aggression and it is quite clear that both men and
women are subject to mental fatigue, this work examined only intrafamily
aggression by women. Future research should examine effects of nature on
aggression by men and other forms of aggression (e.g., road rage and gang
These qualifications notwithstanding, domestic violence is an important
topic in and of itself, and findings with regard to domestic violence have
far-reaching implications. A substantial literature has established that com-
pared with children from nonviolent families, children of violent families are
more likely to grow up to be violent. This increased risk for violent behavior
includes not only children who were victims of abuse but also those who wit-
nessed abuse (Bandura, 1973, 1978; DuRant et al., 1994; Rice & Remy,
1998; Wissow et al., 1992; Wolfe, Jaffe, Wilson, & Zak, 1985). Thus, identi
fying possible avenues to reducing domestic violence may pay benefits for
generations to come. By reducing intrafamily aggression and thus children’s
socialization into aggressive and violent behaviors, green neighborhood
spaces may indirectly reduce aggression in future generations.
This work has implications for understanding and preventing aggression
and for our understanding of the psychological effects of natural
UNDERSTANDING AND PREVENTING AGGRESSION
One contribution of this work is to suggest a potential explanation for a
number of poorly understood phenomena in the environment-behavior litera
ture on human aggression. Mental fatigue might help account for the relation
ships found between crowding and aggression (Ani & Grantham-McGregor,
564 ENVIRONMENT AND BEHAVIOR / July 2001
1998; Nijman & Rector, 1999; Palmstierna, Huitfeldt, & Wistedt, 1991) and
noise and aggression (Donnerstein & Wilson, 1976; Gaur, 1988; Geen &
McCown, 1984; Sherrod, Moore, & Underwood, 1979), and for urban-rural
differences in aggression (Fingerhut, Ingram, & Feldman, 1998). Noise and
crowding both seem likely to place demands on attention (Cohen &
Spacapan, 1978), and urban environments tend not only to be noisier and
more crowded than rural environments but also less green than rural environ
ments. Thus, urban environments seem likely to be more attentionally fatigu
ing and less attentionally restorative in general than rural environments.
Future research might examine whether these phenomena are indeed fatigue
This work may also offer insight into some phenomena in human aggres
sion that do not necessarily involve the physical environment. For example,
both the extremely high rates of aggression and violence in poor families
and the link between stressful life events and aggression (Guerra,
Huesmann, Tolan, Van Acker, & Eron, 1995; Hammond & Yung, 1991;
Patterson, Kupersmidt, & Vaden, 1990; Spencer, Dobbs, & Phillips, 1988;
Straus et al., 1980) might be explained at least in part by mental fatigue. As
described in the introduction, poverty is likely to place relatively high, un-
remitting demands on attention. And stressful life events such as moving to a
new home or having a family member become seriously ill can involve sub-
stantial amounts of problem solving, contingency planning, and other atten-
tionally demanding, mentally fatiguing forms of cognition. If the relatively
high rates of aggression associated with poverty and stressful life events are
indeed partially attributable to mental fatigue, future research should find
links between poverty and mental fatigue as well as links between stressful
life events and mental fatigue.
This work also suggests a number of possible interventions for addressing
aggression and violence in the inner city. Specifically, efforts to improve con
flict behavior might involve preventing, detecting, and treating attentional
fatigue. For example, conflict behavior might be improved by preventing
attentional fatigue through reducing the attentional demands of the environ
ment by means of soundproofing, reducing crowding, and increasing safety.
Similarly, providing insurance against the drastic life changes to which the
poor are most susceptible might also help prevent fatigue and fatigue-related
aggression. DSB and other tests of attentional functioning might help detect
fatigue and let individuals know when they are most at risk for aggressive or
violent behavior. Finally, strategies for treating attentional fatigue, including
taking green breaks and getting more sleep, might help prevent
Kuo, Sullivan / AGGRESSION AND VIOLENCE 565
UNDERSTANDING THE EFFECTS OF NATURAL
ENVIRONMENTS ON HUMAN BEHAVIOR
This work contributes to our understanding of the psychological effects of
natural environments in a number of ways. First, the findings provide strong
evidence for a potential effect of nature that has been largely unexplored—
reducing aggression and violence. Previous research on the effects of nature
has focused on its effects on mood, recovery from stress, everyday function
ing, and attention (e.g., Cimprich, 1993; Hartig et al., 1991; Hull & Michael,
1995; Ulrich et al., 1991), and only two previous studies have hinted at a
potential effect of nature on aggression (Mooney & Nicell, 1992; Rice &
Remy, 1998). This study demonstrates a link between nature and reduced
aggression in an experimental design and provides clear support for the pro
posed mechanism of attentional restoration. In doing so, it extends attention
restoration theory and shows that the theory has implications for a concern as
important and socially relevant as levels of aggression and violence in
A second contribution is to raise an interesting question with regard to the
benefits of residential nature. In these data, the vegetation around apartment
buildings was significantly related to measures of attentional functioning but
not to measures of stress or positive mood. This is consistent with the previ-
ous literature: Other studies have found significant relationships between
residential vegetation and measures of attention (R. Kaplan, 2001 [this
issue]; Tennessen & Cimprich, 1995), and to date we are unaware of any stud-
ies demonstrating links between residential nature and either stress or posi-
tive mood. Are there in fact no relationships between residential nature and
stress or residential nature and mood? Perhaps these relationships exist and
the procedures in this study simply failed to uncover them. It also seems pos
sible that mood and stress are simply not affected by highly habituated forms
of nature. This seems a fascinating question for future research.
A third contribution of this work concerns the density and extent of nature
necessary to convey benefits. It might seem implausible that a few trees and
grass in relatively small areas outside public housing apartment buildings
could have any clear effects on residents’ levels of aggression. Yet this low
dose of vegetation has been shown to have far-reaching and positive effects
on a number of other important outcomes, including residents’ management
of major life issues (Kuo, 2001) and neighborhood social ties (Kuo et al.,
1998; Kweon et al., 1998). Future research might explore how the benefits of
contact with nature vary as a function of the density of vegetation.
A final contribution of this work is to suggest that the geographic distribu
tion of natural areas matters. Although large central or regional parks are
566 ENVIRONMENT AND BEHAVIOR / July 2001
clearly important components of urban design, the results of this study sug
gest that a few major parks are not enough. All residents of RTH live within 2
miles of one of the most extensive examples of urban nature in North Amer
ica—Lake Michigan and the parks along Lake Shore Drive in Chicago. Yet
the proximity to these tremendous natural resources is apparently insufficient
to keep all residents of RTH at similar levels of attentional functioning. Per
haps, as Rachel Kaplan (1985) suggested, cities should be designed with
nature at every doorstep.
1. Given that residents do have some choice of apartment within Robert Taylor Homes
(RTH), it seemed possible (although not likely) that better functioning and therefore potentially
less aggressive residents might self-select into greener buildings. As a check on that possibility,
participants were asked what criteria were important to them in choosing a place to live: Of 118
responses, 93% were clearly unrelated to levels of vegetation (47% of respondents “just needed a
place;” 12% desired safety or cleanliness; 10% were concerned about access to work, school, or
family; 9% were concerned about cost; 8% were concerned about space or number of bedrooms;
6% wanted an apartment on a “low floor,” perhaps because of the frequency of elevator malfunc-
tions; and 1 participant mentioned sense of community). Seven percent of respondents expressed
concerns that might be interpreted as related to levels of vegetation (e.g., location, neighbor-
hood, area, and environment), and 1 participant of the 145 specifically reported that a “natural
setting” was important to her. However, analyses indicated that these participants lived in no
greener areas on average than the remainder of the participants in this study. Thus, the level of
nearby nature does not seem to be an important criterion in residents’ selection of apartments
within RTH; moreover, it appears that the level of choice residents have in selecting an apartment
is sufficiently low that even residents who might strongly value access to nature are no more
likely to be assigned to a green area.
2. Eligibility requirements for public housing and some other forms of public aid favor single
mothers. This creates a pressure for families not to list adult males as official residents (and for
these unofficial residents not to participate in studies about life at RTH).
Ani, C. C., & Grantham-McGregor, S. (1998). Family and personal characteristics of aggressive
Nigerian boys: Differences from and similarities with Western findings. Journal of Adoles
cent Health, 23, 311-317.
Baker, C. F. (1984). Sensory overload and noise in the ICU: Sources of environmental stress.
Critical Care Quarterly, 6(4), 66-80.
Bandura, A. (1973). Aggression: A social learning analysis. Englewood Cliffs, NJ: Prentice
Kuo, Sullivan / AGGRESSION AND VIOLENCE 567
Bandura, A. (1978). Social learning theory of aggression. Journal of Communication, 28(3),
Barnett, O. W., & Hamberger, L. K. (1992). The assessment of maritally violent men on the Cali
fornia Psychological Inventory. Violence and Victims, 7, 15-28.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psy
chological research: Conceptual, strategic, and statistical considerations. Journal of Person
ality and Social Psychology, 51, 1173-1182.
Baum, A., & Koman, S. (1976). Differential response to anticipated crowding: Psychological
effects of social and spatial density. Journal of Personality and Social Psychology, 34,
Bolger, K., Thomas, M., & Eckenrode, J. (1997). Disturbances in relationships: Parenting, fam
ily development, and child maltreatment. In J. Garbarino (Ed.), Understanding abusive fami
lies: An ecological approach to theory and practice (pp. 86-98). San Francisco: Jossey-Bass.
Brady, K. T., Myrick, H., & McElroy, S. (1998). The relationship between substance use disor
ders, impulse control disorders, and pathological aggression. American Journal on Addic
tions, 7, 221-230.
Canin, L. H. (1991). Psychological restoration among AIDS caregivers: Maintaining self-care.
Unpublished doctoral dissertation, University of Michigan.
Caprara, G. V., & Renzi, P. (1981). The frustration-aggression hypothesis vs. irritability.
Recherches de Psychologie Sociale, 3, 75-80.
Chang, Y. C. (1994). Parenting stress and social support of mothers who physically abuse their
children in Hong Kong. Child Abuse and Neglect, 18, 261-269.
Cherek, D. R., Moeller, F. G., Dougherty, D. M., & Rhoades, H. (1997). Studies of violent and
nonviolent male parolees: II. Laboratory and psychometric measurements of impulsivity.
Biological Psychiatry, 41, 523-529.
Chicago Housing Authority. (1995). Statistical profile: The Chicago Housing Authority 1994 to
1995. Chicago: Author.
Cimprich, B. (1993). Development of an intervention to restore attention in cancer patients.
Cancer Nursing, 16, 83-92.
Coccaro, E. F., Bergeman, C. S., Kavoussi, R. J., & Seroczynski, A. D. (1997). Heritability of
aggression and irritability: A twin study of the Buss-Durkee aggression scales in adult male
subjects. Biological Psychiatry, 41, 273-284.
Cohen, S., & Spacapan, S. (1978). The aftereffects of stress: An attentional interpretation. Envi
ronmental Psychology and Nonverbal Behavior, 3, 43-57.
Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information-processing
mechanisms in children’s social adjustment. Psychological Bulletin, 115, 74-101.
Dodge, K. A., & Crick, N. R. (1990). Social information-processing bases of aggressive behav
ior in children. Personality and Social Psychology Bulletin, 16, 8-22.
Dodge, K. A., & Schwartz, D. (1997). Social information processing mechanisms in aggressive
behavior. In D. M. Stoff, J. Breiling, & J. D. Maser (Eds.), Handbook of antisocial behavior
(pp. 171-180). New York: John Wiley.
Donnerstein, E., & Wilson, D. W. (1976). Effects of noise and perceived control on ongoing and
subsequent aggressive behavior. Journal of Personality and Social Psychology, 34, 774-781.
DuRant, R. H., Pendergrast, R. A., & Cadenhead, C. (1994). Exposure to violence and victimiza
tion and fighting behavior by urban Black adolescents. Journal of Adolescent Health, 15,
Evans, G. W., & Lepore, S. J. (1997). Moderating and mediating processes in environment-
behavior research. In G. T. Moore (Ed.), Advances in environment, behavior, and design, Vol.
568 ENVIRONMENT AND BEHAVIOR / July 2001
4: Toward the integration of theory, methods, research, and utilization (pp. 255- 285).
Edmond, OK: Environmental Design Research Association.
Fingerhut, L. A., Ingram, D. D., & Feldman, J. J. (1998). Homicide rates among U.S. teenagers
and young adults: Differences by mechanism, level of urbanization, race, and sex, 1987
through 1995. Journal of American Medical Association, 280, 423-427.
Garbarino, J., & Sherman, D. (1980). High-risk neighborhoods and high-risk families: The
human ecology of child maltreatment. Child Development, 51, 188-198.
Gaur, S. D. (1988). Noise: Does it make you angry? Indian Psychologist, 5, 51-56.
Geen, R. G., & McCown, E. J. (1984). Effects of noise and attack on aggression and physiologi
cal arousal. Motivation and Emotion, 8, 231-241.
Guerra, N. G., Huesmann, L. R., Tolan, P. H., Van Acker, R., & Eron, L. D. (1995). Stressful
events and individual beliefs as correlates of economic disadvantage and aggression among
urban children. Journal of Consulting and Clinical Psychology, 63, 518-528.
Hammond, W. R., & Yung, B. R. (1991). Preventing violence in at-risk African-American youth.
Journal of Health Care for the Poor and Underserved, 2, 1-16.
Hartig, T., Mang, M., & Evans, G. W. (1991). Restorative effects of natural environment experi
ences. Environment and Behavior, 23, 3-26.
Hinshaw, S. P. (1987). On the distinction between attentional deficits/hyperactivity and conduct
problems/aggression in child psychopathology. Psychological Bulletin, 101, 443-463.
Hull, R. B., & Michael, S. E. (1995). Nature-based recreation, mood change, and stress restora-
tion. Leisure Sciences, 17, 1-14.
Hynan, D. J., & Grush, J. E. (1986). Effects of impulsivity, depression, provocation, and time on
aggressive behavior. Journal of Research in Personality, 20, 158-171.
Kant, R., Smith-Seemiller, L., & Zeiler, D. (1998). Treatment of aggression and irritability after
head injury. Brain Injury, 12, 661-666.
Kaplan, R. (1984). Wilderness perception and psychological benefits: An analysis of a continu-
ing program. Leisure Sciences, 6, 271-290.
Kaplan, R. (1985). Nature at the doorstep: Residential satisfaction and the nearby environment.
Journal of Architectural and Planning Research, 2, 115-127.
Kaplan, R. (2001). The nature of the view from home: Psychological benefits. Environment and
Behavior, 33, 507-542.
Kaplan, S. (1987). Mental fatigue and the designed environment. In J. Harvey & D. Henning
(Eds.), Public environments (pp. 55-60). Edmond, OK: Environmental Design Research
Kaplan, S. (1995). The restorative benefits of nature: Toward an integrative framework. Journal
of Environmental Psychology, 15, 169-182.
Kavoussi, R. J., & Coccaro, E. F. (1998). Divalproex sodium for impulsive aggressive behavior
in patients with personality disorder. Journal of Clinical Psychiatry, 59, 676-680.
Kuo, F. E. (1992, March). Inner cities and chronic mental fatigue. Presentation at the Environ
mental Design Research Association Conference, Denver, CO.
Kuo, F. E. (2001). Effective life functioning in the inner city: Impacts of environment and atten
tion. Manuscript submitted for publication.
Kuo, F. E., Sullivan, W. C., Coley, R. L., & Brunson, L. (1998). Fertile ground for community:
Inner-city neighborhood common spaces. American Journal of Community Psychology, 26,
Kweon, B. S., Sullivan, W. C., & Wiley, A. (1998). Green common spaces and the social integra
tion of inner-city older adults. Environment and Behavior, 30, 832-858.
Lezak, M. D. (1983). Neuropsychological assessment. New York: Oxford University Press.
Kuo, Sullivan / AGGRESSION AND VIOLENCE 569
Lohr, V. I., Pearson-Mims, C. H., & Goodwin, G. K. (1996). Interior plants may improve worker
productivity and reduce stress in a windowless environment. Journal of Environmental Hor
ticulture, 14, 97-100.
Luengo, M. A., Carrillo-de-la-Pena, M. T., Otero, J. M., & Romero, E. (1994). A short-term lon
gitudinal study of impulsivity and antisocial behavior. Journal of Personality and Social
Psychology, 66, 542-548.
Markovitz, P. (1995). Pharmacotherapy of impulsivity, aggression, and related disorders. In
E. Hollander (Ed.), Impulsivity and aggression (pp. 263-287). New York: John Wiley.
Martinko, M. J., & Zellars, K. L. (1998). Toward a theory of workplace violence and aggression:
A cognitive appraisal perspective. In R. W. Griffin (Ed.), Dysfunctional behavior in organi
zations: Violent and deviant behavior (pp. 1-42). Greenwich, CT: JAI.
McNair, D., Lorr, M., & Droppleman, L. F. (1981). EITS manual for the Profile of Mood States.
San Diego, CA: Educational and Industrial Testing Service.
Mesulam, M. (1985). Principles of behavioral neurology. Philadelphia: F. A. Davis.
Miles, I., Sullivan, W. C., & Kuo, F. E. (1998). Prairie restoration volunteers: The benefits of par
ticipation. Urban Ecosystems, 2, 27-41.
Mooney, P., & Nicell, P. L. (1992). The importance of exterior environment for Alzheimer resi
dents: Effective care and risk management. Healthcare Management Forum, 5(2), 23-29.
Neff, J. A., Holamon, B., & Schluter, T. D. (1995). Spousal violence among Anglos, Blacks, and
Mexican Americans: The role of demographic variables, psychosocial predictors, and alco-
hol consumption. Journal of Family Violence, 10, 1-21.
Nijman, H.L.I., & Rector, G. (1999). Crowding and aggression on inpatient psychiatric wards.
Psychiatric Services, 50, 830-831.
Ovitt, M. A. (1996). The effect of a view of nature on performance and stress reduction of ICU
nurses. Unpublished master’s thesis, University of Illinois.
Palmstierna, T., Huitfeldt, B., & Wistedt, B. (1991). The relationship of crowding and aggressive
behavior on a psychiatric intensive care unit. Hospital and Community Psychiatry, 42,
Patterson, C. J., Kupersmidt, J. B., & Vaden, N. A. (1990). Income level, gender, ethnicity, and
household compositions as predictors of children’s school-based competence. Child Devel
opment, 61, 485-494.
Pihl, R. O., & Zacchia, C. (1986). Alcohol and aggression: A test of the affect arousal hypothesis.
Aggressive Behavior, 12, 367-375.
Rabiner, D. L., Lenhart, L., & Lochman, J. E. (1990). Automatic versus reflective social problem
solving in relation to children’s sociometric status. Developmental Psychology, 26, 1010-
Rice, J. S., & Remy, L. L. (1998). Impacts of horticultural therapy on psychosocial functioning
among urban jail inmates. Journal of Offender Rehabilitation, 26, 169-191.
Roth, F. (1986). A practice regimen for diagnosis and treatment of child abuse. Child Welfare, 54,
Rubin, K. H., Bream, L. A., & Rose-Krasnor, L. (1991). Social problem solving and aggression
in childhood. In D. J. Pepler & K. H. Rubin (Eds.), The development and treatment of child
hood aggression (pp. 219-248). Hillsdale, NJ: Lawrence Erlbaum.
Rule, B., Taylor, B., & Dobbs, A. (1987). Priming effects of heat on aggressive thoughts. Social
Cognition, 5, 131-143.
Scholte, R.H.J., van Aken, M.A.G., & van Lieshout, C.F.M. (1997). Adolescent personality fac
tors in self-ratings and peer nominations and their prediction of peer acceptance and peer
rejection. Journal of Personality Assessment, 69, 534-554.
570 ENVIRONMENT AND BEHAVIOR / July 2001
Schwartz, D. (1994). The measurement of inhibitory attention and psychological effectiveness
among adolescents. Unpublished master’s thesis, University of Michigan.
Sherrod, D. R., Moore, B. S., & Underwood, B. (1979). Environmental noise, perceived control,
and aggression. Journal of Social Psychology, 109, 245-252.
Spencer, M. B., Dobbs, B., & Phillips, D. (1988). African-American adolescents: Adaptational
processes and socioeconomic diversity I behavioral outcomes. Journal of Adolescence, 11,
Stanford, M. S., Greve, K. W., & Dickens, T. J. (1995). Irritability and impulsiveness: Relation
ship to self-reported impulsive aggression. Personality and Individual Differences, 19,
Stewart, M. A. (1985). Aggressive conduct disorder: A brief review. Aggressive Behavior, 11,
Straus, M. A. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT)
Scales. Journal of Marriage and the Family, 41, 75-88.
Straus, M. A. (1995). Manual for the Conflict Tactics Scales. Durham, NH: University of New
Hampshire, Family Research Laboratory.
Straus, M. A., & Gelles, R. J. (1986). Societal change and change in family violence from 1975
to 1985 as revealed by two national surveys. Journal of Marriage and the Family, 48,
Straus, M. A., & Gelles, R. J. (1988). How violent are American families? Estimates from the
National Family Violence Resurvey and other studies. In G. T. Hotaling, D. Finkelhor, J. T.
Kirkpatrick, & M. A. Straus (Eds.), Family abuse and its consequences: New directions in
research (pp. 14-36). Newbury Park, CA: Sage.
Straus, M. A., Gelles, R. J., & Steinmetz, S. K. (1980). Behind closed doors: Violence in the
American family. Garden City, NY: Doubleday.
Tennessen, C., & Cimprich, B. (1995). Views to nature: Effects on attention. Journal of Environ-
mental Psychology, 15, 77-85.
Thackray, R. I., Bailey, J. P., & Touchstone, R. M. (1979). The effect of increased monitoring
load on vigilance performance using a simulated radar display. Ergonomics, 22, 529-539.
Tuinier, S., Verhoeven, W.M.A., & Van Praag, H. M. (1996). Serotonin and disruptive behavior:
A critical evaluation of the clinical data. Human Psychopharmacology Clinical and Experi
mental, 11, 469-482.
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 Psy
chology, 11, 201-230.
Warm, J. S., & Dember, W. N. (1986). Awake at the switch. Psychology Today, 20(4), 46-53.
Wissow, L. W., Wilson, M. E., Roter, D., Larson, S., & Hope, I. B. (1992). Family violence and
the evaluation of behavioral concerns in a pediatric primary care clinic. Medical Care, 30,
Wolfe, D. A., Jaffe, P., Wilson, S. K., & Zak, L. (1985). Children of battered women: The relation
of child behavior to family violence and maternal stress. Journal of Consulting and Clinical
Psychology, 53, 657-665.
Kuo, Sullivan / AGGRESSION AND VIOLENCE 571