ABSTRACT 

 

 

Title of Thesis: THE BRAIN DOES NOT LIE: A CASE STUDY OF 
PSYCHOPHYSIOLOGY AND LANDSCAPE IN SOUTH 
CLIFTON PARK 

 

  

 Audrey Claire Seiz, Master of Landscape Architecture, 
2023 

 

Thesis Directed By:  Byoung-Suk Kweon, Associate Professor, Department of 
Plant Science and Landscape Architecture 

Researchers have long explored how humans respond psychologically and 

physiologically to distinct landscapes and natural features. Walking in nature and viewing 

photographs of natural landscapes have been shown to reduce stress measured through 

physiological responses of blood pressure, salivary cortisol concentration, and pulse rate. 

Exposure to natural landscapes has also been shown to improve feelings of relaxation and 

positive emotion. The increased popularity of virtual reality (VR) in landscape architecture 

provides an additional visualization tool to immerse a participant in a landscape at human scale. 

Little research has focused on the potential impact of visualization through VR, studied the 

impact of urban nature, or compared the impact of landscape design using the same site. This 

study explores how employment of psychophysiological measures provides objective 

assessment of humans' landscape perception in response to the restorativeness of a virtual place. 

Twenty students were recruited to view an actual site in South Clifton Park, Baltimore 

City. Utilizing VR, participants observed the site as it exists currently and reimagined 

using the tenets of Attention Restoration Theory (ART), Stress Reduction Theory (SRT), 



and community vision. Psychological response was analyzed using the Perceived 

Restorativeness Scale (PRS-16), a survey designed to evaluate a place’s restorativeness through 

principles of ART, and physiological response was analyzed using electroencephalogram 

(EEG), the non-invasive measurements of the electrical brain activity. Findings indicated that 

perceived restorativeness increased in the designed site for the factors Being Away/Fascination 

and Compatibility; however, no significant difference was identified for the factor Extent. 

Regarding EEG data, alpha brain frequencies (broadband alpha, low alpha, and high alpha) were 

not significantly different when viewing the vacant versus designed site within the frontal or 

parietal lobes; however, beta brain frequencies (broadband beta, low beta, and high beta) 

demonstrated a marginally significant effect of sex in the frontal and parietal lobes with male 

beta brain frequencies decreasing when viewing the designed site and female beta brain 

frequencies increasing. Finally, frontal alpha asymmetry, a measure of approach-withdrawal 

motivation, demonstrated a marginally significant decrease when viewing the designed site, 

indicating increased withdrawal motivation in the designed site. The present research seeks to fill 

a gap in understanding objective indicators of restorativeness of a place and explore the power of 

VR as a tool for visualizing place.   

 

 

 
 
 
 
 
 
 
 
 



 
 

 
 

THE BRAIN DOES NOT LIE: A CASE STUDY OF PSYCHOPHYSIOLOGY AND 
LANDSCAPE IN SOUTH CLIFTON PARK 

 
 

by 
 

Audrey Claire Seiz 
 

 

 

Thesis submitted to the Faculty of the Graduate School of the  
University of Maryland, College Park, in partial fulfillment  

of the requirements for the degree of 
Master of Landscape Architecture 

2023 
 
 
 
 
 
 
 
 
 
 
 
 

Advisory Committee: 
 
Dr. Byoung-Suk Kweon, Associate Professor, University of Maryland, College Park 
Dr. Christopher Ellis, Associate Professor, University of Maryland, College Park 
Dr. Hyuk Oh, Assistant Research Professor, University of Maryland, College Park 
 

 

 



 

 

 

 

 

 

 

 

 

 

 
© Copyright by  

Audrey Claire Seiz 
2023 

 



ii 
 

Acknowledgements 

My deepest thanks to my committee chair, Dr. Byoung-Suk Kweon, who put so 

much time, energy, and care into helping with this study- I would not have been able to 

do this without you!  

I would also like to thank my committee members, Dr. Christopher Ellis and Dr. 

Hyuk Oh, for their guidance, patience, and dedication throughout this process.  

Additionally, a huge thank you to Dr. Doris Minor-Terrell and the Broadway East 

Community & CDC for welcoming me to work in their community and providing critical 

design insight.  

Next, thank you to my classmates for their encouragement, and the study 

participants that made this all possible.  

Finally, I cannot begin to express my gratitude to my husband, Cai, and my mom, 

Sharon, for their endless love and support.  

 

 

 

 

 

 

 

 

 



iii 
 

Table of Contents 

Acknowledgements ............................................................................................................. ii 
Table of Contents ............................................................................................................... iii 
List of Tables ..................................................................................................................... vi 
List of Figures ................................................................................................................... vii 
List of Abbreviations ......................................................................................................... ix 
Chapter 1: Introduction ....................................................................................................... 1 

Study Intention ................................................................................................................ 1 
Study Purpose, Research Questions, and Scope ............................................................. 1 

Chapter 2: Literature Review .............................................................................................. 3 
Two Theories Supporting the Restorative Capacity of Nature ....................................... 3 

Attention Restoration Theory ...................................................................................... 3 
Stress Reduction Theory .............................................................................................. 5 

Virtual Reality and Virtual Reality Induced Symptoms and Effects .............................. 6 

Brain Anatomy, Frequencies, and Measurement ............................................................ 7 

Psychological Responses to Environment- Attention Restoration Theory ..................... 9 
Perceived Restorativeness ........................................................................................... 9 

Physiological and Emotional Responses to Environment- Stress Reduction Theory ... 11 

Physiological Metrics with a Stress Stimulus............................................................ 11 

Physiological Metrics without a Stress Stimulus ...................................................... 12 
Emotion ..................................................................................................................... 13 

Electroencephalogram- A Psychophysiological Metric Relating to Stress Reduction 
and Attention Restoration Theories............................................................................... 15 

Images and Videos..................................................................................................... 15 

Outdoor Immersion.................................................................................................... 17 

Virtual Reality ........................................................................................................... 18 
Frontal Alpha Asymmetry ......................................................................................... 19 

Landscape Architecture’s Place in Neurourbanism and Environmental Justice ........... 20 

Study Framework and Hypothesis ................................................................................ 21 

Chapter 3: Methods ........................................................................................................... 22 
Site Location, Inventory, and Analysis ......................................................................... 22 

Site Selection with Community Leader ..................................................................... 22 



iv 
 

Site Inventory ............................................................................................................ 23 

Site Analysis .............................................................................................................. 31 

Experimental Data Collection ....................................................................................... 33 
IRB Application ......................................................................................................... 33 

Participant Recruitment ............................................................................................. 33 

Digital Modeling and Stereoscopic Images ............................................................... 34 

Data Collection Procedure ......................................................................................... 37 
Raw EEG Data Processing ........................................................................................ 40 

Statistical Analysis .................................................................................................... 41 

Chapter 4: Results ............................................................................................................. 43 
Design Program ............................................................................................................. 43 

Vision, Goals, and Strategies ..................................................................................... 43 

Site Design .................................................................................................................... 44 

Goal 1- Promote the multifunctional use of space .................................................... 44 

Goal 2- Provide accessibility, order, and safety ........................................................ 45 
Goal 3- Encourage interaction with nature ................................................................ 47 

Goal 4- Increase tree canopy ..................................................................................... 48 

Goal 5- Bolster fascination and complexity .............................................................. 48 

Experimental Results..................................................................................................... 49 
Participant Demographics and Previous Virtual Reality Experience ........................ 49 

Perceived Restorativeness ......................................................................................... 50 

Electroencephalogram Data ....................................................................................... 53 

Chapter 5: Discussion ....................................................................................................... 65 
Site Design .................................................................................................................... 65 

Implications ............................................................................................................... 65 

Limitations and Future Study .................................................................................... 66 

Experimental Results..................................................................................................... 66 
Implications ............................................................................................................... 66 

Limitations and Future Study .................................................................................... 69 

Chapter 6: Conclusion....................................................................................................... 71 
Appendices ........................................................................................................................ 72 

Appendix I: IRB Application Approval ........................................................................ 72 

Appendix II: Flyer for Study Participation ................................................................... 77 



v 
 

Appendix III: Consent Form ......................................................................................... 78 

Appendix IV: Questionnaire 1 ...................................................................................... 81 

Appendix V: Questionnaire 2 & 3................................................................................. 82 
Appendix VI: Compensation Form ............................................................................... 84 

Appendix VII: EEG Data Processing Methodology ..................................................... 85 

References ......................................................................................................................... 86 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



vi 
 

List of Tables 

Table 1: Summary of brain frequencies 

Table 2: Baltimore City and South Clifton Park income, age, and education demographics 

Table 3: Site opportunities and constraints 

Table 4: Image matrix for the experimental study by participant 

Table 5: Site design goals and strategies 

Table 6: Results of Varimax rotation factor analysis for the vacant site 

Table 7: Results of Varimax rotation factor analysis for the designed site 

Table 8: Descriptive statistics for the vacant and designed site by factor 

Table 9: Results of the paired t-test showing the difference in perceived restorativeness 
between the two stimulus environments   

Table 10: Summary of extreme outliers of each brain frequency in frontal electrodes 

Table 11: Descriptive statistics of brain frequencies for the vacant and designed site in 
frontal electrodes 

Table 12: Results of the paired t-test showing the difference in brain frequencies between 
the two stimulus environments in frontal electrodes 

Table 13: Results of the repeated measures ANOVA per brain frequency in the frontal 
electrodes 

Table 14: Summary of extreme outliers of each brain frequency in parietal electrodes 

Table 15: Descriptive statistics of brain frequencies for the vacant and designed site in 
parietal electrodes 

Table 16: Results of the paired t-test showing the difference in brain frequencies between 
the two stimulus environments in parietal electrodes 

Table 17: Results of the repeated measures ANOVA per brain frequency in the parietal 
electrodes 

Table 18: Descriptive statistics of FAA for the vacant and designed site 

Table 19: Results of the paired t-test showing the difference in FAA between the two 
stimulus environments   

Table 20: Results of the repeated measures ANOVA for FAA 

 



vii 
 

List of Figures 

Figure 1: Milgram and Kishino (1994) Virtuality Continuum 

Figure 2: Neighborhood context map 

Figure 3: Growth map of the City of Baltimore 

Figure 4: HOLC map of Baltimore City 

Figure 5: Watersheds, topography, and tree canopy in South Clifton Park 

Figure 6: Racial demographics of Baltimore City and South Clifton Park 

Figure 7: Bus routes and bus stops in South Clifton Park 

Figure 8: Vacant lots and buildings in South Clifton Park 

Figure 9: Community assets in South Clifton Park 

Figure 10: Site context map and site images 

Figure 11: Site hydrology, topography, and surrounding storm drains 

Figure 12: Land ownership and parcel boundaries  

Figure 13: Soil and tree canopy  

Figure 14: Site analysis diagram 

Figure 15: Location and view direction of experimental images 

Figure 16: Vacant and designed views at Location 1 and Location 2 

Figure 17: Vacant and designed views at Location 3 and Location 4 

Figure 18: Study room and room setup 

Figure 19: Experimental study design conducted for each participant 

Figure 20: Map of the 32 channel locations on the EEG cap 

Figure 21: Site plan of South Clifton Park greenspace 

Figure 22: Image 1- View of the main plaza 

Figure 23: Section A-AI of greenspace grading north to south 

Figure 24: Image 2- View from E. North Avenue 

Figure 25: Image 3- View of formal play area and lawn 

Figure 26: Image 4- View of vegetated areas and walking paths 



viii 
 

Figure 27: Box plot showing median, interquartile range, minimum and maximum value, 
and extreme outliers for each perceived restorativeness factor by stimulus environment 

Figure 28: Mean perceived restorativeness factor score by stimulus environment 

Figure 29: Graphs of beta, beta I, and beta II mean PS for the frontal electrodes F7 and F8 
in the vacant and designed stimulus environments 

Figure 30: Graphs of beta and beta I mean PS for the parietal electrodes P3 and P4 in the 
vacant and designed stimulus environments 

Figure 31: Graphs of beta, beta I, and beta II mean PS for the parietal electrodes P7 and 
P8 in the vacant and designed stimulus environments 

Figure 32: Box plot showing median, interquartile range, minimum and maximum value, 
and outliers and extreme outliers for FAA by stimulus environment 

 

 

 

 

 

 

 

 

 

 



ix 
 

List of Abbreviations 

ANOVA- Analysis of Variance 

ART- Attention Restoration Theory 

BP- Blood pressure  

DSB- Digit Span Backward 

DSF- Digital Span Forward  

EEG- Electroencephalogram 

FAA- Frontal alpha asymmetry 

HR- Heart rate  

HRV- Heart rate variability 

Hz- Hertz  

NCPC- Necker Cube Pattern Control 

PANAS- Positive Affect Negative Affect Scale 

PRS- Perceived Restorativeness Scale 

PS- Power spectral 

POMS- Profile of Mood States 

SART- Sustained Attention to Response Task  

SDMT- Symbol Digit Modalities Test  

SRT- Stress Reduction Theory 

VR- Virtual reality 

VRISE- Virtual reality-induced symptoms and effects 

ZIPERS- Zuckerman Inventory of Personal Reactions 

 



1 
 

Chapter 1: Introduction 

Study Intention 

 In Baltimore City, the stark divisions between neighborhoods, often the 

longstanding remnants of racist redlining policy, are marked by changes in the number of 

parks, play spaces, and trees lining city streets. Baltimore City is not alone; however, 

with so many urban areas facing greenspace inequity and environmental injustice. This is 

not the future for our designed landscapes, and this study embraces the challenging work 

needed to untangle us from the status quo and create an equitable, inclusive, community-

driven future using research as a powerful tool to explore, document, and advocate for 

change.  

Study Purpose, Research Questions, and Scope 

 Through previous conversations with community members living in East 

Baltimore neighborhoods of Broadway East, South Clifton Park, and Oliver, the critical 

topics for these communities often centered around housing vacancy, neighborhood 

crime, air quality, and solid waste dumping. Residents discussed the impacts of the built 

environment on their personal finances (high cooling costs), physical health (prevalence 

of asthma), and emotional health (high stress levels). These conversations sparked an 

interest in exploring how to quantitatively measure the physiological and psychological 

impacts of environments on humans and investigate how the employment of these 

psychophysiological measures provides objective assessment of humans’ landscape 

perception in response to the restorativeness of a place. This study asks three research 

questions: 



2 
 

• Does viewing a virtual vacant site versus the site as a designed greenspace alter 

brain activities and/or perception of restorativeness? 

• How do the two different stimulus environments impact brain activities and 

perception of restorativeness?  

• Do the two different stimulus environments impact approach-withdrawal 

motivation? 

To answer these research questions, this study developed a design for a vacant site in 

South Clifton Park, Baltimore based on site inventory, community vision, and the 

principles of Attention Restoration Theory (ART) and Stress Reduction Theory (SRT). 

Virtual reality (VR) visualizations of the design, and the existing site, were used as part 

of the human subject study of 20 University of Maryland students. The study collected 

original data on brain activities using an electroencephalogram (EEG) and perceived 

restorativeness through surveys. Although community engagement on the site design was 

not within the scope of this study, it would be critical throughout the design process if 

this site were to be developed.  

 

 

 

 

 

 



3 
 

Chapter 2: Literature Review 

Two Theories Supporting the Restorative Capacity of Nature 

Researchers have long been using psychophysiological measures to examine the 

restorative capacity of nature on human beings. The two main theories underpinning a 

substantial amount of this research in restoration and nature are Attention Restoration 

Theory (ART) and Stress Reduction Theory (SRT). ART and SRT both stem from an 

evolutionary assumption that, in general, because humans evolved for most of their 

existence in natural environments, humans are adapted psychologically and 

physiologically to the natural world versus an urban environment (Kaplan & Kaplan, 

1989; Ulrich, 1983). This assumption is strongly associated with E.O. Wilson’s biophilia 

hypothesis, which contends that humans have a predisposition to attend to and 

emotionally connect with the nature owing to a series of innate, phylogenetic values 

(Kellert & Wilson, 1993; Wilson, 1984). Both ART and SRT also consider the element of 

stress as playing a role within the restoration models (Kaplan, 1995; Ulrich, 1983). The 

mechanism through which restoration is achieved for humans in natural environments; 

however, establishes the division between these theories. Both theories also provide 

attributes of environments that promote restoration based on the underlying mechanisms 

of the theory (Hartig et al., 1997a).  

Attention Restoration Theory 

ART establishes that a natural space has restorative potential because of its impact 

on human cognition. Utilizing a two-pronged approach to attention established in James 

(1892), ART posits that directed attention requires effort while fascination requires no 

effort (Kaplan, 1995). Directed attention is a finite resource for humans that is crucial for 



4 
 

information-processing, and over time utilization of this resource can cause mental 

fatigue that impacts elements of human perception, thought, action, and emotion (Kaplan, 

1995). Since directed attention is limited, ART offers the concept of fascination, 

involuntary human attention, as a vehicle to lower cognitive effort and rebuild directed 

attention capacity (Kaplan, 1995; Hartig et al., 1991). Fascination can come from a 

variety of sources including natural environments. Kaplan (1995) deemed fascination in 

natural environments “soft fascination”, which promotes reflection to aid directed 

attention recovery (p. 172).  

In addition to fascination, three other attributes of the environment must be 

present to establish a restorative experience- being away, extent, and compatibility 

(Kaplan & Kaplan, 1989; Kaplan, 1995). Being away is described as the feeling of 

breaking away from routine environments, extent relates to the connectedness of features 

within an environment at scale to enhance the feeling of being somewhere else, and 

compatibility references the relatedness of the environment to a human’s goals for using 

a space (Kaplan & Kaplan, 1989; Kaplan, 1995; Hartig et al., 1997b). When all four 

attributes are present, then a restorative experience in an environment is possible that 

facilitates recovery from mental fatigue (Kaplan, 1995).  

As it relates to stress, ART focuses on the aspects that generate an inevitable 

stress response, not the aftereffects of that response. The first aspect is “harm”, either 

direct or indirect, in the form of a threat, and the second aspect is “resource inadequacy” 

(Kaplan, 1995, p. 177). “Resource inadequacy” suggests that a human may not always 

have the appropriate resources to manage a situation that he or she is facing or expected 

to face (Kaplan, 1995). Kaplan (1995) argues that psychological resources, such as 



5 
 

directed attention, are often critical resources that when depleted can lead to a stress 

response. Research evaluating ART, therefore, examines the impacts of environments on 

directed attention and cognition, discussed in Psychological Responses to Environment- 

ART.    

Stress Reduction Theory 

 SRT asserts that a human’s analysis of an environment originally depends on 

emotion (affect) not cognition (Ulrich, 1983). The theory describes a process where an 

immediate emotional reaction (like or dislike) to an environment is derived by our initial 

emotional state prior to a change in environment as well as generalized visual cues 

(Hartig et al., 1991; Ulrich, 1983). This initial emotional reaction causes arousal that then 

sparks the cognitive processing of the environment that could further alter arousal and 

emotional response (Ulrich, 1983). Arousal and emotional response are proceeded by a 

behavior or motivation, which is often oriented towards approach or withdrawal from an 

environment (Ulrich, 1983). SRT documents a series of visual cues that impact a 

human’s immediate reaction to a natural environment. Generally, humans prefer complex 

environments with some ordering that has clear sightlines, a continuous and even ground 

surface texture, hidden vistas, and a water feature (Ulrich, 1983).  

 SRT argues that an environment can provide emotional and physiological 

restoration after interacting with a stress-inducing stimulus (Hartig et al., 1991; Ulrich, 

1983). Specifically, viewing a natural environment following a stressful stimulus alters a 

human’s emotional state to facilitate immediate positive emotional reactions that in turn 

reduce arousal while still maintaining interest (Hartig et al., 1991; Ulrich, 1983). 

Research exploring the theory of SRT, therefore, should capture metrics of human 



6 
 

emotion and physiological traits before and after a stress stimulus, discussed in 

Physiological and Affective Responses to Environment- SRT.  

Virtual Reality and Virtual Reality Induced Symptoms and Effects 

 Virtual reality (VR) is expanding in access throughout a multitude of disciplines; 

however, within the literature, a universal definition of virtual reality is difficult to 

identify (Kardong-Edgren et al., 2019; Portman et al., 2015; Wohlgenannt et al., 2019). 

Milgram & Kishino (1994) developed a spectrum that is still widely referenced with real 

environments on one end and virtual environments on the opposing end that exclusively 

contains computer-generated 

simulations (Figure 1). Wohlgenannt 

et al. (2019) proposes that the three 

components of a VR experience are “(tele-)presence, interactivity, and immersion” 

leading to a comprehensive definition, adopted for this study, that states, “VR leverages 

immersive technologies to simulate interactive virtual environments or virtual worlds 

with which users become subjectively involved and in which they feel physically 

present” (p. 457). Within landscape architecture practice, VR technology is more readily 

being used as a visualization tool for presenting final designs (Hill, 2019); however, the 

use of the technology is trailing behind other related fields, such as architecture and 

environmental planning (Portman et al., 2015).  

 Side effects from virtual reality use have been well documented in the literature 

(Souchet et al., 2022). VR-induced symptoms and effects (VRISE) include physiological 

symptoms, such as motion sickness (cybersickness), visual fatigue, muscle fatigue, 

nausea, dizziness, and disorientation, as well as psychological symptoms, such as stress 

Figure 1: Milgram & Kishino (1994) Virtuality 
Continuum  



7 
 

or mental overload (Kourtesis et al., 2019; Souchet et al., 2022). Time spent and motion 

in virtual reality seems to predictably impact VRISE with increase duration and/or 

motion increasing the likelihood of VRISE in users (Kourtesis et al., 2019). In addition, 

hardware and software capacities, such as refresh rates, latency, and display flickering, 

impact susceptibility to VRISE (Kourtesis et al., 2019; Chang et al., 2020). Individual 

factors of age, gender, and VR experience have also been explored as potential predictors 

of VRISE; however, only frequent VR experience has been demonstrated to consistently 

decrease VRISE (Chang et al., 2020). 

Brain Anatomy, Frequencies, and Measurement 

The cerebral cortex is the outermost layer of the human brain, which contains a 

complex organization of nerve cells (neurons) that create the characteristic tissue 

enfolding (Vachha et al., 2022). It is divided into right and left hemispheres, and each 

hemisphere consists of four major lobes with differing functions- the frontal lobe, the 

parietal lobe, the temporal lobe, and the occipital lobe (Borden et al., 2016). Although the 

intricacies of the lobe functions are still being explored, there are commonly accepted 

capacities that are impacted by damage to these lobes. The frontal lobe is the largest in 

the hemisphere at the front of the skull and is involved in the capacity for motor function, 

including speech production, muscle movement, controlled behavior, emotion regulation, 

including personality and social interactions, and executive function, including decision 

making and selective attention (Firat, 2019). The parietal lobe is located behind the 

frontal lobe and manages sensory and spatial perception and processing as well as 

language skills (Berlucchi & Vallar, 2018). The temporal lobe sits below the frontal and 

parietal lobes and provides auditory and olfactory processing, language formation as well 



8 
 

as learning and memory consolidation, which allows humans to form long-term memory 

(Kiernan, 2012). Finally, the occipital lobe rests behind the parietal lobe and facilitates 

visual input processing and recognition (Nehmad, 1998).  

When a human encounters a sensory input, the neurons in the brain send 

communication via electrical signals within a neuron and chemical signals between 

neurons (Lovinger, 2008). These electrical signals oscillate at distinct rate creating 

waves, and researchers have categorized the electrical signals into five brain frequencies 

based on the rate of these oscillations per second- delta, theta, alpha, beta, and gamma 

(Table 1) (Grassini et al., 2019). Dominance of 

certain brain frequencies is associated with 

cognitive states that relate to human experience. 

As such, increased power of delta activity is not 

normally seen in adults during waking; 

however, is present during sleep, theta activity 

reflects a deeply relaxed and meditative mind, 

alpha activity represents a relaxed but wakeful mind, beta activity signifies an alert and 

attentive mind, and gamma activity often indicates a cognitively engaged or hyperactive 

mind (Khosla et al. 2020). In addition, alpha and beta brain frequencies are further 

categorized into low and high frequencies. Low alpha is associated with attentional 

processing and high alpha cognitive processing, specifically memory (Schomer & Lopes 

de Silva, 2018). Low beta is oriented toward concentration, anxiety, and performance 

whereas high beta relates to stress, anxiety, and arousal (Schomer & Lopes de Silva, 

2018).  

Table 1: Summary of brain frequencies 
Name  Frequency (Hertz) 
Delta 0.5-4 Hz 

Theta 4-8 Hz 

Broadband Alpha 8-13 Hz 

     Low Alpha      8-10 Hz 

     High Alpha      10-13 Hz 

Broadband Beta 13-30 Hz 

     Low Beta       13-20 Hz 

     High Beta      20-30 Hz 

Gamma >30 Hz 



9 
 

Brain frequencies are measured and visualized using an electroencephalogram 

(EEG), a noninvasive instrument composed of a series of electrodes fitted onto a cap that 

adheres to a human’s scalp (Abhang et al., 2016). Although the number of electrodes 

present on a cap can vary, the current placement of electrodes is standardized using a 

10/20 international electrode placement system that organizes electrodes by brain region 

(a series of letters) and hemisphere (right hemisphere is even numbers and left 

hemisphere is odd numbers) (Abhang et al., 2016). With the addition of a conductive 

element on the electrode, often gel if the electrode is dry, it measures the neural electrical 

activity in the brain area below the electrode, providing a direct opportunity to measure 

brain functioning (Abhang et al., 2016). EEG represents a psychophysiological metric 

because it is measuring brain electrical activity (physiological) that is associated with 

cognition (psychology).  

Psychological Responses to Environment- Attention Restoration Theory 

Perceived Restorativeness 

 The Perceived Restorativeness Scale (PRS) is the predominant metric for 

analyzing a participant’s self-reported feeling about the restorativeness of an environment 

with questions targeting ART’s four attributes of a restorative experience (Berto, 2014). 

Initially devised by Hartig et al. (1991), PRS has been adapted and modified to fully 

target the attributes of fascination, being away, compatibility, and extent through distinct 

types of statements (Berto, 2005; Hartig et al., 1996; Hartig et al., 1997a; Hartig et al., 

1997b; Pasini et al., 2014; Purcell et al., 2001). Although variations of this metric do 

exist, there are apparent trends between perceived restorativeness and environment. 

Studies utilizing imagery of a series of urban versus natural environments reported 



10 
 

natural environments were rated as more restorative when compared to urban 

environments (Berto, 2005; Mahamane et al., 2020; Purcell et al., 2001). In addition, 

when experiencing environments in-situ, participants demonstrated higher perceived 

restoration for natural settings (Hartig et al., 1997a; Hartig et al., 1997b; Stigsdotter et al., 

2017).  

Nature in urban environments can also impact the perceived restorativeness of a 

human’s experience. Studies investigated viewing urban settings with and without 

greenspace found that urban greenspace was perceived as more restorative than urban 

areas without greenspace (Hernández & Hidalgo, 2005; Lee et al., 2015; Wang et al., 

2016). Research comparing a natural, urban greenspace, and urban street reported higher 

perceived restorativeness for natural setting, urban greenspace, and urban street, 

respectively; however, human preference plays a vital role in perceived restoration, 

specifically in urban greenspace environments (Korpela, 2013; Wilkie & Clouston, 

2015).  

Little research has explored virtual nature and its impact on perceived 

restorativeness. Schutte et al. (2017) used VR to display images of an Australian natural 

landscape and a small town, and results indicated the virtual natural setting was 

significantly more restorative than the urban environment. One additional study 

compared exposure to nature outdoors versus virtual nature, and the study determined 

that both virtual and outdoor statistically increased restorativeness relative to the control 

environment with no nature present. This indicated that virtual nature could facilitate a 

restorative experience for humans (Browning et al., 2020).  



11 
 

Overall, perceived restorativeness seems to improve when viewing or 

experiencing natural environments compared to urban streets. In addition, greenspace in 

urban environments improves perceived restorativeness; however, not to the same degree 

as natural settings. Further research is needed to understand how virtual nature influences 

perceived restorativeness, but initial studies indicate that virtual nature does increase 

perceived restorativeness.  

Physiological and Emotional Responses to Environment- Stress Reduction Theory 

Physiological Metrics with a Stress Stimulus 

 A human’s physiological response to stress and stress recovery are fundamental 

factors for research investigating SRT. Skin conductance, or sweat production, is one 

involuntary, physiological, autonomous nervous system metrics that is activated when the 

body needs to act due to a stress-inducing stimulus- fight or flight (Ulrich et al., 1991). 

Once a threat is no longer perceived, the body regulates its response back to baseline 

conditions that conserve energy (Jiang et al., 2014; Ulrich et al., 1991). Several studies 

provided participants with a stress stimulus to see how stress recovery changed 

depending on the environment. Ulrich et al. (1991) found that watching a video of a 

natural setting returned skin conductance to baseline levels after 10 minutes, a recovery 

that was not seen in participants that watched an urban setting video. This study also 

found that natural settings restored cardiovascular and muscle tension responses (Ulrich 

et al., 1991). Jiang et al. (2014) identified that stress recovery in different densities of 

urban trees demonstrated a gender-specific effect with female stress recovery unimpacted 

by tree density while males showed an improved skin conductance response and salivary 

cortisol levels at low and high tree densities. Wang et al. (2016) evaluated the impact of 



12 
 

urban settings after a stress stimulus and determined that viewing a lawn (with or without 

people), a lake or a walkway significantly improved skin conductance response and stress 

recovery when compared to viewing an urban road.  

Physiological Metrics without a Stress Stimulus 

 Many research studies have investigated stress responses to environments without 

the use of stress stimulus, but skin conductance findings are varied depending on the 

research study conditions. Generally, skin conductance decreased, a sign of stress 

reduction, when viewing a natural setting compared to an urban area (Elsadek et al., 

2020; Elsadek et al., 2021); however, Grassini et al. (2022) found that urban 

environments in fact had the lower skin conductance. When in-situ, vegetation in an 

urban setting decreased skin conductance (Elsadek et al., 2019), but did not impact skin 

conductance when comparing an urban and natural environment (Reeves et al., 2019). 

One study examining tree density in urban environments found that skin conductance 

increased regardless of the environment (Wang et al., 2020a).  

Cardiovascular metrics such as heart rate (HR), heart rate variability (HRV), 

blood pressure (BP), and salivary cortisol are also frequently utilized in research as 

metrics of stress response. Significant research indicates that whether viewing or 

immersed in urban greenspace or natural settings, HR and BP decrease when compared to 

a fully urban space (Elsadek et al., 2020; Hassan et al., 2018; Jiang et al., 2019; Lanki et 

al., 2017; Lee et al., 2009; Liu et al., 2021; Park et al., 2010; Song et al., 2014; Sonntag-

Öström et al., 2014; Tsunetsugu et al., 2013). However, Ottosson & Grahn (2005) and 

Gladwell et al. (2012) found no significant change in heart rate or blood pressure when 

experience or viewing nature. HRV indicated that the fight or flight response decreased, 



13 
 

and body relaxation increased when viewing or immersed in natural versus urban settings 

(Elsadek et al., 2019; Elsadek et al., 2021; Lee et al., 2011; Park et al., 2010; Song et al., 

2014; Tsunetsugu et al., 2013) except for Gidlow et al. (2016), which showed no specific 

trend. Salivary cortisol markedly decreased when viewing or immersed in natural settings 

versus urban environments, a sign of lowered stress (Lee et al., 2009; Lee et al., 2011; 

Park et al., 2010; Triguero-Mas et al., 2017); however, in an urban greenspace versus 

urban settings without greenspace, no significant difference between salivary cortisol was 

discovered (Tyrväinen et al., 2014).  

Limited virtual nature studies have been conducted; however, those completed 

have focused on the metrics of skin conductance, HR, and HRV. When compared to a 

control or urban environment, skin conductance and heart rate decreased in virtual natural 

environments (Huang et al., 2020; Valtchanov et al., 2010), a trend that was not identified 

in Yu et al. (2018). However, in studies comparing virtual nature to images or immersion 

in real nature, the findings were diverse. Browning et al. (2020) found that skin 

conductance increased in both outdoor or virtual nature in comparison to a control. Jo et 

al. (2021) found that HRV improved more significantly within virtual forest versus 

images of a forest; however, Knaust et al. (2021) found that HR was unimpacted by 

condition but both virtual and images of nature decreased skin conductance with no 

difference between the two nature conditions. More research examining VR can help 

clarify the physiological impacts of virtual nature.  

Emotion 

 Three surveys are frequently used in research to measure affect or emotion- 

Positive Affect Negative Affect Scale (POMS), Positive Affect Negative Affect Scale 



14 
 

(PANAS), and Zuckerman Inventory of Personal Reactions (ZIPERS). Many studies 

exploring human self-reported mood have determined that viewing or immersion in 

natural environments or urban greenspace versus urban environments increase positive 

affect and decrease negative affect (Elsadek et al., 2019; Elsadek et al., 2020; Elsadek et 

al., 2021; Hartig et al., 1991; Hartig et al., 2003; Reeves et al., 2019; Song et al., 2014). 

Triguero-Mas et al. (2017) determined that negative affect was significantly lower in both 

natural outdoor space with and without water when compared to an urban environment. 

Multiple studies found that viewing or immersion in urban greenspace improved mood 

with and without a stress stimulus (Stigsdotter et al., 2017; Tyrväinen et al., 2014; Ulrich 

et al., 1991). Three studies; however, did not find significant differences in affect when 

viewing fully urban, natural, or urban greenspace (Karmanov & Hamel, 2008; 

Olszewska-Guizzo et al., 2021; Tennessen & Cimprich, 1995). Research with VR often 

found a benefit of virtual nature on positive mood (Browning et al., 2020; Schutte et al., 

2017; Valtchanov et al., 2010), and after a stress stimulus, Yu et al. (2018) found an 

increase in both positive affect and decrease in negative affect when viewing a forest 

versus urban setting.  

 Regarding physiological response, skin conductance is an impactful metric for 

assessing stress recovery in distinct environments after a stressful event. In addition, 

without that stimulus, cardiovascular metrics are significantly affected by green 

environments and decrease HR and BP. Emotion, overall, is impacted by environments 

with virtual and non-virtual greenspace increasing positive emotion, decreasing negative 

emotion or both.  



15 
 

Electroencephalogram- A Psychophysiological Metric Relating to Stress Reduction 

and Attention Restoration Theories 

Images and Videos 

 Ulrich (1991) was a foundational study exploring the impact of natural versus 

urban images on brain activities. The study of 18 participants, utilizing an 

electroencephalogram (EEG) with four electrodes positioned in the central parietal, found 

that alpha was significantly higher when participants viewed vegetation or water in 

comparison to urban images (Ulrich, 1991). This study indicated that natural spaces 

increased a relaxed but awake cognitive state based on an increase in alpha brain activity 

(Ulrich, 1991). Ulrich’s study only investigated alpha power, but these results spurred 

further research with mixed results regarding patterns of brain activities and environment. 

Numerous studies identified significant results associated with alpha power when 

juxtaposing environments. When compared to a control, alpha power brain activity 

increased in the medial prefrontal cortex (Chang et al., 2008) as well as high alpha and 

low theta brain activities in central and occipital areas (Grassini et al., 2022). Elsadek et 

al. (2021) found significantly higher alpha power in the frontal electrode F3 and 

prefrontal electrode AF4 and marginally significant higher alpha power in the frontal 

electrode F4 and prefrontal electrode AF3 when comparing images of nature to built 

environments. Jiang et al. (2019) found a marginally significant increase in high and low 

alpha power when comparing an urban image to a series of natural environments with no 

significant effects on other brain activities (beta, theta, delta, or gamma); however, it is 

unclear from the study the location and number of electrodes used to determine the raw 

EEG data. Window views of a greenspace versus urban setting resulted in marginally 



16 
 

higher alpha power in prefrontal electrode AF3 and occipital electrode O1 and 

significantly higher alpha power in prefrontal electrode AF4 and occipital electrode O2 

(Elsadek et al., 2020). Grassini et al. (2019) reported an increase in low alpha power 

when comparing built and natural environments; however, high alpha, beta, theta, and 

delta all decreased across 64 electrodes. In a study exploring forest types and tree density, 

Chiang et al. (2017) found that interior forest significantly increases alpha frequency in 

the prefrontal region compared to forest edge and forest exterior, a signal that unification 

of vegetation may improve cognitive relaxation. 

Few studies have explored impacts of beta power on natural versus built 

environment or examined solely urban or natural settings. As previously mentioned, 

Grassini et al. (2019) determined a decrease in beta power in natural environments 

compared to urban spaces across all electrodes; however, Olszewska-Guizzo et al. (2018) 

reported no significant impact of “contemplative” versus “non-contemplative” 

environments on frontal beta power. The distinction between these two environmental 

types was difficult to categorize from the provided photographs. Wang et al. (2020a) did 

not find any difference between alpha or beta brain activities when comparing urban 

bamboo forest settings, yet the number and location of electrodes is unclear.  

 The results of these studies demonstrate that substantial work is still needed to 

understand that impacts of natural and built environments on brain activities. In addition, 

a lack of homogenization within the research regarding electrodes of interest or brain 

regions of interest as well as EEG equipment makes identifying brain activity trends 

challenging. More research is also needed to target the effects of urban nature on brain 

activities within brain regions.  



17 
 

Outdoor Immersion 

 Like the indoor EEG studies, outdoor EEG results varied widely between studies. 

Two studies compared the impacts of natural setting only, and the first study found that 

posterior alpha power was significantly lowered during nature exposure as compared to 

pre- or post-exposure (Hopman et al., 2020). The second study exploring solely natural 

exposure found that active forest walking increased alpha whereas Qigong, an exercise 

practice, in the forest decreased alpha and beta in the pre-frontal region (Yi et al., 2021).  

 Two studies contrasted urban and natural environments. Hassan et al. (2018) 

identified a significant increase in both high alpha and high beta at the end of a 15-minute 

walk in a bamboo forest versus urban environment, and similarly Reeves et al. (2019) 

found an increased high beta band activity, with no significant change for low beta, 

alpha, theta, or delta, in the wetland compared to the urban environment. The unexpected 

increase in beta power was qualified as related to attentional requirements or data 

collection challenges associated with movement in an outdoor environment (Reeves et 

al., 2019).  

Several studies examined solely urban settings as environmental conditions. Deng 

et al. (2020) and Herman et al. (2021) found no significant difference between urban 

conditions for alpha, beta, delta, gamma, and theta power. When viewing a green façade, 

participants had significantly higher alpha power in the occipital and frontal lobes 

(Elsadek et al., 2019). Neale et al. (2020), interested in low beta, high beta, and alpha, 

found that low beta increased when walking in a busy urban setting compared to an urban 

quiet or urban greenspace; however, alpha increased in a busy urban setting compared to 

the two other environments, which is counter to the proposed hypothesis that alpha would 



18 
 

in fact decrease. No significant differences were found between the urban greenspace and 

urban quiet setting, emphasizing that the quantity of greenspace may play a role in the 

effect on brain activities (Neale et al., 2020). Several outdoor studies used a specific 

brand of EEG that did not provide raw data, but instead provided “mental states;” 

however, it is unclear how this data was interpreted (Al-barrak et al., 2017; Aspinall et 

al., 2015; Bailey et al., 2018; Lin et al., 2020).  

Studies of outdoor immersion in environments examining EEG are still critically 

needed. Research currently spans a wide range of environmental locations with limited 

repetition of sites. In addition, numerous studies published do not utilize EEG equipment 

that provides raw data for analysis, which makes result interpretation difficult to interpret.  

Virtual Reality 

 VR has been used as a tool in tandem with a variety of EEG studies, often 

associated with rehabilitation (Teo et al., 2016); however, few studies have used VR to 

explore environments and EEG. Gao et al. (2019), focused on alpha activity, found there 

was not a significant difference in alpha power between environments ranging from zero 

to greater than 70 % tree canopy; however, only one electrode was utilized for the study, 

and it is unclear where the electrode was positioned on the forehead. Wang et al. (2020b) 

found that twenty minutes of exercise viewing virtual nature or viewing virtual abstract 

art significantly increased alpha power when comparing the pre-test to the post-test, 

indicating that exercise, not environment, led to increased cognitive relaxation. One study 

exploring the impact of urban greenspace vegetation on structure and degree of greening 

provided inconclusive EEG results (Zhu et al., 2021). Two studies (Hu & Roberts, 2020; 

Rounds et al., 2020) focused on built development and design; however, Hu & Roberts 



19 
 

(2020) provided no raw EEG data as part of the study. Rounds et al. (2020) examined 

theta bands and found that contrasting architectural design elements, including twisting 

buildings, increased posterior parietal theta power, yet posterior parietal theta power 

decreased when buildings contained green facades. More data is needed to understand the 

associations between VR, EEG, and environment.  

Frontal Alpha Asymmetry 

 Frontal alpha asymmetry (FAA) in relationship to emotion has been studied for 

many decades (Allen et al., 2004). Normalized FAA is currently measured by subtracting 

a frontal electrode on the right hemisphere from that same electrode on the left 

hemisphere and dividing that by the sum of that same right and left electrode, which 

provides a value between -1 and 1 (Allen et al. 2004). Several studies and reviews have 

determined that this normalized FAA has been demonstrated association with state and 

trait emotion as well as approach-withdrawal motivation with greater left frontal activity, 

or a positive normalized FAA, indicating approach motivation and greater right frontal 

activity, or negative normalized FAA, indicating withdrawal motivation (Coan & Allen, 

2003; Coan & Allen, 2004; Davidson, 1993; Smith et al., 2017). Although frequently 

associated with positive emotion, it is noteworthy that approach motivation can be 

associated with negative emotions, such as anger (Harmon-Jones, 2003). 

 This indication of emotion and arousal, through normalized FAA, relates to SRT 

and a human’s initial emotional response that informs cognition and behavior; however, 

there is minimal research examining FAA and environment. Olszewska-Guizzo et al. 

(2018) determined no effect of “contemplative” versus “non-contemplative” 

environment, and two additional studies, Olszewska-Guizzo et al. (2020) and Olszewska-



20 
 

Guizzo et al. (2021), inappropriately measured frontal alpha asymmetry. Additional study 

could reveal the connection between environment and approach-withdrawal motivation 

through FAA.  

Landscape Architecture’s Place in Neurourbanism and Environmental Justice 

 Neurourbanism is an evolving discipline that engages a multidisciplinary group of 

researchers from architecture, urban planning, neuroscience, and public health to 

investigate the impacts of urban cities on human mental health and brain function (Adli et 

al., 2017; Interdisciplinary Forum Neurourbanism, n.d.; Ndaguba et al., 2022). In 

addition, this discipline seeks to provide evidence-based design practices to promote 

equity within urban cities (Buttazzoni et al., 2022). This second objective speaks directly 

to the importance of environmental justice in urban spaces, a movement sparked by 

African American communities disproportionally impacted by environment health 

hazards (Bullard, 1983; Bullard & Wright, 1990). Distributive justice, one of four 

components of environmental justice according to Kuehn (2000), states that humans 

should have equal access to assets within communities. One of these assets, identified in 

environmental justice research, is access to high-quality urban greenspace (Jennings et 

al., 2012; Kephart, 2022). Although several analyses of the framework of Neurourbanism 

have mentioned the value of greenspace on mental health and cognition, (Buttazzoni et 

al., 2021; Interdisciplinary Forum Neurourbanism, n.d.; Küçük & Yüceer, 2022), with 

specific reference to ART, landscape architecture is not mentioned as a collaborator in 

the Neurourbanism discipline. As designers of public and private space in urban 

environments, landscape architecture research plays a crucial role in the effort to research 



21 
 

design’s impact on mental health and advocate for equity in access to healthful 

greenspaces for all urban communities.  

Study Framework and Hypothesis 

 Based on the above review, this study seeks to fill the following gaps in the 

literature: 

1. No comparative studies exploring an existing site that is redesigned, 

2. Few studies examining brain frequencies solely in urban settings, 

3. Little data about FAA and nature, and 

4. Limited VR studies. 

Using VR, a comparative study was developed that collected EEG and perceived 

restorativeness data from participants after viewing a vacant and designed site to explore 

restorativeness, cognitive states, and approach-withdrawal motivation within two urban 

stimulus environments. The hypotheses for this study are as follows: 

1. The two stimulus environments will differently impact brain activities and 

perception of restorativeness, 

2. Perceived restorativeness factors will increase when viewing the designed 

versus vacant site, 

3. Alpha frequency will increase, and beta frequency will decrease when 

viewing the designed versus vacant site, and 

4. Approach motivation will increase when viewing the designed versus vacant 

images. 



22 
 

Chapter 3: Methods 

Below is an explanation of the methods that informed the site selection and design 

as well as the methods used to collect data about feelings of perceived restorativeness and 

brain frequencies associated with viewing the vacant and designed site in virtual reality 

(VR).  

Site Location, Inventory, and Analysis 

Site Selection with Community Leader 

 Prior to selecting a site for the study, I met with the President of the Broadway 

East Community & CDC- Dr. Doris Minor-Terrell. I had previously worked on a series 

of greening projects throughout Broadway East and proposed the study idea to Dr. 

Minor-Terrell to ask for her expert opinion on a site 

location within her community advocacy network that 

could benefit from conceptual site design. Dr. Minor-

Terrell recommended the vacant lots addressed 1738-

1752 E. North Avenue in the South Clifton Park 

neighborhood. South Clifton Park is located just south 

of Clifton Park and north of Baltimore City’s Inner 

Harbor and Patterson Park (Figure 2). Dr. Minor- 

Terrell’s justification for this site selection was 

threefold: 

1. The site is in high visibility area, 

2. The site is accessible to two early childhood education institutions, and  

3. The site is located near future planned redevelopment. 

Figure 2: Neighborhood context map 



23 
 

Site Inventory 

 After speaking with Dr. Doris Minor-Terrell to determine the site location, site 

inventory was performed at the neighborhood and site scale. Open-source GIS data of 

Baltimore City, additional online resources, as well as site visits to examine physical, 

biological, and cultural attributes all informed the final site design.  

Neighborhood Inventory 

History 

Baltimore City was founded through town charter in 1729 (Baltimore City 

Department of Planning, 2006). Although, Baltimore had relatively slow growth between 

1730 and the early 1750s, Dr. John Stevenson was one the first settlers to utilize the 

potential of Baltimore’s port through his export of flour to Ireland (Baltimore City 

Department of Planning, 2006). Over the next 100 years, Baltimore’s port was a central 

economic driver for the city and its millwork industry. The port was also a location of 

critical ship building during Revolutionary War and War of 1812 (Baltimore City 

Department of Planning, 2006). 

Beginning in the mid-1800s, Baltimore City took hold as one of the leaders in the 

canning industry, fertilizer industry, and manufacturing of chrome, copper, and steel due 

to technological advances such as steam power (Baltimore City Department of Planning, 

2006). Between 1850 and 1900, Baltimore’s population grew from 169,000 to 508,957 

(Baltimore City Department of Planning, 2006). This population growth required the 

annexation of more land for working class housing, and the 1888 annexation incorporated 

more land to the north and west of the existing boundary, including South Clifton Park 



24 
 

(Figure 3) (Baltimore City Commission for 

Historical and Architectural Preservation, 2002). 

This diverse, residential neighborhood 

thrived in the late 1800s and early 1900s as a prime 

example of rowhome architecture of its time—so 

much so that the majority of South Clifton Park is 

now designated as a historic district along with 

Broadway East (Baltimore City Commission for 

Historical and Architectural Preservation, 2002). In 

1937; however, the federally run Home Owners’ 

Loan Corporation (HOLC) developed a “residential security map” that categorized 

neighborhoods in Baltimore City as first, second, third, and fourth grade (Badger, 2016). 

These grades were designated based on racist and classist policies, and South Clifton 

Park was designated as third grade due to its 

working class and African American 

population (Badger, 2016). The federal 

government no longer supported loans in the 

South Clifton Park neighborhood, leading to 

chronic disinvestment and white flight from 

the neighborhood (Figure 4) (Badger, 2016).  

 

 

 

Figure 3: Growth map of the City of 
Baltimore – the black line is the 1888 
boundary, and the pink dot is South 
Clifton Park (Department of Public 
Works, 1977) 

Figure 4: HOLC map for Baltimore City- the pink 
dot is South Clifton Park (Nelson et al., n.d.) 



25 
 

Today, this 60-acre neighborhood in East Baltimore City is surrounded by seven 

neighborhoods: East Baltimore Midway, Darley Park, Oliver, Broadway East, Berea, 

Clifton Park, and Four-by-Four.  

Watersheds, Topography, and Tree Canopy 

Approximately 58-acres of South Clifton Park lie within the Inner Harbor 

watershed; however, two-acres on the northwest side of the neighborhood are part of the 

Jones Falls watershed. No above ground water sources are present in the South Clifton 

Park neighborhood. The highest point elevation is approximately 176 feet, and the lowest 

elevation is 94 feet located in the northwest and southcentral portions of the 

neighborhood, respectively.  

South Clifton Park contains no City-run public run parks; however, a southern 

entry Clifton Park (a 267-acre city park) adjoins South Clifton Park to the north (The 

Cultural Landscape Foundation, n.d.). The tree canopy coverage in South Clifton Park is 

12%, approximately 7 -acres, compared to City’s tree canopy coverage of 28% (Figure 5) 

(TreeBaltimore, 2007). 

 
Figure 5: Watersheds, topography, and tree canopy in South Clifton Park 



26 
 

Demographics 

The total population of South Clifton Park in 2020 was 567 compared to 

Baltimore City’s population of 585,708 (Baltimore City Department of Planning, 2020). 

Ninety percent of the residents are Black compared to 58% of Baltimore City’s 

population (Baltimore City Department of Planning, 2020). In addition, South Clifton 

Park’s population has decreased by -23% from 2010 to 2020 compared to the -6% 

decrease in overall population in Baltimore City (Figure 6) (Baltimore City Department 

of Planning, 2020). 

Median household 

income is $40,032 

compared to Baltimore 

City’s median income of 

$50,379 (Data USA, 2019; 

U.S. Census Bureau, 2019). 

Within the South Clifton 

Park population, 72% of residents are adults (18 years or older) compared to 83% in 

Baltimore City (Data USA, 2019; U.S. Census Bureau, 2019). Of those adults, 37% have 

completed a high school education and 2% have completed a college education compared 

to 24% and 15% in Baltimore City, respectively (Table 2) (U.S. Census Bureau, 2019). 

Table 2: Baltimore City and South Clifton Park income, age, and education demographics 
Demographics Baltimore City South Clifton Park 
Median Household Income $50,379 $40,032 

Percent of Population Adults (18+) 83% 72% 

Percent of Population Children (<) 17% 28% 

Percent of Population Completed High School 24% 37% 

Percent of Population Completed College 15% 2% 

Figure 6: Racial demographics of Baltimore City and South Clifton 
Park 



27 
 

South Clifton Park has a lower median income and education completion percentage for 

high school and college compared to Baltimore City at-large.  

Regarding transportation, 57% of South Clifton Park residents do not have a have 

vehicle, and 58% of residents utilize public transit to commute and 39% of the population 

use a personal car, truck, or other vehicle to commute (U.S. Census Bureau, 2019). Four 

bus routes run along South Clifton Park- three routes running north/south and one 

running east west within Baltimore City providing opportunities for public transportation 

use for residents in the neighborhood (Figure 7) (Maryland Transit Administration, n.d.).  

 

Figure 7: Bus routes and bus stops of South Clifton Park 

Vacancy 

South Clifton Park has a total of 589 property parcels designed for the 

neighborhood. Of those 589 parcels, there are a total of 243 vacant lots (41% of the total 

number of parcels), 89 vacant building notices (15% of the total number of parcels), and 

six recently rehabbed vacant houses (2% of the total number of parcels) (Figure 8) 

(Baltimore City Department of Housing & Community Development, n.d.). 



28 
 

 

Figure 8: Vacant lots and buildings in South Clifton Park  

Crime 

In the past year, 97 crimes were committed in South Clifton Park. Most crimes are 

concentrated along E. North Avenue. Both violent crimes and non-violent crimes have 

been committed in the neighborhood with most violent crimes being aggravated assault 

and most non-violent crimes being larceny. For comparison, 2,546 crimes were 

committed in Baltimore’s Eastern Police District. Crime in South Clifton Park accounts 

for approximately 4% of the crime in the Eastern District and represents 6% of its 

geographic area (Baltimore Police Department, n.d.)  

Community Assets 

South Clifton Park has a series of community assets throughout the neighborhood. 

Based on site visits and a review of Google Maps, these assets include three faith 

institutions, three schools, one library (temporarily closed), one urban farm, and two 

community gardens. In addition, a future development for the Equality Equation is 

planned on the east side of South Clifton Park- an organization that provides 

disadvantaged community members with workforce development training and access to 

government contracted work opportunities (Figure 9) (The Equality Equation, 2021).  



29 
 

 

Figure 9: Community assets in South Clifton Park 

Site Specific Inventory 

Site Context 

The site, addressed 1738-1752 E. North Avenue, is in the southwestern portion of 

the South Clifton Park neighborhood. The site is 0.22-acres and is adjoined by: 

• An alley and Rising Sun Baptist Church to the north, 

• N. Regester Street and a parking lot to the west, 

• E. North Avenue (Bypass 40) and Gompers School to the southwest,  

• E. North Avenue and rowhomes (vacant and occupied) to the south, 

• E. North Avenue and vacant land to the southeast, 

• McDonogh Street and rowhomes to the east.  

Gompers School is on Maryland’s National Register of Historic Properties. The building 

was built between 1904 and 1906 and serves as a monument to a “progressive” reform in 

Baltimore City’s education system during the early 1900s that elevated the standards of 

public education and the facilities where students were taught (Figure 10) (Department of 

Planning Maryland Historical Trust, 1984).  



30 
 

 

Figure 10: Site context map and site images (original images) 

Hydrology and Topography 

Within the site area, there is about a 10-foot change in elevation from the high 

point in the northwestern portion of the site to the low 

point in the southeastern corner of the site. Most 

water flows to the southeast where there are two 

stormwater drains present- one in E. North Avenue 

and one in McDonogh Street (Figure 11). 

Zoning, Land Use, and Ownership 

Zoning on the site is Residential-8 District, 

which permits community-managed open space gardens, parks, and playgrounds 

(Baltimore City Department of Legislative Reference, 2022). Currently, the site contains 

two vacant building structures (1738-1740 E. North Avenue) and six vacant lots (1742-

1752 E. North Avenue). Five of the parcels are privately owned, two are owned by a 

Figure 11: Site hydrology, topography, 
and surrounding storm drains 



31 
 

corporation, and one is owned by the Baltimore City 

Housing Authority (Baltimore City Department of 

Housing & Community Development, n.d.) (Figure 

12).  

Vegetation and Soil 

Identification of plant material occurred during 

visits to the site and may not include all plant species 

present on site. Most of the site is covered in grassy lawn weeds, including crabgrass, 

Bermuda grass, broad-leafed dock, dandelion, and purple deadnettle. The trees on the 

site’s northern edge are Tree of Heaven, or Ailanthus altissima, a deciduous, invasive 

species throughout much of the United States that is often found in disturbed, urban areas 

(Fryer, 2010) and provide a tree canopy over approximately 20% of the site. In addition, 

several of the trees on site are enwrapped by English Ivy, or Hedera helix. Soil in 

northern and western portions of the site are Hydrologic Soil Group C with slow 

infiltration rates, and soil in the southern and western portions of the site are Hydrologic 

Soil Group D with very slow infiltration rates and 

high runoff potential (United States Department of 

Agriculture Natural Resources Conservation Service, 

n.d.) (Figure 13). 

Site Analysis 

Opportunities and Constraints 

Utilizing the data collected from the inventory, 

site visits and conversations with Dr. Minor-Terrell, the following opportunities and 

constraints were developed for the site: 

Figure 12: Land ownership and parcel 
boundaries 

Figure 13: Soils and tree canopy 



32 
 

Table 3: Site opportunities and constraints 
Opportunities  Constraints 

• Incorporate bioretention to facilitate 
stormwater infiltration 

• Minimal outdoor activation currently 
around the site 

• Utilize topography that exists on site as 
design feature 

• Increase tree canopy within the 
neighborhood 

• Create the first programmed public park 
within the neighborhood 

• Incorporate stakeholder groups near the site 
that could become future stewards/users of 
the park 

• Utilize vegetation as a screen for unwanted 
views 

• Site safety and sightlines 
• Budget for future park maintenance 
• Invasive species 
• Rowhome demolition timeline 
• Private ownership of most of the site 

parcels 
• Substantial topographic change within a 

small site area 
• N. Regester and E. North Avenue are high 

traffic roads 

These opportunities and constraints led to a series of valuable design decisions when 

placed spatially on the site. First, bioretention must be located in the southeastern corner 

of the site to capture stormwater within the site at the lowest elevation. Additionally, 

vegetation screening could be used along the western portion of the site to block views of 

the adjoining parking lot. Also, lighting within the park and sightlines from E. North 

Avenue and N. Regester Street could help improve safety as a site user or law 

enforcement officer. Finally, easy pedestrian accessibility for nearby stakeholders 

requires access from the southwest, northwest, and northeast corners of the site (Figure 

14).  



33 
 

 

Figure 14: Site analysis diagram 

Experimental Data Collection 

IRB Application 

 An application for the study was submitted to the Institutional Review Boards and 

approved after expedited review on August 30th, 2022. An amendment was made to 

change the study location, and that amendment was approved on November 4th, 2022. 

Finally, an amendment was made to allow Ph.D candidate, Kyle Pietro, and Assistant 

Research Professor, Dr. Hyuk Oh, in the Department of Kinesiology to conduct the post-

processing of deidentified EEG data, and that amendment was approved on December 

2nd, 2022 (see Appendix I). 

Participant Recruitment 

 UMD students were recruited to be participants of the study beginning in 

November 2022. A total of 21 participants were recruited using the following methods: 

• Flyering throughout UMD’s Plant Sciences building (see Appendix II), 

• Email blasts to all landscape architecture students, 



34 
 

• Conversations with students and word of mouth, and 

• An announcement in an urban agriculture course. 

Students that were interested in participating in the study reached out via email and were 

provided with additional information about the study and offered dates to sign up for 

study appointment. Initially participants were scheduled within two weeks between 

Tuesday, November 29th, 2022 and Friday December 9th, 2022; however, concerns 

expressed about additional noise in the EEG recording required halting data collection on 

December 1st and December 2nd. Importantly, the concerns regarding additional noise did 

not alter the EEG data collection procedure. The study data collection was extended 

through Thursday, December 15th, 2022 to accommodate study participant rescheduling.  

Digital Modeling and Stereoscopic Images 

 One digital model was built for this study using Rhinoceros 3D (Rhino). The 

model contained layers for the site context, site as it is currently, and site redesigned. The 

model was brought into Lumion 12 for rendering, and four images were captured from all 

sides of the site for both the vacant and designed 

models in the same location using Lumion LiveSync 

for Rhino (Figure 15).  

All images were exported as large format 

(8192x8192), stereoscopic, 360-degree panoramas. To 

create a boundary for the image and discourage 

participants from looking around in the VR headset, the images were all brought into 

Adobe Photoshop, and a black border was placed on the left and right sides of the image. 

Figure 15: Location and view direction 
of experimental images 



35 
 

Figure 16 and Figure 17 exemplify the two-dimensional version of the stereoscopic 

images that participants saw in the experimental study. 

 

Figure 16: Vacant and designed views at Location 1 and Location 2 



36 
 

 

Figure 17: Vacant and designed views at Location 3 and Location 4 

 



37 
 

Data Collection Procedure 

 Data collection occurred in UMD’s Hornbake Library on the ground floor in 

Room 0302E. The study room was arranged so that the participant faced away from the 

door towards the wall, and a sign was posted that read, “Study in progress! Please wait” 

(Figure 18).  

 

Figure 18: Study room and room setup (original images) 

The day before participants were scheduled to complete the study, an email was sent to 

remind them of the date, time, and location of the appointment as well as the approximate 

study length (1.5 hours). Before a participant entered the study room, the eight 

stereoscopic images were loaded into an Oculus Quest 2 VR headset using Pigasus VR 

Media Player, and the ANT Neuro eegoTMsports EEG recording platform was set-up. An 

image matrix was created to ensure that there was counterbalance of the two stimulus 

environments (vacant versus designed) and four image views (Table 4).  

Table 4: Image matrix for the experimental study by participant- V 
indicates images of the vacant site and D indicates designed images 
Sex Image Order Sex Image Order 
Female V1, V2, V3, V4 

D1, D2, D3, D4 

Male V1, V2, V4, V3 

D1, D2, D4, D3 

Female D4, D1, D2, D3 

V4, V1, V2, V3 

Male D2, D4, D1, D3  

V2, V4, V1, V3 



38 
 

Female V2, V1, V4, V3 

D2, D1, D4, D3 

Male V2, V1, V3, V4 

D2, D1, D3, D4 

Female D3, D2, D4, D1 

V3, V2, V4, V1 

Male D1, D4, D3, D2 

V1, V4, V3, V2 

Female V2, V3, V1, V4 

D2, D3, D1, D4 

Male V3, V1, V4, V2 

D3, D1, D4, D2 

Female D2, D3, D4, D1 

V2, V3, V4, V1 

Male D3, D2, D1, D4 

V3. V2, V1, V4 

Female V4, V3, V2, V1 

D4, D3, D2, D1 

Male V4, V3, V1, V2 

D4, D3, D1, D2 

Female D1, D3, D2, D4 

V1, V3, V2, V4 

Male D4, D1, D3, D2 

V4, V1, V3, V2 

Female V1, V4, V2, V3 

D1, D4, D2, D3 Total Male Participants: 8 

Female D4, D2, D3, D1 

V4, V2, V3, V1 
 

Female V3, V4, V1, V2 

D3, D4, D1, D2 
 

Female D3, D4, D2, D1 

V3, V4, V2, V1 
 

 
Total Female Participants: 12 

 

 

One participant initially recruited was removed from the final data analysis because it 

was not possible to collect EEG data due to dreadlocks.  

 An overview of the experimental study design is provided in Figure 19.  

 

Figure 19: Experimental study design conducted for each participant 



39 

Upon entering the study room, participants were given an overview of the study process 

and had an opportunity to ask any questions. Participants were also reminded to try to 

limit any head or body movement when viewing the images; this statement was reiterated 

before each image was shown during data collection. Participants then reviewed and 

completed a Consent Form followed by Questionnaire 1, which was a general 

information form focused on age, sex, and experience with VR (see Appendix III and 

IV). All questionnaires were completed online using an iPad via Qualtrics.  

Participants were next fitted with a 32 channel, ANT Neuro waveguardTMoriginal 

electroencephalogram (EEG) cap based on the circumference of the participant’s head 

(22-24” = Large, 20-22” = Medium, and 18.5-20” = Small). After selecting the size and 

placing on the cap, all 32 electrodes were prepared 

with highly conductive electrolyte gel using a blunt 

needle, and the impedance was checked and re-

gelled to confirm that impedance of electrodes 

were below 25kOhm. Figure 20 shows the location 

of the channels on the EEG cap.  

The VR headset was then fitted on top of 

the EEG cap, and EEG data was continuously recorded at 500 Hz throughout the 

remainder of the study. Annotations within the eegoTM software were utilized to identify 

the start and end of any event that occurred during the recording. Participants were asked 

to take two minutes to settle in and adjust to the set-up before the study commenced.  

To begin, two, one minute baseline conditions were collected- first with eyes open 

viewing a green image displaying the text “Study will begin shortly” and second with 

Figure 20: Map of the 32 channel 
locations on the EEG cap 



40 
 

eyes closed. Then participants were shown a series of four images from a stimulus 

environment (vacant or designed) per Table 4 above. Each image was viewed for one 

minute, and in between each experimental image, a green image displaying the text 

“Please wait” was presented. Participants were asked to confirm that they were viewing 

the “Please wait” image before proceeding to the next experimental image. Once all four 

experimental images were viewed, the VR headset was removed, and participants 

completed Questionnaire 2, which asked about VR side effects and perceived 

restorativeness using Perceived Restorativeness Scale (PRS)-16 from Hartig et al. 

(1997b) (see Appendix V). The VR headset was then re-fitted, and participants completed 

the same process including the acclimation, baseline conditions, and viewing of four 

experimental images that were not seen in the previous data collection phase per Table 4 

above. The VR headset was removed, and participants completed Questionnaire 3, which 

was identical to Questionnaire 2.  

 Finally, the recording was stopped, the EEG cap was removed, a Compensation 

Form was filled out, and participants were given a $20 gift certificate for their time (see 

Appendix VI). After the appointment, a copy of the Consent Form was sent to the email 

address provided in the Compensation Form.  

Raw EEG Data Processing 

Data processing was required to convert the raw EEG data into a usable format 

for statistical analysis. The brain frequencies of interest were broadband alpha, low alpha 

(alpha I), high alpha (alpha II), broadband beta, low beta (beta I), and high beta (beta II). 

EEG data processing was completed by Kyle Pietro, and an account of the EEG 

processing methodology is present in Appendix VII.  



41 
 

Statistical Analysis 

 For the PRS data, a Varimax Rotation factor analysis was conducted separately by 

stimulus environment (vacant and designed) to reduce the 16 PRS questions into common 

factors. The mean of each PRS factor by stimulus environment was determined by 

averaging the PRS score for each of the PRS questions associated with a factor per 

participant. Using SPSS Statistics, an outlier analysis identified outliers and extreme 

outliers within each PRS factor with an extreme outlier is identified as outside of the 

range of 3rd quartile + 3*interquartile range or 1st quartile – 3*interquartile range (Parke, 

2013). Extreme outliers were removed from the analysis, and notably if an outlier was 

found in one of the two stimulus conditions, then it was removed from both conditions 

for consistency. Descriptive statistics including central tendency, dispersion, skewness, 

and kurtosis were determined for each factor in each stimulus environment. In addition, 

paired t-tests were performed to determine if there were statistically significant 

differences between means by factor in the vacant versus designed stimulus 

environments.  

 After processing, EEG data was aggregated into mean power spectral (PS) for 

each frequency of interest per participant by frontal and parietal lobe. Specifically, F3 

and F4 as well as F7 and F8 were aggregated for the frontal lobe, and P3 and P4 as well 

as P7 and P8 were aggregated for the parietal lobe. In addition, normalized frontal alpha 

asymmetry (FAA) was calculated for the electrode pairings F3 and F4 and F7 and F8 

(Allen et al., 2004). An outlier analysis, descriptive statistics, paired t-tests, and a 

repeated measures analysis of variance (ANOVA) with sex as a between-subjects factor 



42 
 

was completed for each brain frequency of interest in the frontal and parietal lobes; 

however, only alpha was examined in the repeated measures ANOVA for FAA.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



43 
 

Chapter 4: Results 

Design Program 

Vision, Goals, and Strategies 

 The vision for the site is that it will be a beloved greenspace in South Clifton Park 

providing opportunities for exploration, restoration, and community building for 

neighboring institutions, residents, and visitors. The design goals of the site respond to 

the existing conditions, opportunities and constraints, components of Attention 

Restoration Theory (ART) and Stress Reduction Theory (SRT), as well as conversations 

with Dr. Minor-Terrell regarding the community’s site goals. Table 5 outlines these goals 

and specific programmatic strategies suggested to attain them.  

Table 5: Site design goals and strategies 
1. Promote the multifunctional use of the space 

• Present a series of flexible spaces throughout the site, specifically for gathering. 
• Locate larger gathering spaces near existing community institutions. 

2. Provide accessibility, order, and safety 
• Incorporate navigable, even pathways that are universally accessible. 
• Maintain sightlines and layer planting appropriately to encourage proposed uses of 

the space.  
• Display path lighting for visibility within the site at night.  

3. Encourage interactions with nature 
• Provide opportunities for structured and unstructured play with natural materials. 
• Offer seating surrounded by plant material.  
• Incorporate educational signage within planting beds.  

4. Increase tree canopy  
• Plant canopy trees on the site. 
• Provide understory canopy cover to create a diversity of canopy experience. 

5. Bolster fascination and complexity 
• Create diverse planting material and natural elements within the site.  
• Encourage views and experience in nature throughout the site.  

 



44 
 

Site Design 

Goal 1- Promote the multifunctional use of space 

 The site plan for this greenspace illustrates the multifunctional use of the space, a 

key goal for the South Clifton Park community (Figure 21).  

 

Figure 21: Site plan of South Clifton Park greenspace 

Flexible spaces included two paved gathering areas, one covered and one uncovered, 

which are located on the southwestern and northwestern edges of the site with the most 

direct access to potential future users of the space – Rising Sun Baptist Church and the 

future Gompers School redevelopment (Figure 22).  



45 
 

 

Figure 22: Image 1- View of the main plaza 

These gathering areas could be used for individuals or a group to host events, like 

concerts or markets, or simply as a place to enjoy an afternoon break. In addition, the 

greenspace offered play space, informal lawn area, bench seating, and walking paths 

throughout.  

Goal 2- Provide accessibility, order, and safety 

 Within the greenspace, all paths were paved and graded to a maximum of a 5% 

slope to allow for ADA accessibility without the need for handrails or stairs (Figure 23).  



46 
 

 

Figure 23: Section A-AI of greenspace grading north to south 

Planting beds were structured to provide clear distinction between the paths and natural 

spaces, a crucial element for preferential landscape noted in Stress Reduction Theory 

(SRT) (Ulrich, 1983). Safety, a priority issue for the community, was addressed with the 

addition of path and overhead lighting throughout the site as well as limited large 

plantings along the southern and southeastern boundaries of the site to enable sightlines 

from E. North Avenue to the south and N. Regester Street to the west. Sightlines into and 

within landscapes were also noted in SRT to help enhance landscape preference (Ulrich, 

1983) (Figure 24).  



47 
 

 

Figure 24: Image 2- View from E. North Avenue 

Goal 3- Encourage interaction with nature 

 Play and exploration were supported throughout the site using interpretive signage 

to engage visitors in the ecology present in this urban setting. Bench seating throughout 

the site was also predominantly placed in areas surrounded by planting to encourage 

passive or unstructured interactions with nature. A formal play area was located on the 

southeastern edge of the site with opportunities to climb, jump, and balance, and a lawn 

connected to the formal play space to support informal discovery or leisure (Figure 25). 



48 
 

 

Figure 25: Image 3- View of formal play area and lawn 

Goal 4- Increase tree canopy  

 The tree canopy coverage in the South Clifton Park neighborhood is 

approximately 16% lower than the coverage in the average of Baltimore City. The design 

for the site added 10 canopy trees and four understory trees, which increased the tree 

canopy coverage from approximately 20% to 40%.  

Goal 5- Bolster fascination and complexity 

 Complexity within environment is wide ranging according to SRT, but for natural 

settings tends to associate with a variety of planting materials and textures (Ulrich, 1983). 

Within the design, a collection of various planting materials with a variation in size, 

texture, and structure created that complexity for the user. Regarding fascination, a key 

component for restorativeness of spaces in Attention Restoration Theory (ART), this 

effortless attention is inherent and described as “soft fascination” in natural space 



49 
 

(Kaplan, 1995). The design offered a variety of views of nature throughout the 

greenspace to facilitate a restorative experience for users (Figure 26).  

 

Figure 26: Image 4- View of vegetated areas and walking paths 

Experimental Results 

Participant Demographics and Previous Virtual Reality Experience 

 A total of 12 females and eight males (N=20) participated in the study ranging in 

age from 18-44 years old with an average age of 25 years old. Seventeen participants 

(85%) had used virtual reality before, specifically 11 females and six males. Five of the 

17 participants (29%) experienced virtual reality symptoms, including nausea, 

disorientation, visual fatigue, and/or discomfort. On a scale from 0 (none at all) to 6 

(completely), the average previous enjoyment for these participants was 4.5.  



50 

Perceived Restorativeness 

Factor Analysis 

The factor analysis for the vacant and designed lots identified three factors based 

on the PRS questions. Question 6 (There is much to explore and discover here) and 

Question 12 (I can do things I like here) were excluded from the vacant site factor 

analysis because the factor loading was less than 0.7. The three factors, Being 

Away/Fascination (6 items), Extent (4 items), and Compatibility (4 items), were found to 

be highly reliable (α=0.941; α=0.915; α=0.929, respectively) (Table 6).  

Table 6: Results of Varimax 
rotation factor analysis for the 
vacant site 

Component 
1 2 3 

Q1 0.825 0.069 -0.039 

Q2 0.787 0.355 -0.056 

Q3 0.815 0.383 0.237 

Q4 0.732 0.496 0.053 

Q5 0.770 0.508 0.114 

Q6 0.661 0.351 0.337 

Q7 0.778 0.485 -0.059 

Q8 -0.144 0.229 0.896 

Q9 0.115 -0.155 0.921 

Q10 -0.015 0.141 0.926 

Q11 0.287 -0.046 0.864 

Q12 0.519 0.596 -0.131 

Q13 0.258 0.912 0.109 

Q14 0.378 0.819 0.055 

Q15 0.360 0.866 -0.246 

Q16 0.458 0.709 0.255 
Extraction Method: Principal Component Analysis 
Rotation Method: Varimax with Kaiser Normalization 
a. Rotation converged in 5 iterations 

Being Away/Fascination is highlighted in red 
Extent is highlighted in blue 
Compatibility is highlighted in green 

Question 2 (Spending time here gives a good break from my day-to-day routine) was 

excluded from the designed site factor analysis because the factor loading was less than 



51 

0.7, and Question 10 (There is a great deal of distraction) was excluded from the 

designed site factor analysis because it was the only question within a factor. The three 

factors, Being Away/Fascination (6 items), Extent (3 items), and Compatibility (5 items), 

were found to be highly reliable (α=0.921; α=0.897; α=0.920, respectively) (Table 7). 

Table 7: Results of Varimax rotation 
factor analysis for the designed site 

Component 
1 2 3 4 

Q1 0.705 0.355 -0.201 -0.264 

Q2 0.586 0.512 0.192 -0.258 

Q3 0.784 0.402 0.076 -0.178 

Q4 0.836 0.396 0.045 -0.047 

Q5 0.820 0.307 -0.080 0.064 

Q6 0.852 0.141 -0.078 -0.111 

Q7 0.793 0.013 -0.311 0.126 

Q8 -0.111 -0.066 0.925 0.034 

Q9 -0.070 0.165 0.888 0.012 

Q10 -0.113 0.013 0.187 0.944 

Q11 -0.048 -0.237 0.862 0.207 

Q12 0.252 0.801 -0.125 -0.019 

Q13 0.051 0.950 -0.088 0.036 

Q14 0.446 0.799 0.049 0.075 

Q15 0.333 0.796 -0.148 -0.048 

Q16 0.435 0.719 0.290 -0.140 
Extraction Method: Principal Component Analysis 
Rotation Method: Varimax with Kaiser Normalization 
a. Rotation converged in 6 iterations 

Being Away/Fascination is highlighted in red 
Extent is highlighted in blue 
Compatibility is highlighted in green 

Outlier Analysis and Descriptive Statistics 

Participant 8 was an extreme outlier for the Being Away/Fascination factor for the 

designed site (Figure 27).  

PR
S 

Sc
or

e 

PR
S 

Sc
or

e 

PR
S 

Sc
or

e 

Extent_V Compatbility_VFascination_V



52 

Figure 27: Box plot showing median, interquartile range, minimum and maximum value, and extreme 
outliers for each perceived restorativeness factor by stimulus environment 

Participant 8 was removed from further analysis of both the vacant and designed site for 

the Being Away/Fascination factor. Descriptive statistics of skewness and kurtosis 

determined that each factor represented normal distribution (Table 8).  

Table 8: Descriptive statistics for the vacant and designed site by factor 

N Range Minimum Maximum Mean 
Std. 

Deviation Variance Skewness 
Excess 

Kurtosis 

Statistic Statistic Statistic Statistic Statistic 
Std. 

Error Statistic Statistic Statistic 
Std. 

Error Statistic 
Std. 

Error 
V_Being Away/ 
Fascination 

19 4.84 0.83 5.67 3.044 0.394 1.718 2.950 0.198 0.524 -1.421 1.014 

D_Being Away/ 
Fascination 

19 1.83 4.17 6.00 5.298 0.123 0.535 0.286 -0.363 0.524 -0.626 1.014 

V_Compatibility 20 4.75 0.00 4.75 2.075 0.344 1.537 2.363 0.597 0.512 -0.945 0.992 
D_Compatibility 20 4.00 2.00 6.00 4.560 0.259 1.156 1.337 -0.561 0.512 -0.551 0.992 
V_Extent 20 4.75 1.25 6.00 4.388 0.333 1.488 2.214 -0.627 0.512 -0.570 0.992 
D_Extent 20 3.67 2.33 6.00 4.617 0.285 1.276 1.629 -0.448 0.512 -1.302 0.992 

Paired T-Test 

Table 9 shows the results of the paired t-test performed to compare the mean 

scores of each perceived restorativeness factor between the two stimulus environments.  

Table 9: Results of the paired t-test showing the difference in perceived restorativeness 
between the two stimulus environments   

95% Confidence 
Interval 

Mean Lower Upper t df 
One-Sided 

p 
Two-Sided 

p 
V_Being Away/ 
Fascination -  
D_Being Away/ 
Fascination 

-2.254 -3.032 -1.476 -6.089 18 <0.001 <0.00001****  

V_Compatibility - 
D_Compatibility 

-2.485 -3.349 -1.621 -6.019 19 <0.001 <0.00001****  

V_Extent – 
D_Extent 

-0.229 -0.681 0.223 -1.061 19 0.151 0.302 

**** p < 0.00001 

PR
S 

Sc
or

e 

PR
S 

Sc
or

e 

PR
S 

Sc
or

e 

Extent_D Compatibility_D
 

Fascination_D
 



53 
 

The mean score of the factor Being Away/Fascination for the vacant lot (M= 3.044, SD= 

1.718) was significantly lower than the mean score of the factor Being Away/Fascination 

for the designed site (M= 5.298, SD= 0.535); t (18) =-6.089, p<0.001. In addition, the 

mean score of the factor Compatibility for the vacant lot (M= 2.075, SD= 1.537) was 

significantly lower than the mean score of the factor Compatibility for the designed site 

(M= 4.560, SD= 1.156); t (19) =-6.019, p<0.001. The mean score of the factor Extent; 

however, did not significantly differ between the vacant (M= 4.388, SD= 1.488) and 

designed site (M= 4.617, SD= 1.276); t (19) =-6.019, p=0.151 (Figure 28).  

  

Figure 28: Mean perceived restorativeness factor score by stimulus environment 

Electroencephalogram Data 

Outlier Analysis and Descriptive Statistics- Frontal Lobe 

Table 10 establishes the extreme outliers of each brain frequency for the mean 

frontal F3 and F4 electrodes and the mean frontal F7 and F8 electrodes.  

Table 10: Summary of extreme outliers of each brain frequency in frontal electrodes 

Mean F3 & F4 

 Alpha Alpha I Alpha II Beta Beta I Beta II 

Vacant 10 10 10 10 10 10 

Designed 10 10 10 10 10 10 

Mean F7 & F8 

 Alpha Alpha I Alpha II Beta Beta I Beta II 

Vacant 10  10 10 10 10 

3.044 5.298 2.075 4.560 4.388 4.617 

Ex
te

nt
 



54 
 

Designed       

 
Extreme outliers identified in either of the stimulus environments were removed from 

both stimulus environments to enable comparisons between the two environments. 

Descriptive statistics of skewness and kurtosis determined that each brain frequency for 

the mean of F3 and F4 electrodes and the mean of F7 and F8 electrodes represented 

normal distribution (Table 11). 

Table 11: Descriptive statistics of brain frequencies for the vacant and designed site in frontal electrodes 

  

N Range Minimum Maximum Mean 

 
Std. 

Deviation Variance Skewness 
Excess 

Kurtosis 

Statistic Statistic Statistic Statistic Statistic 
Std. 

Error 

 
 

Statistic Statistic Statistic 
Std. 

Error Statistic 
Std. 

Error 
V_alpha_F3F4 19 0.0344 0.0044 0.0387 0.0161 0.0019 0.0083 0.0001 1.2152 0.5238 1.9471 1.0143 

D_alpha_F3F4 19 0.0300 0.0049 0.0348 0.0164 0.0019 0.0082 0.0001 0.9327 0.5238 0.2704 1.0143 

V_alpha I_F3F4 19 0.0101 0.0017 0.0118 0.0062 0.0007 0.0029 0.0000 0.4123 0.5238 -0.7500 1.0143 

D_alpha I_F3F4 19 0.0110 0.0019 0.0129 0.0064 0.0007 0.0030 0.0000 0.5600 0.5238 -0.1689 1.0143 

V_alpha II_F3F4 19 0.0236 0.0026 0.0262 0.0094 0.0013 0.0058 0.0000 1.5044 0.5238 2.8095 1.0143 

D_alpha II_F3F4 19 0.0184 0.0027 0.0211 0.0096 0.0013 0.0056 0.0000 0.9853 0.5238 -0.1379 1.0143 

V_beta_F3F4 19 0.2756 0.0054 0.2810 0.0973 0.0201 0.0878 0.0077 0.9121 0.5238 -0.4931 1.0143 

D_beta_F3F4 19 0.3348 0.0065 0.3413 0.1126 0.0250 0.1090 0.0119 1.0916 0.5238 -0.1943 1.0143 

V_beta I_F3F4 19 0.1100 0.0025 0.1125 0.0330 0.0071 0.0310 0.0010 1.3412 0.5238 1.1557 1.0143 

D_beta I_F3F4 19 0.1048 0.0026 0.1074 0.0366 0.0078 0.0339 0.0012 1.0923 0.5238 -0.1786 1.0143 

V_beta II_F3F4 19 0.1806 0.0028 0.1834 0.0635 0.0133 0.0582 0.0034 0.8625 0.5238 -0.6315 1.0143 

D_beta II_F3F4 19 0.2553 0.0038 0.2591 0.0751 0.0175 0.0764 0.0058 1.2049 0.5238 0.3847 1.0143 

V_alpha_F7F8 19 0.0418 0.0139 0.0558 0.0272 0.0030 0.0130 0.0002 1.3278 0.5238 0.9019 1.0143 

D_alpha_F7F8 19 0.0448 0.0085 0.0533 0.0269 0.0028 0.0122 0.0001 0.7860 0.5238 0.3643 1.0143 

V_alpha I_F7F8 20 0.0199 0.0047 0.0246 0.0112 0.0012 0.0055 0.0000 1.1735 0.5121 0.7453 0.9924 

D_alpha I_F7F8 20 0.0205 0.0037 0.0242 0.0111 0.0012 0.0054 0.0000 1.0418 0.5121 1.1172 0.9924 

V_alpha II_F7F8 19 0.0276 0.0064 0.0340 0.0156 0.0018 0.0079 0.0001 1.1776 0.5238 0.8790 1.0143 

D_alpha II_F7F8 19 0.0233 0.0045 0.0278 0.0152 0.0015 0.0067 0.0000 0.4457 0.5238 -0.3544 1.0143 

V_beta_F7F8 19 0.6666 0.0242 0.6909 0.1577 0.0380 0.1658 0.0275 2.1376 0.5238 5.1912 1.0143 

D_beta_F7F8 19 0.4496 0.0202 0.4698 0.1532 0.0300 0.1307 0.0171 1.3356 0.5238 0.8539 1.0143 

V_beta I_F7F8 19 0.1549 0.0121 0.1670 0.0499 0.0101 0.0440 0.0019 1.7215 0.5238 2.3526 1.0143 

D_beta I_F7F8 19 0.1347 0.0082 0.1429 0.0495 0.0089 0.0387 0.0015 1.5162 0.5238 1.5996 1.0143 

V_beta II_F7F8 19 0.5083 0.0120 0.5203 0.1066 0.0287 0.1249 0.0156 2.3522 0.5238 6.2531 1.0143 

D_beta II_F7F8 19 0.3101 0.0119 0.3220 0.1025 0.0215 0.0935 0.0088 1.3289 0.5238 0.7411 1.0143 



55 
 

Paired T-Test and Repeated Measures ANOVA- Frontal Lobe 

 A paired t-test revealed no significant difference between means of the vacant 

versus design brain frequencies in the frontal electrodes (Table 12).  

Table 12: Results of the paired t-test showing the difference in brain frequencies between the two 
stimulus environments in frontal electrodes 

 

 
 

95% Confidence 
Interval     

  Mean Lower Upper t dt 
One-

Sided p 
Two-

Sided p 
V_alpha_F3F4 - 
D_alpha_F3F4 

-0.00034 -0.00377 0.00309 -0.207 18 0.419 0.838 

V_alpha I_F3F4 - 
D_alpha I_F3F4 

-0.00017 -0.00116 0.00083 -0.351 18 0.365 0.730 

V_alpha II_F3F4 - 
D_alpha II_F3F4 

-0.00016 -0.00253 0.00221 -0.144 18 0.443 0.887 

V_beta_F3F4 - 
D_beta_F3F4 

-0.01525 -0.05571 0.02521 -0.792 18 0.219 0.439 

V_beta I_F3F4 - 
D_beta I_F3F4 

-0.00358 -0.01662 0.00945 -0.577 18 0.285 0.571 

V_beta II_F3F4 - 
D_beta II_F3F4 

-0.01158 -0.03919 0.01603 -0.881 18 0.195 0.390 

V_alpha_F7F8 - 
D_alpha_F7F8 

0.00021 -0.00408 0.00450 0.103 18 0.459 0.919 

V_alpha I_F7F8 - 
D_alpha I_F7F8 

0.00007 -0.00143 0.00157 0.096 19 0.462 0.924 

V_alpha II_F7F8 - 
D_alpha II_F7F8 

0.00038 -0.00241 0.00317 0.289 18 0.388 0.776 

V_beta_F7F8 - 
D_beta_F7F8 

0.00448 -0.06926 0.07823 0.128 18 0.450 0.900 

V_beta I_F7F8 - 
D_beta I_F7F8 

0.00038 -0.01625 0.01700 0.047 18 0.481 0.963 

V_beta II_F7F8- 
D_beta II_F7F8 

0.00419 -0.05299 0.06137 0.154 18 0.440 0.879 

 
A repeated measures analysis of variance (ANOVA) was also conducted to explore the 

impact of sex on brain frequencies of participants between the two stimulus 

environments. Table 13 provides the results of the repeated measures ANOVA per brain 

frequency for the mean of F3 and F4 electrodes and the mean of F7 and F8 electrodes.  

 



56 
 

Table 13: Results of the repeated measures ANOVA per brain frequency in the frontal electrodes 

Brain 
Frequency Source 

Type III Sum 
of Squares df 

Mean 
Square F Sig. 

Alpha_F3F4 
  

Stimulus Environment 0.0000006 1       0.0000006  0.021 0.887 

Stimulus Environment x Sex 0.0000010 1       0.0000010  0.037 0.850 

Error (Stimulus Environment) 0.0004540 17       0.0000267      

Alpha I_F3F4 Stimulus Environment 0.0000002 1       0.0000002  0.082 0.778 
 

Stimulus Environment x Sex 0.0000001 1       0.0000001  0.026 0.873 

  Error (Stimulus Environment) 0.0000384 17       0.0000023      

Alpha II_F3F4 Stimulus Environment 0.0000001 1       0.0000001  0.007 0.935 
 

Stimulus Environment x Sex 0.0000005 1       0.0000005  0.041 0.842 

  Error (Stimulus Environment) 0.0002174 17       0.0000128      

Beta_F3F4 Stimulus Environment 0.0009154 1       0.0009154  0.259 0.618 
 

Stimulus Environment x Sex 0.0032887 1       0.0032887  0.930 0.348 

  Error (Stimulus Environment) 0.0601379 17       0.0035375      

Beta I_F3F4 Stimulus Environment 0.0000501 1       0.0000501  0.133 0.720 
 

Stimulus Environment x Sex 0.0001844 1       0.0001844  0.490 0.493 

  Error (Stimulus Environment) 0.0063989 17       0.0003764      

Beta II_F3F4 Stimulus Environment 0.0005298 1       0.0005298  0.326 0.576 
 

Stimulus Environment x Sex 0.0018828 1       0.