Document [original]

Journal of Environmental Psychology 96 (2024) 102308

Available online 24 April 2024

Mobile EEG for neurourbanism research - What could possibly go wrong? A

critical review with guidelines

☆,☆☆

Klaus Gramann

Department of Biological Psychology and Neuroergonomics, Technische Universtat Berlin, Germany

ARTICLE INFO

Handling Editor: L. McCunn

Keywords:

Neurourbanism

Neuro-architecture

EEG

Mobile EEG

Real world neuroscience

ABSTRACT

Based on increasing incidents of mental ill-health associated with living in dense urban environments, the field of

Neurourbanism developed rapidly, aiming at identifying and improving urban factors that impact the health of

city dwellers. Neurourbanism and the closely related field of Neuro-Architecture have seen a surge in studies

using mobile electroencephalography (EEG) to investigate the impact of the built and natural environment on

human brain activity moving from the laboratory into the real world. This trend predominantly arises from the

ready availability of affordable and portable consumer hardware, which not only guarantees operational

simplicity but also frequently incorporates automated data analysis functions. This significantly streamlines the

process of EEG data acquisition, analysis, and interpretation, seemingly challenging the necessity of specialized

expertise in the method of EEG or neurosciences in general. As a consequence, numerous studies in the field of

Neurourbanism have used such off-the-shelf systems in laboratory and real-world experimental protocols

including active movement of participants through the environment. However, the recording and analysis of EEG

data entails numerous requisites, the disregard of which may culminate in errors during data acquisition, pro-

cessing, and subsequent interpretation, potentially compromising the scientific validity of the outcomes. The

often relatively low number of electrodes offered by affordable and portable consumer EEG systems further

restricts specific analyses approaches to the low-dimensional EEG data. Crucially, a large part of Neurourbanism

studies used black-box analyses provided by such consumer systems or incorrectly applied complex data-driven

analyses methods that are incompatible with the recorded low-dimensional data. The current manuscript de-

lineates the prerequisites concerning EEG hardware and analytical methodologies applicable to stationary and

mobile EEG protocols, whether conducted within a controlled laboratory environment or in real-world settings. It

conducts a comprehensive review of EEG studies within the domain of Neurourbanism and Neuro-Architecture,

assessing their adherence to these prerequisites. The findings reveal severe deficiencies in the utilization of

hardware and data processing methods, thereby rendering these studies unsuitable for scientific scrutiny.

Consequently, the present paper provides guidelines for the selection of EEG hardware and analytical strategies

for researchers engaged in mobile EEG recordings, be it within a laboratory or real-world context, aimed at

steering future investigations in the field of Neurourbanism and Neuro-Architecture.

1. Background

Over the last century, a rapidly growing urbanization took place

leading a large part of the world’s population to move to and take

advantage of the resources of urban centers. This trend, however, was

accompanied by increased risks of physical and mental ill-health for city

dwellers (Kennedy & Adolphs, 2011). The adverse effects of

urbanization have spurred the emergence of Neurourbanism as a

research discipline that concentrates on the welfare of urban residents

by identifying and investigating the factors of urban living that influence

their health (Adli et al., 2017). Neurourbanism proposes an interdisci-

plinary approach using a wide variety of methods to gain a better un-

derstanding of the human response to the built environment. The

promise of “neuro” in Neurourbanism is to incorporate neuroscientific

☆

The author would like to thank Lara Kl¨

affling, Isabelle Sander, Anna Wunderlich, Chris Hilton, Zak Djebbara, and two anonymous reviewers for their input and

constructive comments preparing this manuscript.

☆☆

During the preparation of this work the author used ChatGPT in order to edit sentences and to improve

readability of the manuscript. After using this tool/service, the author reviewed and edited the content as needed and takes full responsibility for the content of the

publication.

E-mail address: [email protected].

Contents lists available at ScienceDirect

Journal of Environmental Psychology

journal homepage: www.elsevier.com/locate/jep

https://doi.org/10.1016/j.jenvp.2024.102308

Received 10 January 2024; Received in revised form 20 April 2024; Accepted 22 April 2024

Journal of Environmental Psychology 96 (2024) 102308

measures besides the existing assessment tools to provide more “objec-

tive” insights into the subjective responses to specific aspects of the

urban environment and how these relate to well-being. With the initial

work of Ulrich (Ulrich, 1981) and Eberhard (Eberhard, 2009a; 2009b),

the use of neuroscientific methods to specifically investigate the brain

responses of humans to the built environment has seen a constant in-

crease with electroencephalography (EEG), functional magnetic reso-

nance imaging (fMRI), and functional near-infrared spectroscopy

(fNIRS) as the leading methods in this field (Ancora et al., 2022).

1.1. Neurourbanism from the lab to the real world

Studying the brain response to the built environment in situ, how-

ever, is limited by the restrictions inherent in many established neuro-

imaging methods mainly due to the weight of the systems and their

susceptibility to movement-artifacts (Makeig et al., 2009). Further,

different brain imaging methods assess different aspects of brain activ-

ity. fMRI and fNIRS measure hemodynamic activity (i.e., changes in

blood flow in different brain regions) with high spatial resolution but

relatively slow temporal resolution due to the sluggish nature of blood

flow. In contrast, EEG measures the electrical activity of cortical neurons

with high temporal resolution but relatively low spatial resolution due

to the unknown mixture of volume- and capacitive conducted signals

that mix at the sensor level (Gramann, Jung, et al., 2014; Mehta &

Parasuraman, 2013). The use of MRI is expensive and access often

limited, the systems are heavy, and the recorded signal is susceptible to

movements requiring stationary protocols with participants lying supine

in the scanner not allowed to move at all (Gramann et al., 2011).

Experimental protocols using MRI are thus restricted to watching images

or videos of the built environment while the associated brain dynamics

are recorded (e.g., Kim et al., 2010; Kühn et al., 2021). It is possible

though to combine information about the individual living conditions

and responses to urban or other stressors beyond the immediate

response to urban stimulus material (Dimitrov-Discher et al., 2023;

Spiers & Maguire, 2007). EEG is relatively inexpensive, the systems are

smaller and often portable but the recordings are also susceptible to

motion artifacts. Thus, Neurourbanism studies using EEG as a method to

assess neural responses to the built or natural environment often use

stationary protocols (e.g., Grassini et al., 2019; Mahamane et al., 2020).

These traditional EEG protocols reduce behavioral responses of partic-

ipants to a bare minimum - often button presses at the end of a trial - to

avoid movement-related artifacts from contaminating the feeble signals

of interest (Gramann, Ferris, et al., 2014). Such highly controlled

laboratory-based studies provide the advantage of control over all fac-

tors of interest, which especially for research in urban contexts might

otherwise not be controllable, including traffic, ambient noise, people,

or weather conditions (Vallet & Van Wassenhove, 2023).

However, human beings naturally interact with their built and social

surroundings and experience the world not only from a fixed position

but often by moving through it using different modes of transportation.

Traditional stationary human brain imaging protocols are often criti-

cized as being artificial due to their sensorial, social, and contextual

deprivation due to the restriction of participants’ movements (Sha-

may-Tsoory & Mendelsohn, 2019). It further seems questionable

whether watching 2D representations of the real world while sitting in a

sparsely illuminated lab or lying in a scanner share comparable afford-

ance with the built environment in situ (Djebbara & Kalantari, 2023).

Criticisms frequently arise against laboratory protocols for yielding

ecologically invalid results, particularly when the targeted cognitive or

behavioral processes lack clear definitions (Holleman et al., 2020), and

this might be especially the case for the field of Neurourbanism. In

Neurourbanism research, however, it should be considered that the

environment can be experienced in a variety of stationary (sitting in a

park) or mobile contexts (walking, bicycling, driving a car, public

transportation). All these modes of experience are important and worth

investigating and are likely to impact the user’s experience of the

surrounding environment itself.

To enhance ecological validity and address the limitations associated

with stationary laboratory protocols, researchers have increasingly

explored human brain activity in relation to the built or natural envi-

ronment in actively behaving participants in the laboratory or in real-

world settings. Over the past decade, there has been a notable rise in

publications within the field of Neurourbanism (Ancora et al., 2022),

accompanied by a surge in real-world mobile EEG investigations in

general (Niso et al., 2023). The emergence of low-cost, compact, and

portable EEG systems seemingly capable of recording neural activity

during active movement inside or outside traditional laboratories has

opened new avenues for studying human neural responses to the built or

natural environment, offering the potential to overcome the ecological

validity constraints associated with conventional laboratory research.

The affordable cost, ready availability off-the-shelf, ease of application,

and automated data analysis that is sold with these systems suggest that

researchers from various disciplines can utilize EEG to gain deeper in-

sights into the human response to the environment even without

neuroscience training or specific EEG expertise. As a consequence, the

number of Neurourbanism studies using EEG and mobile EEG in the

laboratory or the real-world increased significantly over the last decade.

1.2. The pitfalls of (mobile) EEG

While the benefit of mobile neuroscientific methods for Neuro-

urbanism research, specifically mobile brain imaging methods seems

obvious, the necessary expertise for recording scientifically valid data in

the lab or the real world, as well as the subsequent analyses and inter-

pretation of complex and high-dimensional data, is often less empha-

sized. Leveraging commercially available EEG systems with modest

costs, assuring easy setup and operational simplicity, along with algo-

rithms facilitating automated analyses of EEG data and classification of

human states, appears to present a facile solution to mitigate the need

for specialized expertise. Nevertheless, publications in the field of

Neurourbanism based on such systems often seem to encounter chal-

lenges in conducting methodologically sound and replicable neurosci-

entific studies. Incorrect application of the methods and assumptions of

what the system can and cannot do in combination with inadequate or

even black box data analysis approaches can lead to drastic misinter-

pretation of the outcome. With an increasing number of predatory

journals that favor profit over scientific quality, such results will still be

published and can have lasting effects on the field.

The present review thus addresses critical issues in the new and fast-

growing field of Neurourbanism studies that use EEG or mobile EEG

protocols providing an overview of the method and analyses. In the end,

guidelines will be provided for EEG protocols in general and specifically

for mobile EEG protocols in Neurourbanism research as these require

additional considerations regarding hardware and analyses approaches.

In the subsequent sections, the primary challenges associated with

employing EEG recordings in the laboratory or real-world experiments

are delineated. Commencing with a brief overview of the origin of the

EEG signal, an examination of crucial considerations regarding the

recording hardware and the analytical approaches follows. The focus

will be on two main aspects that are critical for EEG studies in general

and specifically for real-world neuroscientific studies using mobile EEG:

i) The recording equipment, encompassing amplifier specifications, and

details on the quantity, type, and application of electrodes, and ii) EEG

analyses, including data preprocessing and downstream extraction of

brain electrical features. These aspects are evaluated based on publica-

tion guidelines and recommendations for EEG studies (Keil et al., 2014).

After describing the methodological foundations regarding the EEG

recording technology and analysis approaches, publications in the

domain of Neurourbanism using EEG will be analyzed with respect to

the described methodological issues with a special focus on EEG studies

conducted in the real world or using mobile EEG methods. Summarizing

the methodological requirements and the results from the literature

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

review, guidelines will be provided allowing to narrow down equipment

options and analytical protocols for Neurourbanism studies with more

ecologically valid protocols.

1.3. EEG origins, volume conduction, and capacitive conduction

The EEG signal mainly originates from the synchronized electrical

activity of thousands to millions of pyramidal neurons located in the

human cortex with a perpendicular orientation of these neurons to the

surface of the cortex (Lopes da Silva, 2013). The EEG signal is based on

postsynaptic activity in larger populations of these pyramidal neurons

resulting in the summed activity of excitatory (EPSPs) or inhibitory

(IPSPs) postsynaptic potentials. The specific architecture and the syn-

chronous activation of these neurons produce potentials that are strong

enough to be sensed outside the brain volume. The movement of

charged particles throughout the brain volume is known as volume

conduction. Beyond the brain volume, the potentials extend through

capacitive conductance across the protective layers of the meninges (the

protective membranes surrounding the brain), skull, and skin to reach

the electrodes (Jackson & Bolger, 2014; Nunez & Srinivasan, 2006). At

the electrode level, the signals are then conducted through electrolyte to

the conductive layer of the electrode, where they are detected and

transmitted to the amplifier for amplification of the miniscule signal.

The recorded EEG data typically consist not only of brain activity.

Brain signals are characterized by their small amplitudes, typically

measured in microvolts (one-millionth of a volt), that have to pass

through the less conductive layers of the human meninges, skull, and

skin, undergoing spatial dispersion and filtering of high-frequency

components. Additional biosignals such as eye and muscle move-

ments, which produce signals with much larger amplitudes, will mix

with the signals originating in the brain. These physiological non-brain

signals do not encounter the same degree of distortion as brain signals

because they don’t traverse the less conductive layers of the meninges

and skull. Crucially, participants’ movements can introduce systematic

activation patterns into the recorded signal. These must be identified as

movement-related activity that mix with the brain dynamics of interest.

As such, experimental conditions with and without movement might

systematically differ regarding physiological non-brain activity. In

addition to physiological non-brain signals which can be informative

with respect to participants behavioral and psychological state, elec-

tronic and mechanical artifacts contribute to the recorded signal. Elec-

tronic devices are ubiquitous in the environment introducing

interference into the recordings that might lack both spatial stability

with respect to the sensors and consistency in their electromagnetic

spectrum. Finally, the weight of the amplifier and electrode cables can

lead to movement of the system or parts of the system that introduces

mechanical artifacts to the recording.

Ecologically more valid protocols allowing for active behaviors in

the real world will likely include all of the above described physiological

and non-physiological contributions. Active movements of the eyes and

muscles, potential system and cap displacements due to head move-

ments of the participants, cable sway during walking as well as external

electronic devices in the environment can lead to a different signal being

recorded compared to traditional EEG protocols in the laboratory. It is

thus imperative that the activity from non-brain sources including bio-

logically relevant activity like eye movement and muscle contraction as

well as mechanical and electrical artifacts are dissociated from the brain

signal of interest during data analysis. This can be achieved, to a

reasonable degree, by good-quality EEG equipment and adequate

analytical approaches.

2. EEG recording equipment

2.1. Amplifiers specifications

Over the last decade, a wide range of mobile and wireless EEG

amplifiers became available on the market (for an excellent overview of

some recent systems see Niso et al., 2023). For Neurourbanism re-

searchers focused on capturing EEG data in the real world, crucial

amplifier aspects include the amplifier resolution in bits, the sampling

rate in Hertz, and effective common mode rejection (CMR). Addition-

ally, in mobile EEG systems, factors such as the system weight, wireless

protocols, and integrated motion sensors are important. Bateson et al.

(2017) provide a categorization scheme for EEG amplifier technologies

describing the amplifier mobility in a range from 0 to 5 describing

systems with no mobility that are off-body mounted with participant

tethered via cabling to the EEG equipment (mobility score of 0) up to

systems with a head-mounted amplifier that includes acquisition, stor-

age, and analysis equipment within the headset (mobility score of 5) (see

also Table 1).

Most commercially available systems offer amplifiers with a resolu-

tion ranging from 12 to 24-bit, enabling digitization of a broad signal

range of the recorded analog brain activity. Higher resolutions are

preferable since they can resolve the signal range in more detail. Simi-

larly, for the temporal representation of the signal, higher sampling rates

allowing better temporal resolution of the analog signal. As per the

Nyquist-Shannon sampling theorem, the sample rate sets the limit for

the highest representable frequency in a digitized signal, which is half

the sampling frequency. If the frequencies of interest in the recorded

EEG signal for a specific application are below 64 Hz, a sampling fre-

quency of 128 Hz would prove sufficient and most commercially

available systems typically enable digitization of the analog signal at

128 Hz or higher. The effective CMR, in contrast, varies drastically be-

tween available systems ranging from 75 dB to 140 dB. The CMR in an

EEG amplifier refers to the amplifier’s ability to minimize signals that

are present in both the active and reference electrodes, which are typi-

cally noise or interference. CMR thus helps to eliminate unwanted

background signals and ensures that the EEG amplifier primarily am-

plifies the brain’s electrical activity while rejecting common noise

sources. For EEG applications, a CMR of 80 dB or higher is often

considered good, but the specific requirements can depend on the nature

of the recording environment. Because real-world environments come

with uncontrollable and often changing ambient electrical noise, higher

CMR values are desirable for a mobile EEG amplifier. In terms of weight,

the majority of available mobile EEG systems are compact and light-

weight, enabling extended data recording sessions without causing

participant discomfort, because they offer participants greater freedom

of movement between tasks. These systems typically employ various

industrial wireless protocols for transferring data to mobile devices.

Additionally, some systems provide the option to store recorded data

locally, necessitating subsequent offline synchronization with other

potential data sources. Furthermore, the incorporation of motion sensors

into the amplifier systems offers researchers the ability to utilize motion-

related information for signal quality control and, in some cases, to

identify and remove artifacts associated with movement (e.g., Gwin

et al., 2010).

2.2. Quantity, types, and placement of electrodes

Frequently overlooked but equally significant as amplifier specifi-

cations are the considerations surrounding the quantity and types of

electrodes employed in mobile EEG recordings. The number of elec-

trodes can vary substantially among different systems. In cases where

the primary objective is to reliably extract specific EEG features that are

known a priori, the suitable count and positioning of electrodes are

determined by the minimum requirements for such extraction. Howev-

er, in the realm of basic research, particularly within emerging disci-

plines that often explore novel phenomena previously uninvestigated

like in the case of Neurourbanism, our understanding of how the human

brain responds to experimental manipulations may be limited. This

limitation extends to replicating EEG features established under

controlled lab conditions, which might undergo alterations due to

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

Table 1

Publications from 1945 to November 2023 using EEG or mobile EEG in Neurourbanism studies. Participant and system mobility scores according to Bateson et al.

(2017). Participant mobility values indicate 0 =lying, sitting, or standing still; 1 =lying, sitting, or standing with localized movement; 2 =constrained walking/-

cycling; 3 =unconstrained walking/cycling; 4 =walking and carrying, climbing stairs, and constrained running; 5 =unconstrained running and vigorous physical

exercise or sport. System mobility values indicate 0 =all equipment off-body mounted and participant tethered via cabling to EEG equipment; 1 =waist-mounted (or

back-mounted) with additional equipment located in a rucksack; 2 =all equipment waist-mounted; 3 =head-mounted EEG system, with additional equipment located

in a rucksack or off-body tethering participant via limited-range wireless link; 4 =head-mounted requiring smartphone/tablet; 5 =head-mounted with acquisition,

storage, and analysis equipment integrated within the headset. Replicability scores ++ = replicable; + = partially replicable; – =not replicable.

study

Authors Year Publisher Participant

Mobility

Score

System

Mobility

Score

#Electrodes System Data

Preprocessing

Approach

Replicability Parameter

Space

1 Al-barrak et al. 2017 PLoS One – PloS 1 4 1 NeuroSky Black Box – Frequency

domain

2 Allahverdy &

Jafari

2016 Iran J Public

Health – School of

Public Health Iran

1 0 16 G-tec Manual – Frequency

domain

3 Asim et al. 2023 Building and

Environment -

Elsevier

0 4 4 Muse ICA – Frequency

domain

4 Aspinall et al. 2015 Br J Sports Med –

BMJ Group

3 4 14 Emotiv Black Box – Frequency

domain

5 Banaei et al. 2017 Frontiers Hum

Neuroscience –

Frontiers

3 1 128 Brain Products ICA ++ Frequency

domain,

Cluster

level

6 Baumann &

Brooks-Cederqvist

2023 Heliyon – Cell

Press

0 4 4 Muse Manual +Frequency

domain

7 Chen et al. 2016 PeerJ – PeerJ 1 4 14 Emotiv Automated +Frequency

domain

8 Cho et al. 2022 J Urban Health –

Springer

2 4 14 Emotiv Black Box – Frequency

domain

9 Choi et al. 2016 Complementary

Therapies in

Medicine – Elsevier

1 0 4 Biopac Black Box – Frequency

domain

10 Ding et al. 2022 PLoS One – PloS 1 4 1 NeuroSky Black Box – Frequency

domain

11 Djebbara et al. 2019 Proc Natl Acad Sci

USA – Natl Acad

Sci USA

3 1 64 ANT ICA ++ Time

domain ERP

12 Djebbara et al. 2021 Sci Rep – Nature

Springer

3 1 64 ANT ICA ++ Frequency

domain

13 Ducao et al. 2020 Int J Com WB –

Springer

3 4 1 NeuroSky Black Box – Frequency

domain

14 Elsadek et al. 2021 Health Environ Res

Design J – Sage

Publishing

1 4 14 Emotiv Automated +Frequency

domain

15 Ergan et al. 2019 J Comput Civ Eng –

American Society

of Civil Engineers

1 4 14 Emotiv Manual – Frequency

domain

16 Erkan 2018 Architectural

Science Review –

Taylor & Francis

1 4 1 NeuroSky Manual – Frequency

domain

17 Erkan 2023 Open House

International –

Emerald

Publishing

3 3 19 Not named NA – Frequency

domain

18 Gao et al. 2019 Int J Environ Res

Pub Health – MDPI

1 4 1 NeuroSky Black Box – Frequency

domain

19 Grassini et al. 2022 Frontiers Psychol –

Frontiers

1 0 64 Bittium ICA ++ Frequency

domain

20 Grassini et al. 2019 J Environ Psychol –

Elsevier

1 0 64 Bittium ICA ++ Frequency

domain,

Time

domain

21 Grima Murcia et al. 2019 Integrated

Computer-Aided

Engineering – IOS

Press

1 0 64 NeuroScan Automated +Time

domain,

Source level

22 Hagerhall et al. 2015 Nonlinear

Dynamics Psychol

Life Sci – EBSCO

Information

Services

1 0 6 Cephalon Manual – Frequency

domain

23 Hassan et al. 2018 Evidence-Bas

Cemplement

Alternat Medicine

– Hindawi

2 4 1 NeuroSky Black Box – Frequency

domain

(continued on next page)

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

Table 1 (continued)

study

Authors Year Publisher Participant

Mobility

Score

System

Mobility

Score

#Electrodes System Data

Preprocessing

Approach

Replicability Parameter

Space

24 Herman et al. 2021 Sustainability 1 4 4 Muse Black Box +Frequency

domain

25 Higuera-Trujillo et

al.

2020 Building Research

& Information –

Taylor & Francis

1 4 9 Advanced

Brain

Monitoring

ICA +Frequency

domain

26 Hollander & Foster 2016 Architectural

Science Review –

Taylor & Francis

3 4 1 NeuroSky Black Box – Frequency

domain

27 Hu & Roberts 2020 Urban Science –

MDPI

1 4 5 Emotiv Black Box – Frequency

domain

28 Hu et al. 2021 Sci Rep – Nature

Springer

1 0 96 Brain Products Regression +Frequency

domain,

Time

domain

29 Imperatori et al. 2023 Frontiers Psychol –

Frontiers

1 0 31 Micromed ICA ++ Network,

Source level

30 Jiang et al. 2020 Indoor and Built

Environ – Sage

Publishing

1 4 1 NeuroSky Black Box – Frequency

domain

31 Jung et al. 2023 J Environ Psychol –

Elsevier

1 4 14 Emotiv ICA – Frequency

domain

32 Karandinou &

Turner

2017 Int J of Parallel,

Emergent and

Distributed

Systems – Taylor &

Francis

3 4 14 Emotiv Black Box – Frequency

domain

33 Li et al. 2021 Building and

Environment -

Elsevier

1 4 NA Emotiv ICA +Frequency

domain

34 Li et al. 2020 Energy and

Buildings –

Elsevier

1 4 NA Not named NA – Frequency

domain

35 Li et al. 2023 Forests – MDPI 2 4 8 Kingfar Black Box – Frequency

domain

36 Li et al. 2021 Int J Environ Res

Pub Health – MDPI

1 0 2 Biopac Automated +Frequency

domain

37 Li et al. 2021 Urban Forestry

Urban Green –

Emeral Publishing

Ltd

1 4 9 Advanced

Brain

Monitoring

Black Box – Frequency

domain

38 Lin et al. 2020 J Urban Health –

Springer

2 4 14 Emotiv Black Box – Frequency

domain

39 Mahamane et al. 2020 Frontiers in

Psychol – Frontiers

1 4 14 Emotiv ICA +Frequency

domain,

Time

domain

40 Manohare et al. 2023b Applied Acoustics

– Elsevier

1 4 14 Emotiv Black Box – Frequency

domain

41 Manohare et al. 2023a Noise and Health –

Wolters Kluwer

1 4 14 Emotiv ICA +Frequency

domain

42 Mavros et al. 2022 Sci Rep – Nature

Springer

2 4 24 MBrainTrain ICA ++ Frequency

domain

43 Mostafavi et al. 2023 J Building

Engineering –

Elsevier

1 4 22 MBrainTrain ICA ++ Frequency

domain

44 Naghibi Rad et al. 2019 Front Behav

Neurosci –

Frontiers

1 0 32 ANT Automated +Time

domain

45 Neale et al. 2020 Cities & Health –

Taylor & Francis

3 4 14 Emotiv ICA +Frequency

domain

46 Neale et al. 2017 J Urban Health –

Springer

3 4 14 Emotiv Black Box – Frequency

domain

47 Nie et al. 2022 Landsc. Archit.

Front. – Higher

Education Press

1 4 7 Emotiv ICA – Frequency

domain

48 Olszewska-Guizzo

et al.

2022 Frontiers in

Psychiatry –

Frontiers

1 0 16 Brain Products ICA +Frequency

domain

49 Olszewska-Guizzo

et al.

2018 Frontiers

Psychiatry –

Frontiers

1 4 8 Neuroelectrics Manual +Frequency

domain

50 Olszewska-Guizzo

et al.

2018 Int J Environ Res

Pub Health – MDPI

1 4 14 Emotiv ICA ++ Frequency

domain

(continued on next page)

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

participants’ movements during mobile EEG recordings. Consequently,

in nascent research domains, the parameters of interest, their spatial

distribution across the scalp, and their temporal dynamics are largely

unknown or have not yet been adequately replicated to attain validation

as reliable parameters. In such circumstances a higher density of the

montage with a greater number of electrodes uniformly distributed

across the entire scalp becomes indispensable. This approach facilitates

the extraction of potential EEG features that co-vary with the experi-

mental manipulation. Importantly, it further permits various analytical

methods to distinguish brain activity from non-brain activity. Thus,

more electrodes with a uniform distribution across the scalp are better

even though they require longer preparation times and are more

cumbersome for participants to carry.

Besides the density of the electrode montage, the kind of electrodes

used can have a major impact on the recorded signal quality. In Neu-

rourbanism research, commonly employed EEG electrode types include

wet electrodes, sponge electrodes, or dry electrodes. Only in Wet elec-

trodes are coated with conductive materials, typically Silver/Silver

Chloride (Ag/AgCl), utilizing hypertonic electrolyte to build a bridge

between the skin and the sensor coating. Sponge electrodes are also

usually Ag/AgCl or gold-coated electrodes embedded in sponges satu-

rated with saline water that provides good contact with and serving as

Table 1 (continued)

study

Authors Year Publisher Participant

Mobility

Score

System

Mobility

Score

#Electrodes System Data

Preprocessing

Approach

Replicability Parameter

Space

51 Olszewska-Guizzo

et al.

2020 Int J Environ Res

Pub Health – MDPI

1 0 16 Brain Products ICA +Frequency

domain

52 Olszewska-Guizzo

et al.

2021 J Environ Psychol –

Elsevier

1 0 16 Brain Products ICA +Frequency

domain

53 Olszewska-Guizzo

et al.

2022 Sci Rep – Nature

Springer

1 0 16 Brain Products ICA +Frequency

domain

54 Qi et al. 2022 J Environ Psychol –

Elsevier

1 4 1 NeuroSky Black Box – Frequency

domain

55 Qin et al. 2013 Urban Forestry

Urban Green –

Emeral Publishing

Ltd

1 0 2 ADInstruments Black Box – NA

56 Reece et al. 2022 Int J Environ Res

Pub Health – MDPI

1 0 32 Brain Products ICA +Frequency

domain

57 Reeves et al. 2019 Front Psychol –

Frontiers

1 4 14 Emotiv ICA +Frequency

domain

58 Roe et al. 2013 Environ Sci –

Hikari Ltd

1 4 12 Emotiv Black Box – Frequency

domain

59 Rounds et al. 2020 Frontiers Hum

Neuroscience –

Frontiers

1 0 64 Brain Products ICA – Frequency

domain

60 Shan et al. 2019 Energy and

Buildings –

Elsevier

1 4 14 Emotiv ICA – Frequency

domain

61 Shemesh et al. 2021 Architectural

Science Review –

Taylor & Francis

1 4 5 Emotiv Black Box – Frequency

domain

62 Shemesh et al. 2017 Architectural

Science Review –

Taylor & Francis

1 4 14 Emotiv Black Box – NA

63 Shemesh et al. 2022 J Environ Psychol –

Elsevier

1 4 5 Emotiv Black Box – Frequency

domain

64 Song et al. 2022 Front Psychol –

Frontiers

1 4 1 NeuroSky Black Box – Frequency

domain

65 Tilley et al. 2017 Int J Environ Res

Pub Health – MDPI

3 4 14 Emotiv Black Box – Frequency

domain

66 Ulrich 1981 Environ Behav –

Sage Publishing

1 0 2 Not named Black Box – Frequency

domain

67 Vecchiato et al. 2015 Cogn Process –

Springer

1 1 19 EB Neuro ICA +Frequency

domain

68 Vecchiato et al. 2015 Frontiers Psychol –

Frontiers

1 1 19 EB Neuro ICA ++ Frequency

domain

69 Vijayan & Embi 2019 Int J Built Environ

Sustain – Elsevier

1 4 5 Emotiv Black Box – Frequency

domain

70 Wang et al. 2021 Int J Environ Res

Pub Health – MDPI

1 4 1 NeuroSky Black Box – Frequency

domain

71 Xiong et al. 2023 Urban Forestry

Urban Green –

Emeral Publishing

Ltd

1 4 14 Emotiv ICA +Frequency

domain

72 Yang et al. 2011 Int J Environ Res

Pub Health – MDPI

1 3 16 Siga Medical Black Box – Frequency

domain

73 Zhang et al. 2021 Applied Sciences –

MDPI

1 4 14 Emotiv ICA – Frequency

domain

74 Zhang et al. 2021 Int J Environ Res

Pub Health – MDPI

1 4 14 Emotiv Black Box – Frequency

domain

75 Zou et al. 2021 J Comput Civ Eng –

American Society

of Civil Engineers

1 4 14 Emotiv ICA +Frequency

domain

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

the conductive medium between skin and sensor. Finally, dry electrodes

require direct contact with the skin beneath the electrode and do not use

electrolyte or other fluids to improve conductivity (Lun-De, LinMcDo-

well, Wickenden, Gramann, JungKo, & Chang, 2012; Niso et al., 2023).

Electrodes can further be passive or active (some form of built-in

amplification at the electrode level) and their cables can be shielded

or non-shielded. However, none of these latter technical specifications

seem to lead to differences in data quality when medical-grade wet

electrodes are used with appropriate preparation of the recording sites

(Scanlon et al., 2021). In general, all sensor types can provide

high-quality data in stationary protocols with appropriate preparation of

the skin and no active movement of participants. Wet electrodes need

longer preparation times and participants’ hair becomes disheveled due

to the application of electrolyte or saline water. Additionally, over time,

the conductive medium can dry out, resulting in a decline in the quality

of the recorded signals. In contrast, dry electrodes generally require less

preparation time and involve fewer cleaning demands when compared

to wet electrodes. However, dry electrodes tend to exhibit noticeable

increases in single-trial and average noise levels, particularly when

impedance levels are higher (Chi et al., 2010) and they more often lead

to headaches due to the more extensive pressure that is exerted to

establish a good skin contact (Mathewson et al., 2017; Zander et al.,

2017). Dry electrodes are sometimes also restricted to areas of the head

that have less hair to allow a very good and stable connection to the skin

and are specifically susceptible to movement of the sensor surface over

the skin since no flexible bridge in form of electrolyte is used.

Another important aspect for mobile EEG recordings is the applica-

tion and placement of electrodes. Electrodes are typically incorporated

into elastic caps that provide predetermined positions. These positions

are standardized across various head sizes, offering a nomenclature for

describing activation patterns or effects in a topographic fashion (Oos-

tenveld & Praamstra, 2001). Some commercial systems provide alter-

native contraptions to hold the electrodes in place, including headbands

or flexible arms that can be moved to different sites. The cables con-

necting the electrode to the amplifier can be loosely arranged or directly

incorporated into the cap design prohibiting cable sway. During active

behaviors of participants, the way electrodes and cables are fixed is

important as specific movements like head turning or walking can lead

to movement of the cap, the amplifier, or the electrode cables that then

lead to movement-related artifacts in the recorded signal (Gorjan et al.,

2022; Gwin et al., 2010; Wunderlich & Gramann, 2021). Rigidity of the

electrode holders will lead to movement of the electrodes with

increasing acceleration in movements leading to stronger

micro-movement of electrodes over the skin that can lead to impedance

changes associated with drastic signal distortions. Likewise, the longer

and looser the cables are, the more participants’ movement will lead to

cable sway which itself can impact the recorded data. In addition, strong

cable sway can lead to micro-movement of the electrode itself and lead

to the above-described impedance changes introducing noise to the

signal. This is especially the case for dry electrodes that are highly

susceptible to artifacts during active behaviors of participants.

3. EEG analysis approaches

3.1. Data preprocessing

Beyond the selection of the equipment, the multitude of data analysis

methods presents a broad spectrum of analytical possibilities. However,

this diversity also introduces a heightened risk of making erroneous

decisions, and it’s worth noting that some of these decisions are closely

connected to the characteristics of the chosen equipment. In general,

due to the volume and capacitive conducted nature of the EEG signal, a

varying number of brain and non-brain activities as well as mechanical

and electrical artifacts will be recorded at any given moment. This

strongly argues for the processing of the recorded data before the

extraction of features to avoid interpreting EEG parameters as brain

activity when in fact the activity does not or only partially originated

from the brain. Especially for mobile EEG recordings that come with

increased eye movement, neck, and facial muscle activity as well as

potential movement-related mechanical artifacts, an objective and

replicable preprocessing of the data is important. Commercial systems

often provide only a low number of electrodes and biased distribution of

electrode locations together with proprietary algorithms that were

validated in stationary protocols and that are unlikely generalizable to

mobile settings. When proprietary analysis algorithms are employed, the

preprocessing algorithms and the specifications integral to feature

extraction cannot undergo critical evaluation, and the results become

non-replicable. Hence, results from black-box analysis methods lacking

transparency in algorithmic details and the absence of statistics

regarding cleaned or interpolated data should be excluded from scien-

tific consideration.

EEG cleaning approaches often focus on the most prominent physi-

ological non-brain contributions to the recorded EEG signal, i.e. eye

movement and muscle activity. Usually, several preprocessing steps are

applied, including filtering of the signal to remove slow drifts or high-

frequency activity that is characteristic of muscle activity. Often, addi-

tional time domain cleaning is applied to remove muscle or eye

movement-related activity. This can be done in an automated fashion

according to predefined criteria or by visual inspection. More specific

artifact rejection approaches allow for regressing out eye movement

activity (Croft & Barry, 2000) or using spatial filters to filter different

non-brain sources from the recorded data without sacrificing samples

(Jiang et al., 2019). Regarding data cleaning in mobile EEG recordings, a

large number of Neurourbanism studies use Independent Component

Analysis (ICA; Makeig et al., 1995), a blind source separation method

that allows for the removal of artifactual activity from the recorded

signal (Jung et al., 2000). ICA decomposes the data into a matrix of

statistically independent source time series with a weight matrix for

each resultant independent component (IC) indicating how much each

IC contributes to each channel. Using the IC activity time course and

spatial distribution allows for differentiating different sources of activity

like brain, muscles and eye movement activity. ICs can then be removed

from the unmixing matrices and the thus “artifact-cleaned” data can be

further processed. While ICA developed into one of the most widely used

blind source separation methods for the analyses of EEG data (Delorme

& Makeig, 2004), the application of ICA to EEG data requires certain

assumptions to be met. Some of the theoretical assumptions include that

the extracted processes are stationary, meaning that the statistical

properties of the sources and the mixing matrix remain constant over

time. Another significant assumption within the ICA model, which bears

practical implications for data decomposition, pertains to the necessity

of prior knowledge concerning the quantity of underlying sources. In

particular, the majority of ICA algorithms partition the data into a

quantity of independent components (ICs) equivalent to the number of

electrodes, leading to, for instance, 64 ICs in the context of a mobile EEG

recording featuring 64 electrodes. Applying ICA to a dataset with 14

electrodes yields only 14 sources to account for the recorded data. If

more than 14 sources were active during the recording, the resulting ICs

would mix several activation patterns across multiple ICs to elucidate

the signal. Generally, ICA is a viable analysis method that has shown to

be effective for mobile EEG recordings. However, like any analytical

approach, it relies on various model assumptions that should be un-

derstood. (e.g., for filtering of EEG data see Widmann, Schr¨

oger, &

Maess, 2015) and it is not evident yet whether ICA, as a method for

reducing artifacts, diminishes artifacts equally across conditions that

contain different kinds or intensities of movement. There might be

consistent variations in non-brain activity present in different parts of

the experiment that render it challenging to attribute differences in the

recorded activity between experimental conditions solely to a neural

source.

Revisiting the physiological underpinnings of the EEG signal, the

significance of the number of electrodes becomes apparent for the use of

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

ICA as data cleaning tool. Physiological non-brain sources like eye,

facial, and neck muscles contribute significantly through capacitive

conduction to the recorded signal. With sufficient number of electrodes,

2 or more ICs will result from the decomposition reflecting vertical and

horizontal eye movements while several additional ICs are required to

account for the activity of over 40 muscles of the head and neck

(Kamibayashi & Richmond, 1998; Westbrook et al., 2023). Beyond po-

tential mechanical artifacts linked to movement and potential electrical

artifacts from the environment, the available degrees of freedom to

explain the activation of an unknown number of brain processes is

markedly diminished. It seems plausible that ICA should not be used

with less than 20 channels and researchers need to carefully consider the

context and characteristics of their data when applying ICA and inter-

pret the results accordingly since low electrode densities might not allow

for separating brain from non-brain activity. Nevertheless, merely hav-

ing a high number of electrodes does not ensure favorable ICA outcomes,

as issues such as non-stationarities or condition-specific artifacts persist

regardless of the channel count. Further considerations are the obtru-

siveness and practicability of high-density montages that can restrict

natural head movements due to the additional weight and in some cases

bulks of cable attached to the head. In case the brain activity feature of

interest is well known and can be recorded with only very few channels,

there might be no need for high-density montages and blind source

separation analyses at all. The minimum number of electrodes necessary

to record and analyze a known feature of interest is the best solution to

reduce the obtrusiveness and increase the practicability of the recording.

3.2. EEG feature extraction

After data cleaning and isolating the time periods of interest in the

recorded signal, the relevant features are extracted for further statistical

analyses. A feature is commonly defined as a distinctive characteristic or

property extracted from biosignals in psychophysiological research,

such as the amplitude of an ERP or the spectral power at designated

electrode sites. Generally, the EEG signal can be analyzed in the time

domain, the frequency domain, and the time-frequency domain (Cohen,

2014; Gramann & Plank, 2019) both on the sensor or the source level

after appropriate source reconstruction. Time and time-frequency

domain analyses require events, usually derived from controlled stim-

ulus presentations, that allow for extracting stereotypical brain activity

associated with the onset of the events, e.g., every time an image is

presented on a screen or a response button is pressed. These

event-related fluctuations in the EEG signal can then be averaged and

analyzed in the time domain as event-related potentials (ERPs; Luck,

2005) or, in the time-frequency domain, as event-related spectral

perturbation (ERSPs; Onton & Makeig, 2006). By averaging all time

intervals around the onset of events, invariant electrocortical responses

to the respective event or class of events are extracted and variable

activation in the signal is averaged out. Since event presentation in the

real world is difficult to control, the majority of Neurourbanism and

Neuro-Architecture studies do not investigate event-related activity.

However, innovative analysis approaches allow for investigating

event-related activity without stimulus presentation using the behavior

of participants itself as event input. For example, eye blinks can be used

to extract blink-related potentials from the EEG (Wascher et al., 2014,

2022; Wunderlich & Gramann, 2021) that allow for the comparison of

blink-related ERPs in different conditions (e.g., built vs green environ-

ment). The electrocardiogram (ECG) can be used to extract R-peaks from

the ECG to compute heartbeat-evoked potentials for further investiga-

tion (e.g., Luft & Bhattacharya, 2015). In case mobile EEG is combined

with eye-tracking, fixation-based ERPs for specific object classes that

attract fixations can be investigated also in mobile protocols (e.g.,

Ladouce et al., 2022).

As an alternative, the transformation of data from the time domain to

the frequency domain reveals insights into the energy of specific fre-

quency bands across the entire signal spectrum, contingent upon

experimental manipulations. EEG, representing neural oscillatory ac-

tivity, can be analyzed by decomposing data into weighted sine and

cosine functions with different frequencies, phases, and amplitudes. This

decomposition is achievable through methods like the Fast Fourier

Transform (FFT) converting the signal from the time to the frequency

domain. Frequency-domain analyses do not necessitate the controlled

presentation of events and can be applied to any time period of interest.

Examples include comparing power levels during baseline activity

against those in experimental periods or directly contrasting different

time intervals contingent on experimental conditions, such as alter-

nating between experiences of urban as compared to green environ-

ments in designated experimental blocks. Alterations in power within

defined frequency bands, exhibiting a specific topography, have been

linked to many psychological constructs, with emotion (for a recent

review see Suhaimi et al., 2020) and meditative states (e.g., Cahn &

Polich, 2013) attracting strong interest from the Neurourbanism com-

munity. The specific topographic distribution of the frequency band of

interest can provide important insight into the involved sensory mo-

dalities or multimodal integration processes. For example, shifts in vi-

sual attention are accompanied by modulations in the posterior alpha

band (8–12 Hz) that can be used for monitoring EEG correlates of visual

attention (Van Gerven & Jensen, 2009).

Additionally, in moving beyond understanding the brain activity in a

single region at the time, the EEG signal can be transformed into the

Hilbert space, which provides phase information relevant to timing and

synchronization of different brain activities. This allows for instance to

understand the coordination of neural activity in different brain regions

during different tasks such as perception (Lachaux et al., 1999; Rodri-

guez et al., 1999). Particularly due to the rhythmic nature of brain

function, the very shape of the signals is rich in information (Buzs´

aki,

2006). For instance, the Hilbert transformed signal can be used to obtain

the amplitude envelope of EEG signals. This can be particularly useful

for studying amplitude modulation, which reflects the waxing and

waning of rhythmic brain activity, in synchronization with the envi-

ronment (Charalambous & Djebbara, 2023).

A crucial consideration for mobile EEG applications in Neuro-

urbanism research however is the question whether findings regarding

spectral power modulations observed in stationary protocols can be

extrapolated to mobile recordings without restrictions. In the time

domain, it becomes evident that specific movement artifacts (e.g.,

blinks, jaw clenches) that are time-locked to an event (onset of car

honking, increasing social density) will easily lead to incorrect inter-

pretation of the results if the artifact is not successfully identified and

removed. For the frequency domain, movement-related brain activity is

a particularly intriguing question since movement might not only pro-

duce artifactual activity related to an event but because numerous pat-

terns of movement-related activity overlap in their frequency bands with

frequencies that mirror cognitive and affective processes under inves-

tigation. For example, walking is associated with systematic modula-

tions in a wide frequency range including alpha, beta, and low gamma

activity over frontal, central, and parietal areas (e.g., Seeber, Scherer,

Wagner, Solis-Escalante, & Müller-Putz, 2014). Head rotations lead to

neck muscle activity that contains strong posterior alpha modulations in

addition to typical muscle spectra peaking beyond 20 Hz (Grassini,

Segurini, & Koivisto, 2022) and spatial updating of position and orien-

tation including head and body rotations affect theta power as well as

alpha power in widespread brain regions (Do, Lin, & Gramann, 2021).

Considering the fact that memory, attention, and affective states all are

associated with modulations in the theta and alpha range (Klimesch,

Doppelmayr, Russegger, Pachinger, & Schwaiger, 1998; Onton,

Delorme, & Makeig, 2005; Smith, Reznik, Stewart, & Allen, 2017),

active behavior might add modulations of the frequency bands of in-

terest that add to the ongoing cognitive and affective modulatory pro-

cesses. In addition, the control of potentially confounding factors

possible in the lab is mostly absent in the real world, and sensory pro-

cessing of uncontrolled events will take place in the frequency bands of

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

interest that might be interpreted as power modulation regarding ma-

nipulations of the construct under investigation (Barnes et al., 2023).

With this general overview on the requirements of the EEG equip-

ment for recording human brain activity in Neurourbanism, the analyses

approach and extraction of parameters, and their final interpretation,

the following section will use these aspects to analyze published papers

in the field of Neurourbanism using EEG.

3.3. A review of publications in the field of neurourbanism using (mobile)

EEG

A literature review was conducted through the Web of Science (WoS)

database and Google Scholar on November 25, 2023, focusing on sci-

entific journal publications exclusively in English. The search in WoS

encompassed empirical studies indexed from 1948 onwards, limited to

full-length peer-reviewed journal articles. Exclusions were made for

reviews, abstracts, and conference proceedings and the main search

terms for abstract queries included the following:

AB=((“built environment” OR “architectural space” OR “interior

space” OR “environmental design” OR “physical environment” OR

“urban” OR "urbanism" OR “urban density” OR "neurourbanism" OR

"neuroarchitecture" OR “designed space” OR “urban design” OR “green

space” OR “urban landscape”) AND ("EEG" OR “mobile EEG” OR

“ambulatory EEG” OR “Mobile Brain/Body Imaging”)).

After the initial search, the results were refined by excluding the

search terms “Epilepsy”, “Seizure”, and “Evolutionary Economic

Geography”.

The search returned 99 publications of which 55 publications re-

ported empirical EEG research and matched the selection criteria. Based

on these results, an additional search was conducted in Google Scholar

based on the literature in the identified papers resulting in 75 peer-

reviewed papers using EEG in the laboratory or the real world to

assess the human brain response to built and natural environments.

Nearly 75% of these papers have been published within the last five

years. All studies were subsequently classified regarding the mobility

score of the EEG system and the participant mobility score according to

the classification scheme from Bateson et al. (2017). An additional focus

was set on the number of electrodes in the selected studies which,

together with the system and participant mobility score, allows for a

critical evaluation of the reported results regarding potential limitations

through movement-related artifacts and restrictions regarding the

analysis approach. System scores were derived from the hardware

description in the respective studies and participant mobility scores

were evaluated based on the description of the study protocol in the

selected studies. Since a number of the identified studies used

head-mounted virtual reality (VR) allowing restricted or unrestricted

movements of the head, it should be noted that the classification from

Bateson and colleagues did not further differentiate between walking

and head movements with the latter also contributing movement-related

artifacts to the recordings (see e.g., Gramann et al., 2010; Jungnickel &

Gramann, 2016). For better comparison with previously published

classifications the classes were kept as suggested by Bateson (Bateson

et al., 2017).

3.4. Evaluation of selected studies regarding use of hardware

The studies were evaluated according to the publication guidelines

and recommendations for studies using electroencephalography and

magnetoencephalography (Keil et al., 2014). Only a subset of the rec-

ommended key points was used to allow for a more liberal evaluation of

the reviewed studies. All studies were investigated regarding the infor-

mation on the amplifiers used, including the amplifiers’ type (amplifier

name was provided), sampling rate (online sampling rate was stated),

and the use of online filters (online filters were described, without

having to state the type of filter or roll-off and cut-off parameters).

Regarding the specification of electrodes, the criteria included

information about the number of electrodes, sensor types (the type of

EEG sensor was described, e.g., wet or dry sensor), and sensor locations

(sensor locations were specified, including reference electrode(s),

providing general location systems like 10% system etc. were considered

sufficient). Of note, several studies used consumer market systems that

could be researched online regarding these criteria. However, individual

configurations might have changed standard setups and thus these

criteria were considered not fulfilled in case the information was not

provided.

The analyses of the selected publications showed that many studies

failed to provide sufficient details on amplifier specifications for their

recordings, such as the name of the system, the sampling rate and filter

settings during recording, and sometimes even the number of electrodes.

Nearly all studies (94.7%) reported the amplifier make and 68% of the

reviewed studies reported the sample rate during recording. Only 14.7%

of all studies reported online filter settings of the amplifier (see Fig. 1).

Of the 75 studies overall, only two (2.6%) studies failed to report the

number of electrodes. The majority of studies (72%) also reported the

electrode positions but only 32% of the studies indicated the sensor type

(e.g., dry or wet electrodes). Approximately half of the studies (48%)

reported which electrode or combination of electrodes was used as the

reference. Regarding the data processing criteria (discussed later), the

review revealed less than half of the studies (46.7%) to provide infor-

mation about the filter settings during analyses. Only 18.7% of all

studies reported criteria for data rejection and only 9.3% reported

whether and how they interpolated artifactual channels.

Overall, a wide range of different systems from diverse companies

were used varying in scores for system portability, participant porta-

bility as well as the number of electrodes in the respective study (see

Fig. 2).

Of the 73 studies that provided information on the number of elec-

trodes, the majority (82.2%) used montages with less than 20 electrodes

while 6.8% of the studies used between 22 and 32 electrodes. Further 6

studies used 64 electrodes (8.2%) and 2 studies used 96 or 128 elec-

trodes (each 1.4%). Most of the amplifiers were manufactured by Emotiv

(38.7%), NeuroSky (14.7%), and Brain Products (10.7%). In terms of

system mobility, as defined by (Bateson et al., 2017), 24% of the

amplifier systems had a mobility score of 0, while 7% had a system

mobility score of 1, and 3% had a score of 3. With 66%, the majority of

EEG systems demonstrated high mobility with a system mobility score of

4. Regarding the participant mobility score, the high percentage of

mobile systems used in the studies was not necessarily accompanied by a

high participant mobility score in the respective studies. A large number

of studies using EEG systems with the amplifiers mounted in the cap still

restricted participants’ movements using protocols that required par-

ticipants to sit still while watching images or movies presented on

screens or head-mounted virtual reality displays. Participant mobility

scores indicated that 3% of studies involved participants sitting or lying

with no movements. The largest portion of all identified studies allowed

only minimal movements with participants sitting or lying and

responding by button presses or providing other output at the end of

experimental trials (76%). Only 7% of the identified studies allowed

constrained walking (instructed slow walking or walking on a tread-

mill), while 14% permitted unconstrained walking (walking at preferred

speed in or outside the laboratory).

From Fig. 3, it becomes visible that most studies that used EEG sys-

tems with a high electrode density used stationary protocols. Only three

studies with 64 or more electrodes (with two of the studies reporting

different analyses approaches for the same dataset) allowed participants

to walk through the (virtual) environment (see Fig. 4).

3.5. Evaluation of selected studies regarding data preprocessing and

feature extraction

The following section provides an overview on data preprocessing

approaches based on a reduced set of guidelines from the committee

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

report for studies using electroencephalography and magnetoencepha-

lography (Keil et al., 2014). The reduced key point for data processing

and feature extraction included information regarding rereferencing

(information about the new reference would fulfil the criteria even in

case the original reference electrode for recording was not mentioned),

the detection of noisy channels and the interpolation of channels (the

number of channels that were removed for each participant and the

interpolation algorithm used for estimating missing channels were re-

ported) and artifact rejection approaches (artifact rejection procedures

are described, including the type and proportion of artifacts rejected).

Rereferencing approach were not considered here since the absence of a

clear description for rereferencing was only possible in case rerefer-

encing was reported in the first place (see Fig. 1). Less than half of the

studies (46.7%) provided information about the filter settings and the

criteria for data rejection and channel interpolation were provided by

only 18.7% and 9.3%, respectively.

Overall, concerning the analysis approaches in the selected studies,

only 10 out of 75 studies (13.3%) provided sufficient information to

replicate their analysis, while 23 studies (30.7%) offered sufficient de-

tails on algorithms and treatment of artifactual data allowing for a

partial replication of their results. Alarmingly, 56% of the surveyed

studies, i.e. 42 of 75 peer-reviewed and published studies lacked

essential details in the methods section or throughout the paper failing

to provide information on amplifier make and settings, sample rates

during recording or offline down sampling, reference electrodes, filter

settings, artifact rejection criteria, or the amount of data removed before

feature extraction. This prohibits meaningful assessment of the accuracy

of analyses and the validity of the results and interpretation of the data

due to insufficient information. These papers are thus unsuitable for

consideration in a scientific context. Importantly, among these studies,

33 studies (44%) used proprietary output from commercially available

amplifier systems that lack the necessary information to reproduce the

results and that do not allow for critical evaluation of the data or ana-

lyses approaches and how artifactual data was handled during the

preprocessing.

The use of black-box output features was observed mainly for sys-

tems with electrode densities below 20 electrodes. This can be attributed

mainly to the use of the NeuroSky and Emotiv EEG systems that provide

automated classification of EEG data according to different user states.

With 33 studies out of 75, 44% of all identified studies relied on

black-box analyses outcome. Beyond the use of non-replicable black box

output parameters, 40% (30 studies) of all studies used ICA as an

analysis approach to preprocess the data. The use of ICA for some kind of

data preprocessing, mostly artifact rejection, is a critical aspect for a

large number of the identified studies with less than 20 electrodes due to

the limited dimensionality of the data that does not allow dissociating all

active brain and non-brain activity patterns. Importantly, the descrip-

tion of the exact use of ICA and subsequent removal of ICs varied

drastically. Most of the studies did not provide sufficient information on

which ICA algorithm was used, whether the dimensionality was reduced

before decomposition, how the resulting ICs were classified as artifac-

tual or functional data or how many ICs were removed, which would be

necessary for a replication of the analyses. In one case, ICA was

computed on data with 4 electrodes, in a second case ICA was computed

on data that was reduced from 14 to 2 electrodes before ICA decom-

position without stating how the decomposed data was subsequently

used. The largest portion of these studies did not report at all what the

outcome of the decomposition was or what was done with the decom-

posed data. Only 6 studies (8%) used automated or semi-automated data

preprocessing before computation of the final features and 6 studies

(8%) reported visual inspection of the data for artifact rejection before

feature extraction.

With more than 90%, the majority of the studies identified in the

present literature search extracted features in the frequency domain.

Among those, 8.3% of the studies used both time and frequency domain

features and only 2.8% of the studies used only time domain features

(ERPs) and 2 studies used source reconstruction or network analyses on

the recorded data. The main focus was on alpha power comparing

different conditions with 87% of all studies using alpha power alone

(alpha or frontal alpha asymmetry) or in combination with other fre-

quency bands.

Fig. 1. Evaluation criteria applied to reviewed studies. Information on respective category provided in the reviewed studies. Dark filled area displaying the per-

centage of studies not providing the respective information, light filled are reflecting percentage of studies providing the required information.

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

4. Discussion

The present study investigated which EEG systems were used and

how the acquired data was analyzed in Neurourbanism research in the

laboratory or the real world. A focus regarding aspects of EEG recordings

was on the mobility grades regarding the system and the participant

mobility as well as the number of electrodes used to acquire the signals.

In addition, the processing of the recorded data and the subsequent

feature extraction were considered.

It is important to emphasize that Neurourbanism is an emerging field

that faces numerous challenges, including the need for interdisciplinary

collaboration and the search for protocols that repesent human re-

sponses to the real environments in an ecologically valid fashion. The

results of this research field should eventually inform urban planners

and policy makers to make informed decisions about the design of our

surroundings. In this sense, the results of the present review paint a

deeply concerning picture of the scientific rigor in the emerging field of

Neurourbanism. More than half of the research papers cannot be

Fig. 2. Distribution of EEG systems and setups used in the analyzed Neurourbanism studies categorized according to the CoME score (Bateson et al., 2017).

Companies are arranged alphabetically and in counterclockwise direction. Right panel displays distribution of system mobility scores (upper panel) and participant

mobility scores (lower panel) across all included studies.

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

Fig. 3. EEG devices used in the identified studies displayed according to their system mobility score and participant mobility score. The number of electrodes is

displayed with increasing electrode density indicated by increasing diameter of the respective sphere. Categories indicate electrode number to be smaller or

equivalent to the category label (e.g., ≤2 electrodes). Color coding indicates whether the study was conducted in the lab or in the real world. Colors based on

colorspace (Zeileis et al., 2019). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 4. Data analyses approaches for EEG devices used in the identified studies according to their system mobility score and participant mobility score. The number

of electrodes is displayed with increasing electrode density indicated by increasing diameter of the respective sphere. Color coding indicates the analyses approach

with (pink) Black Box output, (green) ICA, (cyan) automated preprocessing (other than ICA) providing toolbox names and settings, (purple) manual analyses based

on visual inspection. Colors based on colorspace (Zeileis et al., 2019). (For interpretation of the references to color in this figure legend, the reader is referred to the

Web version of this article.)

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

replicated and an additional 30% of the residual papers allow only

partial replicability, rendering more than 80% of all papers unsuitable

for a critical scientific evaluation. Numerous papers lack objectivity in

data acquisition, analysis, and result interpretation. Consequently, over

50% of the studies uncovered in this literature review should be cate-

gorically excluded from the realm of credible scientific discourse.

Alarmingly, these deficient papers were still published despite failing to

meet even minimal standards for reporting physiological data collection

and analysis. This underscores the imperative need for multidisciplinary

expertise in studies of this nature, as relying solely on commercially

available systems without scientific competence and critical scrutiny is

clearly inadequate. Furthermore, this situation highlights the role of

publishers in disseminating misinformation for profit, which seems not

exclusive to any single publishing entity. Notably, Elsevier published 16

of the 75 reviewed studies (26.2%) lacking essential data recording and

analysis information or relying on opaque methodologies that hinder

replication. They were followed by MDPI at 19.5%, Taylor and Francis at

11.9%, and Springer at 9.5%. Concerning the proportion of non-

replicable studies among all studies issued by a specific publisher,

several publishers demonstrate a 100% rate of non-replicability due to

the fact that the sole study they published was not replicable. When

focusing on publishers that released at least five papers, included in this

analysis, Springer had the highest non-replicability rate at 80% (4 out of

5), followed by Taylor and Francis at 71.4% (5 out of 7), Elsevier at

68.8% (11 out of 16), and MDPI at 61.5% (8 out of 13). What is

particularly alarming is that all of these publishers also offer journals in

the field of Neurosciences, which could have facilitated the inclusion of

subject matter experts in the editorial process. Regrettably, this does not

appear to have occurred.

All studies that were identified in the present work used EEG

amplifier technology with good to very good hardware specifications

that would allow for recording high quality EEG data. Nonetheless, even

though the acquired raw data might have been of very good quality, the

raw data was not used in more than half of the studies. The majority of

studies used portable consumer-grade mobile amplifier systems

reflecting a general trend towards small and lightweight EEG systems

even in case these systems were used in stationary protocols that do not

necessarily require mobile EEG systems. It seems rather the low-cost

aspect than the general mobile use case that these systems provide

compared to medical grade or research-grade amplifier systems that

render them attractive to researchers confirming previous observations

in this direction (Niso et al., 2023). In combination with automated

analyses approaches, these systems appear as an attractive alternative

for non-experts in the field. Furthermore, exactly these low-cost EEG

systems often come with a low number of electrodes with the NeuroSky

and Emotive Epoc systems with only 1 or 14 electrodes, respectively,

making up the largest portion of systems used in Neurourbanism

research. Recording EEG data with only one electrode does not allow for

any analytical exclusion of physiological non-brain activity like eye

movement or facial muscle contraction and additional

movement-related artifacts in case of ambulating participants. This is

also still a critical issue for systems with 14 electrodes. Still, none of the

reported studies that can be considered non-replicable discussed this

problem.

Several studies, including some of the studies that were replicable,

further claimed that the Emotiv Epoc system was validated for use in the

real world by citing a paper from Debener et al. (2012), one of the first

ambulatory EEG papers in the field. This study, however, did not use the

original electrode system of the Emotiv Epoc but replaced the electrodes

using medical grade wet electrodes connecting them to the Emotiv

amplifier. This is not comparable to the relative inflexible electrode

holder system or the quality of the electrodes that come with the

off-the-shelf system. As such, the paper by Debener and colleagues did

not validate the use of the Emotiv Epoc system in real world settings or

protocols with active movements of participants rendering this argu-

ment invalid. Moreover, the only study to the knowledge of the author

that directly compared a medical grade EEG system with the Emotiv

Epoch system in actively moving participants demonstrated signifi-

cantly decreased data quality for the Emotiv Epoc system (Duvinage

et al., 2012). As such, data recorded with any of these low-density sys-

tems in mobile protocols should be considered with caution and only

appropriate data analyses approaches should be used. ICA is not

necessarily one of them.

Overall, the electrode density varied significantly across studies. The

majority of high-density studies used stationary laboratory protocols

and only three of these studies used mobile protocols in the laboratory.

Low-density studies used both laboratory as well as real-world protocols

with only a minor subset of studies allowing participant movement

through the environment. All higher density studies relied on automated

data analyses approaches including ICA as a data cleaning tool. In case

of high-density recordings, the dimensionality of the acquired data is

high enough to account for multiple physiological and mechanical as

well as electronic artifacts. In case of low-density recordings, however,

this is not the case and the analyses of EEG with 14 or less electrodes

might not be adequate to clearly separate brain from non-brain activity.

Nonetheless, the majority of studies using low-density recordings used

either the output of proprietary black-box algorithms or ICA as a data

cleaning tool. What was even more concerning for the studies with low-

density recordings that used ICA as preprocessing tool, however, was the

description or rather the absence of a description of which ICA algorithm

was used, how the data was prepared before ICA decomposition, and

how the decomposed data was subsequently used. Only a few studies

used toolboxes to automatically label ICs as reflecting artifactual ac-

tivity, including MARA (Winkler et al., 2011), SASICA (Chaumon et al.,

2015), or ICLabel (Pion-Tonachini et al., 2019). A notable observation is

that the majority of studies either failed to disclose their approach to IC

selection or omitted any discussion about the treatment of decomposed

data. This raises concerns regarding the level of transparency and un-

derstanding of the analytical methods employed in these studies.

Regarding the extraction and interpretation of features from the

recorded physiological data, the present review showed that the ma-

jority of studies used parameters from the frequency domain with alpha

being the most often investigated parameter. Parameter extraction is a

crucial step in operationalizing the research question involving the

identification and quantification of specific features or characteristics

within the data that are relevant to the physiological processes under

investigation. However, challenges and problems can arise during the

parameter extraction process when the presence of noise and artifacts in

the physiological data distort the accuracy of parameter measurements.

This is especially problematic, when the experimental protocol leads to

non-brain sources contributing to the frequency band of interest. This is

the case for eye and head movement-related activity as well as gait-

related artifacts in mobile protocols that contribute to a broad fre-

quency range and can distort the signal of interest. Only a small fraction

of studies used mobile protocols with the minority providing sufficient

electrode density to allow for identification and subsequent removal of

eye and body movement-related activity while also missing to report the

exact approach that was implemented. None of the low-density studies

addressed movement-related activity to potentially impact the extracted

parameters at all. The majority of studies identified in the present

literature review, however did not use mobile protocols and as such, the

extraction and interpretation of frequency parameters might be

considered a reliable approach.

Most of the studies in the field of Neurourbanism that recorded EEG

data used frequency domain parameters referring to studies from other

research domains that describe power modulations in specific frequency

bands to covary with the experimental manipulation. Selecting previ-

ously established features from different research areas and interpreting

these features in the Neurourbanism context is a solid scientific

approach. Critically, however, in real-world protocols, there is no con-

trol over the onset of events in the environment. Unexpected occur-

rences, such as individuals abruptly crossing paths or unpredictable

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

variations in traffic noise, environmental sounds like chirping birds or

rustling leaves, can introduce uncontrollable factors that impact the

moment-to-moment dynamics of human brain activity. As a conse-

quence, the shorter the recording and analyses periods of EEG are, the

more likely variation in human brain dynamics will increase due to the

impact of uncontrollable events. A direct comparison with features

extracted in highly controlled laboratory settings that do not allow for

any kind of movement should thus be considered with caution.

The method of stimulus presentation also warrants further scrutiny

in this context. Brain activity, inherently responsive to sensory inputs, is

particularly relevant in the study of Neurourbanism, a field defined by

environmental sensory stimuli. As noted by Eberhard (2009b), the

interplay between the brain, behavior, and environment is central to

Neurourbanism. This interplay is underscored by the difference in

behavioral responses to simulated urban environments (e.g., videos)

versus actual urban settings. Considering the profound impact of envi-

ronmental characteristics on behavior, it is necessary to incorporate

complex and contextually rich stimuli in research designs and analyses

beyond simplistic representations (e.g., ‘green space’ vs. ‘urban space’)

to include multifaceted elements like spatial dimensions, visual

boundaries, interactivity, temporal patterns, and other nuanced envi-

ronmental features.

In summary, a significant portion of EEG studies within the Neuro-

urbanism field exhibited notable shortcomings in scientific rigor and

should be excluded from scientific consideration. The principal issues

identified in these studies encompassed inadequate reporting of hard-

ware specifications and data acquisition setups, the utilization of a

limited number of electrodes for signal recording, often in conjunction

with undisclosed proprietary analysis outputs, or inadequate use of In-

dependent Component Analysis (ICA) for data cleaning. When data

acquisition and analysis approaches lack objectivity, the generation of

reliable and valid results becomes untenable, significantly compro-

mising result interpretation. It is worth noting that individuals lacking

expertise in neuroscience occasionally rely on erroneous assumptions

concerning EEG, further exacerbated by the inappropriate application of

analytical methods to the data that was recorded with questionable

quality. All these misconceptions might stem from the available off-the-

shelf and black-box EEG systems falsely conveying confidence that

“anyone can use EEG”. Should these findings be further extended,

generalized, or even suggested as the basis for urban development pol-

icies, it becomes essential to implement mechanisms that deter unwar-

ranted extrapolations and their associated consequences. The

observation that journals haphazardly publish such research, often

prioritizing profit motives over rigorous quality assessments, un-

derscores a significant departure from the established standards of sci-

entific rigor within the scholarly publishing realm, which is expected to

serve as a safeguard against such practices.

It is crucial for Neurourbanism research to fully embrace inter- and

transdisciplinary approaches to harness the range of perspectives and

expertise needed for studying our complex and ever-evolving urban

environments. Addressing the diverse questions in this field requires

contributions from many areas including but not limited to urban

planning, architecture, psychology, medicine, neuroscience, geography,

data science, in different combinations tailored to specific research in-

quiries. These perspectives should be systematically extended by

including citizen scientists who not only stand to benefit from this

research but also contribute unique knowledge to it. Each group offers

valuable insights and skills that can enhance experimental efforts and

ensure scientific rigor. This is particularly important given the chal-

lenges Neurourbanism faces, such as budgetary limitations, reaching

specific vulnerable populations, and dealing with uncontrollable factors

in living labs, among others. This field is still in its infancy providing

significant opportunities for breakthroughs that could fundamentally

transform urban living. This potential highlights the necessity for

interdisciplinary collaboration to gain new insights while maintaining

scientific integrity in this vital research area.

5. Conclusion - guidelines for (mobile) EEG protocols in the

laboratory or the real world

From the reported problems in the reviewed studies, it seems

imperative that the interdisciplinary character of Neurourbanism

research becomes indeed interdisciplinary, securing expertise from all

fields involved in the research question addressed. In case EEG is used to

investigate the human brain response to the natural or built environ-

ment, an expert from the neuroscience with experience in EEG should be

part of the research team as well as experts from the other domains

addressed in the specific project.

Specifically, with respect to the use of (mobile) EEG in Neuro-

urbanism research, the following decision steps and guidelines are

derived to help researchers in this field to identify the best EEG system

and analysis approaches for their study protocols (see Fig. 5).

5.1. Amplifiers

In general, most research-grade as well as all consumer-grade mobile

amplifiers are adequate for EEG recordings in terms of data quality.

However, it’s important to note that many EEG amplifiers on the con-

sumer market are not optimized for protocols involving participants’

free movement. With respect to amplifier specifications, the following

points should be considered.

●Most EEG amplifiers that are commercially available provide suffi-

cient bit rates as well as sample rates allowing for the acquisition of

EEG data of good quality. If the frequency domain of interest is

limited from 1 to 60 Hz, sampling rates of 128 Hz are sufficient. In

case of explorative analyses without upper frequency band limita-

tion, higher sampling rates of 250 Hz or 500 Hz are advised.

● The CMR should be at least 80 dB or higher.

● If mobile recordings are planned, the weight of the amplifier system

should be low and the wireless protocol should allow reliable data

recording with missed samples being indicated by the system.

● In case no other data streams are synchronized with the EEG, the

data can be stored on local memory cards in case the recording does

not need any timing information.

● In case additional data streams are recorded together with EEG (e.g.,

eye-tracking, ECG), synchronization protocols like Lab Streaming

Layer (LSL) (https://github.com/sccn/labstreaminglayer) or other

synchronization protocols can be used to allow multimodal data

fusion during recording and analysis. These recording environments

are powerful tools but require expertise to avoid erroneous streaming

and synchronization.

5.2. Number and distribution of electrodes

Further challenges arise in (mobile) EEG studies in case of low

electrode numbers and uneven distribution of electrodes over the scalp.

Insufficient numbers of electrodes and uneven distribution restrict the

analyses approaches and feature extraction to the provided recording

sites, particularly affecting investigations into potentially yet unknown

EEG features. The following guidelines might be considered.

● Never use only one electrode for scientific investigation of human

brain activity.

● Always use only one electrode as the reference electrode during

recording. The data can always be re-referenced offline to any

desired reference setting. The reference electrode should sit tight in

the cap and have good contact to the skin (e.g., mastoid references

incorporated in the cap are often artifact prone due to the loose fit of

the caps next to the ear slits).

● As in the case of stationary protocols, there is no set minimum

number of electrodes required for mobile EEG recordings. In general,

the number of electrodes in non-stationary settings should be derived

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

based on the model assumptions included in the experimental pro-

tocol, the analytical approaches applied to the data, and the features

of interest.

● Systems with 14 electrodes might be sufficient in case of stationary

protocols when the parameter of interest can be derived from this set

of electrodes at the specified electrode positions.

● When participants are allowed to move through the environment,

more electrodes are generally preferable due to the increasing

number of non-brain sources significantly contributing to the

recorded signal. A higher number of electrodes increases the degrees

of freedom for blind source separation approaches like ICA to isolate

additional activation patterns stemming from movement (physio-

logical sources like eye movements, facial and neck muscle activity

as well as mechanical artifacts like cable sway, electrode movements

etc.) and simulation studies have shown that 64 channels allow for

good source separation in mobile protocols (Klug & Gramann, 2021).

● In case the parameters of interest are known in advance, electrodes

should be placed at those locations where the feature is expected

based on previous work. In case of explorative studies, the electrodes

should be distributed across the entire scalp using standardized

electrode locations (e.g., Oostenveld & Praamstra, 2001).

● High numbers of electrodes with even distribution (equidistant

montages) allow for a wider range of analytical approaches (e.g.,

ICA, source reconstruction, network analyses).

5.3. Kind of electrodes

In case of stationary protocols without participant movement, most

electrode types can provide good-quality data with appropriate prepa-

ration of the recording sites. The use of dry electrodes in real-world

mobile EEG studies may pose issues with the restriction of electrode

locations to sites with less hair often in proximity to the eyes and close to

facial muscles. Moreover, dry electrodes are especially artifact-prone

during active participant movements due to changes in impedance

with movements of the electrode over the skin.

● When mobile protocols are employed, wet electrodes provide less

artifact vulnerability and prohibit, to a certain degree, artifacts

related to micro-movements of the electrodes.

● With appropriate preparation of the recording sites, passive and

active wet electrodes with or without shielding can provide compa-

rable quality (see Scanlon et al., 2021)

● In case of mobile EEG recordings, cables should be fixed in the cap or

arranged in a way that cable sway is inhibited as far as possible to

avoid movement-related artifacts.

5.4. Data preprocessing

Usually, EEG data is preprocessed to eliminate unwanted, artifactual

data. This can be done in a variety of ways each influencing the data in a

specific way. The data processing pipelines for stationary data differ

from the preprocessing approach for mobile data (Klug & Gramann,

2021). Specific pipelines to preprocess mobile EEG data exist (e.g., Klug

et al., 2022). When only time domain features are of interest, filter

settings can be used that suppress high and low frequency aspects of the

data (e.g., low-pass filter of 40Hz and high-pass filter of 1 Hz) elimi-

nating aspects of muscle activity and slow drifts, respectively. In case

frequency domain parameters are of interest, filters that interfere with

the frequency bands of interest should be avoided.

● Filter settings depend on the feature of interest; in case of explorative

analyses and no specific feature being targeted, filters should be

avoided.

● When ICA is used, the model assumptions should be known and met.

The quality of the ICA outcome depends on the quality of the

recorded data (“garbage in, garbage out”), the number of potentially

active sources, the number of electrodes, and the mobility of

participants.

● ICA should not be used with less than 24 electrodes since it is

arguably unlikely that the summed number of brain and non-brain

sources is lower than 24. However, a high number of electrodes

does not guarantee a good ICA solution as additional factors can

impact the data decomposition.

● If lower electrode densities (e.g., 14 electrodes) are used, it is advised

to remove only ICs reflecting eye movement as these activity patterns

might be well dissociated from other sources due to their high energy

in the EEG signal. This, however, does not warrant that no brain

activity will be removed too.

● Removing ICs, i,e. eye and muscle activity-related ICs, should have

an objective basis that can be reported and replicated (e.g., classifi-

cation like MARA, ICLabel etc).

● For all recordings but especially for low density recordings, the

likelihood of removing functional brain activity increases with

increasing number of ICs that are removed from the data.

● In general, the minimum number of electrodes necessary to record

and analyze a known feature of interest is the best solution. Param-

eters that are established regarding their origin and analytical

extraction can be recorded from only a few electrodes located over

the area of interest without requiring blind source separation ap-

proaches. In case the features of interest are unknown, however,

higher density montages allow a wider range of analytical methods

for explorative analyses but come with increasing weight for par-

ticipants and prolonged preparation times for experimenters.

Fig. 5. Flowchart of potential decision points for planning Neurourbanism experiments with EEG. Please note that different decision are co-dependent (e.g., “neural

processes of interest” with the sample rate of the amplifier and the location of electrodes, data preprocessing). Decision aspects are provided but are not limited to the

examples. Further information is provided below for selected decision points.

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

The interested reader can find very good insights on general aspects

of data processing for neural time series in the textbook by Mike X

Cohen.

•Cohen, M. X. (2014). Analyzing neural time series data: theory and

practice. MIT press

5.5. Feature extraction and interpretation

Ultimately, the interpretation of EEG data hinges on features

observed within specific experimental protocols. These features transi-

tion into parameters when their empirical observations are numerically

quantified to represent specific aspects of interest, as seen in posterior

alpha power to change dependent on the level of relaxation of partici-

pants. A psychophysiological indicator, in this context, emerges when a

parameter consistently provides reliable information about a targeted

phenomenon. In Neurourbanism research, an indicator might be an EEG

parameter viewed as a stable representation of an underlying cognitive

or emotional state replicated across various protocols and conditions.

Nevertheless, such stability for specific parameters is not yet observed in

the field.

●Parameters that rely on proprietary algorithms that lack trans-

parency, preventing a thorough evaluation of processing parameters

or the reproduction of results should be excluded from scientific

investigations.

● Parameters should be interpreted with caution without “cherry-

picking” previous studies that support the authors’ claims.

● Parameters that were established in stationary protocols would be

interpreted with caution if the experimental protocol included

movement.

For interested readers, the following textbooks provide very good

overviews on EEG analyses in the time and frequency domain.

•Time domain: Luck, S. J. (2014). An introduction to the event-related

potential technique. MIT press).

•Frequency domain: Gable, P., Miller, M., & Bernat, E. (Eds.). (2022).

The Oxford handbook of EEG frequency. Oxford University Press.

These guidelines, while not comprehensive, represent key consider-

ations for (mobile) EEG studies drawn from over 15 years of research in

mobile brain imaging using EEG. They are intended to assist researchers

venturing into this dynamic area of study.

Funding

This research was funded under the Excellence Strategy of the Fed-

eral Government and the L¨

ander by the Berlin University Alliance.

CRediT authorship contribution statement

Klaus Gramann: Writing – review & editing, Writing – original

draft, Visualization, Resources, Methodology, Investigation, Funding

acquisition, Formal analysis, Conceptualization.

Declaration of competing interest

The author declares he has no conflicts of interest related to this work

to disclose.

References

Adli, M., Berger, M., Brakemeier, E.-L., Engel, L., Fingerhut, J., & Gomez-Carrillo, A.

(2017). Neurourbanism: Towards a new discipline. The Lancet Psychiatry, 4,

183–185. https://doi.org/10.1016/S2215-0366(16)30371-6

Al-barrak, L., Kanjo, E., & Younis, E. M. G. (2017). NeuroPlace: Categorizing urban places

according to mental states. PLoS One, 12(9), Article e0183890. https://doi.org/

10.1371/journal.pone.0183890

Allahverdy, A., & Jafari, A. H. (2016). Non-auditory effect of noise pollution and its risk

on human brain activity in different audio frequency using electroencephalogram

complexity. Iranian Journal of Public Health, 45(10), 1332–1339.

Ancora, L. A., Blanco-Mora, D. A., Alves, I., Bonif´

acio, A., Morgado, P., & Miranda, B.

(2022). Cities and neuroscience research: A systematic literature review. Frontiers in

Psychiatry, 13, Article 983352. https://doi.org/10.3389/fpsyt.2022.983352

Asim, F., Chani, P. S., Shree, V., & Rai, S. (2023). Restoring the mind: A

neuropsychological investigation of university campus built environment aspects for

student well-being. Building and Environment, 244, Article 110810. https://doi.org/

10.1016/j.buildenv.2023.110810

Aspinall, P., Mavros, P., Coyne, R., & Roe, J. (2015). The urban brain: Analysing outdoor

physical activity with mobile EEG. British Journal of Sports Medicine, 49(4), 272–276.

https://doi.org/10.1136/bjsports-2012-091877

Banaei, M., Hatami, J., Yazdanfar, A., & Gramann, K. (2017). Walking through

architectural spaces: The impact of interior forms on human brain dynamics.

Frontiers in Human Neuroscience, 11, 477. https://doi.org/10.3389/

fnhum.2017.00477

Barnes, L., Davidson, M. J., & Alais, D. (2023). The speed and phase of locomotion dictate

saccade probability and simultaneous low-frequency power spectra [Preprint].

Neuroscience. https://doi.org/10.1101/2023.06.22.546202

Bateson, A. D., Baseler, H. A., Paulson, K. S., Ahmed, F., & Asghar, A. U. R. (2017).

Categorisation of mobile EEG: A researcher’s perspective. BioMed Research

International, 2017, 1–15. https://doi.org/10.1155/2017/5496196

Baumann, O., & Brooks-Cederqvist, B. (2023). Multimodal assessment of effects of urban

environments on psychological wellbeing. Heliyon, 9(6), Article e16433. https://doi.

org/10.1016/j.heliyon.2023.e16433

Buzs´

aki, G. (2006). Rhythms of the brain. Oxford University Press.

Cahn, B. R., & Polich, J. (2013). Meditation states and traits: EEG, ERP, and

neuroimaging studies. Psychology of Consciousness: Theory, Research, and Practice, 1

(S), 48–96. https://doi.org/10.1037/2326-5523.1.S.48

Charalambous, E., & Djebbara, Z. (2023). On natural attunement: Shared rhythms

between the brain and the environment. Neuroscience & Biobehavioral Reviews, 155,

Article 105438. https://doi.org/10.1016/j.neubiorev.2023.105438

Chaumon, M., Bishop, D. V. M., & Busch, N. A. (2015). A practical guide to the selection

of independent components of the electroencephalogram for artifact correction.

Journal of Neuroscience Methods, 250, 47–63. https://doi.org/10.1016/j.

jneumeth.2015.02.025

Chen, Z., He, Y., & Yu, Y. (2016). Enhanced functional connectivity properties of human

brains during in-situ nature experience. PeerJ, 4, Article e2210. https://doi.org/

10.7717/peerj.2210

Chi, Y. M., Jung, T.-P., & Cauwenberghs, G. (2010). Dry-contact and noncontact

biopotential electrodes: Methodological review. IEEE Reviews in Biomedical

Engineering, 3, 106–119. https://doi.org/10.1109/RBME.2010.2084078

Choi, J.-Y., Park, S.-A., Jung, S.-J., Lee, J.-Y., Son, K.-C., An, Y.-J., & Lee, S.-W. (2016).

Physiological and psychological responses of humans to the index of greenness of an

interior space. Complementary Therapies in Medicine, 28, 37–43. https://doi.org/

10.1016/j.ctim.2016.08.002

Cohen, M. X. (2014). Analyzing neural time series data: Theory and practice. MIT Press. htt

ps://books.google.de/books?id=rDKkAgAAQBAJ.

Croft, R. J., & Barry, R. J. (2000). Removal of ocular artifact from the EEG: A review.

Neurophysiologie Clinique/Clinical Neurophysiology, 30(1), 5–19. https://doi.org/

10.1016/S0987-7053(00)00055-1

Debener, S., Minow, F., Emkes, R., Gandras, K., & De Vos, M. (2012). How about taking a

low-cost, small, and wireless EEG for a walk? Psychophysiology, 49(11), 1617–1621.

https://doi.org/10.1111/j.1469-8986.2012.01471.x

Delorme, A., & Makeig, S. (2004). Eeglab: An open source toolbox for analysis of single-

trial EEG dynamics including independent component analysis. Journal of

Neuroscience Methods, 134(1), 9–21. https://doi.org/10.1016/j.

jneumeth.2003.10.009

Dimitrov-Discher, A., Gu, L., Pandit, L., Veer, I. M., Walter, H., Adli, M., & Kn¨

oll, M.

(2023). Stress and streets: How the network structure of streets is associated with

stress-related brain activation. Journal of Environmental Psychology, 91, Article

102142. https://doi.org/10.1016/j.jenvp.2023.102142

Ding, N., Zhong, Y., Li, J., Xiao, Q., Zhang, S., & Xia, H. (2022). Visual preference of plant

features in different living environments using eye tracking and EEG. PLoS One, 17

(12), Article e0279596. https://doi.org/10.1371/journal.pone.0279596

Djebbara, Z., Fich, L. B., & Gramann, K. (2021). The brain dynamics of architectural

affordances during transition. Scientific Reports, 11(1). https://doi.org/10.1038/

s41598-021-82504-w. Article 1.

Djebbara, Z., Fich, L. B., Petrini, L., & Gramann, K. (2019). Sensorimotor brain dynamics

reflect architectural affordances. Proceedings of the National Academy of Sciences, 116

(29), 14769–14778. https://doi.org/10.1073/pnas.1900648116

Djebbara, Z., & Kalantari, S. (2023). Affordances and curvature preference: The case of

real objects and spaces. Annals of the New York Academy of Sciences, 1527(1), 14–19.

https://doi.org/10.1111/nyas.15038

Do, T. T. N., Lin, C. T., & Gramann, K. (2021). Human brain dynamics in active spatial

navigation. Scientific Reports, 11(1), 13036.

Ducao, A., Koen, I., Guo, Z., Frank, J., Willard, C., & Kam, J. (2020). Multimer: Modeling

neurophysiological experience in public urban space. International Journal of

Community Well-Being, 3(4), 465–490. https://doi.org/10.1007/s42413-020-00082-

Duvinage, M., Castermans, T., Dutoit, T., Petieau, M., Hoellinger, T., Saedeleer, C. D.,

Seetharaman, K., & Cheron, G. (2012). A P300-based quantitative comparison between

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

the Emotiv Epoc headset and a medical EEG device. Biomedical engineering/765:

Telehealth/766: Assistive technologies. Biomedical engineering. Austria: Innsbruck.

https://doi.org/10.2316/P.2012.764-071

Eberhard, J. P. (2009a). Applying neuroscience to architecture. Neuron, 62(6), 753–756.

https://doi.org/10.1016/j.neuron.2009.06.001

Eberhard, J. P. (2009b). Brain landscape: The coexistance of neuroscience and architecture.

Oxford University Press.

Elsadek, M., Shao, Y., & Liu, B. (2021). Benefits of indirect contact with nature on the

physiopsychological well-being of elderly people. HERD: Health Environments

Research & Design Journal, 14(4), 227–241. https://doi.org/10.1177/

19375867211006654

Ergan, S., Radwan, A., Zou, Z., Tseng, H., & Han, X. (2019). Quantifying human

experience in architectural spaces with integrated virtual reality and body sensor

networks. Journal of Computing in Civil Engineering, 33(2), Article 04018062. https://

doi.org/10.1061/(ASCE)CP.1943-5487.0000812

Erkan, ˙

I. (2018). Examining wayfinding behaviours in architectural spaces using brain

imaging with electroencephalography (EEG). Architectural Science Review, 61(6),

410–428. https://doi.org/10.1080/00038628.2018.1523129

Erkan, I. (2023). A neuro-cognitive perspective on urban behavior of people with

different moods. Open House International. https://doi.org/10.1108/OHI-10-2022-

0252

Gable, P., Miller, M., & Bernat, E. (Eds.). (2022). The Oxford handbook of EEG frequency.

Oxford University Press.

Gao, Z., Zhu, G., & Qiu. (2019). Exploring psychophysiological restoration and individual

preference in the different environments based on virtual reality. International

Journal of Environmental Research and Public Health, 16(17), 3102. https://doi.org/

10.3390/ijerph16173102

Gorjan, D., Gramann, K., De Pauw, K., & Marusic, U. (2022). Removal of movement-

induced EEG artifacts: Current state of the art and guidelines. Journal of Neural

Engineering, 19(1), Article 011004. https://doi.org/10.1088/1741-2552/ac542c

Gramann, K., Ferris, D. P., Gwin, J., & Makeig, S. (2014). Imaging natural cognition in

action. International Journal of Psychophysiology, 91(1), 22–29. https://doi.org/

10.1016/j.ijpsycho.2013.09.003

Gramann, K., Gwin, J. T., Bigdely-Shamlo, N., Ferris, D. P., & Makeig, S. (2010). Visual

evoked responses during standing and walking. Frontiers in Human Neuroscience, 4.

https://doi.org/10.3389/fnhum.2010.00202

Gramann, K., Gwin, J. T., Ferris, D. P., Oie, K., Jung, T.-P., Lin, C.-T., Liao, L.-D., &

Makeig, S. (2011). Cognition in action: Imaging brain/body dynamics in mobile

humans. Reviews in the Neurosciences, 22(6). https://doi.org/10.1515/RNS.2011.047

Gramann, K., Jung, T.-P., Ferris, D. P., Lin, C.-T., & Makeig, S. (2014). Toward a new

cognitive neuroscience: Modeling natural brain dynamics. Frontiers in Human

Neuroscience, 8. https://doi.org/10.3389/fnhum.2014.00444

Gramann, K., & Plank, M. (2019). The use of electroencephalography in

Neuroergonomics. In Neuroergonomics (pp. 11–15). Elsevier. https://doi.org/

10.1016/B978-0-12-811926-6.00002-6.

Grassini, S., Revonsuo, A., Castellotti, S., Petrizzo, I., Benedetti, V., & Koivisto, M. (2019).

Processing of natural scenery is associated with lower attentional and cognitive load

compared with urban ones. Journal of Environmental Psychology, 62, 1–11. https://

doi.org/10.1016/j.jenvp.2019.01.007

Grassini, S., Segurini, G. V., & Koivisto, M. (2022). Watching nature videos promotes

physiological restoration: Evidence from the modulation of alpha waves in

electroencephalography. Frontiers in Psychology, 13, Article 871143. https://doi.org/

10.3389/fpsyg.2022.871143

Grima Murcia, M. D., Ortiz, M. J., L´

opez-Gordo, M. A., Ferr´

andez, J. M., S´

anchez

Ferrer, F., & Fern´

andez, E. (2019). Neural representation of different 3D

architectural images: An EEG study. Integrated Computer-Aided Engineering, 26(2),

197–205. https://doi.org/10.3233/ICA-180591

Gwin, J. T., Gramann, K., Makeig, S., & Ferris, D. P. (2010). Removal of movement

artifact from high-density EEG recorded during walking and running. Journal of

Neurophysiology, 103(6), 3526–3534. https://doi.org/10.1152/jn.00105.2010

Hagerhall, C. M., Laike, T., Küller, M., Marcheschi, E., Boydston, C., & Taylor, R. P.

(2015). Human physiological benefits of viewing nature: EEG responses to exact and

statistical fractal patterns. Nonlinear Dynamics, Psychology, and Life Sciences, 19(1),

1–12.

Hassan, A., Tao, J., Li, G., Jiang, M., Aii, L., Zhihui, J., Zongfang, L., & Qibing, C. (2018).

Effects of walking in bamboo forest and city environments on brainwave activity in

young adults. Evidence-based Complementary and Alternative Medicine. , Article

e9653857. https://doi.org/10.1155/2018/9653857, 2018.

Herman, K., Ciechanowski, L., & Przegali´

nska, A. (2021). Emotional well-being in urban

wilderness: Assessing states of calmness and alertness in informal green spaces (IGSs)

with muse—portable EEG headband. Sustainability, 13(4), 2212. https://doi.org/

10.3390/su13042212

Higuera-Trujillo, J. L., Llinares Mill´

an, C., Monta˜

nana I Avi˜

n´

o, A., & Rojas, J.-C. (2020).

Multisensory stress reduction: A neuro-architecture study of paediatric waiting

rooms. Building Research & Information, 48(3), 269–285. https://doi.org/10.1080/

09613218.2019.1612228

Hollander, J., & Foster, V. (2016). Brain responses to architecture and planning: A

preliminary neuro-assessment of the pedestrian experience in Boston, Massachusetts.

Architectural Science Review, 59(6), 474–481. https://doi.org/10.1080/

00038628.2016.1221499

Holleman, G. A., Hooge, I. T. C., Kemner, C., & Hessels, R. S. (2020). The ‘real-world

approach’ and its problems: A critique of the term ecological validity. Frontiers in

Psychology, 11, 721. https://doi.org/10.3389/fpsyg.2020.00721

Hu, M., & Roberts, J. (2020). Built environment evaluation in virtual reality

environments—a cognitive neuroscience approach. Urban Science, 4(4), 48. https://

doi.org/10.3390/urbansci4040048

Hu, M., Simon, M., Fix, S., Vivino, A. A., & Bernat, E. (2021). Exploring a sustainable

building’s impact on occupant mental health and cognitive function in a virtual

environment. Scientific Reports, 11(1), 5644. https://doi.org/10.1038/s41598-021-

85210-9

Imperatori, C., Massullo, C., De Rossi, E., Carbone, G. A., Theodorou, A., Scopelliti, M.,

Romano, L., Del Gatto, C., Allegrini, G., Carrus, G., & Panno, A. (2023). Exposure to

nature is associated with decreased functional connectivity within the distress

network: A resting state EEG study. Frontiers in Psychology, 14, Article 1171215.

https://doi.org/10.3389/fpsyg.2023.1171215

Jackson, A. F., & Bolger, D. J. (2014). The neurophysiological bases of EEG and EEG

measurement: A review for the rest of us. Psychophysiology, 51(11), 1061–1071.

https://doi.org/10.1111/psyp.12283

Jiang, X., Bian, G.-B., & Tian, Z. (2019). Removal of artifacts from EEG signals: A review.

Sensors, 19(5), 987. https://doi.org/10.3390/s19050987

Jiang, M., Hassan, A., Chen, Q., & Liu, Y. (2020). Effects of different landscape visual

stimuli on psychophysiological responses in Chinese students. Indoor and Built

Environment, 29(7), 1006–1016. https://doi.org/10.1177/1420326X19870578

Jung, D., Kim, D. I., & Kim, N. (2023). Bringing nature into hospital architecture:

Machine learning-based EEG analysis of the biophilia effect in virtual reality. Journal

of Environmental Psychology, 89, Article 102033. https://doi.org/10.1016/j.

jenvp.2023.102033

Jung, T.-P., Makeig, S., Westerfield, M., Townsend, J., Courchesne, E., & Sejnowski, T. J.

(2000). Removal of eye activity artifacts from visual event-related potentials in

normal and clinical subjects. Clinical Neurophysiology, 111(10), 1745–1758. https://

doi.org/10.1016/S1388-2457(00)00386-2

Jungnickel, E., & Gramann, K. (2016). Mobile brain/body imaging (MoBI) of physical

interaction with dynamically moving objects. Frontiers in Human Neuroscience, 10.

https://doi.org/10.3389/fnhum.2016.00306

Kamibayashi, L. K., & Richmond, F. J. R. (1998). Morphometry of human neck muscles.

Spine, 23(12). https://journals.lww.com/spinejournal/fulltext/1998/06150/morph

ometry_of_human_neck_muscles.5.aspx.

Karandinou, A., & Turner, L. (2017). Architecture and neuroscience; what can the EEG

recording of brain activity reveal about a walk through everyday spaces?

International Journal of Parallel, Emergent and Distributed Systems, 32(sup1), S54–S65.

https://doi.org/10.1080/17445760.2017.1390089

Keil, A., Debener, S., Gratton, G., Jungh¨

ofer, M., Kappenman, E. S., Luck, S. J., Luu, P.,

Miller, G. A., & Yee, C. M. (2014). Committee report: Publication guidelines and

recommendations for studies using electroencephalography and

magnetoencephalography. Psychophysiology, 51(1), 1–21. https://doi.org/10.1111/

psyp.12147

Kennedy, D. P., & Adolphs, R. (2011). Stress and the city. Nature, 474(7352), 452–453.

https://doi.org/10.1038/474452a

Kim, T.-H., Jeong, G.-W., Baek, H.-S., Kim, G.-W., Sundaram, T., Kang, H.-K., Lee, S.-W.,

Kim, H.-J., & Song, J.-K. (2010). Human brain activation in response to visual

stimulation with rural and urban scenery pictures: A functional magnetic resonance

imaging study. Science of The Total Environment, 408(12), 2600–2607. https://doi.

org/10.1016/j.scitotenv.2010.02.025

Klimesch, W., Doppelmayr, M., Russegger, H., Pachinger, T., & Schwaiger, J. (1998).

Induced alpha band power changes in the human EEG and attention. Neuroscience

letters, 244(2), 73–76.

Klug, M., & Gramann, K. (2021). Identifying key factors for improving ICA-based

decomposition of EEG data in mobile and stationary experiments. European Journal

of Neuroscience, 54(12), 8406–8420. https://doi.org/10.1111/ejn.14992

Klug, M., Jeung, S., Wunderlich, A., Gehrke, L., Protzak, J., Djebbara, Z., Argubi-

Wollesen, A., Wollesen, B., & Gramann, K. (2022). The BeMoBIL Pipeline for

automated analyses of multimodal mobile brain and body imaging data.

Neuroscience. https://doi.org/10.1101/2022.09.29.510051 [Preprint].

Kühn, S., Forlim, C. G., Lender, A., Wirtz, J., & Gallinat, J. (2021). Brain functional

connectivity differs when viewing pictures from natural and built environments

using fMRI resting state analysis. Scientific Reports, 11(1), 4110. https://doi.org/

10.1038/s41598-021-83246-5

Lachaux, J.-P., Rodriguez, E., Martinerie, J., & Varela, F. J. (1999). Measuring phase

synchrony in brain signals. Human Brain Mapping, 8(4), 194–208. https://doi.org/

10.1002/(SICI)1097-0193(1999)8:4<194::AID-HBM4>3.0.CO;2-C

Ladouce, S., Mustile, M., Ietswaart, M., & Dehais, F. (2022). Capturing cognitive events

embedded in the real world using mobile electroencephalography and eye-tracking.

Journal of Cognitive Neuroscience, 34(12), 2237–2255. https://doi.org/10.1162/jocn_

a_01903

Li, H., Dong, W., Wang, Z., Chen, N., Wu, J., Wang, G., & Jiang, T. (2021). Effect of a

virtual reality-based restorative environment on the emotional and cognitive

recovery of individuals with mild-to-moderate anxiety and depression. International

Journal of Environmental Research and Public Health, 18(17), 9053. https://doi.org/

10.3390/ijerph18179053

Li, J., Jin, Y., Lu, S., Wu, W., & Wang, P. (2020). Building environment information and

human perceptual feedback collected through a combined virtual reality (VR) and

electroencephalogram (EEG) method. Energy and Buildings, 224, Article 110259.

https://doi.org/10.1016/j.enbuild.2020.110259

Li, J., Wu, W., Jin, Y., Zhao, R., & Bian, W. (2021). Research on environmental comfort

and cognitive performance based on EEG+VR+LEC evaluation method in

underground space. Building and Environment, 198, Article 107886. https://doi.org/

10.1016/j.buildenv.2021.107886

Li, H., Xie, H., & Woodward, G. (2021). Soundscape components, perceptions, and EEG

reactions in typical mountainous urban parks. Urban Forestry and Urban Greening, 64,

Article 127269. https://doi.org/10.1016/j.ufug.2021.127269

Li, Y., Zhang, J., Jiang, B., Li, H., & Zhao, B. (2023). Do all types of restorative

environments in the urban park provide the same level of benefits for young adults?

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

A field experiment in nanjing, China. Forests, 14(7), 1400. https://doi.org/10.3390/

f14071400

Lin, W., Chen, Q., Jiang, M., Tao, J., Liu, Z., Zhang, X., Wu, L., Xu, S., Kang, Y., &

Zeng, Q. (2020). Sitting or walking? Analyzing the neural emotional indicators of

urban green space behavior with mobile EEG. Journal of Urban Health, 97(2),

191–203. https://doi.org/10.1007/s11524-019-00407-8

Lopes da Silva, F. (2013). EEG and MEG: Relevance to neuroscience. Neuron, 80(5),

1112–1128. https://doi.org/10.1016/j.neuron.2013.10.017

Luck, S. J. (2005). An introduction to the event-related potential technique. MIT Press.

Luft, C. D. B., & Bhattacharya, J. (2015). Aroused with heart: Modulation of heartbeat

evoked potential by arousal induction and its oscillatory correlates. Scientific Reports,

5(1), Article 15717. https://doi.org/10.1038/srep15717

Lun-De, L., Lin, C.-T., McDowell, K., Wickenden, A. E., Gramann, K., Jung, T.-P., Ko, L.-

W., & Chang, J.-Y. (2012). Biosensor technologies for augmented brain–computer

interfaces in the next decades. Proceedings of the IEEE, 100(Special Centennial Issue),

1553–1566. https://doi.org/10.1109/JPROC.2012.2184829

Mahamane, S., Wan, N., Porter, A., Hancock, A. S., Campbell, J., Lyon, T. E., &

Jordan, K. E. (2020). Natural categorization: Electrophysiological responses to

viewing natural versus built environments. Frontiers in Psychology, 11. https://www.

frontiersin.org/articles/10.3389/fpsyg.2020.00990.

Makeig, S., Bell, A. J., Jung, T.-P., & Sejnowski, T. J. (1995). Independent component

analysis of electroencephalographic data. Advances in Neural Information Processing

Systems, 8.

Makeig, S., Gramann, K., Jung, T.-P., Sejnowski, T. J., & Poizner, H. (2009). Linking

brain, mind and behavior. International Journal of Psychophysiology, 73(2), 95–100.

https://doi.org/10.1016/j.ijpsycho.2008.11.008

Manohare, M., Rajasekar, E., & Parida, M. (2023a). Analysing the change in brain waves

due to heterogeneous road traffic noise exposure using electroencephalography

measurements. Noise and Health, 25(116), 36–54.

Manohare, M., Rajasekar, E., & Parida, M. (2023b). Electroencephalography based

classification of emotions associated with road traffic noise using Gradient boosting

algorithm. Applied Acoustics, 206, 109306. https://doi.org/10.1016/j.

apacoust.2023.109306

Mathewson, K. E., Harrison, T. J. L., & Kizuk, S. A. D. (2017). High and dry? Comparing

active dry EEG electrodes to active and passive wet electrodes. Psychophysiology, 54

(1), 74–82. https://doi.org/10.1111/psyp.12536

Mavros, P., J W¨

alti, M., Nazemi, M., Ong, C. H., & H¨

olscher, C. (2022). A mobile EEG

study on the psychophysiological effects of walking and crowding in indoor and

outdoor urban environments. Scientific Reports, 12(1). https://doi.org/10.1038/

s41598-022-20649-y. Article 1.

Mehta, R. K., & Parasuraman, R. (2013). Neuroergonomics: A review of applications to

physical and cognitive work. Frontiers in Human Neuroscience, 7. https://doi.org/

10.3389/fnhum.2013.00889

Mostafavi, A., Cruz-Garza, J. G., & Kalantari, S. (2023). Enhancing lighting design

through the investigation of illuminance and correlated color Temperature’s effects

on brain activity: An EEG-VR approach. Journal of Building Engineering, 75, Article

106776. https://doi.org/10.1016/j.jobe.2023.106776

Naghibi Rad, P., Shahroudi, A. A., Shabani, H., Ajami, S., & Lashgari, R. (2019).

Encoding pleasant and unpleasant expression of the architectural window shapes: An

ERP study. Frontiers in Behavioral Neuroscience, 13, 186. https://doi.org/10.3389/

fnbeh.2019.00186

Neale, C., Aspinall, P., Roe, J., Tilley, S., Mavros, P., Cinderby, S., Coyne, R., Thin, N.,

Bennett, G., & Thompson, C. W. (2017). The aging urban brain: Analyzing outdoor

physical activity using the Emotiv affectiv suite in older people. Journal of Urban

Health, 94(6), 869–880. https://doi.org/10.1007/s11524-017-0191-9

Neale, C., Aspinall, P., Roe, J., Tilley, S., Mavros, P., Cinderby, S., Coyne, R., Thin, N., &

Ward Thompson, C. (2020). The impact of walking in different urban environments

on brain activity in older people. Cities & Health, 4(1), 94–106. https://doi.org/

10.1080/23748834.2019.1619893

Nie, W., Jia, J., Wang, M., Sun, J., & Li, G. (2022). Research on the impact of panoramic

green view index of virtual reality environments on individuals’ pleasure level based

on EEG experiment. Landscape Architecture Frontiers, 10(2), 36. https://doi.org/

10.15302/J-LAF-1-020059

Niso, G., Romero, E., Moreau, J. T., Araujo, A., & Krol, L. R. (2023). Wireless EEG: A

survey of systems and studies. NeuroImage, 269, Article 119774. https://doi.org/

10.1016/j.neuroimage.2022.119774

Nunez, P. L., & Srinivasan, R. (2006). Electric fields of the brain: The neurophysics of EEG

(2nd ed.). Oxford University Press.

Olszewska-Guizzo, A., Escoffier, N., Chan, J., & Puay Yok, T. (2018). Window view and

the brain: Effects of floor level and green cover on the alpha and beta rhythms in a

passive exposure EEG experiment. International Journal of Environmental Research and

Public Health, 15(11), 2358. https://doi.org/10.3390/ijerph15112358

Olszewska-Guizzo, A., Fogel, A., Escoffier, N., Sia, A., Nakazawa, K., Kumagai, A., Dan, I.,

& Ho, R. (2022). Therapeutic garden with contemplative features induces desirable

changes in mood and brain activity in depressed adults. Frontiers in Psychiatry, 13,

Article 757056. https://doi.org/10.3389/fpsyt.2022.757056

Olszewska-Guizzo, A. A., Paiva, T. O., & Barbosa, F. (2018). Effects of 3D contemplative

landscape videos on brain activity in a passive exposure EEG experiment. Frontiers in

Psychiatry, 9. https://www.frontiersin.org/articles/10.3389/fpsyt.2018.00317.

Olszewska-Guizzo, A., Sia, A., Fogel, A., & Ho, R. (2020). Can exposure to certain urban

green spaces trigger frontal alpha asymmetry in the brain?—preliminary findings

from a passive task EEG study. International Journal of Environmental Research and

Public Health, 17(2), 394. https://doi.org/10.3390/ijerph17020394

Onton, J., Delorme, A., & Makeig, S. (2005). Frontal midline EEG dynamics during

working memory. Neuroimage, 27(2), 341–356.

Onton, J., & Makeig, S. (2006). Information-based modeling of event-related brain

dynamics. Progress in Brain Research, 159, 99–120. https://doi.org/10.1016/S0079-

6123(06)59007-7. Elsevier.

Oostenveld, R., & Praamstra, P. (2001). The five percent electrode system for high-

resolution EEG and ERP measurements. Clinical Neurophysiology, 112(4), 713–719.

https://doi.org/10.1016/S1388-2457(00)00527-7

Pion-Tonachini, L., Kreutz-Delgado, K., & Makeig, S. (2019). The ICLabel dataset of

electroencephalographic (EEG) independent component (IC) features. Data in Brief,

25, Article 104101. https://doi.org/10.1016/j.dib.2019.104101

Qi, Y., Chen, Q., Lin, F., Liu, Q., Zhang, X., Guo, J., Qiu, L., & Gao, T. (2022).

Comparative study on birdsong and its multi-sensory combinational effects on

physio-psychological restoration. Journal of Environmental Psychology, 83, Article

101879. https://doi.org/10.1016/j.jenvp.2022.101879

Qin, J., Zhou, X., Sun, C., Leng, H., & Lian, Z. (2013). Influence of green spaces on

environmental satisfaction and physiological status of urban residents. Urban

Forestry and Urban Greening, 12(4), 490–497. https://doi.org/10.1016/j.

ufug.2013.05.005

Reece, R., Bornioli, A., Bray, I., & Alford, C. (2022). Exposure to green and historic urban

environments and mental well-being: Results from EEG and psychometric outcome

measures. International Journal of Environmental Research and Public Health, 19(20).

https://doi.org/10.3390/ijerph192013052. Article 20.

Reeves, J. P., Knight, A. T., Strong, E. A., Heng, V., Neale, C., Cromie, R., &

Vercammen, A. (2019). The application of wearable technology to quantify health

and wellbeing Co-benefits from urban wetlands. Frontiers in Psychology, 10, 1840.

https://doi.org/10.3389/fpsyg.2019.01840

Rodriguez, E., George, N., Lachaux, J.-P., Martinerie, J., Renault, B., & Varela, F. J.

(1999). Perception’s shadow: Long-distance synchronization of human brain

activity. Nature, 397(6718), 430–433. https://doi.org/10.1038/17120

Roe, J. J., Aspinall, P. A., Mavros, P., & Coyne, R. (2013). Engaging the brain: The impact

of natural versus urban scenes using novel EEG methods in an experimental setting.

Environmental Sciences, 1, 93–104. https://doi.org/10.12988/es.2013.3109

Rounds, J. D., Cruz-Garza, J. G., & Kalantari, S. (2020). Using posterior EEG theta band

to assess the effects of architectural designs on landmark recognition in an urban

setting. Frontiers in Human Neuroscience, 14. https://www.frontiersin.org/articles/10

.3389/fnhum.2020.584385.

Scanlon, J. E. M., Jacobsen, N. S. J., Maack, M. C., & Debener, S. (2021). Does the

electrode amplification style matter? A comparison of active and passive EEG system

configurations during standing and walking. European Journal of Neuroscience, 54

(12), 8381–8395. https://doi.org/10.1111/ejn.15037

Seeber, M., Scherer, R., Wagner, J., Solis-Escalante, T., & Müller-Putz, G. R. (2014). EEG

beta suppression and low gamma modulation are different elements of human

upright walking. Frontiers in human neuroscience, 8, 485.

Shamay-Tsoory, S. G., & Mendelsohn, A. (2019). Real-Life neuroscience: An ecological

approach to brain and behavior research. Perspectives on Psychological Science, 14(5),

841–859. https://doi.org/10.1177/1745691619856350

Shan, X., Yang, E.-H., Zhou, J., & Chang, V. W. C. (2019). Neural-signal

electroencephalogram (EEG) methods to improve human-building interaction under

different indoor air quality. Energy and Buildings, 197, 188–195. https://doi.org/

10.1016/j.enbuild.2019.05.055

Shemesh, A., Leisman, G., Bar, M., & Grobman, Y. J. (2021). A neurocognitive study of

the emotional impact of geometrical criteria of architectural space. Architectural

Science Review, 64(4), 394–407. https://doi.org/10.1080/00038628.2021.1940827

Shemesh, A., Leisman, G., Bar, M., & Grobman, Y. J. (2022). The emotional influence of

different geometries in virtual spaces: A neurocognitive examination. Journal of

Environmental Psychology, 81, Article 101802. https://doi.org/10.1016/j.

jenvp.2022.101802

Shemesh, A., Talmon, R., Karp, O., Amir, I., Bar, M., & Grobman, Y. J. (2017). Affective

response to architecture – investigating human reaction to spaces with different

geometry. Architectural Science Review, 60(2), 116–125. https://doi.org/10.1080/

00038628.2016.1266597

Smith, E. E., Reznik, S. J., Stewart, J. L., & Allen, J. J. (2017). Assessing and

conceptualizing frontal EEG asymmetry: An updated primer on recording,

processing, analyzing, and interpreting frontal alpha asymmetry. International

Journal of Psychophysiology, 111, 98–114.

Song, R., Chen, Q., Zhang, Y., Jia, Q., He, H., Gao, T., & Qiu, L. (2022).

Psychophysiological restorative potential in cancer patients by virtual reality (VR)-

based perception of natural environment. Frontiers in Psychology, 13, Article

1003497. https://doi.org/10.3389/fpsyg.2022.1003497

Spiers, H. J., & Maguire, E. A. (2007). Decoding human brain activity during real-world

experiences. Trends in Cognitive Sciences, 11(8), 356–365. https://doi.org/10.1016/j.

tics.2007.06.002

Suhaimi, N. S., Mountstephens, J., & Teo, J. (2020). EEG-based emotion recognition: A

state-of-the-art review of current trends and opportunities. Computational Intelligence

and Neuroscience, 1–19. https://doi.org/10.1155/2020/8875426, 2020.

Tilley, S., Neale, C., Patuano, A., & Cinderby, S. (2017). Older people’s experiences of

mobility and mood in an urban environment: A mixed methods approach using

electroencephalography (EEG) and interviews. International Journal of Environmental

Research and Public Health, 14(2). https://doi.org/10.3390/ijerph14020151. Article

Ulrich, R. S. (1981). Natural versus urban scenes: Some psychophysiological effects.

Environment and Behavior, 13(5), 523–556. https://doi.org/10.1177/

0013916581135001

Vallet, W., & Van Wassenhove, V. (2023). Can cognitive neuroscience solve the lab-

dilemma by going wild? Neuroscience & Biobehavioral Reviews, 155, Article 105463.

https://doi.org/10.1016/j.neubiorev.2023.105463

K. Gramann

Journal of Environmental Psychology 96 (2024) 102308

Van Gerven, M., & Jensen, O. (2009). Attention modulations of posterior alpha as a

control signal for two-dimensional brain–computer interfaces. Journal of

Neuroscience Methods, 179(1), 78–84. https://doi.org/10.1016/j.

jneumeth.2009.01.016

Vecchiato, G., Jelic, A., Tieri, G., Maglione, A. G., De Matteis, F., & Babiloni, F. (2015).

Neurophysiological correlates of embodiment and motivational factors during the

perception of virtual architectural environments. Cognitive Processing, 16(S1),

425–429. https://doi.org/10.1007/s10339-015-0725-6

Vecchiato, G., Tieri, G., Jelic, A., De Matteis, F., Maglione, A. G., & Babiloni, F. (2015).

Electroencephalographic correlates of sensorimotor integration and embodiment

during the appreciation of virtual architectural environments. Frontiers in Psychology,

6. https://doi.org/10.3389/fpsyg.2015.01944

Vijayan, V. T., & Embi, M. R. (2019). Probing phenomenological experiences through

electroencephalography brainwave signals in neuroarchitecture study. International

Journal of Built Environment and Sustainability, 6(3), 11–20. https://doi.org/

10.11113/ijbes.v6.n3.360

Wang, Y., Qu, H., Bai, T., Chen, Q., Li, X., Luo, Z., Lv, B., & Jiang, M. (2021). Effects of

variations in color and organ of color expression in urban ornamental bamboo

landscapes on the physiological and psychological responses of college students.

International Journal of Environmental Research and Public Health, 18(3), 1151.

https://doi.org/10.3390/ijerph18031151

Wascher, E., Arnau, S., Gutberlet, M., Chuang, L. L., Rinkenauer, G., & Reiser, J. E.

(2022). Visual demands of walking are reflected in eye-blink-evoked EEG-activity.

Applied Sciences, 12(13), 6614. https://doi.org/10.3390/app12136614

Wascher, E., Heppner, H., & Hoffmann, S. (2014). Towards the measurement of event-

related EEG activity in real-life working environments. International Journal of

Psychophysiology, 91(1), 3–9. https://doi.org/10.1016/j.ijpsycho.2013.10.006

Westbrook, K. E., Nessel, T. A., Hohman, M. H., & Varacallo, M. (2023). Anatomy, head

and neck: Facial muscles. In StatPearls. StatPearls Publishing http://www.ncbi.nlm.

nih.gov/books/NBK493209/.

Widmann, A., Schr¨

oger, E., & Maess, B. (2015). Digital filter design for

electrophysiological data–a practical approach. Journal of neuroscience methods, 250,

34–46.

Winkler, I., Haufe, S., & Tangermann, M. (2011). Automatic classification of artifactual

ICA-components for artifact removal in EEG signals. Behavioral and Brain Functions, 7

(1), 30. https://doi.org/10.1186/1744-9081-7-30

Wunderlich, A., & Gramann, K. (2021). Eye movement-related brain potentials during

assisted navigation in real-world environments. European Journal of Neuroscience, 54

(12), 8336–8354. https://doi.org/10.1111/ejn.15095

Xiong, X., Jin, H., Hu, W., Zeng, C., Huang, Q., Cui, X., Zhang, M., & Jin, Y. (2023).

Benefits of Jasminum polyanthum’s natural aromas on human emotions and moods.

Urban Forestry and Urban Greening, 86, Article 128010. https://doi.org/10.1016/j.

ufug.2023.128010

Yang, F., Bao, Z. Y., & Zhu, Z. J. (2011). An assessment of psychological noise reduction

by landscape plants. International Journal of Environmental Research and Public Health,

8(4), 1032–1048. https://doi.org/10.3390/ijerph8041032

Zander, T. O., Andreessen, L. M., Berg, A., Bleuel, M., Pawlitzki, J., Zawallich, L.,

Krol, L. R., & Gramann, K. (2017). Evaluation of a dry EEG system for application of

passive brain-computer interfaces in autonomous driving. Frontiers in Human

Neuroscience, 11. https://doi.org/10.3389/fnhum.2017.00078

Zeileis, A., Fisher, J. C., Hornik, K., Ihaka, R., McWhite, C. D., Murrell, P., Stauffer, R., &

Wilke, C. O. (2019). colorspace: A Toolbox for Manipulating and Assessing Colors and

Palettes (arXiv:1903.06490). arXiv. http://arxiv.org/abs/1903.06490.

Zhang, S., Feng, X., & Shen, Y. (2021). Quantifying auditory presence using

electroencephalography. Applied Sciences, 11(21), Article 10461. https://doi.org/

10.3390/app112110461

Zhang, Z., Zhuo, K., Wei, W., Li, F., Yin, J., & Xu, L. (2021). Emotional responses to the

visual patterns of urban streets: Evidence from physiological and subjective

indicators. International Journal of Environmental Research and Public Health, 18(18),

9677. https://doi.org/10.3390/ijerph18189677

Zou, Z., Ergan, S., Fisher-Gewirtzman, D., & Curtis, C. (2021). Quantifying the impact of

urban form on human experience: Experiment using virtual environments and

electroencephalogram. Journal of Computing in Civil Engineering, 35(3), Article

04021004. https://doi.org/10.1061/(ASCE)CP.1943-5487.0000966

K. Gramann