Space Syntax Lab Seminars 2025-26

Exploring Urban Activity with Gravity-Based Betweenness and Social Media Evaluation

Urban activity intensity is central to evaluating the social sustainability of cities. While traditionally explained by the physical built environment, the rise of social media has introduced user-generated content as an additional factor shaping human activity by reflecting perceived place attractiveness. This study investigates the interplay between street network configuration and social media evaluations in shaping urban activity intensity within the urban core of Beijing. 

To capture this interplay, we propose a Gravity-Based Betweenness Centrality (GBC) model that integrates spatial interaction mechanisms, route choice behaviour, and search radius to quantify the effects of social media evaluations. Using Baidu heatmap data, we assess model performance with Random Forest and interpret correlations between variables and activity intensity through explainable machine learning. Results show that the GBC model outperforms baseline models by 1.2–2.1%. Moreover, GBC variables show a positive and temporally varying correlation with activity intensity at the local spatial scale. Importantly, findings reveal that positive social media evaluations can mitigate the negative effects of low physical accessibility, allowing less central areas to attract urban activity. 

Our approach provides insights for understanding contemporary urban dynamics in the digital era and supports more human-centred and sustainable planning strategies. 

Speaker

Linghao Wang is a PhD candidate in the Department of Architecture at the National University of Singapore. He holds a Master’s degree from Southeast University in Nanjing and a Bachelor’s degree from Wuhan University of Technology. His research approaches the city as a complex system, with particular interests in street network analysis, urban morphology, social sensing, and the modelling of spatiotemporal activity patterns. His current work focuses on leveraging large language models to extract place semantics from social media data and examining how these place semantics interact with urban centrality to shape urban activity patterns. Through this integrative framework, his research aims to bridge urban morphology, data-driven methods, and human activity analysis, contributing to a more nuanced understanding of the relationships between urban form, place meaning, and urban dynamics. 

How to join

Register for the event via Zoom

The Closed Open Kitchens: Spatial and Visual Analysis of Iranian Kitchens over the Last Century

Over the past century, the domestic kitchen in Iran has transformed from a peripheral, concealed space into a more open and spatially connected zone; yet paradoxically, it remains isolating. This study investigates the spatial and visual evolution of kitchens in Tehran through a mixed-methods approach combining space syntax analysis and ethnographic insight. 19 case studies, spanning five distinct historical periods from the Qajar era to contemporary housing, were analysed using Convex Map Analysis and Visibility Graph Analysis in DepthmapX, complemented by interviews and archival materials. 

Findings indicate that, despite increasing visual exposure—particularly in the contemporary period—kitchens continue to exhibit high spatial segregation, as evidenced by consistently elevated Real Relative Asymmetry values. While open-plan layouts and broader visual fields afford greater awareness of domestic life, they fail to mitigate the sense of isolation reported by many occupants, especially women. Ethnographic accounts reveal divergent desires: some seek integration and control over adjacent spaces, while others avoid the kitchen altogether, preferring more versatile or private environments. 

By framing the kitchen as both a spatial artefact and a lived experience, this study challenges binary narratives of “open” versus “closed” kitchens. Instead, it exposes the complex interplay between spatial structure, domestic routines, and gendered agency. These insights contribute to broader debates on domestic architecture, visual control, and social inclusion, offering implications for more adaptable, inclusive housing design. 

Speaker

Dr Ramin Dehbandi is a researcher and lecturer in architecture and construction management, currently affiliated with Nottingham Trent University. His research bridges spatial analysis, gendered domesticity, and socio-technical systems in the built environment. Ramin’s recent publications explore topics such as the spatial isolation of Iranian kitchens, housing quality rating systems, and virtual design studio pedagogy. His award-winning work combines ethnographic methods with space syntax analysis to examine how architectural design reflects and reinforces social structures. He has received recognition from the Royal Institute of Chartered Surveyors and the TILT Student-Staff Co-Creation Fund for his contributions to research and pedagogy. 

How to join

Register for the event via Zoom

Let the Nurses Talk: Inpatient Spatial Design and Nurse Communication

This seminar describes the long, multidisciplinary process of enacting and testing an evidence-based design process for an inpatient ward in the United States, from design through construction to post-occupancy evaluation, with a focus on nurse communication. Care team communication is fundamental to most healthcare environments, but the dynamic ways that care team workers use space complicate knowing how best to design for communication in hospital settings. Moreover, disciplinary and professional knowledge in architecture is not necessarily constituted to frame solutions to such problems.

Working from nurses’ accounts of what was missing in the current space, the design team developed inpatient floorplan variations and then tested them using space syntax analysis. Design and construction proceeded based on the floorplan layout that better integrated visual fields and created more visual connections between nurse work areas. The post-occupancy comparative case study of the new and old inpatient units used floorplan-based measures of visibility, behaviour mapping, and social network analysis not only to test the efficacy of the new design, but also to gain insight into the relationship between floorplan layout and care team social networks of communication. Greater unit visibility appeared to create conditions for co-presence that enhanced care team communication. 

Speaker

Dr Julie Zook conducts research and scholarship at the interface of medicine and architecture. She is co-editor, with Prof Kerstin Sailer, of the open-access book The Covert Life of Hospital Architecture (UCL Press), which uses sociospatial analysis to uncover hospital design principles that can improve patient safety, care team collaboration, and corridor-system wayfinding. Her work on architecture and medicine also reflects on the historical and contemporary significance of social science to architectural thought and decision-making. 

How to join

Register for the event via Zoom

Travel-Time Accessibility and the Geography of Retail Concentration

Retail activities concentrate unevenly within metropolitan space, clustering in the most accessible areas. Yet much of the empirical literature measures “accessibility” through proximity to transport infrastructure or network topology alone, overlooking the role of actual travel times and service levels. This research addresses that gap by asking whether retail density is better explained by effective market reach, namely access to purchasing power. This idea is operationalised by combining gravity-based accessibility to income with network centrality measures (integration/choice analogues) and indicators of local land-use mix. 

The work estimates travel times for private and public transport using API-based routing and GTFS schedules, and then computes accessibility indices to income (and, for comparison, to population). Retail establishments, the dependent variable, are geocoded from administrative records. To capture non-linear effects and interactions, the study fits XGBoost models and uses GeoShapley to distinguish intrinsic feature effects from geography-dependent feature impacts. Accessibility to income via public transport emerges as a strong predictor of retail density, with land-use diversity also ranking highly. Accessibility to income by private transport and betweenness-like centrality add explanatory power, while population-based accessibility contributes comparatively little. 

These findings imply that commercial concentration reflects market reach shaped by service levels at least as much as corridor topology, motivating planning diagnostics focused on transit coverage, frequency, and integration. 

Speaker

Dr José B. Paiva Neto recently completed a PhD in Transport Engineering at COPPE/UFRJ, where he also earned an MSc, after graduating in Civil Engineering and undertaking international training at Cork Institute of Technology (Ireland). His work sits at the interface of urban form, accessibility, and public transport planning, with an emphasis on land-use/transport interaction as policy-evaluation tools. In parallel to academic research, he has contributed to university-linked applied projects and consulting engagements that provide technical assessments of transport services and mobility plans, translating analysis into practical inputs for planning and decision-making. He values reproducible, collaborative analytics and is motivated by impact-oriented, equity-aware planning practice. 

How to join

Register for the event via Zoom

Concealing Reproductive Tasks: An Inequality Genotype in Middle-Income Housing in Mexico

Social behaviour is reflected in spatial configurations. Traditionally, reproductive work and caregiving tasks have been gendered, with women often bearing a disproportionate share and these responsibilities frequently going unnoticed. With the rise of middle-income housing in Mexico, an important question arises: how do these housing layouts support, or hinder, equitable sharing of reproductive and care tasks, particularly in kitchens and laundry areas? 

To investigate this, the paper focuses on five Mexican cities known for significant middle-income housing production (Guadalajara, Mexico City, Monterrey, Querétaro and Toluca) and selects 25 case studies, consisting of five model homes from each city. Applying space syntax and the justified plan graphs, the paper analyses these homes both mathematically and graphically to map spatial inequalities. The findings reveal that laundry areas are poorly integrated, similar to bathrooms, whereas public and private spaces meet in the kitchen, which may enable more equitable domestic interactions and a fairer distribution of household responsibilities. 

Speakers

Dr Lucía Elizondo Jiménez is a Professor at the Tecnológico de Monterrey, where she has taught since 2002. She holds a PhD in Humanistic Studies and a degree in Architecture from the Tecnológico de Monterrey, as well as a Master’s in Design Studies from Harvard University’s Graduate School of Design. She is a member of the Sustainable Territorial Development Research Group, and her research focuses on housing appropriation and self-production. Her work has been presented at academic forums in Mexico and abroad and published in international journals. She serves on the boards of Cómo Vamos NL and the Sustainable Development Commission of Consejo Nuevo León, and she received the National Housing Award for Research from Mexico’s Secretariat of Territorial and Urban Development (SEDATU).

Dr Ruben Garnica-Monroy is an Associate Professor at the School of Architecture, Art and Design of the Tecnológico de Monterrey. He holds a PhD in Urban and Environmental Studies from El Colegio de México and an MSc in the Built Environment: Advanced Architectural Studies from University College London (1997). He studies urban form and how it affects inhabitants in aspects such as mobility, land-use patterns and health. His approach is characterised by mixed methodologies, including modelling and analysis using space syntax. He has received funding from national and international institutions: the IBM Faculty Award from IBM's Global University Programs (2015); the Newton Mobility Fund from the British Academy (2015); CONACYT – Convocatoria 2015 Problemas Nacionales (2017–2020); the CITRIS Seed Fund Program (UC Berkeley) – Tecnológico de Monterrey (2019); the Border 2020 Program (US EPA), in coordination with the North American Development Bank (2020–2021); and the Ruta Azul Challenge by Tecnológico de Monterrey (2024). 

Dr Lucía Martín López is an Associate Professor in the Department of Architectural Graphic Expression at Rey Juan Carlos University. She holds a PhD in Architecture from the Universidad Politécnica de Madrid and specialises in development cooperation for human settlements in the Global South. She coordinates the PENT(H)A research group, which explores emerging processes and new techniques at the intersection of history and architecture. She previously taught at several Mexican universities for ten years and has published extensively on housing design in Latin America and Europe. Her research was recognised with the Research Article Award at the 14th Spanish Biennial of Architecture and Urbanism (2016–2017).

How to join

Register for the event via Zoom

Evaluating Graph Structure: The Need for a New Measure of Quantitative Spatial Description

Current space syntax theory lacks quantitative methods for comparing the graph structures underlying spatial configurations at both building and urban scales. In his final work, presented at the 12th Space Syntax Symposium (2019), Bill Hillier addressed this gap by classifying four theoretical structure types and arguing that comparisons of these types, along with their associated functional effects, could provide a meaningful way to evaluate how spatial layout influences social functioning. His aim was to establish a more testable basis for space syntax theory, enabling its broader application across design and planning contexts as a scientific approach. 

Building on this foundation, the present research develops a framework for quantitatively classifying and comparing graph structures at the building scale, while also examining how local spatial changes affect global structure. Specifically, the study revisits Hillier’s concepts of traversability and mean local choice, developing measures to evaluate these properties across the four structure types. These measures allow real-world cases to be quantitatively compared against theoretical structures, providing a basis for interpreting their potential functional effects as well as the semantic meanings embedded in configuration. Ultimately, the research redefines the four structure types as four fundamental syntactic rules for generating configuration, each carrying a distinctive meaning, which can be quantitatively differentiated through variations in structural property values. 

Speaker

Chenyang Li is an EPSRC-funded PhD candidate at the Space Syntax Laboratory, UCL. His research focuses on the spatial types and structures of complex public buildings, with particular emphasis on museums and art galleries. He is currently a postgraduate teaching assistant for the MSc/MRes Space Syntax: Architecture & Cities programme and the MPhil/PhD Architectural Space & Computation programme. Before joining The Bartlett, Chenyang worked as an architect at Hunan Architectural Design Institute and was a guest lecturer at Hunan University and Changsha University of Science and Technology for Space Syntax workshops.

How to join

Register for the event via Zoom

Stress and Streets: How Street Networks are Associated with Stress-related Brain Activity? 

Rapid urbanisation has intensified concerns about how city environments shape mental health, particularly stress vulnerability. While previous research has highlighted the roles of green space and built form, the influence of street networks remains under-explored. This lecture presents findings from our cross-disciplinary study Stress and Streets, which integrates Space Syntax analysis of street configurations with fMRI evidence of stress-related brain activity. We examined 42 Berlin residents who underwent a standardised social stress task inside an MRI scanner, validated through cortisol levels, heart rate, and subjective ratings. Street network characteristics—including global and local integration, connectivity, and normalised angular choice—were computed within a 1,500-metre radius of participants’ homes. 

Results show that higher neighbourhood-level integration, indicating more accessible and well-connected street layouts, was associated with lower activation in stress-responsive brain regions such as the dorsal anterior cingulate cortex, insula, and thalamus, all part of the salience network involved in detecting threats. This suggests that living in more integrated street networks may buffer neural reactivity to acute stress. No associations were found at the point-address scale or for connectivity and angular choice. The study demonstrates how urban morphology can be quantified and linked to neural stress processing, providing a novel methodological bridge between urban design, neuroscience, and public health. Implications include rethinking street planning not only for mobility and sustainability but also as a pathway to support community resilience and mental well-being. 

Speaker

Dr Gu is a postdoctoral researcher at the Chair of Urban Design and Planning, Department of Architecture, TU Darmstadt. Her research focuses on healthy urban design, with particular interest in how macro-scale design factors and ground-level interventions shape community health and well-being. Her work has been published in the Journal of Environmental Psychology and Environment and Planning B: Urban Analytics and City Science. Her current research investigates design strategies for urban public spaces as mental health–supportive resources, aiming to inform evidence-based guidelines for practice. 

How to join

Register for the event via Zoom

Second-Generation VGA: Area-Weighted Overlap for Better Movement Prediction

Traditional Visibility Graph Analysis (VGA) has long been a cornerstone of space syntax, yet its reliance on binary co-visibility limits its ability to fully capture how people see, understand and move through space. This talk introduces Second-Generation VGA — a major step forward that replaces simple binary visibility with area-weighted isovist overlap. Using custom-built software (Blink), we explore two new measures: symmetric overlap, which treats visibility between locations equally, and asymmetric overlap, which incorporates direction and transitional perception into the model. These refinements enable VGA to represent spatial experience more realistically, capturing subtle visual cues that influence movement. 

Our case study in Barnsbury, London, compares these weighted measures against conventional VGA and observed pedestrian movement, and the results are striking: both symmetric and asymmetric overlaps produce significantly stronger correlations with real-world activity, with asymmetric overlap providing the most accurate predictions to date. By integrating perceptual weighting into VGA, this approach bridges the gap between spatial configuration and human behaviour; it not only boosts predictive power but also opens new opportunities for modelling urban movement with greater realism and precision. This work was first presented at SSS14 and has since been extended in a journal article published in Archnet-IJAR: International Journal of Architectural Research, and the seminar will also introduce OmniVista 25, an online, ultra-fast, web-based VGA analysis tool developed at Syntax North. 

Speaker

Dr Nick Dalton is an Associate Professor of Computing at Northumbria University. His research has significantly advanced the field of space syntax. He contributed to the development of early analytical tools, including Axman, and played a key role in the methodological shift from axial analysis to angular and segment-based approaches. His work integrates computational methods with architectural theory, enabling more precise modelling of urban movement and spatial networks. Dalton works at the intersection of architecture and computing, developing algorithms and software that bridge disciplinary boundaries and extend the analytical capabilities of space syntax research. 

How to join

Register for the event via Zoom