Conference Program

Keynote speakers

Semir Zeki

• Professor of Neuroaesthetics
• Fellow of Royal Society, Founder Fellow of Academy of Medical Sciences.
• Member of American Philosophical Society, European Academy of Sciences and Arts, International Honorary Member of American Academy of Arts and Sciences
• University College London, Cell & Developmental Biology

We are proud to announce that Professor Semir Zeki will be joining VSAC 2026 as a keynote speaker. This marks a significant highlight of this year’s programme, and we are honoured to welcome him to the conference at King’s College London.

Professor Zeki is a Fellow of the Royal Society and a Founder Fellow of the Academy of Medical Sciences. His pioneering research has shaped our understanding of visual perception, aesthetic experience, and the brain’s role in acquiring knowledge. Widely recognised for his influential work in visual science and neuroaesthetics, he has helped redefine how we think about the relationship between perception, art, and beauty.

His keynote at VSAC 2026 will explore how the brain gives rise to aesthetic experience, making him a distinguished and fitting speaker for the conference.

We are delighted to host this distinguished lecture as part of VSAC 2026🌟

Alan Blackwell

• Professor of Interdisciplinary Design at the University of Cambridge;
• Cognition Scientist and Professor at the Computer Laboratory;
• Co-Director of Cambridge Global Challenges;
• Fellow of Darwin College

Alan Blackwell is Professor of Interdisciplinary Design in the Cambridge Department of Computer Science and Technology. Following an early career as an AI engineer, responsible for the design of novel user interfaces and programming languages, he completed a PhD in cognitive neuroscience, investigating the topic of Metaphor in Diagrams. His subsequent research interests have included computational arts (including music notation and perception, choreography, and live coding), and the role of notation in wide range of technical, craft and design practices. In addition to these areas of critical technical practice, he applies methods from STS and Anthropology for intercultural perspectives on design, computation and AI, as described in his recent book Moral Codes: Designing Alternatives to AI.

We are proud to announce that Professor Alan Blackwell will be joining VSAC 2026 as a keynote speaker at King’s College London.

Professor Blackwell is Professor of Interdisciplinary Design in the Cambridge Department of Computer Science and Technology. His work spans AI, cognitive neuroscience, computational arts, and critical perspectives on design and technology. He is also the author of Moral Codes: Designing Alternatives to AI.

Bringing together design, computation, creativity, and anthropology, his research offers a distinctive perspective on the relationship between technology, culture, and human experience.

We are delighted to welcome Professor Blackwell to VSAC 2026 and look forward to his keynote lecture🌿

Symposia

Composition, hierarchical organization, and aesthetics of images: Interdisciplinary perspectives and recent developments

Johan Wagemans^1,2,3
1University of Leuven, Department of Brain & Cognition, Faculty of Psychology & Educational Sciences, ²Leuven AI Institute, ³Leuven Brain Institute

Composition is a central topic of interest in art and aesthetics, both theories and practice, but much harder to operationalize, measure and manipulate, and therefore largely understudied in empirical and computational aesthetics. Recent progress in machine learning and generative AI tools has enabled a more systematic and quantitative approach than ever before. To become really useful and relevant for everyone working on composition, however, we need to ground this work more strongly in theories of perceptual organization and rely on high-quality images and high-quality data on human aesthetic appreciation. This symposium brings all these relevant disciplines and recent contributions together.

Composition, Hierarchical Organization, and Aesthetics of Images: Background and Preview
Johan Wagemans^1,2,3
1University of Leuven, Department of Brain & Cognition, Faculty of Psychology & Educational Sciences, ²Leuven AI Institute, ³Leuven Brain Institute

The composition of an image or artwork, the way the different elements are put together, can be strong or weak, rich or poor, good or bad, and is therefore essential for its aesthetic quality. Exactly what determines the value of a composition depends on many factors, as will be illustrated by various visual examples from Arnheim’s discussions of famous paintings and from photography textbooks and websites. I will clarify the underlying principles of perceptual organization (perceptual grouping, figure-ground organization), how they deliver the visual building blocks, how they establish the most obvious spatial relationships and in some cases also a strong hierarchical organization between parts and wholes, and how all of this can create balance as well as interesting ambiguities and tensions in the overall image. I will also review some empirical research on the role of these principles and phenomena in determining the aesthetics of images (professional photography and images of paintings), emphasizing the notion of visual rightness, but also highlighting the difficulties and limitations of this work. When discussing the complementary strengths and weaknesses of recent work in computational aesthetics, it will become clear that the different disciplines contributing to this area of research have much to learn from one another and that uniting the forces in innovative, interdisciplinary approaches will be needed to make substantial progress. I will argue that the time to do so is now. A preview of the other talks in this symposium will help to pave the way and clarify their interrelationships.

Empirically Defining, Manipulating and Deriving Composition
Lisa Koßmann^1,2
1University of Leuven, Department of Brain & Cognition, Faculty of Psychology & Educational Sciences, ²Leuven AI Institute

Despite its prominence in art theory and education, composition has received surprisingly little attention in empirical aesthetics. Across three extensive studies, we operationalize composition as a determinant of aesthetic preference and introduce a novel method for annotating it within complex natural and artistic imagery. First, in two large online rating studies (N~1300 each), we provide a foundation for positioning Composition within empirical aesthetics by investigating the concept’s relationship with important aesthetic measures, namely Pleasure, Interest, Order, Complexity, and Layout. Second, through three ranking studies, we manipulated and validated Balance, Emphasis, and Repetition independently, testing their respective impacts on aesthetic preference. Our open access stimulus set spans 63 motifs, each with a baseline version and versions with more or less of the targeted principle, which we validated through a within-triplet ranking task (N=37). With the same task (N=152) we demonstrated that stronger compositional presence significantly increases both preference and perceived composition quality relative to weaker or baseline versions. Finally, in a two-alternative forced-choice task (N=348) with all pairwise comparisons across the three levels and motifs, participants had to choose either the better-composed image or the one they preferred. Our Composition manipulations again consistently influenced participants’ rankings and choices regarding image composition and personal preferences. However, for Composition to have impact on aesthetics it must be perceptually derived from images. To investigate this process, our third study introduces a novel annotation tool that allows participants to annotate regions for 3 Composition and 5 Perceptual Grouping principles. Heatmaps and overlays illustrate the hierarchical nature of these regions and enable us to substantiate and quantify Composition in an innovative way. Throughout this work, we show that Composition has a real, empirically measurable effect on aesthetic appreciation and that participants are sensitive to Composition and can extract it from rich images.

Extracting Composition: Behavioral and Computational Approaches to Visual Structure
Doreen Hii^1,2 & Xiaochang Liu^1,2
1University of Leuven, Department of Brain & Cognition, Faculty of Psychology & Educational Sciences, ²Leuven AI Institute

Image composition is central to visual aesthetics but remains difficult to formalize computationally and empirically. Furthermore, composition often consists of several nested sub-compositions. This contribution explores how compositional structure can be extracted from both behavioral and image-based data. We show that sub-compositions can be identified through either eye-tracking data or interactive user prompts. First, we show how eye-movement patterns can reveal hierarchical organization in visual composition. The aggregated saccadic data is transformed into a graph representation after preprocessing with an edge bundling algorithm. Sub-compositions are extracted by clustering sub-graphs based on node proximity (spatial coordinate differences) and edge weight (connection frequency across participants). The extracted sub-compositions align with human hierarchical perceptual grouping. Second, we review computational approaches for extracting compositional cues directly from images. We discuss recent developments in computer vision methods for identifying compositional structure. These include methods based on visual cues such as symmetry, pose estimation, and semantic segmentation. We also outline preliminary work on an interactive framework for composition line extraction. Together, these approaches illustrate how behavioral and computational perspectives can help characterize composition as a formalizable property of images. Extracting these visual structures opens the door for evaluating abstract aesthetic concepts such as symmetry, balance, and emphasis across different levels of the compositional hierarchy.

Understanding Image Composition through Perceptual Organization and Computational Modeling
Fatemeh Behrad¹, Gonzalo Muradas Odriozola¹, & Li-Wei Chen^1
1University of Leuven, Department of Brain & Cognition and Department of Electrical Engineering

Image composition organizes visual elements through perceptual grouping, where scenes are segmented into meaningful objects and structured by their spatial relationships. Although perceptual grouping is central to composition, existing work lacks a precise operational definition and well-annotated datasets that explicitly capture grouping structure. At the same time, most computational approaches rely on global image representations that do not provide explicit spatial correspondence between meaningful regions. Consequently, the role of perceptual grouping in shaping image composition remains unclear, both conceptually and computationally. In this talk, we present a tutorial on object-centric representations. These models learn representations that are more aligned with human perception, such that each region of an image is associated with a distinct and coherent representation. This property makes them particularly well-suited for studying visual composition, as it enables modeling relationships between regions and provides a clearer account of the spatial structure of a scene. We further highlight recent research directions on (1) deriving hierarchical visual organization from perceptual grouping data, including data collected from human observers, and (2) composition-preserving generation as a testbed for assessing how well compositional structure is maintained under transformation, offering a controlled setting to probe and validate compositional representations. Specifically, we consider a composition-controlled image synthesis framework based on scalable vector graphics (SVG) stimuli that encode structure, where generation is formulated as stimulus-guided image-to-image refinement with semantic and edge-based guidance to enforce structural consistency. This approach improves compositional structure preservation compared to both pixel-based controllable generation methods and existing SVG-based baselines. Together, these perspectives offer a unified view of how composition can be understood, represented, and evaluated, and point toward more principled approaches to modeling visual composition across perception and computation in a consistent and interpretable manner.

Panel Discussion
Aenne Brielmann¹, Aaron Hertzmann², Jeroen Stumpel³, & Ian Verstegen^4
1Liverpool Hope University, UK, ²Adobe Research, USA, ³Utrecht University, The Netherlands, ⁴University of Pennsylvania, USA

The symposium will end with a panel discussion in which we have asked colleagues from different disciplines to comment on the work we have presented from their perspective: What is new and interesting, what is the added value compared to what has been done before, but most of all, what is still missing? What are the current shortcomings and challenges? Three authorities have already committed to being a discussant, and one other is considering it: Aenne Brielmann (empirical and computational aesthetics), Aaron Hertzmann (photography and computer graphics), Jeroen Stumpel (art history), and Ian Verstegen (Arnheim, Gestalt psychology and visual studies). After a brief round of initial impressions, comments and questions, the audience will get the chance to engage in a Q&A on the issues being raised by the speakers and discussants.

Applications of Virtual Reality in Empirical Visual Aesthetics

Alexandra Victoria Alvarez^1,2,3, Maximilian Kenzo Molitor^4,5, Julian Salhofer^6,7,8, & Adriano Tenore^9
1Art Research on Transformation of Individuals and Societies Lab, Empirical Visual Aesthetics Lab, ²University of Vienna, AT, ³University for Continuing Education Krems, AT, ⁴Leibniz-Institut für Wissensmedien, Tübingen, Germany, ⁵University of Tübingen, Tübingen, Germany, ⁶Technische Universität Wien, ⁷University of Applied Sciences St. Pölten, ⁸University for Continuing Education Krems, ⁹Naples, Italy based Artist

As art institutions adopt virtual reality (VR) to expand access and engagement, visual aesthetics researchers are also increasingly incorporating these environments into experimental platforms. This symposium brings together four interdisciplinary perspectives, from cognition and emotion research to architectural education, conservation science, and artistic development, to examine how VR is being designed and applied in empirical projects with institutional collaborators. Together, the session situates VR as a research tool to learn about perception and art engagement across mediaities, increase museum objectives, and provide meaningful experiences for real-world viewers. After a scoping view, the session will conclude with live demonstrations of the VR rooms.

Exploring Emotions in Original, Digital, and Virtual Contexts at the Belvedere Museum
H. Brinkmann^1,2,3,4,5,6, A.V. Alvarez^1,2,3,4,5,6, M. Huskinsky^1,2,3,4,5,6, J. Salhofer^1,2,3,4,5,6, E. Specker^1,2,3,4,5,6, & M. Pelowski,^{1,2,3,4,5,6
1}Art Research on Transformation of Individuals and Societies Lab, Empirical Visual Aesthetics Lab, ²University of Vienna, AT, ³University for Continuing Education Krems, AT, ⁴Center for Cultures and Technologies of Collecting, University for Continuing Education Krems,AT., ⁵University of Applied Sciences St. Pölten, AT., ⁶Belvedere Research Center.

Art has long been noted within psychology for Einfü hlung, or “feeling in” to the affective state of another person or artwork. However today in the Post Digital age, an era marked by the ubiquity of technology, it is still unclear what is lost, translated, or gained when an artwork undergoes technological reproduction. Most research on differences between originals and reproductions has been heterogeneous: focusing on shallow assessments of emotion (e.g. valence and arousal), hedonic variables (e.g. liking and beauty), or evaluation methods (e.g. interestingness) that may fail to capture shifts in emotions within genuineness. Furthermore, the limited studies with inconclusive results often rely on low resolution digital surrogates rather than museum quality artworks and ecologically valid settings. To address this gap, the FWF funded project OrDiV (Original Digital Virtual) project applies emotion-based measures to characterize how the experience of an artwork changes as its mediality shifts. Conducted within the Upper Belvedere museum, OrDiV utilizes seven artworks from the collection featured in the exhibition Provocation and Psyche across the four conditions: original artworks in the exhibition room, high-quality digital reproductions viewed on a computer within the museum’s Oktogon, original artworks activated by phone-based augmented reality (AR), and a virtual reality (VR) identical-twin reconstruction of the original gallery experienced through a headset inside the Oktogon. Museum visitors (n=216), while equipped with mobile eye tracking, reported their felt affective and cognitive responses using a 16-item scale after engagement. By comparing experience types (disengaged, negative, transformative, harmonious, novel), and engagement patterns, we explore the results and implications afforded by virtual technologies in art experiences of the Post Digital age. This talk will also draw on this researcher’s role as PhD student coordinating collaboration between museum research, art history, empirical aesthetics, and VR applications.

Experiencing the Semi-Detached Houses of Le Corbusier: Aesthetic Experience and Learning of Architecture Across Media
M. K. Molitor^1,2,3, L. Peiffer-Siebert^1,2,3, O. Özbek^1,2,3, G. Hochstetter^1,2,3, B. Brucker^1,2,3, J. Maiero,^1,2,3, P. Gerjets^1,2,3, & E. Specker^1,2,3
1Leibniz-Institut für Wissensmedien (IWM), Tübingen, Germany, ²University of Tübingen, Tübingen Germany, ³Center for School Development and Teacher Training (ZSL), Tübingen, Germany

A central goal of school excursions to architectural sites—such as the Semi-Detached Houses by Le Corbusier in Stuttgart, today renovated into a museum—is to enable students to experience architecture firsthand. This is based on the idea that architectural knowledge encompasses not only factual information but also an aesthetic experience (AE). Böhme conceptualizes this AE as the perceived ‘atmosphere’ of architecture, and Coburn et al. have found it to manifest on the dimensions of perceived hominess, coherence, and fascination. In the educational context, school visits to architectural sites are not always possible; slide-based presentations (i.e., PowerPoint) are often used instead. Virtual Reality (VR), however, may offer a more embodied and immersive alternative. Thus, this study investigates whether VR can improve students’ learning and AE of architecture compared to a tablet-based PowerPoint (RQ1), and whether higher interactivity across both media enhances these outcomes (RQ2). Grounded in the Cognitive-Affective Model of Immersive Learning (CAMIL), we hypothesize higher immersivity and higher interactivity would positively influence knowledge acquisition, perceived atmosphere, and perception of aesthetic dimensions (i.e., hominess, coherence, and fascination). A 2 × 2 between-subject design (medium × interactivity) experiment was designed, comparing interactive and non-interactive VR and PowerPoint learning environments. These environments were co-designed in collaboration with the museum Weißenhofmuseum im Haus Le Corbusier, Stuttgart, as well as schoolteachers, VR-developers and researchers in philosophy, art history and psychology. Participants (N = 128) will be school students who will experience a 15-minute learning session in one of the conditions followed by assessments of knowledge and AE. The experiment is currently being finalized, with data collection expected to be completed by the conference date. This talk will present the co-design process and empirical results of the experiment, contributing to the broader discussion on how and if VR can enrich both empirical aesthetics and architectural education.

When Artworks Change: Towards Immersive Exploration of Aging Processes in Oil Paintings on Canvas
J. Salhofer^1,2,3,4, H.Y. Wu^1,2,3,4, H. Brinkmann^1,2,3,4, T. Feilacher^1,2,3,4, E. Weixelbaumer^1,2,3,4, F. Marinovic^1,2,3,4, A. Grebe^1,2,3,4, & M.E. Gröller^1,2,3,4
1Research Unit of Computer Graphics, TU Wien, AT,, ²Department of Media and Digital Technologies, University of Applied Sciences St. Pölten, AT, ³Center for Cultures and Technologies of Collecting, University for Continuing Education Krems, AT, ⁴State Collections of Lower Austria, AT

Understanding how visible changes in materials over time shapes the perception of artworks is a persistent challenge in both empirical aesthetics and art conservation. Although the impact of environmental factors such as relative humidity (RH), temperature, and UV radiation on the condition and longevity of oil paintings on canvas is well established, their effects on perception remain challenging to examine under conditions that are both controlled and ecologically valid. This talk presents an ongoing PhD project that develops a virtual reality (VR) framework for the simulation and visualization of aging-related degradation processes in oil paintings on canvas. At its current stage, the work includes a high-fidelity 3D reconstruction of a selected painting from the State Collections of Lower Austria, produced through macro-photogrammetry and reflectance-based imaging. In addition, early experiments in digital restoration explore how a plausible undamaged reference state can be approximated. Building on these initial results, the talk outlines the next phases of the project: the development of computational models to simulate age- and environment-related degradation phenomena such as crack formation, discoloration, and paint loss, and their integration into an interactive VR environment. By combining physically based approaches with procedural and real-time rendering techniques, the system aims to balance scientific plausibility with computational performance, enabling dynamic exploration of environmental effects over time. Through these contributions, the system will serve as a valuable tool for conservators, curators, and other museum stakeholders, supporting preventive conservation strategies and enhancing public engagement by illustrating the complexities of art preservation. Furthermore, user studies involving eye-tracking and questionnaires will investigate how visual degradation influences aesthetic perception and emotional response to artworks. In doing so, the project contributes to broader discussions on how immersive technologies (such as VR) can support both cultural heritage research and empirical visual aesthetics.

An Artist’s Perspective: Designing a VR room for an Empirical Aesthetics Project
A.V. Alvarez,^1,2, R. Rodriguez- Boerwinkle^1,2, S.L. Miller^1,2, M. Pelowski^1,2, & A. Tenore^1,2
1Yale Center for Emotional Intelligence, Yale University, New Haven, CT, ²ARTIS Lab, Faculty of Psychology, University of Vienna, AT

How does the medium through which we encounter art shape what we feel? This contribution presents the work of a CG AI Lead who served as lead VR environment designer and artistic collaborator within the Art Research on Transformation of Individuals and Societies Lab’s project investigating phenomenal states during art experience. Building on prior museum and screen-based cohorts, this American Psychological Association Division 10–funded study extends the research into virtual galleries by comparing an immersive VR condition with a desktop-based virtual gallery (OGAR), while maintaining experimental structure across conditions. The VR condition was designed as a sequence of six distinct viewing rooms, each presenting a single artwork, connected by a central lobby where participants paused to complete self-report using the 16-item Experience Typing Measure (ExTM) before continuing. The hexagonal lobby, developed in Unreal Engine 5.5 and deployed on Meta Quest 3, mirrors the pacing of the desktop condition. The environment was calibrated with a player height of 1.65m and movement speed of 3 m/s to standardize navigation while preserving embodied exploration. A progressive lighting system marked visited rooms, supporting orientation and memory, while a custom plugin recorded gaze direction, viewing duration per artwork, time spent in each room, and total session time. At the same time, design considerations were challenged by balancing methodological rigor with viewer enjoyment. This collaborator’s artistic practice is informed by anthropology and sociology to create immersive environments as sites where cultural, biological, and technological forces converge. Working across interactive and AI-mediated systems, his eco-punk biology informed perspective centers primal forms of perception and meaning activated through action and context. Within this project, he reflects on the constraints and possibilities of working as an artist in scientific settings where experimental control can put limits on expression, but also trigger creative thought through interdisciplinary collaboration.

Can we know the dancer from the dance?

Qasim Zaidi¹ & Joan Danielle Ongchoco^2
1Graduate Center for Vision Research State University of New York, USA, ²Department of Psychology, University of British Columbia, Vancouver, BC, Canada

For Hegel, the human body was “too expressive for symbolic art, too saturated in […] meaning for classical art, too definitely formed for romantic art.” But this is perhaps what makes dance unique: that the human body is its medium. This symposium will consider dance across formats (from live performances to drawings) and genres (from ballet to Kpop), as it interacts with cognitive processes (from perception, memory, simulation, to social cognition). The goal is to explore not just how we perceive the dancer (or the dance), but also what dance might be able to tell us about broader human experience.

What’s a movement in the mind? Studies in how we parse, remember, and appreciate the human body in motion
Joan Danielle K Ongchoco^1
1Department of Psychology, The University of British Columbia, Canada

Our appreciation for dance seems on the one hand idiosyncratic: there is incredible diversity in terms of dance genres and people’s preferences — and what parts of a dance captures people’s attention and emotions is a complex function of a range of postural, motor, and emotional dimensions. And yet on another hand, our capacity to be moved by a dancer and what they are trying to express seems so fundamental, making dance a powerful medium for human communication and expression. What is built into the human mind that draws us to dance? In the first study, I will talk about how we might be constantly building a ‘movement alphabet’, in which a single movement ‘step’ can be combined into ‘words’ that then form ‘phrases’ — the way we learn a language. This ability to parse regularities from dynamic motion might serve as the foundation for how dancers learn how to dance and how viewers learn to parse a choreography. In the second study, I will show how the memorability of a movement sequence may then be determined by the way the individual ‘steps’ are strung together — but that the most memorable movements are not only consistent across over observers (totalling N=2400), but also the ones that are counterintuitively the most ‘snapshot-like’. Finally, I will demonstrate how our aesthetic preferences for dance may be surprisingly systematic along the dimension of the speed at which it unfolds, such that there may be a ‘sweet spot’ of the amount of motion that aligns with the pace at which people also generally move in the world (with this being the most predictive factor, over and above the expressiveness and scale of the movement). Collectively these studies make a case for bridging the empirical aesthetics of dance with the broader cognitive science of the mind.

Dance beyond movement: Research on choreography as a contextualized investigation of the perceived organization and appreciation of dance
Elisabeth Van der Hulst¹, Jonas Rutgeerts², & Johan Wagemans^1
1Laboratory of Experimental Psychology, Department of Brain and Cognition, KU Leuven, ²Department of Literary and Cultural Studies, KU Leuven, Belgium

Over the past decades, research on dance has extended largely, going way beyond its neuroaesthetic origin. However, choreography, which provides the organizational layer of dance, has been missing from this discussion. In this talk, I provide an overview of a set of studies on the importance of choreography for aesthetic experiences. In a first study, we developed a conceptualization of order and complexity of choreography. This conceptualization was necessary to capture the dynamic component of these elements in dance and was validated by continuous ratings of participants on their perceived complexity in the performance. A combination of inductive and deductive analyses revealed statistically notable diUerences of perceived complexity based on theoretical variability and predictability, with a large eUect of movement variability. In a second study, the aesthetic consequences of these organizational choices were explored. To this aim, eye tracking data and continuous ratings were collected and compared to the previously collected complexity ratings. To extract valuable information from the viewing data, the recently developed Taxonomy for Viewing in Dance (TaViDa) was used. Preliminary analyses reveal a positive correlation between choreographic complexity and aesthetic appreciation. For example, moments of prolonged repetition seem to induce boredom, while a sudden change in one of the choreographic levels often increases appreciation. At the same time, these eUects are expected to be accompanied by diUerences in viewing behavior. Performed analyses have shown that viewing behavior is dependent on perceived complexity, coming analyses will explore if this also results in aesthetic diUerences. Both approaches will be discussed in the context of the observer, as well as the performance. The first refers to individual diUerences such as expertise, the second to the physical context. Both recordings, as well as live performances were employed, facilitating a direct comparison.

Interacting bodies in dance: How visuospatial relations shape aesthetic, emotional, and semantic evaluations
Andrea Orlandi¹, Mohammad Soleyman Nejad², & Matteo Candidi^3
1Social Brain Sciences Lab, ETC Zurich, Switzerland, ²Department of Psychology, University of Manitoba, Winnipeg, MB, Canada, ³Department of Psychology, Sapienza University, Rome, Italy

Dance provides a powerful medium for studying how our brain perceives and evaluates bodies in motion. While much neuroscientific research has focused on single bodies, only recently has attention shifted toward interactions between individuals. From a cognitive neuroscience perspective, two features of interacting bodies have received particular attention: body positioning and synchrony, which seem to facilitate the perception of those bodies as a coherent unit. From a choreographic perspective, however, a wide range of visuospatial features has been explored and intentionally used to shape audience engagement, many of which remain largely unexamined in scientific research. Bringing these perspectives together raises a fundamental question: what combination of visuospatial features of two dancing bodies possibly shapes their aesthetic, emotional, and semantic evaluation? To answer this, we recorded the motion kinematics of a pair of dancers engaged in a visuospatial task and transferred these movements onto 3D avatars. We generated a series of images depicting dyadic body postures that varied along multiple visuospatial dimensions, including shared interpersonal space, centre-of-mass distance, orientation, use of vertical space, and symmetry. Across four studies, 200 participants evaluated these images in terms of aesthetic pleasure, interest, symmetry, emotional valence, and meaning. DiUerent statistical approaches were used to examine how each visuospatial features contributed to these five evaluation dimensions. In particular, aesthetic appreciation seemed to depend on a mosaic of features, with a central role for the extent to which bodies share interpersonal space, which we term “entanglement”. Interactions characterized by greater entanglement are consistently perceived as more aesthetically pleasing, more interesting to watch, and more meaningful than less entangled configurations. Taken together, these findings suggest that knowledge from dance and choreography can inform the neuroscientific study of interacting bodies and social cognition, possibly beyond the dance context.

Audience gaze reflects felt social connection to individual performers during live dance
Albane Arthuis¹ & Guido Orgs^1
1Institute of Cognitive Neuroscience, University College London, UK

In everyday social interactions, gaze behaviour reflects interpersonal relationships and social engagement. We investigated how social connection shapes audience engagement by commissioning a live durational dance performance called How Shall We Begin Again? initiated by artist Jo Fong. The performance lasted 16 hours across two days and consisted of a relay of improvised performances by a total of 50 diUerent people. Each individual performance including sections with and without music. Audience members rated their felt social connection, focus, and prior familiarity with each performer that they saw and described their engagement in answers to open questions. We examined how these social and attentional factors related to gaze behaviour and neural dynamics, using a computer-vision deep learning model to analyse eye-tracking data alongside exploratory analyses of EEG power and inter-brain synchrony. We found a positive relationship between felt social connection, gaze behaviour, and engagement, which was mediated by selfreported focus on individual performers. At the neural level, dancing with music was associated with decreased EEG alpha power and increased delta power, consistent with heightened attentional and aUective engagement. Together, these findings suggest that engagement with live dance emerges from interactions between social connection, attention, and shared contextual cues. Engagement with dance depends not only on the dance that is being performed but also on the dancer that is performing, and the social context in which the performance unfolds.

Embodied perception of dance: a journey from aesthetic preference to emotion processing
Beatriz Calvo-Merino¹, Claudia Corradi¹, Vasiliki Meletaki², Jorge Almansa¹, Alexander Jones³, Jon Silas³, & Tina Forster^1
1Centre for Clinical Social and Cognitive Neuroscience, Department of Psychology and Neuroscience. City St George’s, University of London, UK, ²Penn Center for Neuroaesthetics, University of Pennsylvania, USA, ³Department of Psychology, Middlesex University, London, UK

Recreating dances in air from drawings on land
Qasim Zaidi¹, Tayfun Zaidi², Maryam Vaziri-Pashkam³, & Denise Murphy^4
1Graduate Center for Vision Research, State University of New York, ²Independent choreographer, Glasgow, UK, ³Psychological and Brain Sciences, University of Delaware, ⁴Director of Dance, University of Delaware

‘It’s a Draw’ is a series of drawings with dynamic curls, loops, twists, spirals and vortices, made by Trisha Brown the maverick postmodern dancer while pivoting, rolling and skidding in a freewheeling dance across large-scale paper with charcoal and pastel held by hands and feet. The drawings were not preplanned to represent anything figurative or abstract, they are pure vestiges of improvised actions. Video documentation reveals the movements that generated the marks on the ground which raises the question whether the kinetic nature of the marks is suUicient to decode the movements, or whether decoding requires aids to visualization. Sikkema Malloy Jenkins Gallery kindly provided digital versions of 44 drawings and 2 videos, which enabled us to categorize the drawing movements in the videos. The categorization provides ground-truths for decoding movements from the drawings generated in the videos, and the categories enable us to systematically study the ability of observers to decode a drawing without viewing the generating video. Using the supplied movement categories, four groups of observers were asked to identify the movements that generated specific extended marks, some ambiguous but others resembling silhouettes. Groups 1&2 saw the drawing generated by one or the other of the two videos after seeing the other video of Brown making a drawing, providing models for potentially visualizing complete movements. Control groups 3&4 were restricted to perceptual cues as each saw only one of the same drawings and read a general description of the movements with which they were made. The original drawings are many meters across. Scaled down presentations on a large monitor make it harder to identify the movements generating silhouettes as tracing real limbs, but comparisons between the groups on accuracy of identifying movement categories can reveal how perception guided visualization enables interpretation of art with a kinetic motor component. For a repository of the drawings and the videos https://privateviews.artlogic.net/2/2126340ba3cf9b45dd6b6d/