Source-linked AI summary
Visual-Motion-Induced Modulation of Pedestrian Trajectories Using Spatially Distributed Multi-Display Signage in Public Spaces
Yuri Mikawa, Taiki Fukiage, Yuki Kubota, Takumi Yokosaka, Maki Ogawa, Kazushi Maruya
TL;DR
The paper addresses whether existing multi-display signage can guide pedestrians without additional devices or wearables. It presents vection-inspired lateral stripe motion across spatially distributed displays and evaluates it in laboratory and museum field settings. Full-screen motion significantly biased laboratory trajectories and produced direction-consistent aggregate differences in the field, while partial stripes showed no significant laboratory directional effect.
Problem
Existing MDS is used mainly for advertising, while evidence for using its distributed dynamic visual capability to guide pedestrians nonverbally remains limited.
Method
The study presents laterally moving monochrome stripes across spatially distributed MDS to evoke vection-related postural responses, testing full-screen and partial motion in laboratory and field experiments.
Results
Full-screen motion significantly modulated laboratory walking trajectories and produced direction-consistent differences in aggregate pedestrian positions during unconstrained museum walking, whereas partial stripes showed no significant laboratory directional effect.
Takeaways & Limitations
Effective MDS-based pedestrian modulation may require motion covering a sufficiently large visual area while remaining subtle enough to coexist with advertisements.
Takeaways & Limitations
Experiment 1 cannot directly compare full-screen and partial stripes because they were tested on different days with different participant groups and blocked presentation.
Abstract
from arXiv · showhide
Multi-display signage (MDS), now ubiquitous in urban environments, has the potential to influence human behavior and experience in public spaces. However, despite its unique capability to present spatially distributed dynamic visual stimuli, its current use is mainly limited to advertising. In this study, we propose a perception-based approach for laterally modulating pedestrian trajectories as a nonverbal means of guiding pedestrians in public spaces. The approach is motivated by vection, the illusion of self-motion, and uses laterally moving monochrome stripes, a standard stimulus in vection research, presented across spatially distributed displays to elicit postural responses that may bias pedestrian trajectories. We evaluated the approach through a controlled laboratory experiment and a real-world field deployment involving actual pedestrian flows in a national museum. The laboratory experiment examined whether the MDS setup induced trajectory shifts in the direction predicted by prior research on the behavioral effects of vection. The field deployment investigated whether comparable effects would emerge in aggregate pedestrian behavior during unconstrained movement under conditions closer to those of urban public spaces. In the laboratory, full-screen motion significantly biased walking trajectories in the direction of visual motion, whereas partial-stripe motion produced no significant directional effect. In the field deployment, opposing full-screen motion conditions produced direction-consistent differences in aggregate pedestrian positions. The field results, observed despite the substantial variability in real-world pedestrian flows, extend the controlled laboratory findings and provide ecologically valid evidence supporting practical MDS-based pedestrian modulation in public settings. The results further suggest that sufficient visual-motion coverage may be important.
1. Introduction
The study proposes using spatially distributed visual motion on existing MDS to nonverbally modulate pedestrian trajectories through vection-related postural responses. Laboratory and field results indicate that full-screen motion can bias walking behavior, whereas partial motion may be insufficient.
- Existing MDS offers a spatially distributed, dynamic, non-wearable medium for guiding pedestrians alongside the real environment.This capability extends MDS beyond its predominant advertising use.
- Conventional signs, arrows, and floor markings require active attention and rapid interpretation and may therefore be overlooked or misunderstood.Prior nonverbal vection-based approaches can also require additional floor-mounted devices.
- The proposed approach uses laterally moving monochrome stripes across MDS to evoke vection-related postural responses and bias pedestrian trajectories.The study examines both full-screen motion and advertisement-compatible partial stripes.
- Full-screen stripe motion significantly affected walking trajectories in both controlled laboratory conditions and unconstrained real-world walking.The field effect appeared as direction-consistent differences in aggregate pedestrian positions.
- Partial-stripe motion produced no significant directional effect in the laboratory, suggesting that practical applications may require sufficient visual-motion coverage.The authors propose integrating motion into advertisements or a larger display area.
2. Related Works
The paper builds on vection and visually induced postural responses as nonverbal mechanisms for influencing pedestrian movement. Unlike approaches requiring added infrastructure or wearables, it investigates whether existing MDS can influence natural pedestrian flow.
- 2.1. Vection and Visually Induced Postural Responses: Vection is the illusion of self-motion elicited by a large, coherent visual motion pattern across the retina.Visual signals are integrated with vestibular and somatosensory inputs during self-motion perception.
- 2.1. Vection and Visually Induced Postural Responses: Large-field visual motion can produce vection and body shifts in the same direction as the stimulus, providing the proposed mechanism for trajectory modulation.The study links visually induced postural responses to pedestrian movement through visuo-postural coupling.
- 2.2. Pedestrian Control: Prior pedestrian-guidance methods include explicit signage, floor-based moving stripes, mid-air haptics, spatial audio, and wearable systems.These approaches vary in whether they require additional infrastructure or devices.
- 2.2. Pedestrian Control: The study instead explores leveraging MDS already installed in urban environments without additional hardware installations or wearable devices.This positions MDS as a non-wearable medium for influencing natural pedestrian flow.
3. Visual Stimuli Design
The stimulus design uses laterally moving monochrome stripes distributed across a six-display, three-depth-layer MDS arrangement. It compares full-screen motion with smaller advertisement-compatible stripe regions, trading visual coverage against compatibility with existing content.
- The design uses six vertically oriented displays arranged in two lateral columns and three depth rows with 3.5 m spacing.The same MDS setup and stimulus design were used in both experiments.
- 3.2. Stripe Design: The stimuli follow vection-oriented parameters including low spatial and temporal frequencies, slow angular velocity, high contrast, and peripheral-field presentation.The design rationale is based on prior vection research.
- 3.2. Stripe Design: Full-screen stimuli use black-and-white stripes with 2.5 cycles across each vertical display and a temporal frequency of 2 Hz.For a 55-inch display, the stripe cycle width is 274 mm.
- The three front, central, and rear display layers may enhance vection by providing spatially separated depth information.At the initial laboratory viewing position, the front displays satisfy recommended vection parameter ranges.
- 3.3. Partial Stripe Design: Partial stripes occupy the top and bottom 9% of each display, preserving an advertisement area with an approximately 3:4.37 aspect ratio.This layout is intended to remain compatible with standard advertisement formats.
- 3.3. Partial Stripe Design: The partial design may improve figure–ground perception and advertisement compatibility, but its smaller angular extent is expected to reduce vection strength.The full-screen and partial configurations therefore represent a coverage-versus-compatibility trade-off.
4. Experiment Design
The study combines a controlled laboratory experiment with a naturalistic museum field experiment to test whether MDS visual motion modulates pedestrian trajectories. The laboratory compares full-screen and partial stripes, while the field study reverses full-screen motion direction across days.
- Experiment 1: Experiment 1 tested whether MDS visual motion shifted individually controlled walking trajectories in the direction predicted by vection research.Participants walked toward designated destinations under full-screen and advertisement-compatible partial-stripe conditions.
- Experiment 1: The laboratory layout included a station-like signboard with three destinations, participant starting footprints, and a monitor showing each designated destination.The setup supported controlled destination-directed walking.
- Experiment 2: Experiment 2 tested whether direction-dependent differences emerged in aggregate pedestrian positions during unconstrained museum movement.Visitors walked freely without prescribed starting positions or destinations, and group walking occurred naturally.
- Experiment 2: The field experiment used only full-screen stripes and reversed stripe-motion direction across two days.The spatial and stimulus setup remained identical across the two experiments.
5. Experiment 1: Laboratory Experiment
Experiment 1 evaluated MDS-based trajectory modulation during individually controlled walking toward designated destinations. The laboratory setup included a deliberately asymmetric space, recruited museum workshop attendees, and controlled participant eligibility and positioning.
- 5. Experiment 1: Laboratory Experiment: Experiment 1 tested MDS effects on individually walking participants under designated starting positions and destinations.The laboratory conditions used single-pedestrian walking only.
- 5. Experiment 1: Laboratory Experiment: The six-display MDS array used vertically oriented 55-inch displays spaced 3.5 m apart horizontally and in depth.The array center was 1.4 m high in a 160 m2 conference room with outdoor light blocked.
- 5. Experiment 1: Laboratory Experiment: The room was left-right asymmetric, with a wall on the left and an open side on the right, limiting interpretation of absolute trajectory values.Relative differences among stripe-motion conditions were still evaluated under the same fixed geometry.
- 5. Experiment 1: Laboratory Experiment: Participants walked toward signposted destinations, with starting position, goal line, and destination signage specified before each trial.The signboard simulated railway exits and a restroom in a public-space setting.
- 5. Experiment 1: Laboratory Experiment: The study recruited 30 museum workshop attendees across two non-overlapping days, with 16 participants on Day 1 and 14 on Day 2.Participants were recruited across a broad range of age groups and genders without age or gender restrictions.
5.3. Stimuli
The experiment compared full-screen and advertisement-compatible partial stripe motion across destination conditions. Stripe-motion blocks were shared across participants, while destinations were randomized within blocks.
- 5.3. Stimuli: The partial-stripe pattern was designed to remain compatible with an existing advertisement, unlike the full-screen pattern.The two patterns were presented on separate experimental days.
- 5.3. Stimuli: Day 1 used full-screen stripes with three motion states and three destinations, producing 9 conditions.Day 2 used partial stripes with three motion states and two destinations, producing 6 conditions.
- 5.3. Stimuli: Stripe-motion conditions were presented in shared blocks, while each participant’s destination was displayed immediately before the trial and randomized independently.The MDS stimulus changed only between blocks.
- 5.3. Stimuli: The walking experiment occupied the first 30 minutes of each 1.5-hour workshop session.
5.6. Analysis
Trajectories were detected, projected into floor coordinates, and summarized by signed deviation area from the central line. Full-screen motion significantly affected trajectory area, whereas partial-stripe motion did not.
- 5.6. Analysis: Trajectory areas were computed as signed deviations from the central line, with side-specific signs and rescaling to the original coordinate system.The signed contour areas were summed for each walking trajectory.
- 5.6. Analysis: Figure 6 plots mean X-axis displacement at 0.05 m intervals along Y, with shaded regions representing ±1 SD.Detection errors were larger near the starting position because of lower resolution and footpoint-estimation mismatch.
- 5.6. Analysis: Full-screen post-hoc tests found differences between leftward and rightward motion (p < 0.001) and between stationary and rightward motion (p = 0.046).The leftward-versus-stationary difference was not significant (p = 0.075).
- 5.6. Analysis: Figure 7 reports mean trajectory areas and ±1 SD for full-screen and partial-stripe conditions.The raw trajectory data for every participant and condition are provided in supplementary figures.
5.8. Summary of Experiment 1
Experiment 1 found that full-screen MDS stripe motion significantly modulated pedestrian trajectories, whereas partial-stripe motion showed no significant directional effect. The findings support using sufficiently extensive motion areas for practical MDS-based behavioral modulation.
- 5.8. Summary of Experiment 1: Full-screen stripe patterns significantly modulated trajectories, whereas partial-stripe motion produced no significant directional effect.Destination also effectively modulated walking trajectories in both conditions.
- 5.8. Summary of Experiment 1: Practical implementations may require a sufficiently large motion area to produce measurable pedestrian modulation.
- 5.8. Summary of Experiment 1: Animating the advertisement itself or its background may provide more extensive motion than narrow stripes confined above and below advertisements.
6. Experiment 2: Field Experiment in a National Museum
The museum field experiment tested whether opposing full-screen stripe motion shifted aggregate pedestrian positions during unconstrained walking. All spatial regions showed direction-consistent differences, with the largest displacement before the MDS and evidence that motion effects persisted afterward.
- Design: The direction-consistent position differences occurred despite unrestricted starting positions, unspecified destinations, and group walking in the museum.The field conditions were designed to capture aggregate pedestrian behavior under highly unconstrained movement.
- Design: The experiment compared one continuous full-screen motion direction per day, without a static no-motion baseline, under otherwise constant spatial constraints.Rightward motion was shown on Day 1 and leftward motion on Day 2; the displays operated throughout the seven-hour opening period.
- Environment: The field setup included six vertically oriented displays, an open left side, a wall on the right, LiDAR sensing, and an explanation booth behind the MDS.The spatial arrangement was constrained by the museum’s circulation plan and visitor-flow requirements.
- Results: 13.11 cm was the largest mean displacement, occurring in the Front space; the MDS space showed 5.44 cm and the Rear space 5.91 cm.The Front space contained the largest displacement, while displacement within the MDS space was comparatively attenuated.
- Results: All four spatial regions differed significantly between opposing motion conditions, with p < 0.001 in Full, MDS, and Front spaces and p = 0.017 in Rear.The Front-space effect was moderate (Cohen’s d = 0.59), while the other regions had effect sizes around 0.2.
- Discussion: Trajectory modulation was strongest before the MDS, where all displays were visible, but participants’ displacement appeared to persist after leaving the display area.The authors attribute the smaller MDS-space displacement partly to deliberate resistance to the visual motion and the visible central floor line.
7. Limitations
The paper identifies limitations affecting direct stimulus comparisons, causal attribution between field conditions, and individual-level trajectory analysis.
- Experiment 1: Experiment 1 cannot directly compare full-screen and partial-stripe patterns because they used different participant groups, different days, and blocked presentation.Possible order effects therefore cannot be fully excluded.
- Experiment 2: Experiment 2 assigned opposing motion directions to consecutive days, so residual day-to-day differences cannot be completely excluded.The spatial and operational settings were held constant to reduce variation, but no within-day direction randomization was used.
- Experiment 2: The field study lacked a static baseline because museum requirements required displayed content to remain engaging throughout the exhibition.This prevents direct comparison with a no-motion display condition.
- Experiment 2: LiDAR data could not reliably track individual pedestrian centroids over extended periods, requiring aggregate positional rather than individual-trajectory analysis.Manually assigning consistent person IDs across the point-cloud data was infeasible.
8. Conclusion and Future Work
The study concludes that laterally moving stripes on existing MDS can modulate pedestrian trajectories, with stronger evidence for full-screen than partial-stripe motion. Future designs must balance broad visual coverage with social acceptability and test bidirectional, crowded flows.
- Conclusion: Full-screen stripes significantly modulated laboratory walking trajectories, while partial stripes produced no significant directional effect.The field experiment additionally found small, direction-consistent aggregate position differences under opposing full-screen motion.
- Conclusion: The field differences emerged during unrestricted starting positions, unspecified destinations, and group walking, supporting the approach’s ecological relevance.These conditions more closely resembled unconstrained pedestrian movement in public spaces.
- Design implications: Effective guidance appears to require sufficiently broad visual-motion coverage while keeping the stimuli subtle enough to remain socially acceptable.The paper suggests integrating motion into advertisements or across a large portion of their backgrounds.
- Future work: Future studies should test whether the method promotes consistent side preferences under bidirectional, higher-density, and crowded pedestrian flows.The present experiments were limited to predominantly unidirectional walking.
CRediT authorship contribution statement
The contribution statement assigns conceptualization, analysis, investigation, writing, and related responsibilities across six authors.
- CRediT authorship: YM led conceptualization, data curation, formal analysis, funding, investigation, supervision, visualization, and writing, while TF, YK, TY, MO, and KM contributed primarily to investigation and related project work.The listed contributions also include methodology, resources, data curation, project administration, and manuscript review or editing across the team.