Path preference and path geometry John Zacharias Concordia

Path preference and path geometry John Zacharias, Concordia University, Montréal (Québec) Canada H 3 G 1 M 8 tel: 514 -848 -2424 ext 2058 e-mail: zachar@vax 2. concordia. ca

The research questions: 1. The cognitive map • Do pedestrians make path choices based on cognition of the whole environment, or alternatively, are choices primarily made from information available locally? • Do pedestrians tend to move straight ahead; for example, bisecting the environment? • Do pedestrians choose pathways offering them more path options?

The research questions: 2. Environmental content • Are people primarily drawn to pathways and places with signs of human activity? (Zacharias, J. 2001. Path choice and visual stimuli: signs of human activity and architecture. Journal of Environmental Psychology, 21, 341 -352) • Are people drawn to pathways and places with particular geometry?

Indoor city of Montréal (22 km corridors)

Three intersections in Place Montréal-Trust, Montréal Indoor city

The view down the corridor from the intersection

Experiment I • Participants (n=40) are recruited who do not know the Indoor city • They are asked to freely explore the Indoor city and talk about what they see and why they are making path choices • The choices are recorded by the research assistant who also records their travel account

Experiment I continued • A new group of participants (n=40) is recruited • They sit individually in the lab and explore the same environment represented in virtual reality (VR) • The VR is created using VR Authoring Studio while the choices are recorded manually by the research assistant

Expressed preference for path choices (participants=40; n=668) Motivation n % People Store Design Light Path to new Smell Music Avoid repeated path Avoid dead-end Other 188 186 96 77 69 19 5 9 9 10 27 27 14 11 10 3 1 1

Directional bias in navigation Straight n % Left n % Right n % Real 371 38. 1 268 27. 5 286 29. 3 VR 243 36. 4 132 19. 9 146 22. 0

Search for innovative experience Total % Real environment 5 5. 1 VR environment 11 22. 4 Choose same path second time Choose different path second time Real environment 93 94. 9 VR environment 38 77. 6

Aggregate distribution of paths selected by participants in the real environment (a) and the VR environment

Real vs VR exploration • No difference in path choice is detectable between real and VR exploration 11 intersections, 37 path choices Wilcoxon test: +T = 371. 5; -T = 331. 5; p =. 47 • No difference in path choice is detectable in first path choices either 10 intersections, 33 path choices Wilcoxon test: +T = 287. 5; -T = 307. 5; p =. 39

Boundary relations (BR) • For example, do people choose paths that offer them more options for future path choices – e. g. more visible path choices? • Boundary relations in Real: 214; VR: 166 • 1 < BR < 5; mean = 1. 9 • Real: r =. 071, p =. 48; VR, r = -. 046, p =. 65.

Experiment II • To test the hypothesis that path preference is related to the visible geometry of the intersection, a hypothetical environment is created with different intersection configurations • The 3 D environment is created in Bryce and exported to VR Authoring Studio • Participants are recruited to freely explore the VR environment for 20 individual path choices

VR environment for path choice study

VR exploration • Participants navigate using a mouse and can advance into the hallway as well as select a pathway

Aggregate choices at intersections

Conclusions • Participants navigate in the VR with a modest preference for straight-ahead choices • No left- or right-hand bias is detected that is stronger than the straight-ahead bias • No geometrical configuration resulting in a particular path preference can be detected (in this relatively small sample)