Tag Archives: Unity

IVAS eXPeriment

Thirty-five people took part in the Immersive Virtual Architecture Studio (IVAS) between 26th February and 9th March 2018. The experiment was simply presented as a room scale VR experience where you would have to solve a few jigsaw puzzles inside different rooms. This was one of the most important and exciting parts of my PhD research project. Follows a short explanation of context, purpose and approach.

Soapbox VR room running IVAS with Dean.

Context

Homo Sapiens’ biggest achievement, to a certain extent, as a global civilization, has been to transform and adapt the environment to his needs. The main strength to achieve this outcome is Sapiens “spatial awareness”: the ability to perceive and make sense of his spatial environment and the intrinsic sense of agency that it affords. Sapiens developed this ability following different trait, the most recognizable one being known under the field of “architecture”. For more than five thousand years, using bricks and mortar, he built places to fulfill all the different functions required by society: services, religions, politics and other cultural activities. In the 21st century, Virtual Reality (VR), an inherently spatial technology, offers us the perfect medium to test and apply some architectural principles developed over the centuries to structure and navigate today’s overwhelming digital landscape.

Purpose

The built environment has a significant effect on humans behavior in the physical world (1). How does that translate in VR? The overall aim of this project is to establish the foundations of a framework to support the design of Immersive Virtual Environments. Such a framework will have benefits not only for scientists but in every field VR is disrupting such as game design, industrial design, data visualization and learning applications to name just a few.

Approach

This study is exploring ways to evaluate how different architectural elements affect human’s spatial cognition performance using the IVAS. The following steps will be to apply those findings to support specific cognitive tasks for specific users. This particular iteration of the project is looking at two architectural elements arranged following two spatial characteristics. Those fours conditions are tested using three cognitive tasks. Follows a short description of the setup.

Physical Space – Hardware – Mode of Interaction

For most of our history, natural movement has been the only way to navigate our environment and to experience “architecture”, therefore, it is the primary mode of interaction used in this experiment. To accompany this principle, a room-scale VR environment is set up with a minimum of 9 sqm (3mx3m) of navigable space. In this instance, the IVAS exp. happened in two different rooms, at two different sites: Goldsmiths, Hatcham house, 1 and Soapbox, Old Street 68.
The second mode of interaction is the VR system which is composed of an HTC Vive head-mounted display with two wireless hand-held controllers allowing together 18 degrees of freedom (18 DOF) of movement. The headset is tethered to a powerful laptop that runs the simulation.

Prt Screen from Unity running IVAS A1

Virtual Space – Software – 3D Models

The room with approximately the same dimensions as the physical room is modeled in 3D and will serve as the base for the different conditions (architectural scenes) that will be tested. All other 3D assets are modeled using 3Dsmax before being imported in Unity3D where the interactivity is programmed.

Spatial Conditions

Two architectural elements, wall and columns, were studied following two spatial characteristics: enclosure and complexity (3)

  • A1 : Close Columns
  • A2 : Open Columns
  • B1 : Close Walls
  • B2 : Open Walls
Layout of the different conditions.

Three Tasks involving Spatial Cognition

Solving a Jigsaw Puzzle

This task was design to encourage participants to navigate the space in search for all the items needed to solve the puzzle. A stopwatch was encouraging them to do so as fast as possible – a way to measure performance. VR allows to easily track user’s movement: time, position and rotation. Everybody seems to have enjoyed solving the jigsaw puzzle and were very focused on the task. I had to remind them to explore the space before starting the task. Once the puzzle was solved, the participant was automatically transferred to a transition area where he had to rate two experiential qualities.

Rating of Experiential Qualities (REQ)

The spatial analysis can only be meaningful in regards to an equivalent evaluation from a human experience point of view. Evaluating “lived space” (2) can be done by asking participants to rate their experience with each spatial characteristics. This task brings the qualitative human evaluation into the equation. Using a semantic differential scaling technique, subjects were able to differentiate their appraisal using a five-step Likert-like scale. The rating categories were selected to represent previously mentioned properties: enclosure and complexity.

Perspective Taking Task (PTT)

Once out of the IVAS, participants had to answer a few questions on the online questionnaire before completing this last cognitive task. The main purpose of this task is to measure the memorability of each scene (4). It consists of a sequence of 16 pairs of images. For each pair, one of the images was taken from one of the explored room, the other image was taken from a room not visited. The participant had to identify which image relates to one of the scenes he had experienced.

Perspective Taking Task – All the views
Perspective Taking Task – Pair 03
Perspective Taking Task – Pair 11

Space Syntax Design Analysis

The design analysis using Space Syntax approach will give us an objective measure of each considered spatial characteristics. By combining both “Isovist” and “Visibility Graph” techniques, we obtain a number of measurands (3). In this case, we will be using the following measurands to represent the best predictor variables for the spatial characteristic considered:

The spatial qualities and their related measurands are :

  • Enclosure: “isovist openness” and “jaggedness”;
  • Complexity : “number of vertices”, “vertex density”, “roundness” and “jaggedness”;

Early Observations

A quick glance at the data shows that participants experienced spatial complexity as intended in the scenes designed with the columns. Their average best performance comes out of the scene with the closest room with columns. However, the feeling of openness doesn’t seem to be related to the number of windows in the room. One explanation for this is most probably because there was a texture on the glass. It wasn’t completely transparent. A participant even said: “I didn’t realize that there were glass panel walls!”
This is just a short intro of the kind of conclusions I am working on. This experiment is bringing plenty of good data to dig into, some with positive results some negatives. I have a few pages to fill with that discussion (check further posts).


References:
  1. Arthur E. Stamps. Mystery, complexity, legibility and coherence: A meta-analysis. Journal of Environmental Psychology, 24(1):1–16, 3 2004
  2. Annemarie S. Dosen and Michael J. Ostwald. Lived space and geometric space: comparing people’s perceptions of spatial enclosure and exposure with metric room properties and isovist measures. Architectural Science Review, 60(1):62–77
  3. Jan M. Wiener and Gerald Franz. Isovists as a Means to Predict Spatial Experience and Behavior. pages 42–57. Springer Berlin Heidelberg, 2005.
  4. Barbara Tversky and Bridgette Martin Hard. Embodied and disembodied cognition: Spatial perspective-taking. Cognition, 110(1):124–129, 1 2009.

 

AM XP Pilot – Ready for testing.

This experiment happend all day during the “Human Interactive Conference” on 6th November in NAB at Goldsmiths University of London.

Based on the Method of Loci mnemotechnic, the following experiment is a set of 3 different 3D virtual environment (3DVE) used to remember a sequence of twelveplaying cards. Each 3DVE proposes different architecture styles.  Each style shows specific characteristics that will help us to understand what enhanced the way the user is remembering. As the average memory (digit) span for normal adult is 7 plus or minus 2 (ref to “The Magical Number Seven” G.A. Miller) we will be using twelve cards to avoid using short term memory. The main point of the experiment is to activate long term memory through the association between space,  images (existing memories) and playing cards.

Here is a brief description of the procedure:

1) Warm up (2min)
You have the opportunity to familiarise with the set up and discover the 3D Virtual Environment (3DVE). By navigating that space for a couple of minutes you should be able to memorise it effortlessly.

2) Encoding  (10min)
You have 10 minutes to remember a random sequence of twelve playing cards given one by one at the top of the screen. To get the next card, you  have to associate the current one with an image on a screen. Choose an image from a collection of images organised in 5 categories: animal, action, landscape, people and theme. The method is to use any associative trigger – it can be shapes, numbers, colours, metaphorical or allegorical, what ever memories it evokes – you can find between the image and the playing card. You can then imagine a story in your mind following those images along your way through the space. That story, guided by the surrounding will help you to remember to right sequence of image, each one, associated with a playing card.

3) Consolidation  (2min)
When done associating all the cards, take a couple of minutes to revisit the space and consolidate your memories. Then get out of the memory palace.

4) Retrieving (3min)
You will be presented with a restitution board (webpage) which will show you the twelve playing cards in order. By walking through the space in your mind’s eye, give the sequence back by dropping the playing card in the empty slots. The experiment will end with a score showing the result. This last phase should be repeated  5 times to check how strong the memories are: next day, next week, next month and a year later.

If you want more information about this experiment, please send me an email at info@archimemory.net

XP02.1 – Lets do this in 3D

This a first draft of The ArchiMemory Gallery as a case study to learn the basic of Unity.

You need to run Unity Web Player.

XP02.1 (online 29/11/2013)

Learning skills with a rollerEgg.

By doing this little scene, I have learn about colliders and triggers,

how to script behaviours to move around,

how to count picked up objects and place GUI text that shows the count and the time.

Give it a try. RollerEgg v0.1