Skip to the content

Methodology for assessment of cognitive skills in virtual environments

Pawlik, S 2001, Methodology for assessment of cognitive skills in virtual environments , PhD thesis, University of Salford, UK.

[img]
Preview
PDF (Author version)
Download (25MB) | Preview

    Abstract

    The client briefing of the proposed building design is usually in the form of drawings and artistic impressions being presented to the client. However, very few clients are able to read a technical drawing and the artist impressions are limited and do not aid the client to visualise all aspects of the proposed building. During the client briefing process the client needs to have the experiential quality described, to be able to fully understand the design of the proposed building. Generally, humans perceive and directly experience architectural space by building qualities like texture, form, colour, light, scale, movement. A full-scale model of the proposed building would fully afford the experimental qualities. In reality it would be impractical and not cost effective. However, VR technology allows the creation of an inclusion of space in user's mind, through a minimum of means, but achieves a maximum impact, and affords all the experiential qualities offered by a physical model. A virtual model with a high degree of detail which can be explored by the designer and his clients will therefore be of significant help. However, to give clients the best possible impression of the proposed design it is important to understand how dimensions of those designed spaces are perceived. Therefore, a study was carried out focusing on fundamental investigations into the perception of basic architectural dimensions in order to assess the potential usefulness of VR technology in architecture and the client briefing process. In two experiments, subjects were required to estimate egocentric and exocentric dimensions in Virtual Environments and Real World Setting (RWS). The influence of stimuli orientation was also investigated. In estimating all dimensions a magnitude estimation procedure was employed using a modified free-modulus technique. All participants were pre-tested. Psychometric and visual tests were used for choosing an experimental group with a fair degree of homogenity. Two independent subject groups were used. In addition to dimension estimations recall of simple layout and feeling of space were investigated when evaluating the virtual interface. The general null hypothesis assumed that people perceive space in VE as well as in the real world. It has been shown that the results are statistically significant and therefore one was able to reject the general hypothesis. Overall participants underestimated the dimensions in both experiments by approximately 20%. Results and limitations of the study are discussed. The results of the experiments would indicate that VR technology can be used for simulations of architectural spaces because despite underestimations of dimensions it still performed relatively well if one compares it with results of experiments in the Real World Settings.

    Item Type: Thesis (PhD)
    Contributors: Fernando, TP(Supervisor) and Brandon, PS (Supervisor)
    Additional Information: PhD supervisors: Professor Peter Brandon, Professor Bob Stone, Professor Terence Fernando
    Themes: Subjects / Themes > T Technology > TH Building construction
    Subjects / Themes > T Technology > TA Engineering (General). Civil engineering (General)
    Subjects / Themes > B Philosophy. Psychology. Religion > BF Psychology
    Built and Human Environment
    Subjects outside of the University Themes
    Schools: Colleges and Schools > College of Science & Technology
    Colleges and Schools > College of Science & Technology > School of Computing, Science and Engineering
    Depositing User: Institutional Repository
    Date Deposited: 26 Jun 2009 16:13
    Last Modified: 19 Feb 2014 14:46
    URI: http://usir.salford.ac.uk/id/eprint/2151

    Document Downloads

    More statistics for this item...

    Actions (login required)

    Edit record (repository staff only)