Thursday, June 14, 2007

CASA & UCSB: Finding the invisible landmark

During the past three months I have been working on a new project here at the University of California and Santa Barbara. This project was the result of a very stimulating discussion that started at lunch with Mr. Tom Pingel a Ph.D. candidate here at UCSB. Tom (the person in the video) and I were discussing a great article by Hill et al., (Hill, E. W., Rieser, J. J., Hill, M. M., Hill, M., Halpin, J., & Halpin, R. (1993). How persons with visual impairments explore novel spaces: Strategies of good and poor performers. Journal of visual impairment and blindness, 295-301) and decided that we wanted to replicate this experiment in order to investigate and compare the search strategies of sighted subjects.


In their article, Hill et al., (1993) put forward a simple method for studying mental representations of space and at the same time doing away with some of the problems inherent in the strict classifications of blind and visually impaired subjects based on their visual acuity. Hill was a professor of special education, deeply interested in orientation & mobility and intrigued by the large variation in performance between individuals in spatial tasks. In their study, sixty-five individuals (n congenital = 40; n adventitious = 25) were given seven minutes to freely explore an open bounded space (15 x 15 foot) with the goal of locating four objects and remembering their position. The exploratory behaviour of each subject was videotaped and these were coded into different search pattern or strategies. Subjects were also interviewed regarding the strategies they thought they had used when exploring the space. Adding to the perimeter and gridline strategies, Hill et al., (1993) argued that there a series of strategies (both egocentric and allocentric) that can be implemented in order to facilitate the knowledge of the relation between the different objects in a space. Participants were asked to make a series of heading judgments between the different locations and the accuracy and latency of their response was recorded. Hill et al., (1993) argued that there is much to learn by investigating the accuracy of heading judgments in relation to the frequency and type of exploratory strategies used. In order to investigate this relationship, the authors classified the participants in terms of their performance in the heading task, irrespective of their visual condition by separating the best 15 and worst fifteen performers (top & bottom 25%) and contrasting the strategies used by the individuals in these two groups. Results revealed that the best performers significantly used and verbalized a wider variety of strategies and that the majority of these strategies tended to be object related or allocentric (table 6.1). After locating all the objects “the participants walked back and forth in a straight line among objects, the perimeter and target objects, and/or the home base and the target objects” (Hill et al., 1993) The worst performers did not vary much in the type of strategies used, most of them using egocentric strategies such as the perimeter and grid-line. Many of which tended to randomly wander around the area with no apparent strategy.


In our experiment, instead of using physical objects, we decided to use audio cues to identify the location of objects - in our case the objects were four animals (dog barking, cat meowing, rooster crowing and sheep bleating). Subjects were divided into four groups (sighted, blindfolded, blindfolded outside the room and blind) and asked to enter a specifically constructed empty room where most visual landmarks were eliminated. The room was equipped with a network of lasers whose purpose was to (1) track movement and (2) activate the animals sounds. Like in Hill's experiment subjects were given seven minutes (two trials) to explore the space, locate and remember the position of the animals. They were then asked to perform a series of pointing judgments and to construct a model of the maze.

The two videos below provide further explanation on the experiment, a technical discussion of how the room was constructed and how the lasers were used to detect and code movement as well as activate the animals sounds. The first video is a prototype of the experiment. The second video is the actual room where the subjects were tested:

Experiment prototype:



Actual experiment (room):