Communication Theory

1
Communication Theory

• Lecture 2
• Designing tools for interaction with the
  environment (2)
• Dr. Danaë Stanton Fraser

2
Cognition and Space

• Distributed cognition challenges us to
  investigate the relationship between space and
  the mind
• It is therefore important for us to understand
  how space features in cognition both in the
  laboratory and in the wild
• Fortunately, technologies have begun to provide
  ideal control test beds for understanding the
  cognitive properties of space

3
Humans and animals must adopt strategies to gauge
their constantly altering position within the
environment if they are to successfully negotiate
that great God-given maze which is our human
world (Tolman, 1948, p. 208).
4
The Cognitive Map

• McGee (1982) defined spatial orientation as
• the comprehension of the arrangement of
  elements within a visual stimulus pattern, the
  aptitude for remaining unconfused by the changing
  orientations in which a configuration may be
  presented and the ability to determine spatial
  relations in which the body orientation of the
  observer is an essential part of the problem
  (p.4).

5
3D

• We see our world in three dimensions. Although
  the retina in the eye is a flat surface and
  therefore the images it receives are two
  dimensional, distance cues enable some two
  dimensional images to be perceived as distant in
  a three dimensional world.
• Monocular depth cues include relative size,
  superposition/occlusion and relative height.
  Relative size refers to smaller objects being
  interpreted as further away than larger objects.
  Occlusion is the effect when one object obstructs
  another, causing the overlapping object to be
  perceived as being nearer. The relative height of
  similar objects can enable distance perception,
  for example, objects that are seen as higher in
  the image are perceived as more distant.

6
3D

• The use of two eyes (binocular vision) has
  advantages for depth perception. Our two eyes
  enable us to see two slightly different images of
  an object and we use this disparity to calculate
  the objects orientation in space. The term
  stereopsis is used to explain how the brain adds
  depth from this disparity between the different
  images from the two eyes.

7
3D

• Another type of depth cue arises from autonomous
  movement in space which appears to be crucial to
  the development of an effective internal spatial
  representation. Gibson (1966) stated that as the
  observer moves through space, there is a flow of
  stimulation on the retinas, which leads to a
  better understanding of the three dimensionality
  of our world. When the observer moves....the
  optic array becomes alive with motion (Haber and
  Hershenson, 1973, p.332).

8
Theories of Space

• Kant claims that space and time are the very
  form of the human mind (Ellis, 1991, p.xiii).
  Indeed everyday navigation and technological
  advances aiding exploration (flying, driving)
  involve complex spatial skill.

9
Space

• Psychological space refers to the space of our
  perceptual experience, any space which is
  attributed to the mind...and which would not
  exist if minds did not exist (OKeefe and Nadel,
  1978, p.6-7).
• Physical space refers to the three dimensional,
  Euclidean world in which we live.
• These two types of space overlap and interact
  with one another.
• We can also distinguish between virtual space and
  physical space, as virtual environments do not
  exist in the physical world.
• The focus here is psychological space and
  investigates whether experience of virtual space,
  can be used to supplement physical space.

10
The form of the Cognitive Map

• Tolman (1948) describes a map control room in
  the brain which stimuli enter and are worked
  over and elaborated into a tentative,
  cognitive-like map of the environment (p.192).
• This cognitive map contains routes, paths and
  environmental relationships which are stored and
  can be used when responding to the environment.
  Thus accuracy of response to ones environment is
  intricately linked to the quality of the
  cognitive map formed.

11
The form of the Cognitive Map

• OKeefe and Nadel (1978) drew a distinction
  between the Taxon system (routes) and the
  Locale system (places) in the build up of
  spatial knowledge.
• The Taxon system involves using a series of
  S-R-S (stimulus-response-stimulus) instructions.
  Navigation involves moving from one landmark to
  the next by aligning oneself in relation to the
  landmarks.
• The Locale system is a highly flexible system,
  based on the development and use of internal
  maps.
• OKeefe and Nadel (1978) state that exploration
  is essential for the creation of internal spatial
  cognitive maps and in constantly up-dating them.

12
The form of the Cognitive Map

• Animal behavioural studies
• Most authors agree that humans carry spatial
  representations of their environment in their
  heads, yet there is constant debate concerning
  the type and content of these representations.
• Siegel and White (1975) suggest a three tiered
  process in the development of spatial knowledge
  the use of landmarks, then the adoption of route
  knowledge allowing fairly simple wayfinding, and
  finally internal representations of space,
  allowing more sophisticated methods of
  navigation.

13
Landmarks, routes and maps

• Siegel and White suggest that landmarks may
  constitute meaningful events and the nervous
  system may be continually taking pictures of
  them.
• Routes are built up by connecting a series of
  landmarks. This strategy is egocentric (dependent
  on the bodys location and direction of pointing
  in space) and is efficient as long as the links
  between successive turns are accurate.
• A cognitive map applies to the mental images that
  individuals build up as they become more familiar
  with their surroundings. This type of
  representation is allocentric (not dependent on
  the bodys position in space or direction of
  regard) and thus is extremely flexible

14
Properties of cognitive map

• Pick and Lockman (1981) describe three properties
  of spatial maps reversibility, transitivity and
  enabling detours.
• Lynch (1960) suggests that the cognitive map is a
  product of the number and type of landmarks and
  the number and type of past experiences one has
  had in a particular location.

15
Cognitive Map (contd.)

• Thorndyke and Hayes-Roth (1982) state three
  important points about spatial cognition.
• Firstly, people build up their spatial knowledge
  from a variety of different sources navigation
  through the environment, a wide variety of
  different forms of maps, verbal descriptions and
  photographs. The knowledge gained from each of
  these sources is integrated to form spatial
  knowledge.
• Secondly, dependent on the knowledge they have,
  people use different methods when making spatial
  judgements.
• Thirdly, the accuracy of any spatial judgement is
  dependent not only on the accuracy of spatial
  knowledge but also on the computations performed
  on this knowledge.

16
Virtual Environment (VE)

• Virtual environments (VEs) have as their core the
  simulation by computer of three dimensional
  space.
• The first defining feature of VEs is that they
  can be explored in real time with similar freedom
  to real world exploration.
• The second defining feature is that the user may
  interact with objects and events in the
  simulation.
• Virtual Environments are interesting tools for
  psychology research in spatial cognition, because
  they allow some control over testing spatial
  exploration

17
Application of VEs

• Research using VEs has stemmed from
• Military
• Space
• Aviation
• VEs are now used in a wide variety of settings,
  including
• Education
• Medicine
• Building design
• Applications for those with disabilities

18
Presence in VEs

• VEs consist of three-dimensional, interactive,
  computer generated worlds, running in real time.
• Often, interacting with these worlds provides a
  feeling of presence as every response has a
  consequence, and the egocentric viewpoint gives
  the illusion of looking from within the virtual
  world.
• This may be true for more or less realistic
  technologies (e.g. head-mounted displays vs.
  desktop VEs)

19
Example VE
20
Training in VEs

• Virtual environments are potential useful media
  for training spatial skills
• Interactions with VEs reproduce similar
  visual-spatial characteristics to interactions
  with the real world
• Interactions with VEs can preserve the link
  between motor actions and their perceived effects
  (Regian, Shebilske and Monk, 1992). This may be
  primarily due to the three dimensionality of the
  display, which provides all of the
  transformations in the visual appearances of
  objects that would accompany real movements in
  space. In Gibsons (1979) terms, the optical flow
  patterns that would be experienced in the course
  of real movements are maintained in the displayed
  environment.

21
Training in VEs (contd.)

• VEs enable assessment of the internal spatial
  representations within the same mode as they were
  acquired. A user explores a VE and then can be
  tested on their spatial knowledge within this
  same environment using pointing tasks or route
  tests.
• easily adaptable, allows repeated viewing, and
  provides tight control of cues.
• they allow learning to take place without the
  danger of injury
• a high level of interactivity

22
Training and Education

• Why use 3D environments for training
• Take on different perspectives
• Visualise 3D concepts
• Interact in real time
• Explore dangerous situations in safety
• Independent rehearsal
• Present realistic or abstract scenarios
• Promote different learning styles and teaching
  methods
• possess a high degree of flexibility

23
Training and Education

• Navigation and wayfinding
• simulations of buildings
• spatial orientation measures

24
Example studies

• Experimental work examining the way people
  encode spatial information from exploration of
  virtual environments
• A novel paradigm for investigating configural
  learning
• Transfer of spatial information
• The effect of repeated exposure to virtual
  environments
• Evidence for vertical asymmetry in spatial memory
• Work in progress examining transfer from a
  simulation of a multi-level complex building to
  the equivalent real world environment

25
Developing a Cognitive Map

• The quality of childrens cognitive maps is
  dependent on familiarity with the environment
  (repeated exposure, provision of landmarks,
  locomotion allowing self initiated exploration).
• Children with restricted mobility less chance of
  exploration of environment and thus develop
  poorer spatial cognitive maps
• E.g. Foreman Orencas et al (1989) children with
  physical disabilities significantly worse at
  spatial tasks
• Kozlowski and Bryant (1977) concluded that for
  people to show a good sense of direction it was
  necessary for them (a) to make a conscious effort
  to orientate themselves, and (b) to provide them
  with repeated exposure to the test environment.

26
Why Virtual Environments?

• Virtual environments (VEs) have as their core the
  simulation by computer of three dimensional
  space. The first defining feature of VEs is that
  they can be explored in real time with similar
  freedom to real world exploration. The second
  defining feature is that the user may interact
  with objects and events in the simulation.
• VEs particularly suitable due to Egocentric
  viewpoint, visual flow, safe to explore
  independently..
• Desktop with tailored interfaces

27
1. Background to Shortcut studies

• Few studies looking at the development of
  internal spatial representations in disabled
  children (Foreman et al, 1989b Simms 1987).
  Neither of these studies included a shortcut
  task.
• The ability to take shortcuts demonstrates the
  formation of an effective internal representation
  of space (Chapuis, Durup and Thinus-Blanc, 1987).
• In the present series of studies the experimental
  environment used by Chapuis et al (1987) was
  created as a 3-D simulation and this was used to
  test human participants.

28
The Shortcut study

• 24 able-bodied children with a mean age of 13.6
  years
• 34 physically disabled children, with a mean age
  of 14.1 years, was divided into two sub-groups
  based on their history of mobility (rated by
  their teacher as more mobile when they were
  younger or less mobile when they were younger).

29

• The environment consisted of five pathways
  connecting four rooms that appeared identical
  from the outside.
• N.B. A series of pilot studies had established
  that 4 large cues were optimal for spatial
  orientation within this environment

30
(No Transcript)
31

• children explored a simulated "maze" comprising
  four rooms linked by runways. In a subsequent
  test, they were asked to take shortcuts between
  target room locations.
• For example, they were asked to explore the route
  between room A and room B (all other pathways
  were blocked by no entry barriers). They were
  then asked to explore the path between rooms C
  and D, and then between rooms A and D. In the
  testing phase all the barriers were removed and
  participants were placed in room C and asked to
  find room B by the shortest route available.

32
Results

• In the first shortcut test the probability of
  choosing a correct path by chance alone was 33,
  while the probability was 50 on the second test.
• Approximately 70 of the able bodied group
  selected the shortcut correctly on both tests,
  significantly exceeding chance levels.
• The 'more mobile group,' while not performing
  better than chance on the first test
  (approximately 45 correct), scored better than
  chance on the second test with 80 correct.
• The 'less mobile' group only scored approximately
  45 correct on both tests and therefore did not
  perform better than chance on either test.

33
Conclusion

• These results add further weight to the argument
  that early independent exploration is essential
  for the development of cognitive spatial mapping
  ability in children, and suggest that these early
  influences persist at least into the early
  teenage years.

34
2. Studies of Repeated Exposure

• examining whether repeated exploration of several
  virtual environments promotes better encoding of
  virtual environments in general.
• A series of studies, including 3D vs 2D training
• skills disabled children acquired using virtual
  environments improved with exposure to successive
  environments
• However did not show an improvement in general
  spatial skills (assessed by Money Road test and
  Shepard and Cooper tests adapted for children).

35
3. The AshField study

• Experimental group were 7 physically disabled
  children, 6 boys and one girl. They had a mean
  age of 12.3 years. The control group consisted of
  7 undergraduate students, 2 female and 5 male
  with a mean age of 25.6 years.
• The primary section of Ash Field School in
  Leicester was created to-scale. The environment
  consisted of an entrance door with a corridor
  leading into a central area and nine rooms.

36
(No Transcript)
37
Procedure

• Experimental and control groups were subsequently
  taken to the real Ash Field school. Pointing
  accuracy was measured from two relative locations
  from which the children had completed computer
  pointing tasks in the simulation, along with a
  third untrained location. They were asked to
  estimate the direction of target objects from
  each of these locations using a hand operated
  pointing device.
• Finally, each participant completed two route
  tests. They were taken to a room and were asked
  to move directly to a target room. The first
  route was identical to the one trained within the
  simulation. The second route taken was between
  two different rooms.

38
Results

• Children were more accurate than controls in
  pointing to landmarks that were not directly
  visible from three separate testing sites
  (F(1,12) 67.54, p lt 0.01). They not only
  completed the tasks previously trained in the
  virtual school, but they also completed spatial
  tests that had not been trained in the virtual
  environment equally well.
• their way-finding ability (to adopt the shortest
  route between two locations) was also found to be
  more efficient than that of the control group
  (Mann-Whitney U test, z 2.01, p lt 0.05).

39
Conclusions

• These results support the conclusion that the
  children had acquired flexible, effective
  internal representations of the environment from
  the virtual simulation, enabling them to orient
  themselves from a number of different positions
  within the real environment.
• They add to the accumulating evidence that VE
  training transfers effectively to the real world
  and that this effect is evident even for people
  with physical disabilities whose spatial
  proficiency may be limited.

40
Key findings

• We are accumulating evidence of the positive
  effect of exploration of virtual environments on
  spatial navigational skills.
• We continue to examine whether skills learned in
  virtual environments transfer effectively to real
  world environments.
• The challenge is not only to examine transfer
  from a simulation to it's real world equivalent,
  but also to examine more generally whether
  spatial skills in the real world improve after
  virtual environment experience.
• Ultimate goal is to improve quality of life

41
Other issues

• Active versus passive exploration
• Drivers cognitive mapping skills
• Personal digital assistants and cognitive maps

42
References

• Foreman, N., Stanton, D., Wilson, P and Duffy H.
  (2003). Successful Transfer of Spatial Knowledge
  from a Virtual to a Real School Environment in
  Physically Disabled Children. Journal Of
  Experimental Psychology Applied, Vol. 9, no. 2,
  pp. 67-74
• Gibson, J. J. (1979). The Ecological Approach to
  Visual Perception. Boston Houghton Mifflin.
• Haber, R. N., Hershenson, M. (1973). The
  Psychology of Visual Perception. Holt, Rinehart
  and Winston, Inc.
• Lynch, K. (1960). The Image of the City.
  Cambridge, MA. MIT Press.
• McGee, M. (1982). Spatial Abilities The
  Influence of Genetic Factors. In M. Potegal (ed.)
  Spatial Abilities-Development and Physiological
  Foundation. New York Academic Press.
• O'Keefe, J., Nadel L. (1978). The hippocampus
  as a cognitive map. London Oxford University
  Press.
• Regian, J. W., Shebilske, W. L., Monk, J. M.
  (1992). Virtual Reality An Instructional Medium
  for Visuo-Spatial Tasks. Journal of
  Communication, 4, 136-149.
• Siegel, A. W., White, S. H. (1975). The
  development of spatial representations of
  large-scale environments. Advances in Child
  Development and Behavior, 10, 9-55.
• Stanton, D., Wilson, P. and Foreman, N. (2003).
  Human shortcut performance in a
  computer-simulated maze A comparative study.
  Spatial Cognition and Computation. 3(4)315-329.
• Stanton, D., Wilson, P., and Foreman, N. (2002).
  Effects of early mobility on shortcut performance
  in a simulated maze. Behavioural Brain
  Research.Elsevier, Vol. 136, pp. 61-66
• Thorndyke, P. W., Hayes-Roth, B. (1982).
  Differences in spatial knowledge acquired from
  maps and navigation. Cognitive Psychology, 14,
  560-589.
• Tolman, E. C. (1948). Cognitive maps in rats and
  men. Psychological Review, 55, 189-208.
• Wilson, P. N., Foreman, N., Gillett, R., and
  Stanton, D. (1997). Active versus passive
  processing of spatial information in a computer
  simulated environment. Ecological Psychology,
  9(3), 207-222.