C5 Landscape Projects Research Venues Prospectus Opportunities Personnel
Landscape, Projects, Research, Venues, Prospectus, Opportunities, Personnel
  RCSP Research

Adaptive Learning: Smart Space

Keywords
concept learning, conceptual development, features, perceptual learning, stimulus encoding, cognitive science, architectural space, social space, ecology, environmental hermeneutics, proxemics

Description
Central to the successful description of the environment (social or physical) is recognition of the capacity of various organisms (RCSPs) to describe the same environment differently, i.e. each organism has access to different perceptions that describe its world. This is analogous to people using various languages to describe the same environment; in using varying languages each presents a different picture of the same environment. In this respect flexible feature learning parallels the theory of environmental hermeneutics.
 

Diagram 1
 
  Environmental schematic
 
    Established cognitive science holds that "cognitive processes such as categorization and object recognition operate on a fixed set of perceptual or conceptual features, which are the building blocks for complex object representations." Schyns, Goldstone & Thibaut argue that categorization and object recognition often require the creation of new features. "The featural repertoire, rather than being fixed, is dependent on situational demands, novel categorization requirements, and environmental contingencies." (BBS p. 1) SGT provide an account of feature learning in which the components of a representation have close ties to the categorization history of the organism. (BBS p. 2 ) They write:

"The individual knows what the categories are from external feedback, that is, the consequences of their miscategorizations. . . Flexible features open the possibility that the same input is differently perceived and analyzed before being categorized. Hence a complete theory of conceptual development should not only explain the ways in which object features are combined to form concepts, it should also explain the development of the features participating in the analysis of the input." (BBS p. 5 )
 

Diagram 1
 
  Learning configuration
 
    Two types of category learning should be distinguished. Fixed space category learning occurs when new categorizations are representable with the available feature set. Flexible space category learning occurs when new categorizations cannot be represented with the features available. Whether fixed or flexible, learning depends on the featural contrasts and similarities between the new category to be represented and the individual's existing concepts.

Behavioral and Brain Sciences (1998) 21, 1-54, Philippe G. Schyns, Robert L. Goldstone, Jean-Pierre Thibaut

Methodology
In employing adaptive learning theory as set forth in the theory of flexible feature learning together with theories of environmental hermeneutics and proxemics, this experiment will determine the behavior of multiple autonomous surveillance probes in order to generate data from a multidimensional environment and in doing so provide a multifarious description of the given environment.

1. The five RCSP human controllers (agents) are organized to not be organized in any specific manner but are autonomous agents controlling their individual cars according to an individualized rule set. Each controller/RCSP is enabled with its individualized "vocabulary", i.e. what it recognizes.

2. The intention is to describe the multiple dimensions of a given environment where each RCSP has the opportunity to receive the same stimuli, but through its individualized filtering process it collects different stimuli, and/or perceives the same stimuli differently.

3. Together the stimuli, physical properties, to which the organism is sensitive define a multideminsional space. The space is defined by just those physical dimensions that are important to the organism.

4. Dependent on project goal (what is to be described), different perceptual capabilities are attributed to the various RCSPs. Though a project goal isn't necessary but is allowed to emerge.

5. Some behavior makes sense in terms of the events that preceed it; other behavior makes more sense in terms of the events that follow it. Reflexes are behavior of the first kind. Operant behavior is guided by its consequences where operant refers to an essential property of a goal directed behavior: that it have some effect on the environment.

6. Behavior is guided in some way by the discrepancy between the present state of affairs and some often unknown ideal state.
 

 

[ Home ] [ RCSP Research ]