Architecture of Symbiosis

Symposium:


Session Title:

  • Architecture

Presentation Title:

  • Architecture of Symbiosis

Presenter(s):



Abstract:

  • Herbert Simon [1981] notes that complexity of behavior resides in the hybrid condition of organism and environment. ted  three units of analysis are posited, the organism, the environment and implicitly the behavior as well. This paper examines the relationship between humans and the environments that they create and argues that the nature of these environments is undergoing a fundamental shift, one that suggests that the proper units of analyses are the hybrid conditions rather than individual components that comprise them. Brooks [1991], following Simon, has undertaken the study of intelligence through the agency of robotics. Intelligence is understood in this context as an attribution made on the basis of the relationship between behavior and the environmental conditions. Intelligence is not inferred from behavior, that is: it has no independent or a priori existence that can be discovered by observation, but is completely bound into and considered an attribute of the contextual operations of the agent. Intelligence so conceived does not reside in the sensor-processor functions of the robot, but in the hybrid of “robot within the environment”. There is evidence that this perspective is not limited to silicon and aluminum but may be true as well of humans and the contexts in which they exist.

     

    In the computer game Tetris, irregularly shaped ”bricks” drop from the top of the screen and are to be rotated and translated to form compact walls building up from the bottom of the screen. Kirsh and Maglio [1994] found that many of the operations performed by players, even those that had developed considerable skill in the game, were counter-productive in the sense that they were not instrumental actions toward the goal of constructing walls. The falling shapes were repeatedly rotated in order to perceive advantageous orientations. The time required to enact these rotations on the screen and to visually evaluate the applicability of the resulting orientation is less than the time it takes to do the rotation and evaluation mentally. While this finding may surprise few Tetris players, why should it be so? The answer lies in the genetically determined structure of the nervous system and its development as an adaptive device.

     

    Full text p.33-34

Category: