Abstract code network as a model of perceptual memory

作者:

Highlights:

摘要

According to Structural Information Theory human observers represent important aspects of the information of visual patterns in abstract sequential codes. The theory proposes a syntax for the structural form of these codes. In this paper a generative syntax is presented that is suitable for the generation of a theory based computer model of the human memory. The model consists of many abstract codes. The memory is assumed to be actively used to interpret new information. Our aim is to simulate this interpretation process. The present development of memory data structures and of processes working upon them is a first step.

论文关键词:Structural information theory,Sequential codes,Top down process,Perceptual learning,Associative network,Coding theory of pattern perception,Structural pattern recognition,Pattern interpretation

论文评审过程:Received 5 June 1985, Revised 14 November 1985, Available online 19 May 2003.

论文官网地址:https://doi.org/10.1016/0031-3203(87)90024-0