Recalling Recollections According to Temporal Contexts - Applying a Novel Cognitive Knowledge Representation Approach
In an experiment whose purpose is to model the interruptions phenomenon and its impact on tasks achievement, we show that in some cases, such as temporarily interrupted realisations of cooking recipes, the popular ACT-R knowledge representation approach cannot offer a model that is highly close to the natural human behaviour. We emphasize the incapacity of ACT-R to reproduce correctly the recall of information in a temporal context to achieve properly the action plan for a recipe realisation after a momentarily interruption. We propose an alternative knowledge representation theory, which uses additional knowledge structures that are inspired from the human memory, to faithfully reproduce a usual human behaviour in resuming correctly suspended activities.
Mehdi Najjar Philippe Fournier-Viger Jean-Francois Lebeau André Mayers
Department of Computer Science, University of Sherbrooke 2500, Bld. de lUniversité, Sherbrooke, QC Department of Computer Science, University of Quebec at Montreal C.P. 8888, Succ. Centre-ville, Mont
国际会议
Firth IEEE International Conference on Cognitive Informatics(第五届认知信息国际会议)
北京
英文
424-433
2006-07-17（万方平台首次上网日期，不代表论文的发表时间）