作　　者： ;

机构地区： 广东外语外贸大学

出　　处： 《外语与外语教学》 2011年第6期 26-29,共4页

摘　　要： 借助脑认知、原型和认知负荷模型，计算语言学顿悟研究可实现程序上的应用。通过分析机能固着、心理成规和心理定势可知，“弃旧”是顿悟出现的条件，格式塔分析表明顿悟是知觉场的“平衡-失衡-再平衡”，“迎新”是顿悟产生的根本。本文分析了顿悟的Ns算法，从中得出，语义触发时关键启发信息激活具有瞬时性，循环算法感知具有闭合性，认知折返时错码调试具有回溯性，跨越解码时“原-伪-新”解读模式具有螺旋性。顿悟算法研究证明了程序语言在花园幽径句解码中具有可行性。 ＂Insight＂ study based on computational linguistics can be a program application with the help of theoretical models of brain cognition, prototype and cognitive load. By analyzing theo- ries of functional fixedness, psychological conventions and mind-set, we find that ＂negation of pre-model＂ brings about the action of ＂insight＂. The Gestalt psychology shows ＂insight＂ is a ＂balance-imbalanee-rebalanee＂ conscious process , and ＂establishment of rico-model＂ is the fundamental reason why＂ insight＂ is created. Based on the NS algorithm for ＂insight＂ ,this study identifies the following characteristics：Semantic trigger provides the activation of key heuristic information momentarily;Loop algorithm shows the process of change in a circle; The feature of cognitive half-return presents the backtracking when system debugs error codes in real time; The skipping decoding from premodel to pseudo-model and neo-model exhibits the sign of circular helix. From the algorithm-based ＂insight＂ research, this study draws a conclusion that programming language is feasible for the decoding of garden path sentences.

关 键 词： 计算语言学 认知 句法 花园幽径现象 顿悟

分 类 号： [H0]

领　　域： []