CGSC170 Lecture Notes - Lecture 4: Semantic Network, Search Algorithm, Spreading Activation
14 Nov 2017
School
Department
Course
Professor
Document Summary
Examples: pandemonium model, artificial neural networks. Key features: nodes, links between nodes, activation, thresholds, weights. Operations can be distributed: very abstract. Input output: the focus of network models is not operations input output relationship. Organization can affect: search operations, (cid:862)look up(cid:863, assessing truth value of propositions. The format of the information affects the search process. As target number appears farther to the right reaction time increases. Brain is scanning from left to right: same direction as reading. In the sternberg search task, the number are presented in a string/array: the task biases the form of the representation, representation biases search algorithm. Strength of activation diminishes with each degree of separation. Deter(cid:373)i(cid:374)e (cid:449)hether it is a real (cid:449)ord, or (cid:374)o(cid:374)se(cid:374)se. Lexical priming: words are presented in pairs. If the words are related, youre quicker to determine that they are real words. Why: first word activates node in semantic network, spreading activation related nodes are activated.
