ACT-R is almost a Model of Primate Task Learning: Experiments in Modelling Transitive Inference

M A Wood; J C S Leong; J J Bryson

ACT-R is almost a Model of Primate Task Learning: Experiments in Modelling Transitive Inference

M A Wood, J C S Leong, J J Bryson

Department of Computer Science

Research output: Contribution to conference › Paper › peer-review

Abstract

We present a model of transitive inference (TI) using ACT-R which strengthens the hypothesis that TI is not dependent on underlying sequential ordering of stimuli, but rather on the learning of productions. We nevertheless find a weakness in the ACT-R sub-symbolic learning system and suggest improvements.

Original language	English
Pages	1470-1475
Number of pages	6
Publication status	Published - 2004
Event	The 26th Annual Meeting of the Cognitive Science Society (CogSci 2004) - Chicago Duration: 1 Jan 2004 → …

Conference

Conference	The 26th Annual Meeting of the Cognitive Science Society (CogSci 2004)
City	Chicago
Period	1/01/04 → …

Cite this

@conference{2da1362588684969b40996c3accb0c02,

title = "ACT-R is almost a Model of Primate Task Learning: Experiments in Modelling Transitive Inference",

abstract = "We present a model of transitive inference (TI) using ACT-R which strengthens the hypothesis that TI is not dependent on underlying sequential ordering of stimuli, but rather on the learning of productions. We nevertheless find a weakness in the ACT-R sub-symbolic learning system and suggest improvements.",

author = "Wood, {M A} and Leong, {J C S} and Bryson, {J J}",

year = "2004",

language = "English",

pages = "1470--1475",

note = "The 26th Annual Meeting of the Cognitive Science Society (CogSci 2004) ; Conference date: 01-01-2004",

}

TY - CONF

T1 - ACT-R is almost a Model of Primate Task Learning: Experiments in Modelling Transitive Inference

AU - Wood, M A

AU - Leong, J C S

AU - Bryson, J J

PY - 2004

Y1 - 2004

N2 - We present a model of transitive inference (TI) using ACT-R which strengthens the hypothesis that TI is not dependent on underlying sequential ordering of stimuli, but rather on the learning of productions. We nevertheless find a weakness in the ACT-R sub-symbolic learning system and suggest improvements.

AB - We present a model of transitive inference (TI) using ACT-R which strengthens the hypothesis that TI is not dependent on underlying sequential ordering of stimuli, but rather on the learning of productions. We nevertheless find a weakness in the ACT-R sub-symbolic learning system and suggest improvements.

M3 - Paper

SP - 1470

EP - 1475

T2 - The 26th Annual Meeting of the Cognitive Science Society (CogSci 2004)

Y2 - 1 January 2004

ER -

ACT-R is almost a Model of Primate Task Learning: Experiments in Modelling Transitive Inference

Abstract

Conference

Fingerprint

Cite this