Why Do We SLIP to the Basic Level? A Formal Model

Gosselin, Frederic (2000) Why Do We SLIP to the Basic Level? A Formal Model. PhD thesis, University of Glasgow.

Full text available as:
[img]
Preview
PDF
Download (10MB) | Preview

Abstract

This dissertation introduces a new measure of basic-level performance (Strategy Length & Internal Practicability, SLIP). SLIP implements two computational constraints on the organisation of categories in a taxonomy: the minimum number of feature tests required to place the input in a category (strategy length) and the ease with which these tests are performed (internal practicability). The predictive power of SLIP is compared with that of four other basic-level measures: context model (Medin & Schaffer, 1978; modified by Estes, 1994), category feature-possession (Jones, 1983), category utility (Corter & Gluck, 1992), and compression measure (Pothos & Chater, 1998a), drawing data from the empirical work of Rosch et al. (1976), Murphy and Smith (1982), Mervis and Crisafi (1982), Hoffmann and Ziessler (1983), Corter, Gluck and Bower (1988), Murphy (1991), Lassaline (1990), Tanaka and Taylor (1991), and Johnson and Mervis (1997). Nine experiments further test the validity of the computational constraints of SLIP using computer-synthesised 3-D artificial objects, artificial scenes, and letter strings. The first five experiments examine the two constraints of SLIP in verification. Experiment 1 isolates the effect of strategy length on basic-levelness, and Experiments 2a and 2b that of internal practicability. Experiment 3 examines the interactions between the two factors. Experiment 4 tests, whether, as predicted by SLIP, there is a linear relationship between strategy length and response times. The last four experiments study the two computational constraints in naming. Experiment 5a isolates the effect of strategy length, and Experiment 5b that of internal practicability. Experiment 6 examines the time-course of the effect of strategy length. Finally, Experiment 7 looks at the effect of robustness (i.e., the idea an approximate categorisation is better than none) on the order of feature tests in length 2 strategies.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Additional Information: Adviser: Dave Finnigan
Keywords: Experimental psychology
Date of Award: 2000
Depositing User: Enlighten Team
Unique ID: glathesis:2000-75957
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 19 Nov 2019 17:12
Last Modified: 19 Nov 2019 17:12
URI: http://theses.gla.ac.uk/id/eprint/75957

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year