Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Associative analysis of spatial learning in environments with a distinctive shape.

Horne, Murray R. 2009. Associative analysis of spatial learning in environments with a distinctive shape. PhD Thesis, Cardiff University.

[thumbnail of U584366.pdf] PDF - Accepted Post-Print Version
Download (5MB)

Abstract

The aim of this thesis was to evaluate the proposal by Miller and Shettleworth (2007) that learning about geometric cues in environments with a distinctive shape is governed by a competitive learning rule (e.g., Rescorla & Wagner, 1972). To do this, in all experiments, rats were trained to locate a hidden platform by reference to the shape of a swimming pool. Chapter 2 (Experiments 1 -4) assessed whether a landmark suspended above the platform would overshadow learning about geometric cues. No overshadowing was recorded, even when the salience of the geometric cues was reduced. These findings are inconsistent with the model of Miller and Shettleworth (2007). In Chapter 3 (Experiments 5-7), a blocking paradigm was used. When rats were given extended pre-training with a landmark above the platform, only then did the landmark successfully block learning about geometric cues. However, some unexpected findings suggested that perhaps the spatial cues were competing for attention rather than associative strength. The experiments in Chapter 4 (Experiments 8 and 9) successfully demonstrated superconditioning of geometric cues by an inhibitory landmark providing convincing evidence that learning about geometric cues is governed by the principles of associative learning. Miller and ShettlewortiVs (2007) model however, failed to predict this outcome. Experiments 10-13 in Chapters 5 and 6 showed that associations formed between geometric and non-geometric cues. This outcome provides the basis for a viable explanation for potentiation and for the past failures to find cue competition in the spatial domain. The empirical findings of this thesis show that learning about geometric cues is not entirely void of associative processes as once thought. A number of recent models of spatial learning are discussed.

Item Type: Thesis (PhD)
Status: Unpublished
ISBN: 9781303189296
Date of First Compliant Deposit: 30 March 2016
Last Modified: 13 Jan 2014 11:51
URI: https://orca.cardiff.ac.uk/id/eprint/55831

Actions (repository staff only)

Edit Item Edit Item

Downloads

Downloads per month over past year

View more statistics