The University of Southampton
University of Southampton Institutional Repository

The influence of context on object recognition

The influence of context on object recognition
The influence of context on object recognition

The thesis explores how non-target objects influence object recognition.  In all five experiments, sets of non-target objects are used to generate ‘scene’ contexts and these are presented so that they surround individual target objects.  The foci of investigation are (1) whether scene context effects with multiple objects exist, (2) if they exist are they perceptual or due to response biases, (3) what role does the distribution of attention play in the generation of scene context effects, and (4) what is the time-course of their generation?

Experiments 1-3 found that target objects were named more accurately when non-target objects were semantically related (context-consistent) than semantically unrelated (context-inconsistent).  However the magnitude of the context effect was mediated by visual attention.  A significant effect was only achieved when all objects (targets and non-targets) were within an attended region and not when non-targets fell outside of this region.

Experiment 4 used a paradigm conceptually related to the Reicher-Wheeler paradigm to provide a measure of response bias.  A six-alternative forced-choice response design demonstrated a significant influence of scene context even after the data were corrected for response bias; suggesting a perceptual/representational locus to the scene context effect generated by non-target objects on target objects.

Experiments 4 and 5 also manipulated the time-course of the onset of non-target objects relative to target objects.  The results showed that at least 52msec was required for the presence of non-target objects to influence recognition of target objects.  In other words, the scene context effect for multiple non-target objects requires at least 52msec to accumulate.

In summary, scene context effects for multiple non-target objects on target objects directly influence the representational processes of target recognition.  Furthermore their magnitude is dependent on the distribution of attention across the visual field and the temporal relationship of non-targets and targets.  How these factors influence the modelling of object recognition is also considered.

University of Southampton
Auckland, Mark Edmund
d313c513-77bb-48cd-a9cb-6b559d6a1dcd
Auckland, Mark Edmund
d313c513-77bb-48cd-a9cb-6b559d6a1dcd

Auckland, Mark Edmund (2005) The influence of context on object recognition. University of Southampton, Doctoral Thesis.

Record type: Thesis (Doctoral)

Abstract

The thesis explores how non-target objects influence object recognition.  In all five experiments, sets of non-target objects are used to generate ‘scene’ contexts and these are presented so that they surround individual target objects.  The foci of investigation are (1) whether scene context effects with multiple objects exist, (2) if they exist are they perceptual or due to response biases, (3) what role does the distribution of attention play in the generation of scene context effects, and (4) what is the time-course of their generation?

Experiments 1-3 found that target objects were named more accurately when non-target objects were semantically related (context-consistent) than semantically unrelated (context-inconsistent).  However the magnitude of the context effect was mediated by visual attention.  A significant effect was only achieved when all objects (targets and non-targets) were within an attended region and not when non-targets fell outside of this region.

Experiment 4 used a paradigm conceptually related to the Reicher-Wheeler paradigm to provide a measure of response bias.  A six-alternative forced-choice response design demonstrated a significant influence of scene context even after the data were corrected for response bias; suggesting a perceptual/representational locus to the scene context effect generated by non-target objects on target objects.

Experiments 4 and 5 also manipulated the time-course of the onset of non-target objects relative to target objects.  The results showed that at least 52msec was required for the presence of non-target objects to influence recognition of target objects.  In other words, the scene context effect for multiple non-target objects requires at least 52msec to accumulate.

In summary, scene context effects for multiple non-target objects on target objects directly influence the representational processes of target recognition.  Furthermore their magnitude is dependent on the distribution of attention across the visual field and the temporal relationship of non-targets and targets.  How these factors influence the modelling of object recognition is also considered.

Text
985527.pdf - Version of Record
Available under License University of Southampton Thesis Licence.
Download (2MB)

More information

Published date: 2005

Identifiers

Local EPrints ID: 465719
URI: http://eprints.soton.ac.uk/id/eprint/465719
PURE UUID: c9fab459-0e1b-49f5-9da3-3b69d2fab164

Catalogue record

Date deposited: 05 Jul 2022 02:45
Last modified: 16 Mar 2024 20:20

Export record

Contributors

Author: Mark Edmund Auckland

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×