Dimension-based processing in visual pop-out search [Elektronische Ressource] / vorgelegt von Katharina Mahn

Katharina Mahn

Sujets

• Psychologie

Informations

• Publié par: ludwig-maximilians-universitat_munchen
• Publié le: 01 janvier 2005
• Nombre de lectures: 9
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

Dimension-basedProcessingin
Visual Pop-outSearch
Katharina Mahn
München2005Dimension-basedProcessingin
Visual Pop-outSearch
Katharina Mahn
Dissertation
an der
der Ludwig–Maximilians–Universität
München
vorgelegt von
Katharina Mahn
aus Halle/Saale
München, den 10.10.2005Erstgutachter: Prof. Dr. Hermann Müller
Zweitgutachter: PD Dr. Joseph Krummenacher
Tag der mündlichen Prüfung: 16.12.2005The brain is wider than the sky,
For put them side by side
The one the other will include
With ease – and you beside.
Emily Dickinson
If we manage to to make an action or sensation appear nakedly
simple, it is just because we are concealing all the effort
that went into setting up the moment.
John McCrone1Contents
1 Introduction 4
1.1 Theories of Visual Search . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.2 Target redundancy in visual search . . . . . . . . . . . . . . . . . . . . . . . 18
1.3 Spatial Aspects of Visual Search . . . . . . . . . . . . . . . . . . . . . . . . 26
1.4 Masking and Visual Search . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2 Pop-outversus Compound search 43
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.2 Experiment 1: Visual search for a single feature target . . . . . . . . . . . . 50
2.3 Experiment 2: Visual search for compound targets . . . . . . . . . . . . . . 65
3 Visual FeatureSearchin Neglect 83
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
2Contents
3.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
4 The role ofLeft Lateral FrontopolarCortex 150
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
4.2 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
4.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
4.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
5 General Discussion 167
6 Zusammenfassung 179
7 Appendix 188
7.1 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
7.2 Curriculum Vitae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
7.3 Selbstständigkeitserklärung . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
References 192
31 Introduction
We are constantly confronted with an ever changing visual scenery crowded with
coloured, moving or stationary, bright or dark objects of different shapes and sizes.
Only some of them are relevant in order to achieve behavioural goals. However, in
everyday life, we usually do not feel overwhelmed with information or a multitude
of decisions what to ’see’ or react upon first.
This is in part due to the fact that we can rely on a highly effective interplay
between sensory (bottom-up) input to the brain and top-down processing - that is,
selection of relevant signals guided by already learned and memorized knowledge
about the world. Thus, we move in the environment while constantly searching,
consciously and/or unconsciously, for signals or information relevant to our current
situation and our behavioural goals.
A stimulus can reach significance (or become salient) mainly in two ways: Firstly,
it is simply processed more efficiently than others. These so-called salient stimuli
will, in the presence of other less salient ones, almost always be processed with
4CHAPTER 1. INTRODUCTION
priority. For example, a large object will be seen first amongst small items, a bright
object faster than a dark one, or a coloured item will win over a (non-coloured)
grey one. This ’natural’ salience arises from inherent properties of sensory visual
processes shaped by long-term experience (Lamme, 2003).
Secondly, the stimulus is of a certain behavioural relevance. Standing in front of
a vegetable area in the supermarket and looking for potatoes is successful although
the potatoes might be surrounded by brightly coloured red tomatoes or yellow pep-
pers. Despitenotnecessarilybeingequippedwith’naturally’salientproperties,their
presence can be detected fast and efficiently because the sensory pathways required
to process a particular (non-salient) item are pre-activated in a top-down fashion by
higher rather than early sensory brain areas and thus processing is facilitated.
A third type of mechanism rendering stimuli salient, somehow lying in between
the two mechanisms described above, is referred to as priming. The processing of
a given stimulus is prioritized by an event earlier in time. This might especially be
the case if the previous stimulus shares properties such as size, colour, or location
(Maljkovic & Nakayama, 1996 ) or spatial relation to other items (contextual cuing,
e.g. Chun & Jiang, 1999 ) in the visual field. Stimuli processed earlier leave a
(memory) trace of activation within the processing system and/or modulate the
allocation of attentional resources. The following stimuli, matching the first in at
leastsomecharacteristics(features),mightbenefitfromthispre-activationandtheir
processing is expedited.
5CHAPTER 1. INTRODUCTION
An experimental paradigm which has been extensively used to address the ques-
tion of how a certain stimulus is selected from a visual scene, is the visual search
paradigm. Typically, the observer is asked to report the presence or absence of a
target item within an array of multiple elements. The characteristics of the object
tosearchformightbepre-specified, or, asitisthecaseintheexperimentspresented
below,thetargetitemdiffersfromnon-targetitems(distractors)inasingleattribute
(feature). Ifthetargetdefiningfeatureissufficientlysalientandthedistractoritems
are homogeneous and task-irrelevant, the target phenomenally appears to ’pop out’
of the display. To perform this kind of task, target identity need not necessarily be
known in advance and it may vary unpredictably from trial to trial. However, the
difference of one feature of a single item relative to a homogeneous set of distractor
items, allows for a rapid ’self-definition’ of the target in the ongoing search process.
The ’pop-out’ phenomenon is central to the work presented below. Its under-
lying mechanisms are explored in visual detection and in a compound search task
(Chapter 2), in a group of patients with an visual-spatial deficit: hemi-spatial ne-
glect (Chapter 3), and in patients with damage to the left lateral frontopolar cortex,
a structure previously found to be specifically activated during dimension changes
across a sequence of trials (Chapter 4).
6