Sequence effects in simple cognitive tasks [Elektronische Ressource] : the multiple-weighting-systems hypothesis / Dragan Rangelov. Betreuer: Hermann Müller

Dragan Rangelov

Sujets

• Gesundheit

Informations

• Publié par: ludwig-maximilians-universitat_munchen
• Publié le: 01 janvier 2010
• Nombre de lectures: 8
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

SEQUENCE EFFECTS IN SIMPLE COGNITIVE TASKS:
THE MULTIPLE-WEIGHTING-SYSTEMS HYPOTHESIS
Dragan Rangelov

Dissertation
at the Graduate School of Systemic Neurosciences
of Ludwig-Maximilians-University Munich
Submitted by
Dragan Rangelov
from Bela Palanka, Serbia

Munich, July 20102

















Supervisor: Prof. Dr. Hermann J. Müller
Second expert appraiser: Dr. Michael Zehetleitner
Date of the oral defence: 19.10.2010 3

AKNOWLEDGMENTS
I would like to thank Hermann Müller, who offered me an opportunity to work in an
inspiring scientific environment. His comments and advices shaped and improved my work in
many ways. Thanks to Thomas Töllner for helping me design an electrophysiological study,
as well as collect and analyze the data. He was always a source of useful advices every step
along the way. Many thanks to my colleagues and friends Donatas Jonikaitis, Yi-Huang Su,
and Lana-Marija Ilanković, for sharing the challenges of the PhD life with me. Finally, thanks
to Michael Zehetleitner, for teaching me many small and big things in science, and,
especially, for serving as a great role model both professionally and personally. 4

ABSTRACT
What we did recently influences our current behavior substantially. For example,
responses are speeded if observers are to discriminate an item based on its color twice (e.g.,
color – color discrimination in trials n-1 and n, respectively) compared to when the task-
relevant dimension changes across trials (e.g., orientation – color discrimination). The
dimension repetition effect (or DRE) was explained in two mutually exclusive ways: (i)
selection-based approach assumes that DREs are a consequence of visual selection processes,
while (ii) response-based approach claims these effects to originate from the later processes of
response selection. Importantly, neither of the accounts can fully explain available findings on
DREs: (i) selection-based approach cannot explain strong DREs observed when visual
selection is very easy and fast, while (ii) response-based account cannot explain strong DREs
in paradigms where the sequence of perceptual dimensions is response-irrelevant.
In my work, I focused on developing a theoretical framework which should account for
all the available data by assuming existence of different dimension-sensitive mechanisms that
affect different processing stages (visual selection, perceptual analysis). Additionally, a novel
paradigm was developed in which two tasks, for which DREs were previously reported,
alternated randomly across trials. Across two consecutive trials, the task could either repeat or
change. Independently of the task sequence, the task-relevant dimension could either change
or repeat. Different experiments used different tasks, that could, on the Multiple-Weighting-
Systems (or MSW) hypothesis, engage either the same (e.g., both tasks engaged the visual
selection mechanism) or different DRE mechanisms. Behavioral and electrophysiological data
showed: (i) significant DREs across task changes when the two tasks engaged the same
hypothesized mechanism, (ii) no DREs across task changes when the tasks engaged different
systems. Overall, the data support the MWS hypothesis, in contrast to the single-mechanism
accounts. 5

TABLE OF CONTENTS
Abstract ...................................................................................................................................... 4
1 The Multiple-Weighting-Systems hypothesis: General introduction and discussion ......... 7
1.1 Problem ........................................................................................................................ 8
1.2 The Multiple-Weighting-Systems hypothesis ............................ 11
1.3 Overview of behavioral findings ................................................................................ 12
1.4 Electrophysiological correlates of S and P systems ............... 14 ω ω
1.5 Conclusions ................................................................................................................ 17
1.6 References .................. 18
2 Dimension-specific intertrial priming effects are task-specific: Evidence for multiple
weighting systems ............................................................................................................. 19
2.1 Abstract ...................... 20
2.2 Introduction ................................................................................................................ 21
2.3 General Method .......... 28
2.4 Experiment 1 .............................................................................................................. 32
2.5 Experiment 2 .............. 39
2.6 Experiment 3 .............................................................................................................. 44
2.7 General Discussion .... 49
2.8 References .................................................................................................................. 60
3 Independent dimension-weighting mechanisms for visual selection and stimulus
identification ..................................................................................................................... 63
3.1 Abstract ...................... 64
3.2 Introduction ................................................................................................................ 65
3.3 General Method .......... 74
3.4 Results ........................................................................................................................ 80 6

3.5 General Discussion .................................................................................................... 87
3.6 References .................................................................................................................. 98
4 The Multiple-Weighting-Systems hypothesis: Theory and empirical support ............... 101
4.1 Abstract .................................................................................................................... 102
4.2 Introduction .............. 103
4.3 An integrative approach: The Multiple-Weighting-Systems hypothesis ................. 109
4.4 General Method ........................................................................................................ 121
4.5 Results ...................................................................................................................... 124
4.6 General Discussion .................................. 129
4.7 References ................................................................................................................ 134
5 Published and submitted manuscripts ............. 137 7










1 THE MULTIPLE-WEIGHTING-SYSTEMS HYPOTHESIS: GENERAL
INTRODUCTION AND DISCUSSION 8

1.1 Problem
Adaptive behavior is considered to be the main requirement for complex organisms (such
as humans) to survive in a vast variety of different environments. Controlling what are the
most appropriate responses in a particular environment is the function of the cognitive system.
Conceptually, control of behavior necessitates several processing stages prior to overt actions:
(i) selection of the task relevant item, (ii) perceptual analysis of the selected item, and (iii)
selection of the appropriate response to the analyzed item. To illustrate, the task might be to
turn on the light by using a switch (i.e., target) from a switch board. An initial processing step
would be to select a potential target switch from several other switches. Following target
selection, perceptual analysis processes should determine whether the selected item matches
the target template or not. Finally, the outcome of the perceptual analysis stages is mapped to
possible responses resulting in either turning on the light (if the right switch was identified) or
selecting another switch. In summary, behavioral control could be thought of as a series of
decisions taking place at different processing stages (selection, perceptual analysis,
responding); each decision being controlled by the task rules (stimulus-response mapping)
and properties of the current stimulation.
While one might assume that task rules and current stimulation fully determine the
decision chain from stimulus to behavior, experimental evidence suggests that other factors
also play an important role. Recent experience, for example, has been demonstrated to
profoundly affect processes of visual selection, perceptual analysis and response selection.
Found and Müller (1996) presented stimulus displays containing many items (i.e., multi-item
displays, see Figure 1.1). On some trials, all items were identical (i.e., target-absent displays),
while in a portion of trials there was one item (i.e., singleton target) different from the others
in some respect (e.g., color or orientation). The task was to detect presence/absence of the
singleton target and respond by pressing corresponding response buttons. Across two
consecutive trials (n-1 and n) in both of which the target was present, the dimension of 9

distinction between the target an