% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@PHDTHESIS{Kiefer:968610,
author = {Kiefer, Christian Matthias},
othercontributors = {Shah, Nadim Joni and Grün, Sonja Annemarie and Steland,
Ansgar},
title = {{V}isual processing of naturalistic images: free viewing,
guided visual searching, and memorization},
school = {RWTH Aachen University},
type = {Dissertation},
address = {Aachen},
publisher = {RWTH Aachen University},
reportid = {RWTH-2023-08679},
pages = {1 Online-Ressource : Illustrationen, Diagramme},
year = {2023},
note = {Veröffentlicht auf dem Publikationsserver der RWTH Aachen
University; Dissertation, RWTH Aachen University, 2023},
abstract = {Under natural conditions, eye movements are used to
actively gather visual information from the environment, a
process referred to as ‘active vision’. Despite active
vision forming the basis for visual exploration in everyday
life, research has often focused on simplified, artificial
scenarios with restricted eye movements. As a result, the
mechanisms underlying active vision remain comparatively
unexplored. Particularly, it is unclear how goal-directed
behavior such as during the search for a particular object
influences the visual processing during active vision. In
this dissertation, three distinct modes of active vision
were investigated in a study of 31 subjects: performing a
guided visual search and memorizing objects were compared to
the free viewing of naturalistic images.
Magnetoencephalography was used in conjunction with an
eye-tracking system to simultaneously record the brain
activity and the eye movements during the tasks while the
subjects were viewing naturalistic images with embedded
objects. To answer the question of how visual processing is
affected by goal-directed behavior, causal interactions in
the brain were compared for the free viewing and the visual
searching tasks. First, regions of interest exhibiting
significant fixation-related evoked activity were determined
using spatiotemporal cluster permutation testing. Then,
using generalized partial directed coherence, the
whole-brain causality networks between these regions were
computed. Four regions, namely the supramarginal gyrus, the
superior temporal gyrus, the transverse temporal gyri, and
the posterior insula, formed during free viewing a cluster
in both hemispheres that was highly inter-connected. A
near-identical cluster emerged during searching in the right
hemisphere with the right supramarginal gyrus as a central
node in the network. Based on the results, all four regions
are likely involved in guiding attention and visual
processing. Specifically, the right supramarginal gyrus
might be involved in the integration of visual input related
to objects. Furthermore, it might provide the supplementary
eye field with information for the purpose of updating the
search priority map, which guides the eye movements towards
the search target. Finally, the random convolutional kernel
transformation (ROCKET) classifier for time series was
applied to the fixation-related evoked activity of object
fixations performed during the memorization and the free
viewing tasks. It was possible to classify unseen fixation
data based on the task with $93\%$ accuracy. Furthermore,
with the same method it was also possible to differentiate
between subjects based on resting-state recordings with
$93\%$ accuracy. This appears to be the first proof of
concept that neuronal fingerprinting can be achieved
directly on time series data using ROCKET.},
cin = {535000-5 ; 934020 ; 934010 / 130000},
ddc = {530},
cid = {$I:(DE-82)535000-5_20140620$ / $I:(DE-82)130000_20140620$},
pnm = {DFG project 491111487 - Open-Access-Publikationskosten /
2022 - 2024 / Forschungszentrum Jülich (OAPKFZJ)
(491111487) / GRK 2416 - GRK 2416: MultiSenses-MultiScales:
Neue Ansätze zur Aufklärung neuronaler multisensorischer
Integration (368482240) / HBP SGA3 - Human Brain Project
Specific Grant Agreement 3 (945539) / HMC - Helmholz
Metadata Collaboration $(HMC_20200306)$ / JL SMHB - Joint
Lab Supercomputing and Modeling for the Human Brain (JL
SMHB-2021-2027)},
pid = {G:(GEPRIS)491111487 / G:(GEPRIS)368482240 /
G:(EU-Grant)945539 / $G:(DE-HGF)HMC_20200306$ /
G:(DE-Juel1)JL SMHB-2021-2027},
typ = {PUB:(DE-HGF)11},
doi = {10.18154/RWTH-2023-08679},
url = {https://publications.rwth-aachen.de/record/968610},
}
guest ::