國立虎尾科技大學 |

The Geometry of Observed Action Representation During Natural Vision.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	The Geometry of Observed Action Representation During Natural Vision./
Author:	Nastase, Samuel Alexander.
Description:	1 online resource (141 pages)
Notes:	Source: Dissertation Abstracts International, Volume: 79-07(E), Section: B.
Contained By:	Dissertation Abstracts International79-07B(E).
Subject:	Neurosciences. -
Online resource:	click for full text (PQDT)
ISBN:	9780355650723

The Geometry of Observed Action Representation During Natural Vision.
Nastase, Samuel Alexander.

The Geometry of Observed Action Representation During Natural Vision. - 1 online resource (141 pages)

Source: Dissertation Abstracts International, Volume: 79-07(E), Section: B.

Thesis (Ph.D.)--Dartmouth College, 2018.

Includes bibliographical references

How do we understand the actions of others? Extracting semantic and social content---such as a conspecific's intentions---from spatiotemporal patterns of bodily movement is computationally challenging. To solve this challenge, the brain relies on multiple, hierarchically organized layers of re-representation to effectively disentangle behaviorally relevant information from retinal input. The current study aims to decode the neural representation of observed human actions at multiple stages of this processing hierarchy, from early visual to somatomotor cortex. Using fMRI, we measured neural responses while participants viewed 90 naturalistic video clips depicting humans performing a variety of social and nonsocial behaviors broadly sampling action space (e.g., conversation, cooking, gardening). We then used representational similarity analysis to interpret the organization of these response patterns. At each hub of the action observation network, we formally tested six representational models: a neurally-inspired model of motion energy processing, a model comprising eye movements measured in a separate sample of participants, distributional semantic models for verb and non-verb annotations of the stimuli, and behavioral judgments of sociality and transitivity (object- and goal-relatedness). Finally, we used multidimensional scaling to directly visualize the organization of these representational spaces. Motion and gaze accounted for modest variance in early visual cortex. By contrast, information spaces in lateral occipitotemporal and ventral temporal cortices predominantly represent sociality, transitivity, and verb semantics. Anterior intraparietal and ventral premotor cortex primarily encode transitivity, and to a lesser extent verb semantics. Overall, these findings provide a framework for characterizing how transformations in the geometry of neural representational spaces afford action understanding during natural vision.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355650723Subjects--Topical Terms:

593561
Neurosciences.
Index Terms--Genre/Form:

554714
Electronic books.

The Geometry of Observed Action Representation During Natural Vision.
LDR:03200ntm a2200349Ki 4500 001 918539
005 20181026115418.5
006 m o u
007 cr mn||||a|a||
008 190606s2018 xx obm 000 0 eng d
020 $a 9780355650723
035 $a (MiAaPQ)AAI10745463
035 $a (MiAaPQ)dartmouth:11294
035 $a AAI10745463
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Nastase, Samuel Alexander. $3 1192886
245 1 4 $a The Geometry of Observed Action Representation During Natural Vision.
264 0 $c 2018
300 $a 1 online resource (141 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 79-07(E), Section: B.
500 $a Adviser: James V. Haxby.
502 $a Thesis (Ph.D.)--Dartmouth College, 2018.
504 $a Includes bibliographical references
520 $a How do we understand the actions of others? Extracting semantic and social content---such as a conspecific's intentions---from spatiotemporal patterns of bodily movement is computationally challenging. To solve this challenge, the brain relies on multiple, hierarchically organized layers of re-representation to effectively disentangle behaviorally relevant information from retinal input. The current study aims to decode the neural representation of observed human actions at multiple stages of this processing hierarchy, from early visual to somatomotor cortex. Using fMRI, we measured neural responses while participants viewed 90 naturalistic video clips depicting humans performing a variety of social and nonsocial behaviors broadly sampling action space (e.g., conversation, cooking, gardening). We then used representational similarity analysis to interpret the organization of these response patterns. At each hub of the action observation network, we formally tested six representational models: a neurally-inspired model of motion energy processing, a model comprising eye movements measured in a separate sample of participants, distributional semantic models for verb and non-verb annotations of the stimuli, and behavioral judgments of sociality and transitivity (object- and goal-relatedness). Finally, we used multidimensional scaling to directly visualize the organization of these representational spaces. Motion and gaze accounted for modest variance in early visual cortex. By contrast, information spaces in lateral occipitotemporal and ventral temporal cortices predominantly represent sociality, transitivity, and verb semantics. Anterior intraparietal and ventral premotor cortex primarily encode transitivity, and to a lesser extent verb semantics. Overall, these findings provide a framework for characterizing how transformations in the geometry of neural representational spaces afford action understanding during natural vision.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Neurosciences. $3 593561
650 4 $a Psychology. $3 555998
650 4 $a Cognitive psychology. $3 556029
655 7 $a Electronic books. $2 local $3 554714
690 $a 0317
690 $a 0621
690 $a 0633
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Dartmouth College. $b Psychological and Brain Sciences. $3 1180504
773 0 $t Dissertation Abstracts International $g 79-07B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10745463 $z click for full text (PQDT)