Perception of biological motion by form analysis

Detection of other living beings’ movements is a fundamental property of the human visual system. Viewing their movements, categorizing their actions, and interpreting social behaviors like gestures constitutes a framework of our everyday lives. These observed actions are complex and differences amo...

Ausführliche Beschreibung

Gespeichert in:
1. Verfasser: Lange, Joachim
Beteiligte: Bremmer, Frank (Prof. Dr.) (BetreuerIn (Doktorarbeit))
Format: Dissertation
Sprache:Englisch
Veröffentlicht: Philipps-Universität Marburg 2006
Physik
Schlagworte:
Online Zugang:PDF-Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Zusammenfassung:Detection of other living beings’ movements is a fundamental property of the human visual system. Viewing their movements, categorizing their actions, and interpreting social behaviors like gestures constitutes a framework of our everyday lives. These observed actions are complex and differences among them are rather subtle. However, humans recognize these actions without ma jor efforts and without being aware of the complexity of the observed tasks. In point-light walkers, the visual information about the human body is reduced to only a handful point-lights placed on the ma jor joints of the otherwise invisible body. But even this sparse information does not effectively reduce humans’ abilities to perceive the performed actions. Neurophysiological and neuroimaging studies suggested that the movement of the human body is represented in specific brain areas. Nonetheless, the underlying network is still issue of controversial discussion. To investigate the role of form information, I developed a model and conducted psychophysical experiments using point-light walkers. A widely accepted theory claims that in point-light walkers, form information is decreased to a non-usable minimum and, thus, the perception of biological motion is driven by the analysis of motion signals. In my study, I could show that point-light walker indeed contain useful form information. Moreover, I could show that temporal integration of this information is sufficient to explain results from psychophysical, neurophysiological, and neuroimaging studies. In opposition to the standard models of biological motion perception, I could also show that all results can be explained without the analysis of local motion signals.