Perceived movement of nonrigid motion patterns

Nonrigid materials such as liquids or smoke deform over time. Little is known about the visual perception of nonrigid motion other than that many motion cues associated with rigid motion perception are not reliable for nonrigid motion. Nonrigid motion patterns lack clear borders and their movement c...

Verfasser: Koerfer, Krischan
Lappe, Markus
FB/Einrichtung:FB 07: Psychologie und Sportwissenschaft
Dokumenttypen:Artikel
Medientypen:Text
Erscheinungsdatum:2022
Publikation in MIAMI:26.06.2023
Datum der letzten Änderung:26.06.2023
Angaben zur Ausgabe:[Electronic ed.]
Quelle:PNAS Nexus 1 (2022) 3, 1-11
Schlagwörter:nonrigid motion; visual perception; visual motion field; higher-order motion; curl
Fachgebiet (DDC):150: Psychologie
Lizenz:CC BY 4.0
Sprache:English
Förderung:Finanziert durch den Open-Access-Publikationsfonds der Westfälischen Wilhelms-Universität Münster (WWU Münster).
Förderer: European Commission / Projektnummer: 734227
Förderer: Deutsche Forschungsgemeinschaft / Projektnummer: 274361309
Format:PDF-Dokument
URN:urn:nbn:de:hbz:6-39998453129
Weitere Identifikatoren:DOI: 10.17879/69998694783
Permalink:https://nbn-resolving.de/urn:nbn:de:hbz:6-39998453129
Verwandte Dokumente:
  • ist identisch zu:
  • Onlinezugriff:10.1093_pnasnexus_pgac088.pdf

    Nonrigid materials such as liquids or smoke deform over time. Little is known about the visual perception of nonrigid motion other than that many motion cues associated with rigid motion perception are not reliable for nonrigid motion. Nonrigid motion patterns lack clear borders and their movement can be inconsistent with the motion of their parts. We developed a novel stimulus that creates a nonrigid vortex motion pattern in a random dot distribution and decouples the movement of the vortex from the first-order motion of the dots. We presented three moving vortices that entailed consecutively fewer motion cues, eliminating occlusion, motion borders, and velocity field gradients in the process. Subjects were well able to report the end position and travel path in all cases, showing that nonrigid motion is perceived through an analysis of the temporal evolution of visual motion patterns and does not require borders or speed differences. Adding a coherent global motion did not hamper perception, but adding local noise did, indicating that the visual system uses mid-level features that are on a local scale. We also found that participants judged the movement of the nonrigid motion patterns slower than a rigid control, revealing that speed perception was based on a combination of motion of the parts and movement of the pattern. We propose that the visual system uses the temporal evolution of a motion pattern for the perception of nonrigid motion and suggest a plausible mechanism based on the curl of the motion field.