Deutsch
 
Hilfe Datenschutzhinweis Impressum
  DetailsucheBrowse

Datensatz

DATENSATZ AKTIONENEXPORT

Freigegeben

Poster

Inversion effects are stronger for subordinate than for basic-level action recognition

MPG-Autoren
/persons/resource/persons83877

de la Rosa,  S
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;

/persons/resource/persons83839

Bülthoff,  HH
Department Human Perception, Cognition and Action, Max Planck Institute for Biological Cybernetics, Max Planck Society;

Externe Ressourcen
Volltexte (beschränkter Zugriff)
Für Ihren IP-Bereich sind aktuell keine Volltexte freigegeben.
Volltexte (frei zugänglich)
Es sind keine frei zugänglichen Volltexte in PuRe verfügbar
Ergänzendes Material (frei zugänglich)
Es sind keine frei zugänglichen Ergänzenden Materialien verfügbar
Zitation

de la Rosa, S., & Bülthoff, H. (2015). Inversion effects are stronger for subordinate than for basic-level action recognition. Poster presented at 38th European Conference on Visual Perception (ECVP 2015), Liverpool, UK.


Zitierlink: https://hdl.handle.net/11858/00-001M-0000-002A-4504-9
Zusammenfassung
Previous results showed that actions can be recognized in multiple ways suggesting that several recognition levels exist in action recognition (e.g. a waving action can be recognized as a greeting or a wave). Categorization tasks suggest that the recognition of social interactions is more accurate at the basic-level (e.g. greeting) than at the subordinate level (e.g. waving). What is the origin of the supremacy of basic-level recognition? Here we examined whether basic-level recognition relies to a larger degree on configural processing than subordinate social interaction recognition. To do so we probed basic-level and subordinate recognition performance (RT and discrimination ability (d')) of 20 participants for upright and inverted social interactions. Larger inversion effects are typically associated with stronger configural processing. Participants saw a one image at a time and reported whether it matched a predefined action. Our results showed that - contrary to our initial hypothesis - subordinate recognition of social interactions was significantly more affected by stimulus inversion than basic-level recognition. Moreover, recognition performance was better for subordinate than basic-level recognition. We show that these results can be well explained by a top-down activation of snapshot templates.