Visual and Auditory Components in the Perception of Asynchronous Audiovisual Speech

Research on asynchronous audiovisual speech perception manipulates experimental conditions to observe their effects on synchrony judgments. Probabilistic models establish a link between the sensory and decisional processes underlying such judgments and the observed data, via interpretable parameters...

Descripción completa

Detalles Bibliográficos
Autores: García Pérez, Miguel Ángel, Alcalá Quintana, Rocío
Tipo de recurso: artículo
Fecha de publicación:2015
País:España
Institución:Universidad Complutense de Madrid (UCM)
Repositorio:Docta Complutense
Idioma:inglés
OAI Identifier:oai:docta.ucm.es:20.500.14352/24302
Acceso en línea:https://hdl.handle.net/20.500.14352/24302
Access Level:acceso abierto
Palabra clave:159.9.07
159.95
159.93
Synchrony judgments
Perceived onset
Audiovisual speech
Multisensory integration
Psicología experimental
Psicología cognitiva
Percepción
6106 Psicología Experimental
6104.01 Procesos Cognitivos
6106.09 Procesos de Percepción
Descripción
Sumario:Research on asynchronous audiovisual speech perception manipulates experimental conditions to observe their effects on synchrony judgments. Probabilistic models establish a link between the sensory and decisional processes underlying such judgments and the observed data, via interpretable parameters that allow testing hypotheses and making inferences about how experimental manipulations affect such processes. Two models of this type have recently been proposed, one based on independent channels and the other using a Bayesian approach. Both models are fitted here to a common data set, with a subsequent analysis of the interpretation they provide about how experimental manipulations affected the processes underlying perceived synchrony. The data consist of synchrony judgments as a function of audiovisual offset in a speech stimulus, under four within-subjects manipulations of the quality of the visual component. The Bayesian model could not accommodate asymmetric data, was rejected by goodness-of-fit statistics for 8/16 observers, and was found to be nonidentifiable, which renders uninterpretable parameter estimates. The independent-channels model captured asymmetric data, was rejected for only 1/16 observers, and identified how sensory and decisional processes mediating asynchronous audiovisual speech perception are affected by manipulations that only alter the quality of the visual component of the speech signal.