El tamaño de familiacorrelatos conductales y de respuesta cerebral del papel de la morfología en el reconocimiento léxico

  1. Lázaro López-Villaseñor, Miguel
Supervised by:
  1. Francisco Javier Sáinz Sánchez Director

Defence university: Universidad Complutense de Madrid

Fecha de defensa: 06 June 2008

Committee:
  1. Susana López Ornat Chair
  2. Carlos Gallego López Secretary
  3. José Manuel Igoa González Committee member
  4. Vicenç Torrens García Committee member
  5. Victoria Marrero Aguiar Committee member
Department:
  1. Psicología Experimental, Procesos Cognitivos y Logopedia

Type: Thesis

Abstract

The role played by morphological parsing in lexical access is explored in a series of experiments, in which behavioral and cerebral responses are simultaneously taken. Morphological parsing is systematically manipulated by Family Size, -the number of lexical entries morphologically similar to a base word. The hypothesis to be tested concerns to whether morphological parsing occurs before or after lexical word recognition. In the first experiment, a replica of Schreuder & Baayen’s (1997) main experiment was conducted; with monomorphemic words the results obtained straighforwardly replicates those obtained by Schreuder & Baayen (1997): more greater Family Size of a base word is, faster lexical decision latencies are. In the second and the third experiments, complex words are been used: every word is composed by a stem + a derivational morpheme. In the second and the third experiments, an experimental subject is displayed with derivative words, for which s/he has to decide whether a postmasked visual lexical pattern is a word or not, as in the first experiment. In the second experiment, a subject is first presented with the word stem as a prime, and then with the whole word. In the third experiment, a subject is first presented with the derivative morpheme and then with the whole word. Both, behavioral and event-related potential data show that the integration process on word segments operates simultaneously in order for a lexical candidate to get access to the lexicon. Our results provide evidence against a supralexical view of lexical access (Grainger & Giraudo, 2000) and favor a sublexical model of morphological decomposition and integration, but morphological parsing requires to have access to whole-word representations, even before a lexical decision is being taken.