Processing spatial anaphora: Referent reactivation with overt and null pronouns in american sign language

Abstract

Abstract Unlike English, American Sign Language (ASL) permits phonologically null pronouns in tensed clauses. Null pronouns are licensed by morphological marking of “agreeing” verbs which agree with the spatial loci of the subject and object noun phrases of the sentence. We present two probe recognition experiments which investigated whether overt and null pronouns similarly reactivate their referents during on-line sentence comprehension. Experiment 1 revealed faster response times to probes that were referents of either an overt pronoun or a null pronoun compared to control probes, indicating that both overt pronouns and null pronouns associated with verb agreement reactivate their spatial referents. However, response times to non-referent probes were also faster than to control probes, and it was hypothesiscd that an end-of-sentence probe presentation may have tapped into a sentence integration process in which all possible referents were reactivated. In a second experiment, the Same sentences were presented to a second group of subjects, but probe signs were presented before the end of the sentence, and 1000 msec after the anaphoric element. The main results of the first experiment were replicated (both overt and null pronouns reactivated their referents), and responses to non-referent probes were not significantly different from control probes. Both experiments indicated that (1) there is an important link between spatial verb agreement and the ASL pronominal system, and (2) non-referent inhibition does not have the same processing status in ASL as it does in English.