The Self-Comprehension Hypothesis in Speech Monitoring (SCHISM)

We will investigate the production of speech using a method we have developed called Real-time Speech Exchange (RSE).

RSE allows us to manipulate participants’ auditory feedback to create situations where they say one thing but hear themselves saying something else. Previously, we have shown how speakers believe they said the inserted words.

These results challenge the basic assumption in the psychology of language that an utterance is preceded by a fixed intention containing its full meaning, and against which feedback is checked for accuracy.

Instead, they suggest that intentions are much vaguer and that speakers use feedback more actively to (1) know the meaning of their utterances (‘self-comprehension’); (2) ‘source monitor’ to know if speech was self- or other- generated; and (3) create a feeling of performing the speech act (‘sense of agency’).

Over three years, we will develop these findings by using RSE to study (1) self-comprehension, source monitoring and sense of agency in an advanced ecological experimental setting; (2) interactions between auditory and oral somatosensory feedback during self-comprehension; and (3) the flexibility and contextual anchoring of source monitoring and sense of agency.

This project investigates the validity of prevailing assumptions regarding the foundations of speech production and will significantly contribute to knowledge about how speech is planned and monitored, and how we come to know the content and meaning of our own utterances.