Related Experiment Videos
Characterizing the time course of an implicature: an evoked potentials study
1Institut des Sciences Cognitives, 67 blvd. Pinel, 69675 Bron Cedex, France. noveck@isc.cnrs.fr
Brain and Language
|May 9, 2003
Summary
Scalar implicatures are processed later and require more effort, as shown by reaction times and brain activity (ERP). This supports the view that implicatures are not automatic but part of a demanding decision process.
Area of Science:
- Psycholinguistics
- Cognitive Neuroscience
Background:
- Scalar implicatures, like 'some' implying 'not all', present a challenge in understanding language meaning.
- Two theories, Neo-Gricean and Relevance Theory, offer different explanations for implicature processing.
Purpose of the Study:
- To investigate the cognitive and neural mechanisms underlying scalar implicature production.
- To compare the automaticity versus effortful nature of implicature generation.
Main Methods:
- Evoked Potential (ERP) techniques were used with 19 participants.
- Participants evaluated Underinformative, Patently True, and Patently False sentences.
- Reaction times and electroencephalography data were analyzed.
Main Results:
- Participants responding 'false' to underinformative sentences took longer than those responding 'true'.
- ERP data showed flatter N400s for underinformative sentences compared to true/false ones.
- N400 amplitudes suggest less semantic integration for underinformative statements.
Conclusions:
- Scalar implicatures are not automatically generated but arise from a late-arriving, effortful decision-making process.
- Individual differences in response times correlate with ERP findings.
- The findings support theories emphasizing the effortful nature of pragmatic inference.