How do people decide whether a sentence “sounds right” in their native language by judging its syntactic well-formedness independently of meaning? Sentence acceptability judgments reflect not only the speaker’s grammatical knowledge but also performance factors including information packaging and processing ease. In this talk, I review what we know about the interaction of these factors, including striking cases of “grammatical illusions” and situations where speakers hesitate or cannot reach a clear judgment when these factors conflict. I then present recent experimental evidence showing that speakers use rating scales strategically whereby scale extremes serve as categorical anchors while the midpoint marks maximal uncertainty. These findings highlight sentence evaluation as a dynamic calibration process and shed new light on how people make linguistic decisions under uncertainty.

Seminar hosted by Francesco Vespignani