Context effects in a double-weak theory of speech perception

Area of Science:

• Linguistics
• Auditory Neuroscience
• Speech Science

Background:

• The "double-weak" theory posits abstract symbolic elements, not purely auditory or gestural, as the basis of speech perception and production.
• Speech understanding necessitates considering both articulation and acoustics, as production strategies aim for acoustic goals, while perception is influenced by gestural overlap.

Purpose of the Study:

• To elaborate on the "double-weak" theory of speech perception.
• To explore how speech perception and production are imperfectly related and may function as autonomous subsystems.
• To discuss perceptual models approximating "ideal solutions" for classifying production data with contextual interactions.

Main Methods:

• Theoretical elaboration of the "double-weak" framework.
• Analysis of the relationship between speech production strategies and perceptual context effects.
• Discussion of a family of perceptual models with varying degrees of compensation for gestural overlap.

Main Results:

• Speech perception and production are not perfect inverses; they are constrained symbolic mappings with acoustic and gestural components.
• Context effects in speech perception are largely driven by the consequences of gestural overlap.
• Models with partial compensation for gestural overlap adequately explain many speech perception experiment results.

Conclusions:

• The "double-weak" framework provides a unified view of speech perception and production.
• Real-time speech perception and production may operate as independent subsystems despite mutual accommodation.
• Partial perceptual compensation for gestural overlap explains phenomena like hypo- and hypercorrection.