Exact temporal mechanisms of sentence-level speech‑rate integration

Determine the exact temporal mechanisms by which the human speech perception system incorporates speech‑rate information when processing sentences of arbitrary length, specifying how temporal context is integrated and weighted across preceding phrasal material to influence subsequent phoneme and word perception.

Background

Extensive laboratory work has shown that speech‑rate context influences phoneme and word perception, but reported effects vary: some studies suggest accumulation over several sentences, while others find influence limited to one or two adjacent phonemes. This inconsistency highlights a gap in understanding how timing information is integrated across longer stretches of connected speech.

The paper motivates a need for an algorithmic account of how listeners take up rate information in natural sentences, noting that while distal and proximal effects likely exist and may compete, their generalization beyond short, controlled sentences remains unsettled.