Unimodal and audiovisual stimuli. The compressed timeline of AVConcordant and AVConflicting stimuli is shown. Each unimodal visual utterance (/da/, /fu/ and /du/) was digitized from a recording of a male speaker. All three clips began and ended with the same neutral frame, but were different over the length of the utterance. The release of the consonant was edited to occur at frame 11 for all three visual tokens. A 100ms synthesized syllable, /da/, was created to emulate natural speech. For audiovisual presentation, the speech stimulus was paired with each visuofacial movement and acoustic onset occurred at 360ms.