Directed attention only boosts audiovisual integration for sentence-level contexts

Evidence

The effect of context type [F(1, 20) = 11. 91, p < 0.001] is significant, as in Exp. A: video syllables produce more McGurk than video sentences. Contrary to Exp. A, the effect of context duration [F(1, 20) = 33. 86, p < 0.001] is also significant, with no interaction with context type. The attention factor alone is not significant, but its interaction with context type is significant [F(1, 20) = 11.07, p < 0.005]. Post-hoc analyses with Bonferroni corrections show that while there is no significant difference between the two attention conditions for the “Video syllables” context type, there is a difference for the “Video sentences” condition, with a lower “ba” percentage (a higher McGurk effect) in the “Attention sentence” condition. Interestingly, while the “ba” percentage is higher for the “Video sentences” than for the “Video syllables” condition when attention is put in syllables, there is no more significant difference when attention is put on sentences. — Dijk, Pim van, Physiology, Psychoacoustics and…, 5162