Fig 1. Experimental paradigm and setup.

(A) Paradigm: Two example reference frames for each participant of a pair are shown. The prime for each participant would define a meaning context for the homonym (Flügel—piano/wing) in the upcoming sentence. In this example, one had a prime (Konzert—concert) for one meaning (piano) and the other participant had a prime (Engel—angel) for the other meaning (wing) of the homonym. The sentence meaning was ambiguous and fit to both reference frames until the critical word (Taste—key), which would fit the reference frame of a piano (i.e., it would be congruent to concert), but not the reference frame of a wing (i.e., it would be incongruent to angel). Sentence translation: ‘Jana sees the piano/wing. She touches a key and hears the clear sound.’ (B) Setup: EEG and audio were recorded from both participants concurrently. The prime had to be read silently and was shown in a violet color to distinguish it more easily from the words of the sentence shown in black that had to be read aloud. Each participant started to read aloud Word 1 of the sentence in 50% of the trials. The analysis focused on the ERP of the CW (highlighted in blue) that was measured for the participant (blue head) that perceived the word read aloud by the interacting partner (black head). The offline defined speech onset of the interacting partner was used as onset marker for the listened word. The reaction time (RT) to speak was measured for CW+1, CW+3, and CW+5 in the same participant after the ERP of the CW. In the same trial, the RT to speak the CW, CW+2, and CW+4 was measured in the partner. In the experiment, the prime was either congruent for both participants or congruent for one participant and incongruent for the interacting partner. Presentation time of the prime and mean speech times of each word are shown. (RTs are not included).