Xf205 camcorder

The conversations were recorded in a soundproof room at the Max Planck Institute for Psycholinguistics in Nijmegen, The Netherlands. Participants were seated facing each other at approximately 90 cm distance from the front edge of the seats (Figure 1).
Two video cameras (Canon XE405) were used to record frontal views of each participant, two cameras recorded each participant’s body from a 45 degree angle (Canon XF205 Camcorder), two cameras (Canon XF205 Camcorder) recorded each participant from a birds-eye view while mounted on a tripod, and finally one camera (Canon Legria HF G10) recorded the scene view, displaying both participant at the same time. All cameras were recorded at 25 fps. Audio was recorded using two directional microphones (Sennheiser me-64) for each participant (see the Appendix A for an overview of the set-up). Each recording session resulted in seven video files and two audio files, which were synchronised and exported as a single audio-video file for analysis in Adobe Premiere Pro CS6 (MPEG, 25 fps), resulting in a time resolution of approximately 40 ms, the duration of a single frame. For the coding of facial signals reported in the present study, only the face close-ups were used, one at a time for best visibility of detailed facial signals.