Abstract
People can be recognized by a number of cues in different modalities, including face, body, voice, name, and biographic information. These are often present in conjunction in daily life. A number of studies have shown cross-modal interactions between these information sources during the retrieval stage of recognition tasks. However, it is less known whether there are cross-modal influences during the encoding stage. Our goal was to determine if either face familiarity or face identification were affected by simultaneous presentation of voice or biographical data when subjects were learning new faces. Using a between-subject design, four groups of 10 participants each learned 24 faces seen in dynamic video-clips with their names written on the screen. 12 of these face/names were presented with other information simultaneously and 12 without. The nature of the other information differed between the four groups. The first group heard the voice of the person stating non-biographic information. The second group heard an announcer giving biographic information for each person. The third group heard the voice of the person stating their own biographic information. A fourth group, a control, heard the announcer stating non-biographic information. In the retrieval phase, subjects saw dynamic faces only, and completed a face familiarity task and an identification (face-to-name matching) task. We found that no condition with additional information gave better or faster accuracy during retrieval than learning the face alone, and that this was true for both familiarity and identification tasks. We conclude that, unlike the retrieval stage, the encoding stage in face memory tasks is relatively immune to cross-modal influences from simultaneous voice or biographical information.
Meeting abstract presented at VSS 2017