Effects of Faces and Voices on the Encoding of Biographic Information

Abstract

There are multiple forms of knowledge about people. Whether diverse person-related data interact is of interest regarding the more general issue of integration of multi-source information about the world. Our goal was to examine whether perception of a person's face or voice enhanced the encoding of their biographic data. We performed three experiments. In the first experiment, subjects learned the biographic data of a character with or without a video clip of their face. In the second experiment, they learned the character's data with an audio clip of either a generic narrator's voice or the character's voice relating the same biographic information. In the third experiment, an audiovisual clip of both the face and voice of either a generic narrator or the character accompanied the learning of biographic data. After learning, a test phase presented biographic data alone, and subjects were tested first for familiarity and second for matching of biographic data to the name. The results showed equivalent learning of biographic data across all three experiments, and none showed evidence that a character's face or voice enhanced the learning of biographic information. We conclude that the simultaneous processing of perceptual representations of people may not modulate the encoding of biographic data.

Keywords: facilitation; learning; memory; person; semantic.

Figure 1

Methods. The learning phases for the three different experiments are shown on the left, with each row showing the control and experimental conditions. Each box depicts the auditory and visual components in each trial. The screen showed the biographic data (’bio’) in brief written format, as bullet points (’•a, •b’, etc., representing these points) in a textbox, with the name of the character below. Audio clips are shown with the speaker symbol; these were voiced by either a narrator or the character, describing the same biographic data for that character. Conditions with video clips are shown with silhouettes or faces; the face shows those conditions with a video clip of the character, while the silhouette represents a generic narrator. After one cycle of learning, subjects had a familiarity test, with a textbox showing one bullet point alone (e.g., ’hockey goalie’). This was followed by the second learning phase of one cycle. The final phase tested identification by having the subject match the single bullet point to the correct name in an array of four choices shown below the textbox.

Figure 2

Results. (A). Familiarity accuracy, (B). Familiarity response time and (C). Identification accuracy. Each graph shows results for the three experiments separately (Face, Voice, Face + Voice). The bars compare mean performance of learning the biographic data alone with learning accompanied concurrently with the additional sensory information appropriate for that experiment. Group means are shown, and error bars indicate one standard error.