It is more difficult to retrieve a familiar person's name and occupation from their voice than from their blurred face

Abstract

Damjanovic and Hanley (2007) showed that episodic information is more readily retrieved from familiar faces than familiar voices, even when the two presentation modalities are matched for overall recognition rates by blurring the faces. This pattern of performance contrasts with the results obtained by Hanley and Turner (2000) who showed that semantic information could be recalled equally easily from familiar blurred faces and voices. The current study used the procedure developed by Hanley and Turner (2000) and applied it to the stimuli used by Damjanovic and Hanley (2007). The findings showed a marked decrease in retrieval of occupations and names from familiar voices relative to blurred faces even though the two modalities were matched for overall levels of recognition and rated familiarity. Similar results were obtained in Experiment 2 in which the same participants were asked to recognise both faces and voices. It is argued that these findings pose problems for any model of person recognition (e.g., Burton, Bruce, & Johnston, 1990) in which familiarity decisions occur beyond the point at which information from different modalities has been integrated.

Recalling semantic information about personally known faces and voices

“…When these problems were avoided by employing the Schweinberger, Herholz, and Steif (1997) procedure designed to limit the extent to which the speech content of the extracts could give clues to a speaker's identity, the results unambiguously indicated that more semantic details could be recalled from blurred faces than from voices even though overall recognition performance was similar for both types of stimuli. Although, these two studies (Damjanovic & Hanley, 2007;Hanley & Damjanovic, 2008) confirm the view that access to semantic information is easier from faces, even blurred, than voices, further studies are needed.…”

“…Such results did not support the view that it is more difficult to associate semantic information with people's voice than with people's face. Nevertheless, more recent studies (Damjanovic & Hanley, 2007;Hanley & Damjanovic, 2008) suggested that the latter results were due to methodological problems. The spoken extracts used in the voice condition of the Hanley and Turner study were likely to provide cues as to the occupation of some of the target celebrities.…”

“…Second, as noted in previous papers (e.g., Hanley et al, 1998;Hanley & Turner, 2000), we probably encounter a famous person's face in the media more frequently than we hear their voice. By choosing professors or teaching assistants as target individuals, we think this problem Damjanovic and Hanley (2007) and the Hanley and Damjanovic (2008) studies. Indeed, present findings showed a significant and numerically substantial decrease in retrieval of occupational details and names from familiar voices relative to blurred faces even though the level of overall recognition (e.g., the rates of hits and false alarms) was similar for blurred faces and voices.…”

Hearing Faces and Seeing Voices: The Integration and Interaction of Face and Voice Processing

“…In addition, when care is taken to balance recognition levels through blurring the faces, the capacity to retrieve semantic information about an individual remains substantially weaker when presented with the voice rather than the face. Notably, this remains the case when trying to retrieve semantic details about celebrities (Hanley & Damjanovic, 2009;Hanley, Smith & Hadfield, 1998), personally familiar individuals (Barsics & Brédart, 2011;Brédart, Barsics & Hanley, 2009) or newly learned individuals (Barsics & Brédart, 2012a).…”

“…A series of methodologies combine to support this view. For example, voices are less well recognised, and elicit significantly more 'familiar only' experiences compared to faces (Ellis, Jones & Mosdell, 1997;Hanley, Smith & Hadfield, 1998) and, in fact, recognition performance from faces and voices can only be equated when the faces are substantially blurred (Damjanovic & Hanley, 2007;Hanley & Damjanovic, 2009). In addition, when care is taken to balance recognition levels through blurring the faces, the capacity to retrieve semantic information about an individual remains substantially weaker when presented with the voice rather than the face.…”