Processing of Individual Items during Ensemble Coding of Facial Expressions

Abstract

There is growing evidence that human observers are able to extract the mean emotion or other type of information from a set of faces. The most intriguing aspect of this phenomenon is that observers often fail to identify or form a representation for individual faces in a face set. However, most of these results were based on judgments under limited processing resource. We examined a wider range of exposure time and observed how the relationship between the extraction of a mean and representation of individual facial expressions would change. The results showed that with an exposure time of 50 ms for the faces, observers were more sensitive to mean representation over individual representation, replicating the typical findings in the literature. With longer exposure time, however, observers were able to extract both individual and mean representation more accurately. Furthermore, diffusion model analysis revealed that the mean representation is also more prone to suffer from the noise accumulated in redundant processing time and leads to a more conservative decision bias, whereas individual representations seem more resistant to this noise. Results suggest that the encoding of emotional information from multiple faces may take two forms: single face processing and crowd face processing.

Elucidating the Neural Representation and the Processing Dynamics of Face Ensembles

“…Importantly, the temporal profile described above is not at odds with prior research documenting the ability to rapidly extract summary representations from an ensemble, often in Ͻ100 ms Li et al, 2016). Indeed, we also note a first peak in ensemble decoding at ϳ100 ms (Fig.…”

Limited evidence of hierarchical encoding in the cheerleader effect

“…In addition to correctly recognising the individual faces that had been members of the group, observers also mistakenly "recognised" the set average (i.e., a face that had been created by digitally averaging the individual faces in the group together). Crucially, observers did not falsely recognise "foil averages" (i.e., a face made by averaging a different group of faces), suggesting that the set average is not mistakenly recognised because average faces are inherently familiar 40 , but because the observer mentally summarises the group and extracts its average identity 37,38 .Ensemble coding also facilitates the accurate identification of the average emotional expression shown by a group of faces 34,36,41,42 . Remarkably, even though no individual face shows the average emotional expression of the group, observers can identify the group's average expression with the same precision that is shown when identifying the expression of a single emotional face 41 .…”

“…Remarkably, even though no individual face shows the average emotional expression of the group, observers can identify the group's average expression with the same precision that is shown when identifying the expression of a single emotional face 41 . The ensemble coding of expression occurs rapidly, as observers can accurately identify the average expression shown by a group of 16 faces that has been presented for just 50 ms 35,36,42 . In addition to rapidly extracting the average emotion 35,36,41 and identity 37-39 from groups of faces, observers can accurately summarise other characteristics of faces, including average ethnicity and dominance 43 , gender 36 , and attractiveness 44 .…”

Mean emotion from multiple facial expressions can be extracted with limited attention: Evidence from visual ERPs

“…("visual search"), especially when the attentional resources are limited (Alvarez, 2011;Fischer & 14 Whitney, 2011;Haberman & Whitney, 2009;Li et al, 2016). …”