2015
DOI: 10.3389/fpsyg.2015.01386
|View full text |Cite
|
Sign up to set email alerts
|

Automated facial coding software outperforms people in recognizing neutral faces as neutral from standardized datasets

Abstract: Little is known about people’s accuracy of recognizing neutral faces as neutral. In this paper, I demonstrate the importance of knowing how well people recognize neutral faces. I contrasted human recognition scores of 100 typical, neutral front-up facial images with scores of an arguably objective judge – automated facial coding (AFC) software. I hypothesized that the software would outperform humans in recognizing neutral faces because of the inherently objective nature of computer algorithms. Results confirm… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
51
0

Year Published

2016
2016
2024
2024

Publication Types

Select...
6
1
1

Relationship

1
7

Authors

Journals

citations
Cited by 73 publications
(57 citation statements)
references
References 22 publications
0
51
0
Order By: Relevance
“…FaceReader allows for the simultaneous presence of multiple emotions. A more detailed description of the science behind FaceReader can be found at: http://info.noldus.com/free-white-paper-on-facereader-methodology/, or see [32,33].…”
Section: Categorymentioning
confidence: 99%
“…FaceReader allows for the simultaneous presence of multiple emotions. A more detailed description of the science behind FaceReader can be found at: http://info.noldus.com/free-white-paper-on-facereader-methodology/, or see [32,33].…”
Section: Categorymentioning
confidence: 99%
“…FaceReader TM is a commercial computer vision tool used to objectively analyze the presence of expressions, action units, and emotional overall valence in facial images. FaceReader TM is well established and validated in the scientific literature, with results approaching expert level (Lewinski, 2015;Adams et al, 2016).…”
Section: Facereader Resultsmentioning
confidence: 99%
“…To be sure, measures for coding should avoid incurring inconsistencies arising from human idiosyncrasies, [53] and to the extent that the agreed results reported below are based upon a simple 100% agreement that one set of witnesses or committee exhibited relatively more nonverbal cues (anger, happiness, etc) than the other set of witnesses or committee, the bulk of the coding results do not report as findings any inconsistencies among the coders (and so, a measure such as Krippendorff's Alpha is not used). Studies do not usually discuss differences among coders (although exceptions include: Schubert and colleagues, who comment on a coder's "idiosyncratic tendency to overcode"; [43] and Bucy and Gong, who discuss specific techniques for improving intercoder reliability and precision [9,[55][56][57][58]) and yet-as discussed earlier-receivers of nonverbal messages do not necessarily respond in similar ways, as these signals are conditional on pre-existing attitudes and the situational context of the behaviour, and some individuals are simply more adept than others in discerning the meaning of the signals. Finally, "stereotypical" university undergraduates have been criticized for being "socially compliant" and "more likely to be mercurial in their attitudes because of lack of self-knowledge."…”
Section: Coding Of Nonverbal Behaviour In Parliamentary Hearingsmentioning
confidence: 99%
“…First, humans still outperform computers in interpreting the nuances and context of facial expressions, although the capacity of automation is no doubt rapidly evolving. [57] Second, no software as yet (of which I am aware) automatically codes facial expressions, vocal cues and gestures as a whole package. Third, software that codes all relevant aspects of nonverbal communication may well be around the corner; nonetheless, this does not negate the importance of obtaining human coding of the various categories, as observed in real world settings.…”
Section: Coding Of Nonverbal Behaviour In Parliamentary Hearingsmentioning
confidence: 99%