5
$\begingroup$

I'm often surprised by the human ability to correctly identify other individuals despite significant modifications due to ageing, hairstyle, injury, etc. But, sometimes the addition of a beard and a hat can radically increase the identification error.

I wonder whether all facial regions are equally important when we try to recognise human faces? Do we focus more on some regions than others? I doubt that this is the case, but I'm not certain.

Update:

So far I've read through the following:

1) http://en.wikipedia.org/wiki/Face_perception

2) http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2998394/

So far I haven't found an answer so I decided to try Biology.StackExchange.

Improve this question
$\endgroup$
7
  • $\begingroup$ Have you done any background research of your own? I'm pretty sure that there are quantitative research on this, e.g. on how humans scan faces (i.e. our eye movements) and the amount of time we focus on different facial features. I strongly suspect that the eyes + areas closeby are very important, so I would imagine that changes there will increase errors. $\endgroup$
    – fileunderwater
    Oct 15, 2014 at 22:45
  • $\begingroup$ I have read through en.wikipedia.org/wiki/Face_perception and ncbi.nlm.nih.gov/pmc/articles/PMC2998394 but so far I haven't found a conclusive answer to this question. Hopefully an expert on this stackexchange will be able to answer. $\endgroup$
    – Aidan Rocke
    Oct 15, 2014 at 22:59
  • $\begingroup$ Look for facial recognition software algorithms. They seemed to have figured something out about it. Distance from pupil to pupil etc.... $\endgroup$
    – rhill45
    Oct 16, 2014 at 1:03
  • 3
    $\begingroup$ @rhill45 While interesting, this is something completely different. Even if we have identified software algorithms that are effective in identifying faces, this does not mean that we as humans (with behaviours molded by evolution) use the same cues/features when we recognize faces. $\endgroup$
    – fileunderwater
    Oct 16, 2014 at 7:25
  • 2
    $\begingroup$ And for the record, I don't understand at all why this is closed as opinion-based. This an interesting Q, answerable by quantitative methods (brain activity when looking at facial features, eye scanning of faces etc), and on-topic for the site. It could be closed due to a lack of background research (@AidanRocke, you should add your summary of the Leipold & Rhodes paper to your Q, and what questions that remain after reading this), but I don't see why it would be opinion-based. $\endgroup$
    – fileunderwater
    Oct 16, 2014 at 7:33

1 Answer 1

Reset to default
1
$\begingroup$

The most important features are the eyes, followed by the nose and mouth, respectively.

An elaborate answer can be found on the duplicate question on Cognitive Sciences SE:

Are some facial features more important than others in human facial recognition?

Improve this answer
$\endgroup$
5
  • 1
    $\begingroup$ Will format question link once I am off mobile $\endgroup$
    – AliceD ♦
    Jan 26, 2015 at 14:17
  • 1
    $\begingroup$ Hmm.. Chris did you check who posted that question on cogsci? It is the same OP and the Same Answerer too! lol $\endgroup$
    – One Face
    Jan 26, 2015 at 14:28
  • 1
    $\begingroup$ Good one Chris! Cross-site duplicates are not closed as a policy (from what I gathered in my stumbling through meta), though cross-posting is not encouraged. I think you should elaborate your answer in both sites. Even duplicating your answer is good $\endgroup$
    – One Face
    Jan 26, 2015 at 14:31
  • $\begingroup$ @Crags - thanks! What I understand is that a link to a cross-site is OK. Bio people not at CogSci can still find it that way. I will leave as is. $\endgroup$
    – AliceD ♦
    Jan 26, 2015 at 14:43
  • 1
    $\begingroup$ Link is definitely Ok. Just explain what you want in detail in one site and post the link with summary in the other! Yes, leave as is $\endgroup$
    – One Face
    Jan 26, 2015 at 14:45

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .