Face Recognition

| 4 Comments | 0 TrackBacks

In class, we have been talking about memory and how witness identification is a key role in many cases. I came across this website, which has a couple face recognition tests. It was really interesting to see that my memory isn't as good as I thought it was. Memory, as defined in class is the ability to recover information and keep it stored safely. These tests mostly focus on short term memory. Try one of the three tests, its shocking to see how good, or bad your memory is!

http://www.faceblind.org/facetests/index.php

 

No TrackBacks

TrackBack URL: http://www.psychologicalscience.com/cgi-bin/mt/mt-t.cgi/1171

4 Comments

I did the cambridge face recognition test and got 88%, about 8% better than the average. They don't list a Std. Dev. but since anything over like 65% they said, means you do not have problems recognizing faces it is probably pretty large.


I'd be interested to learn more about the way they go about making up the various alternate faces. For instance how do they judge how different they make each face from the target face. Also I wonder why they chose to do the pixalated distortion as a test as well. What does that test for? There are numerous ways to distort a face to test recognition, what are they testing with their method?

It would seem more practical, in particular as regards this class, to have faces distorted in terms of amount of light on the face, reflection as though seen through a window, with shadows over various parts of the face or with various parts of the face completely obscured like behind a gate, part of a sign, or in a crowd, etc. Of course any information obtained regarding facial processing in general would shed light on these other area's as well.

I completed the Cambridge face recognition test with only 65% accurate. This tells me that I do have difficulties distinguishing faces. But it stated that since I right on the line, I may have some difficulties, but not that much of a problem.

This activity really opened my eyes to what witnesses do go through. It also shows how accurate I may be in identifying a suspect in a crime when I only saw the perpetrator for a split second. I may be able to describe a little bit of the person's features, but my accuracy may not be that grand. I also felt a little pressured to make the correct choice for a suspect. Because of this I tended to look at the facial expression of the faces in the lineup and choose the one that looked the most aggressive. This also shows me how people are charged with crimes they did not commit.

This activity also shows how the lighting of a picture, how clear the picture is, or the people around the suspect affect the witness’s choice of a suspect.

I believe this activity really tests the memory and helps people see how well they actually remember faces. Regarding to what we talked about in class about memory, which is failure to encode. I did not put the material of the faces into my long-term memory. Also, I believe my own biases of the faces did get into the way of who I chose to be a suspect.

I also took the facial recognition test. I was very surprised but it told me,
"Out of 72 faces, you correctly identified 71.
In other words, you got 99% correct."

I thought this test was actually fairly easy. I have always been a very visual learner and I would guess that that would help with being able to recognize the faces correctly.

I came across a very interesting article about a study done by Harvard students about what they think is a new group of people called, "Super-recognizers". They are people that can recognize someone from years ago from far away for example. The article states that science has a big focus on people who struggle with recognizing faces. Research says that 2% of the population actually has "face-blindness," or prosopagnosia, which is condition that causes a great difficulty in recognizing people's faces. This article suggests that the focus on face recognition might be interesting to look at from the opposite view.

http://www.sciencedaily.com/releases/2009/05/090519172204.htm

I thought this test was fun. I took it and got: Out of 72 faces, you correctly identified 70. In other words, you got 97% correct. I think this is interesting too, because I am also tired, seeing that it's 1:40 am. I have always been good at recognizing faces though. I can almost always recognize a face that I may have only met once. I may not know where from, but I can at least know I've seen them before. It used to happen a lot when I worked at HyVee. I would notice that I'd rung someone up before. I sometimes would mention it and they would finally be like "oh yeah!" :)

I also noticed in the beginning when you look at all 6 faces that the faces had similarities to each other. It was almost as if they took one face then just morphed certain parts each time. I think it is pretty cool.

Leave a comment

Recent Entries

Welcome to Psychology & Law!
Familiarize yourself with the blog. You'll quickly notice that all of your assignments are listed here in chronological order.…
Using Movies
In time for Thursday's, please read the following link: http://www.psychologicalscience.com/kim_maclin/2010/01/i-learned-it-at-the-movies.html  as well as the 3 resource links at the…
Book Selection
There are several options for you to choose from to do your book report. They are: Lush Life, The…