Given that we did not have time to discuss this in seminar, please post your thoughts, comments, questions, confusions, insights, analyses, etc, regardint this chapter as a comment to this post.
MacLin & MacLin (in press)
No TrackBacks
TrackBack URL: http://www.psychologicalscience.com/cgi-bin/mt/mt-t.cgi/752
14 Comments
Leave a comment
Sign in to comment.
I still think it would be interesting to look at this from an adopted child from another race. If it is more genetic in certain races being adopted by other certain races. I also think the over all legal implications of this are outstanding. To prove that, for instance, white people not only have a difficult time distinguishing black faces but that they also are more likely to say they have seen a particular black face when they have not. This needs to be brought into every "eye witness account."
What are some different techniques police can use in presenting lineups to eye witnesses? The reason I ask this is because this has become a serious issue for a lot of African American males being falsely accused for crimes they didn't commit. As a result of this, some of them have been taken away from their homes and families in which they were a key parenting figure. Is there ways that they can present lineups without them being bias to ensure that the person makes an accurate description of the suspect?
What are some different techniques police can use in presenting lineups to eye witnesses? The reason I ask this is because this has become a serious issue for a lot of African American males being falsely accused for crimes they didn't commit. As a result of this, some of them have been taken away from their homes and families in which they were a key parenting figure. Is there ways that they can present lineups without them being bias to ensure that the person makes an accurate description of the suspect?
The impact that findings like this have on cases such as the cross-race effect, ambiguous picture line ups were a individual can not differentiate the difference between race. If the brain is thinking unconsciously or consciously bias views then how creditable would a testimony, towards a out-group member who is being falsely identified.
I myself looking and the different rows of morphed face could not differentiate between the middle portion of the lineup were the appearance and characteristics look such like my won features of being Hispanic even though there were no Hispanics in the morphed line ups based on the African American(L), and the Caucasian(R) because to me Hispanic and Latinos have different skin tones.
The thing that I have found consistently interesting in the CGM is the social psychophysics aspect of it. As one who studies more social theories of prejudice and discrimination, and having a modicum of foundation in neurophysiological mechanisms, the CGM is quite attractive and intuitive to me.
The implications of the CGM certainly include a plethora of questions and new opportunities of study, but it also leads to a whole new problem of relaying such information to the general public. For one who understand very little of the mechanisms involved or the whole process by which facial identification occurs, how are they to process such a theory? Will there be repercussions in terms of accepting that we are hard-wired for bias, so what hope do we have for change?
Finally, if we are good at identifying all faces (just with different purpose) what are the thresholds or the amount of time it takes to retrain our brains. Is the phenomena of bias to same race faces primarily in place because of our history of homogeneity in populations? Does it have a social aspect to it (i.e. in a majority group being less able to distinguish a minority group face verses a minority group distinguishing a majority group face)? Lastly, in any model, CGM included, lines indicate pathways, and pathways are usually ripe for intervention strategies. I would be interested to see what kinds of interventions can be made to manipulate the such pathways.
I have never thought about how we are more likely to recognize faces that are the same race as ours and who we surround ourselves with. But I understand how it makes sense.
In other classes we have talked about how we are more likely to put ourselves around people who are more like us, with personality and physical traits. So I can see how those are the faces that we seem to most recognize. This does seem to have issues as well with witness identification in criminal justice.
Like stated before in another comment, and like we discussed in class, I think it would be interesting if children adopted into a family that wasn't their race, if they would more likely recognize the face of their family, or the face of themselves.
I am amazed at how different changing the hair makes a face look. It isn't something that we normally think about as being that defining of a marker. There are so many different hair styles and lengths that I wonder if different lengths would make a difference. If you used long hair styles for both races would the effect be so pronounced? I also noticed that there were only examples with males. Is this just an artifact of your choices or does it reflect the research? If it reflects the research would there be a reason to think that there would be a difference between using female faces as opposed to male faces?
Zach brings up a really good point. Has there been any research done with female faces? I would think that it'd be more difficult to identify women because they are better able than men to change their appearance quickly. This of course can be done with hair style change, make-up, accessories, and the like.
Jeff: There has been research on the CRE (cross race effect) with individual who are adopted. I can't remember the name of the study but they used Asian individuals who had been adopted in France. They were in fact better at recognizing White French individuals than Asian faces. I do think there does need to be more research that looks at this phenomenon.
I also question how helpful or to what extent the CMG theory can explain the cross race effect. I say this because in real criminal cases, there are times where witnesses don't look at an actual police line-up until weeks or even months later. This is a huge confoudning variable. We sometimes mistake people for someone else, even if we have known them a long time and are the same race as ourselves. So what makes us think we can identify a complete stranger, of a different race, 6 weeks after we witness a crime?
Also, I think there needs to be more than 3 people in the study in order to boost the validity and reliability of the CGM theory (no offense). This is the first time I've heard of a theory based on the data for 3 people. It'd be great to rerun the study with a much larger N to see what the results would be.
I've noticed people have been bringing up issues with lineups and the cross race effect. Here is some background information on lineups. As far as different ways of presenting lineups to eyewitnesses, there are two possible ways: simultaneous and sequential. Simultaneous lineups are where the photographs are presented all together, whereas sequential lineups are where each photograph is presented separately. The main issue between these two types of lineups are relative vs. absolute judgments. Simultaneous lineups are thought to elicit more relative judgments because the witness can compare each photograph to the others and to their memory of the suspect, while sequential lineups involve more absolute judgments because the witness has only one photograph at a time to work with. It has been recently thought that the sequential lineup can reduce bias in making incorrect identifications, but there is a controversy about this, as well.
As far as the CGM, I have to agree with what Tom wrote. Presenting this information to the public is difficult. I attended one of the presentations on this topic last year, and I remember people in the audience asking, "So, how do we fix this or what do we do about it, etc.?" I think it may be very hard for people to accept that we may be hard wired for bias because not very many people want to believe that they are biased against another race.
I also agree with Elizabeth. There are issues with delay in making identifications. By this time, there is the potential for unconscious transference to occur, where we can make mistakenly identify someone because they are familiar to us (i.e. we see them in another context). I think that this shows the need for experts to testify about these issues to make jurors aware of them.
A lot of what we talked about in class seemed to correspond to what was in the article, but I found the results of the studies very interesting. When talking about race as a feature I feel like everyone has been in a situation where they have noticed that one different person within a group of like individuals. It is easy to notice a black person in a group of white people or a female in a group of males. They just "pop-out". Another thing that really intrigued me was Valentine's model for facial stimulus class. This type of model does explain to some extent how we catergorize what we know, but it does have some flaws like discussed in the article. It almost makes sense that we would put a certain face into categories of other faces with similar nose size or eye distance.
When it came to the studies with different races morphing together with only the hair changing, I could see that the middle did not seem to be an inbetween beetween white and black, I saw it as more latino or some other race. Being that this is the case it makes it difficult to make these studies 100% valid. Changing the hair makes a big difference in my opinion and I would like to see this study carried out again with other races or even females like other people posted above.
One thing I wanted to point out based on the previous comments so far is that the CGM emphasizes in-group/out-group differences in face processing, not necessarily just same-race/other race differences. I think we briefly brought this up in class when Kim was discussing the research from our lab, but based on the assumptions of this model we would expect to see biases in processing for individuals that may be of the same race, but if they have "out-group" markers than we may process or categorize them based on features (whereas in-group members are processed more holistically or based on the configuration of the face). An example of this would be people with mo-hawks. If you are an average looking white person and see another white person with a mo-hawk, the presence of the out-group marker (the mo-hawk) would force a lower level of processing, thus a categorization of the person based on featural processing. To me, this seems like an attention driven type of thing, where the first thing you notice might influence how the individual is processed (notice mo-hawk (or race/skin tone) first leads to a quick categorization). I don't think a lot of research has been done on this idea, but we proposed the idea of doing a face recognition study using white, bearded exemplars.
Anyway, I just wanted to point out that the racial marker that triggers differences in processing doesn't necessarily have to be a RACIAL thing, it could very well be a "you're not like me and my group" kind of thing.
a couple of clarifying points...
while the CGM can be used to talk about issues related to lineups, this model is not fundamentally an applied one(like lineups, or the cross race effect as it pertains to eyewitnesses)...rather, we are studying this from the perspective of understanding basic face processing.
regarding the comment re: small-n's--it highlights a common concern, and underlying assumption that larger n is automatically better. it's not. n size depends on methodology and unit of analysis. psychophysics is not measuring at the person level, but rather at the response level...such that, when each person makes several hundred (sometimes several thousand) judgments in an experimental task, that is the n-size of interest. psychophysics is not appropriate for all research questions (because of course, your research question drives your method), but when it is appropriate, it gives quite the bang for the buck.
I feel the ambiguous race faces are solid evidence for some kind of featural gating mechanism. We looked at this last year using the face continuums and it is interesting that once you give people a third category label (i.e., Hispanic) they have no problem putting the face into the category and it is no longer "ambiguous" to them. That's what the CGM is all about: once you see a marker, it changes the way the face is processed. You go from seeing the ambiguous face and saying "Wow, what is this guy? He's trippy." to "Oh, there's a Hispanic guy." I think that in our country, Hispanic is a good third category label to insert, but I bet you could find the same thing using some Middle Eastern/South Asian faces.
I think it's important that we keep informing the legal system about these issues. Like Kim said though, we need to do more than just tell people it's their brain and have them just using the excuse without doing anything about it. We need to educate them about ways we can become better at recognizing our own faults and adapting.