My boss has me making face morphs.  We’re using the morphs as practice trials to teach subjects how to use our test interface.   I was told my face morphs rock, so I thought I would share some of the ones we aren’t using in our study.

One of my favorite morphs is the transformation of Yoda into Albert Einstein.  Another one is Audrey Hepburn into George Washington.

I’ve spent the majority of my time at the vision sciences laboratory making faces for two new measures of facial recognition. For the first measure, I made more than 2000 faces based on 50 “parent” faces. The derived faces were based on the same characteristic features as their parent, except that I manipulated gender for each face. Now we have a library of faces that subtly vary in gender. We’ll be using these faces to see how well people can identify gender based on facial features, as well as measure the sensitivity of their gender perception skills.

The second library I made was similar to the first, except I kept gender constant and varied age instead. It’s a lot of fun to scroll through the faces and see them get older and younger.

I’ve also made a third library, although it isn’t face oriented. This library is composed of unfamiliar objects. We might use the objects in this library to develop a measure of how well people can recognize novel objects. Or not. We’ll see!

As of right now, none of the studies I’m involved in have started recruiting participants.

However, next week I’ll start screening subjects for the prosopagnosia family study.  At some point in the not-so-distant-future, I’ll be administering a battery of facial recognition tests for the machine learning study.

I’ve been contributing the development of new measures of how we perceive gender and age through faces by making more than 2000 variations of “gendered” faces and 800 variations of “aged” faces.  It feels good to be directly contributing to the project!

It looks like I will be working on three projects this summer:

1) Investigating genetic and cognitive trends in families with prosopagnosia.

2) Developing new measures of how people determine age and gender based on facial information.

3) Collecting data for a study on machine learning of facial recognition.

I don’t know much about machine learning so I’ll be spending the next few days reading up on neural networks.  It’s really great being in a place where reading journal articles qualifies as work!  (More evidence I should just go to graduate school.)

So far, I’ve been reading journal articles about prosopagnosia (the inability to recognize faces), most of which have been written by my mentors.  Reading their research has been fascinating, as I’ve had the opportunity to take the measures they use in their studies as well as talk to them personally about their work.

I have also been helping debug the battery of tests we will be using for our study this summer on familial trends in prosopagnosia.  In addition to getting to take the test to familiarize myself with the measures, I’ve been helping find problems in the way the test is presented as well as check for multiple browser compatibility.  We’ll be administering the test online, which makes it much easier to distribute.  However, it also raises a whole set of additional problems, so we’re lucky to have Sam Wilmer with us.  Sam is one of Ken’s graduate students, as well as a great programmer.  After I mentioned I know python, he suggested I help code for another study.  It would be pretty cool if I got to work with psychology (my major) and computer science (my minor) this summer!

My studies at Oxford ended on Friday, I flew to NY on Saturday, I moved into my new apartment in Boston on Sunday, and I started work at Harvard’s Vision Sciences Laboratory on Monday.

It’s been a few hectic days.

But it has also been incredibly exciting!  My mentor, Ken Nakayama, is a former Haverford student.  The other researcher I will be working under, Jeremy Wilmer, is a professor at Wellesley and also studied abroad at Oxford for a year.  I will also be working with Sarah Cohan, Professor Nakayama’s lab manager.

It looks like this summer’s research will focus on a family study of people with prosopagnosia.  Prosopagnosia is a relatively rare condition affecting about 1-2% of the population.  It has the fascinating symptom of not being able to recognize faces.  This can cause embarrassment and distress for prosopagnosics, as it is difficult to explain why one can’t recognize one’s mother, or even oneself in a mirror.  Prosopagnosia has been seen to run in families, suggesting there is a genetic component of this disorder.  Our research will investigate differences in cognitive abilities in families of people with prosopagnosia in order to see how cognitive functions group together.  We will also be getting genetic analyses of our participants.  Hopefully, we’ll be able to isolate the gene(s) underlying this disorder!

My role over the next few weeks will involve helping recruit participants and organize the study.  I might also have the opportunity to design my own project and conduct some independent research.

Can’t wait!

-Abby

Abigail Novick ’12 received a Koshland Integrated Natural Sciences Center summer stipend to do research at the Vision Sciences Laboratory at Harvard University.