I'm really happy with my result. 9/10, and in the one I missed, I mistook the programmer for a killer. Better than the other way around, I think! I didn't make any of what I would call guesses, which is the main reason I am so happy with this result. Also, to my knowledge I had never seen any of these people before, which indicates that I made accurate decisions based only on the data supplied by the faces in the pictures.
Since I think I can describe it in fair detail, I'll try to explain the basic process of what I did, in case it might be helpful to others. For what it is worth:
It was basically a process of what I might call "systematically applied intuition". Basically, a balanced questioning of both possibilities was performed on each photograph. First, I looked at the face calmly, just taking in the features as well as the whole face. I tried to ignore clothing because I figured it was the face that would give real information. If an initial impression arose as to whether the person was a killer or programmer, I took note of this, but didn't assume. Then, while still calmly examining the face, I held in my mind the "feeling" or state of mind that represents to me the concept of a serial killer or a skilled programmer. I didn't make it a point to examine the possibilities in a particular order, but I tried to give seeing each possibility an equal effort. I tried to see the person, as he was in the photograph, as a programmer, then as a serial killer. It is like looking through two pairs of lenses in an attempt to get an impression of which "tint" or concept matched better with the demeanor of the person in the photograph. There was also a visual stage as well; I imagined visually the person in the photograph sitting and looking into a computer monitor with code displayed on it and maybe typing things, and - as the counterbalance - I imagined them standing in front of and looking at a sort of torn up human corpse (not repulsed by it at all) and maybe stabbing it with a knife. I tried to see how well the person's face "fit" into each of the two visualized scenarios, and how well the feeling associated with the face matched with each of the two concepts/impressions (killer and programmer). Then, I returned to basically neutral observation, but keeping with me the impressions that I gained from the comparison. At this point I had a decent sense of whether the face "matched" more with my conception of a serial killer or a programmer. At one point an image of a gauge appeared in my mind, with "serial killer" at one end and "programming language inventor" at the other, and a mark on the gauge appeared which seemed to represent the choice the "intuitive" information indicated was correct, as well as the estimated probability of correctness (it was a bar; closer to the center would indicate more of a random guess, further to either edge would represent a confidence in the answer it was approaching). I must emphasize that this was done in a relaxed, natural manner. The steps came naturally as a free flowing process; when I felt I had not yet reached a decision, I returned to the different modes of observation I described and spent some more time on them in an attempt to collect more data and "collapse"/"crystallize" the data better. Despite this long description, the process itself is pretty quick; I'm pretty sure I was done with the quiz within five minutes.
It seems to me that "intuition" of the type I was using is an inductive process of some sort. It's like by asking the mind for for these impressions in a controlled and somewhat concise manner, "feelings" can be produced which maybe come from a sort of "pattern matching" scan of a person's knowledge and experience. For instance, to compare one of the people in the photographs to my internal concept/idea of a programmer, I must observe the face in the photograph to extract information from it (the quality of which may be determined by my applicable knowledge and experience, which guides the extraction process) and I must call to mind my concept of a programmer (the quality of which may be determined by my applicable knowledge and experience); then, by imagining the person in the photo as a programmer and asking myself how well the two images "fit" together (asking is in the form of intent to "see"), I instruct my mind to perform a pattern-matching search to find correlations between the data extracted from the photograph and the data contained in my concept of a programmer. Then, by performing the same process with the serial killer concept/idea, I am provided with two sets of pattern matching data, and I can "see" which of the two concepts - killer or programmer - has a stronger pattern match with the face of the person in the image. If this is true, then it appears that my knowledge and experience base was fairly well suited to this task.
I apologize if it appears that I am gloating or showing off, but as far as I can tell, this isn't the case. I think that more than anything I'm just excited because this simple exercise has allowed me to observe and learn some personally very interesting things about the workings of my mind.
I thank you for you patience if you've read this far. This is only my understanding, as well as I can express it, given for what it is worth. I hope it is interesting and useful to you. I welcome mirrors in case some form of foolishness or mechanicalness is apparent in my post.