Easily. Experiments on people are not the same as experiments in chemistry or physics. If we define a scientific experiment as “controlled, repeatable, objective, and double-blind”, then nearly all psychology experiments become very very difficult.
As I and others pointed out, measuring something like “does this feel creepy” is extremely extremely subjective. It is hard to imagine any controlled, repeatable way to define the question, which makes this very unscientific.
Even if you could objectively measure the feeling of creepiness, it is highly dependent on thousands of impossible-to-control-for factors (like whether the subject walked through a scary neighborhood on the way to the testing center, or whether the subject is in a good mood on the day of the test). You are extremely unlikely to get the same results twice (even if you test the exact same person), which by definition makes it unscientific.
People are not the same. We react differently to the same stimuli. Unlike hydrogen atoms, results achieved on one group of people are not guaranteed to be repeatable on other people. Case in point – the pictures above are “creepy” to me, but apparently not to you. And the sort of people who typically volunteer for psychology experiments are provably NOT representative of the population as a whole.
For something as complicated as this, to draw meaningful conclusions, you’ll need statistically thousands of data points. A typical psychology experiment is lucky to get maybe a few dozen subjects.
Unlike experiments in chemistry or physics, people know they are being experimented on and consciously or unconsciously alter their behavior. IE, “The psychologist asks me if this picture makes me feel creepy. Well, I wasn’t even thinking of that, but since he’s asking a leading question, well, now that I think about it, I guess maybe it does seem a little creepy”. Or alternatively, “That scares me, but I don’t want to admit I’m scared to this psychologist, so I’ll say it’s not creepy”. There is no reason to assume people have the same behavior when they’re being monitored by a psychologist. In fact, there’s a lot of evidence that they behave differently.
Related to the above point – read “Super-Freakonmics”, where the author analyzes a psychology experiment that stood unchallenged for 50 years as proof that people are basically altruistic. But he shows how the results are highly questionable if you think about it, and basically just re-confirm the psychologist’s original viewpoint. It is very difficult to come up with a psychology experiment that directly tests the effect you’re measuring.
People are simply not logical. There are plenty of instances where people make clear “mistakes” in psychology tests. IE, I’ve taken the Myers-Brigg test multiple times and gotten different scores. Or if you show a person a picture and ask “on a scale of 1-10, how creepy is this picture?”, then show them other pictures, and then show the exact same original picture again and ask how creepy it is, it’s highly likely the person will give a different answer than they originally did.
I could go on. Psychology may well be a valid field of study. But claiming psychology experiments are as scientific as experiments in the hard sciences is very dubious.
doubled good psychology experiments have no problem with any of those objections.
Yes psychology must objectify the subjective. Various rating scales are create and validated across multiple populations for just such purposes. Ranking animated faces or face/voice combinations as “trustworthy” or describing them in positive or negative terms is easy to measure and to quantify.
So since it is not all that difficult to measure how “creepy” something is then you need the same controls that you need in medicine or any other biological science. Generally that means using double blind control groups that distribute those factors fairly evenly across both conditions, enough that statistical analyses can be used. Indeed a researcher must be cautious to not overgeneralize findings from one population to all populations - just as do medical researchers and those who study any living systems.
Addressed already
Meaningful results may or may not require large populations. Look at the stats of the studies.
Again controlled double blind studies address that. Often the item in question is buried among a host of other irrelavent questions and repeated to be sure it is valid across time.
ALL knowledge is subject to faulty interpretation by sloppy thinkers; psychology researchers do need to be especially circumspect that they interpret findings critically and cautiously but psychology is certainly not alone among fields of study in having those who think sloppily. They are in chemistry and physics as well.
No people are not logical and that is why we study psychology. How does that rating change with time and repeated exposures? Is it a consistent change among people with similar backgrounds and experiences or not? Is it the same or different across genders or cultures?
Good scientific studies in psychology require careful design and careful interpretation. Some researchers fail to meet those standards, tis true. But that is a limit of those researchers, not of the field.
NPR actually did a piece on this last week, referencing an experiment done by Asif Ghazanfar, a professor in the Neuroscience Institute in the psychology department at Princeton University, using Macaque monkeys.
I wonder if this explains why so many people are creeped out by people who’ve used botox. Botox paralyzes facial muscles, effectively making your face behave a little less naturally. Perceptive people probably notice that something’s a bit off.
But if it affects how the uncanny valley is expressed then isn’t it in fact directly related to the phenomenon? One can then say that the phenomenon isn’t a universal absolute, that its expression depends on cultural context. One can then directly apply the findings in modifying robots or CGI characters for different populations. It wouldn’t just be noise in the data.
This is the best example of the uncanny valley (links to flash video with sound) that I have yet seen. Pretty soon, robots such as these will be walking (and singing!) amongst us. I shudder with dread.
The giant puppet in the YouTube link is a lot less creepy if you mute the sound (though nothing can quite fix the creepiness of the tongue). Ron Mueck’s stuff is creepy bordering on revolting to me; however, I don’t get the same feeling from it that I do from viewing clips of something like The Polar Express.
A) A real person acting a bit like Enzyte Bob while made up like Progressive Flo & doing a minimal lip sync to look mechanical.
B) A complete CGI.
C) Video of a live person processed through a very mild cartoon filter like those live people on the Charles Schwab brokerage commercials.
A bit out of the ordinary? Yes. Off-putting or creepy? Hardly.
The “music” on the other hand … Where’s my machete; that guy needs his neck adjusted.