More on Misinterpretation of Social Psychology

I sort of briefly mentioned in my last post how people (including the researchers themselves) tend to misinterpret social psychology. Here’s a pretty good (and only tangentially autism-related) example:

Researchers at the University of Nebraska recruited seventy-two undergrads to participate in a study. The undergrads were told to look at a little picture of a face at the center of the computer screen and not move their eyes from that point. Periodically, a little black circle would appear either to the right or to the left of the face, and the undergrads were told to press the spacebar as soon as they saw the circle. The eyes in the face randomly alternated between looking to the right and looking to the left, but the undergrads were told that the direction of the eye gaze was completely unrelated to where the circle would appear.

After this test, the undergrads were given two questionnaires designed to elicit information about the subjects’ political beliefs. The questionnaires asked the students for their opinions about abortion, gay marriage, and statements like “it is better to follow authority or it is better to question authority.” Students were also asked to identify themselves as either liberal or conservative. The researchers used an algorithm to rank students along a liberal-conservative axis based on their answers to these questions, then put the most “liberal” half into the “liberal” category and put the most conservative half into the “conservative” category.

Then the researchers performed a statistical analysis examining how the response time (how long it took the students to press the spacebar after the circle appeared on the screen) was related to the direction of the cartoon face’s eye gaze and the students’ political beliefs. They found that, overall, students were faster at detecting the circle when the face was looking in the circle’s direction than when the face was looking away from the circle; this makes sense since people generally instinctively follow eye gaze (even when, as in this case, they are explicitly told not to and that the eye gaze is irrelevant). However, they also found that people categorized as conservatives were much less influenced by the eye gaze of the cartoon face than liberals were; in fact, they showed no statistically significant delay in seeing the target when the face was looking away from it.

This is a neat result, and – assuming that the data aren’t falsified – there’s clearly something going on there. But scientific experiments are really only as useful as the theory that explains their results; otherwise, all you know is the not-very-useful fact that students at the University of Nebraska who respond a certain way to a questionnaire are more or less likely to be influenced by the eye gaze of a cartoon – which they were told was irrelevant – when trying to detect a circle on a screen. We’d like to be able to say that this says something about how one’s political orientation influences your thinking, or vice-versa.

In this case, the researchers suggested that conservatives’ ability to ignore eye gaze (and, since they were explicitly told that eye gaze was irrelevant to their task, I’m going to call this “ability to ignore eye gaze” rather than “failure to follow eye gaze”) was evidence of their “individualist” temperaments (interestingly, they considered the possibility that conservatives may have had more autistic traits overall than liberals, since Autistic individuals tend not to have as strong an instinct to follow eye gaze, but rejected this because there were more women on the conservative side than men and men are more likely to show “autistic traits” than women!). They concluded this even though:

  1. As noted above, the subjects were explicitly told that the eye gaze was irrelevant, so it’s not like they simply chose to ignore it on their own;
  2. The researchers cited absolutely no evidence that conservatives were more “individualist” than liberals, other than the fact that conservatives are more likely to like Ayn Rand;
  3. They didn’t include, in the experiment, any questionnaire designed to directly detect individualism; and
  4. In two of the three items in the questionnaires that are explicitly mentioned in the paper, the “conservative” answer is not the more “individualist” answer but rather the one that affirms the importance of prevailing social norms (for example, I can’t think of an “individualist” reason to oppose gay marriage, and I’m guessing that the “conservative”-coded response to “it is better to follow authority or it is better to question authority” was “yes”). For the third question they mention, which concerned respondents’ feelings about abortion, you can at least imagine a person giving the “conservative” answer on individualist grounds (for example, that the fetus is an individual with rights), but the “liberal” answer is just as likely, if not more, to be based on individualist thinking as well.

Overall, it just baffles me that anyone would try to explain this sort of experimental result based on the supposed “individualism” of one group of people even when part of what got them placed into that group is their anti-individualistic responses to a political questionnaire. Based on my read of the study, it’s just as likely that “conservatives” were able to ignore eye gaze because they were less individualist: they heard the researcher (an authority figure) tell them that eye gaze was irrelevant, and they believed it, so they ignored eye gaze.

Or you could conclude that conservatives have more self-control when it comes to following directions than liberals do. Or, if you wanted to go way out there, it’s even possible that this effect actually does have something to do with where people fall on the autism spectrum: maybe in Nebraska, autism doesn’t correlate with political beliefs among men, but does correlate with political beliefs among women, such that all of the women with autistic traits affecting gaze following were clustered in the “conservative” category and threw off the results. This seems pretty unlikely to me, but it wouldn’t surprise me if someone found it plausible.

Hopefully I’m not the only person who find these researchers’ proposed explanation pretty weak. The paper that the University of Nebraska published on its web site (“The Politics of Attention: Gaze cuing effects are moderated by political temperament” (text-encoded PDF)) appears not to have been published in a peer-reviewed journal yet, and it’s possible that peer reviewers would make them tone down their conclusions before publishing. Nevertheless, it’s not unlikely that the researchers deliberately publicized their findings before going through peer review so that they could come up with an “exciting” spin on their findings and get publicity from popular science bloggers (see, e.g., i09‘s post about this).  These days you don’t even have to get peer reviewed before going to the news with overblown descriptions of your research findings.


Leave a comment

Filed under Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s