Ideas for testing a psychological hypothesis

[Warcold]

I am currently in the first steps of a research on which i should write my bachelor thesis on.
We (the guy going for his PhD and my fellow students) have to come up with a way to test our hypothesis and i thought i'd might try you for some good ideas too 

We are looking for a computer-based way to let people do automatized tasks. This task should be something that (nearly) all people are good in (automatized processes). The test subjects will be 1st/2nd year students mostly, so about 18-20 years of age, both male and female.
What I thought off was (blind)typing, but it was argued this wasnt automatized for a lot of students. But think along these lines. Experiments have been conducted on expert golfers at golf putting, but this of course isnt testable on a computer.
The idea behind the automatized tasks is that the cognitive load (working memory) is near zero.

Preferably the computer task already exists, but experiments can be programmed at the uni too. Performance should be measurable, that's the biggest thing.

Thanks for thinking about this   

[Warcold]

For those of you interested: our study is doing research on stereotype threat, for example:

When women perform math, unlike men, they risk being judged by the negative stereotype that women have weaker math ability. We call this predicament stereotype threat and hypothesize that the apprehension it causes may disrupt women’s math performance. In Study 1 we demonstrated that the pattern observed in the literature that women underperform on difficult (but not easy) math tests was observed among a highly selected sample of men and women. In Study 2 we demonstrated that this difference in performance could be eliminated when we lowered stereotype threat by describing the test as not producing gender differences. However, when the test was described as producing gender differences and stereotype threat was high, women performed substantially worse than equally qualified men did.Athird experiment replicated this finding with a less highly selected population and explored the mediation of the effect. The implication that stereotype threat may underlie gender differences in advanced math performance, even those that have been attributed to genetically rooted sex differences, is discussed.
(abstract of Spencer, S.J., Steele, C.M., & Quinn, D.M. (1999). Stereotype Treat and Women's Math Performance. Journal of Experimental Social Psychology, 35, 4-28.)

and

(http://en.wikipedia.org/wiki/Stereotype_threat)

[Rubino]

Why not a simple application that displays clickable objects on the screen.
Colour code the object, use numbers, letters, shapes or just simple buttons (something they would be familair with). The size and position of the objects can easily be randomised.

They are shown what they need to click on (either constantly or for a short period of time) and then they need to move the mouse and click on the object.

You could extend the idea to use lists.  They are first shown 1 item for a period of time, which they need to do.  Then 2 items.  Then 3, etc
You would be able to adjust the time the list is shown for, the delay between the list of items being remove and being able to start the task.

Using the mouse and clicking on an object is a simple task that your audience probably does hundreds of times a day. 

[Warcold]

Ahh! See? That's why I posted here!
I like the idea very much!

There may be a couple of problems, though. I hope you don't mind mentioning them, I just would like more input on how we might solve them.
- We have to measure the performance (e.g. how fast is the screen cleared). When putting the items randomly on the screen, performance will depend on strategy, and won't really be comparable to the other measurements (even within subjects). You could solve this by putting the items in a logical order, but that would make the task very boring.
While writing this, I think we can maybe do a funny filler task (silly thing in which you dont measure stuff, just keep the subject occupied/amused/distracted) to lighten it up a bit.
- The idea of the lists is good in itself, but more for working memory research. For the research we want to do (no working memory) it is less qualified, cause you would need to use working memory to monitor which items to you need to click and which you dont. -> If I understand your list idea correctly that is.


On second thought, forget about the first problem. What we probably are going to do is present this task here as the 'filler task'.
Experiment could look something like this (my idea... the guy who leads the research has the say in this ofc, not me)
- click task (measure moment 1, presented as a filler (please do this as fast as possible as we prepare your next task)
- math-test or other stereotype laden task (the subject thinks this is what the experiment is about, but in fact we apply the stereotype threat here)
- click task (measure moment 2, presented as another filler)
- done


Thanks rub! let me know if you think i understand your list-idea right: all the clickable items on the screen consist of n groups. the list would say: click all n1, n3 and n7 items, right?

[Rubino]

Yep - depending on what you want to measure allows you to work the tests in different ways.
I have more experience with memory tests when I work on AI problems - but in your situation (depending if you wish to determine sterotype threat or some other type of problem) you can tweak the test accordingly.

Regarding memory vs non-working memory tests - there's a fine line between then as you know as the subject becomes sensitised to the problem they have been presented with.

Otherwise you could do something along the lines of clicking on the flashing buttons on the screen.
Display 4 buttons,  1 of the buttons is flashing, they have to click it, on clicking, x seconds later another starts flashing.  The reaction time can be measured, along with the search pattern (of the mouse from it's rest position to the new button).   With verbal suggestion you could tell the female subjects that women have a slower reaction time to clicking the purple buttons than the yellow (and that they also feel themselves more attacted to the researcher when there are orange buttons on the screen ).  With male subjects you tell them something different - yellow buttons or something.

[Warcold]

Nah, it works even more nifty. You do a first measurement (baseline). You say women are worse in math and that they will get a math test later in the experiment. You do a second test (effect). You say 'the experiment is over, you dont have to do the mathtest'.
Even in this (simplified) case you will see the stereotype effect.

Self-control also appears to be a construct involved. Experiments have shown ppl can't seem to hold on to a firm grip as long when under stereotype threat. Quite amazing i think.

[Rubino]

Interesting - almost like it introduces a general state of ambiguity into the system.
I wonder if it's related to earlier childhood states and protective mechanisms.

[Warcold]

Hmm, not sure about earlier childhood states, that's quite psychoanalytical and not really researchable.

Some things that are proposed to be involved:
- working memory (among others self-control)
- explicit monitoring (the golf putt experiment focuses on that, related to 'choking under pressure')
- anxiety/arousal (negative affect) has been proposed, but mixed evidence for that

one of the central foci in our research are the preventive/promotional strategies...
which i will explain later, cause i really need to get to bed! 

[Rubino]

Cool. Looking forward to it.
Isn't "explicit monitoring" & "anxiety/arousa" the same issue or closely related?