I'm working on some Cognitive Psychology papers on text comprehension these days.

In the experiments those papers describe, they always use their undergraduate students as participants. I have to participate in those kinds of experiments myself to get the credits.

Now, I am not a Psychology student. But I know that to be allowed at studying Psychology, you should be really smart. At least that is the case here in Germany.

When all those experiments are conducted with some kind of elite brainies, can they really be transfered to all kind of people? Im not questioning their relevance, but if the data is really giving you a profile of the mean brain?

For example, when they say: "Calvo and Castillo (1996, 1998, 2001) and Calvo et al. (1999) observed facilitation in naming the inferential target word under 1,250- and 1,500-ms SOA conditions, but not when 500- or 1,000-ms SOAs were used."

This means that the partipant's brains did  react when the stimulus and the measurement were 1,250 - 1.500 milliseconds apart. They did not react significantly when they were only 500 or 1,000 milliseconds apart.  Now we know how fast smart young people react, when they are asked to name a target word. But:

Is this still true when you stop testing only the nerds? When you test  people that are not very bright, it might be that you get  results that don't have anything in common with the results above.

Maybe there is a study on that question, but how would the question be? "Are smart people accurate subjects for psychological studies?" No one can define what "smart" means. We all know where the problem is, but science needs more specific grounds.

I would suggest to differentiate people on their memory ability because that is what came to my mind first when I thought about Psychology students - they know how to learn :-)