Basic Psychological Statistics

Before you left class today, you should have received this practice worksheet that requires you to apply some statistics knowledge to hypothetical quiz scores. Your textbook does cover these topics, and many people have encountered them before in other classes, so I'm hoping for many of you this is a simply a matter of review.

However, because some of you might feel lost or in need of additional help, I'm embedding below a video that I made for last year's students regarding this topic. Be warned, I was cutting it close when I recorded this and you'll hear the bell ringing toward the end. However, it should clear up confusion regarding the various descriptive and inferential statistics you'll need to be familiar with for this unit.

Additionally, if you feel you want more practice matching correlational coefficients with scatterplots, here's the link to the website we used in class today.

Having troubles playing the video? If the player above doesn't work, try here.

Scientists Identify Brain Areas Linked to Emotion in Dead Fish?

Even if this piece of recent psych research didn't reveal such interesting principles about the use of fMRI in researching the functions of specific brain regions, it would be worth sharing just for its hilariousness.

A team of neuroscientists, led by Craig Bennett of the University of California, Santa Barbara recently conducted an fMRI on a dead fish (specifically, an Atlantic salmon) while showing it pictures of people. The result: statistically significant registers of electrical activity in the fish's tiny brain!

Whoa! The brains of dead fish respond when their unliving eyes are forced to stare at pictures of humans?! Not so fast....

The researchers are not claiming some sort of zombie-like post-mortem affection fish display toward humans. Rather, they are using this result to point out that sometimes fMRI's give false positives because of how much data they are able to report. They suggest that the cutoff for statistical significance (normally a p-value of .05) when using fMRI's may need to be lowered. In other words, the currently accepted level of risk that results occurred by sheer chance (p-value) may be too high for the amount of data being collected by these sophisticated techniques: they are likely to show statistically significant activity when, in actuality, it's not there (as in the dead fish).

My favorite part of this research? The cool title the researchers chose for their project: "Neural correlates of interspecies perspective taking in the post-mortem Atlantic salmon: An argument for multiple perspectives comparison." Nice work, fellows!

Click here to see the poster presentation on this research.