Investigation 12: Fruit Fly Behavior
Fruit flies are placed in a two-chamber choice device to test whether they move toward or away from a stimulus (taxis) rather than just wandering randomly, and a chi-square test tells you if the lopsided count is a real preference or chance.
A choice chamber connects two chambers, one with a stimulus (light, a chemical, moisture) and one without, through a tube flies can cross freely. If flies have no preference, they should end up split roughly 50:50 between the two sides, since movement is random relative to the stimulus. If they pile up on one side, that clustering suggests the stimulus is actually influencing where they settle. The setup works because both chambers are identical except for the one variable being tested, so any lopsided split can be attributed to that variable rather than something else about the chamber.
To run the test, flies start at the connecting tube and are released, then counted on each side at set time intervals: maybe every 30 seconds for several minutes. Plotting the counts over time usually shows the distribution swinging around early on, then leveling off into a stable split as the flies finish responding to the stimulus and settle. That final stable count is what gets compared to the 50:50 null hypothesis using chi-square.
This investigation sits in the lab unit alongside other behavior and statistics labs because it's really a chi-square exercise wearing a biology costume: the mechanism is a goodness-of-fit test, same math as the genetics crosses elsewhere in the course. It also sharpens the taxis versus kinesis distinction, a vocabulary point that shows up again in animal behavior discussions. Taxis is oriented: the fly moves directly toward or away from the stimulus source. Kinesis is unoriented: the fly just moves faster or turns more often without heading anywhere in particular relative to the stimulus.
Key terms
- taxis
- Directed movement toward or away from a specific stimulus, oriented relative to its direction.
- kinesis
- An undirected change in movement rate or turning frequency that doesn't orient toward or away from a stimulus.
- chi-square test
- A statistical test comparing observed counts to expected counts to see if a deviation is likely due to chance.
- null hypothesis
- The default assumption, here that flies split 50:50 with no real preference between chambers.
- degrees of freedom
- The number of categories minus one, used to pick the correct critical value for a chi-square test.
Watch it explained
- How it worksHow does a two-chamber setup separate a real preference from random wandering?
- How it worksWalk me through counting flies over time and plotting the distribution as it settles.
- Worked exampleFind the chi-square value for 34 flies on one side and 16 on the other.
- Common mistakeWhy isn't a taxis the same kind of behavior as a kinesis?
Check yourself
End of topic.