Question
Can Goldfish be conditioned to prefer a certain color while feeding?

Hypothesis
During the Goldfish's five day test with each color present in their container, they will eat from the same color feeding square that they were conditioned to eat during the first 10 days of the experiment.

Materials
Straws, Spray paint, Scotch tape, Wardley Essentials Goldfish Flakes, 20 goldfish, 10 one-gallon containers, One five-gallon bucket, Tap water, Jungle Gold Fix Complete Water Conditioner for Goldfish, Turkey baster, Fish net

Background Research
Classical conditioning is an association between a stimulus and a response. In this form of conditioning, this is where no new behaviors are learned. Instead, associations are made between an unconditioned stimulus and a neutral stimulus, so that the response is the same for both stimuli.

The unconditioned stimulus is a stimulus that will naturally bring out a reflexive response. A neutral stimulus is a stimulus that by itself will not elicit an unconditioned response. A conditioned stimulus is a stimulus that is transformed from a neutral stimulus once it elicits an unconditioned response.
(Huitt, et al, 1997)

The retina of a Goldfish contains rods, four cone types in juveniles and three cone types in adults. This allows the Goldfish to be the only animal that can see infrared and ultra violet light.
http://animal-facts-trivia.blogspot.com/2008/09/goldfish-incredible-eyesight.html

Results

Figure 1. Preferred eating for subjects one through 10, using different colored squares

Note: Table 1 and Table 2 are listed at the bottom of the page.



The data shows that only eight goldfish made it to the end of the experiment resulting in subjects four and 10’s death. Subject one was the only successfully conditioned fish. Three of the Goldfish’s data was discredited, subject two, five, and nine, because they would only eat when the goldfish flakes would sink to the bottom of the containers. The data was noted but not taken into consideration with which color feeding square the food sank from because there would have been more room for error due to guessing. The remaining four goldfish, subjects three, six, seven, and eight, ate from more than their assigned colored feeding square, reporting that they did eat more from their assigned squares than other colored squares. (Refer to Table 2 and Figure 1 for the subject’s colored preferences).


Discussion
The data collected during this experiment does not support our hypothesis that the Feeder Goldfish will eat from the same color feeding square that they were conditioned to during their five day test using all colors. The data was not significant because there were only 16 occasions that the feeder goldfish ate out of their assigned colored feeding squares during the five day test out of a total of 40 possible times.
Although our experiment was not accepted and did not have a significant outcome to classical conditioning, we can conclude that the time frame given was not sufficient for the Goldfish to view the neutral stimulus as the condition stimulus. The time period that the Goldfish were allowed to eat each day may not have been sufficient for their needs as well.
Ideas for further research include: using a larger sample size, using colored food instead of feeding squares, and the use of colored straws instead of spray painted ones. Another idea would be to determine how long of a time period it would take for de-learning to take place if future studies conditioning is upheld. Classical conditioning has not been represented with this experiment; therefore this type of research could have been more successful with the advanced tools and research stated above.