P-Value

Article: http://www.scientificamerican.com/article.cfm?id=the-mind-reading-salmon

P-value is the probability of seeing a positive experimental outcome even when your hypothesis is wrong. In science, an experiment with a P-value below 0.05 is rendered statistically significant. Many scientists are beginning to see that this convention is a wrong one. When comparing an ineffective drug to a placebo, you will typically get a statistically significant result, one time out of twenty. Therefore, even if the drug does not work, on average, you will typically get a p-value less than 0.05 one out of twenty times. This was exemplified in a study conducted on a salmon by a team of neuroscientists. When the fish was shown pictures of humans expressing emotions, regions of the salmon's brain lit up. The p-values for the study was less than 0.001. There's no way that the salmon could have responded to the pictures though... it was dead.
I found this article interesting, because it made me rethink many of the statistics we hear. There's a chance that some of the information attained through statistics is false. The study done with the salmon proves this point. With a p-value of less than 0.001, it would be accepted that salmon respond to humans making emotions. However, the salmon was dead so it could not respond to anything. I also found it interesting because it shows how science is always changing. The p-value used to be accepted by basically all scientists and now some scientists are starting to reconsider it. Perhaps a new way to ensure accurate results will arise in the future.