- As scientists trying to get to the truth in whatever matter we're studying, it's critical that we're really conscious about these sources of bias. We have outcomes that we often hope to see from an experiment and it's critical that when we're taking measurements we do so in a way with the checks in place to make sure that with that bias that we have doesn't creep into the resulting data.

- If the experimenter is having a bias in how they are collecting the data treating the samples. In a way, that is a confounding variable because it is a factor that is differing from sample to sample. That is, your technique, or your attention to detail or your level of criticality is varying from sample to sample, in a way that you are not controlling for. So, the presence of your bias can also confound your results. So, if you see something that's different between two treatments, you won't know whether it's real biology, or if it's because you were being biased.

- I think the key thing to mention here is that probably in most cases bias is unintentional. Meaning that, the experimenter is really... trying to do it right, but, but there are choices that the experimenter is making that are not allowing, kind of a neutral measurement.

- If you're aware of your own bias, then you can integrate that into your experiment to have an experimental design that is less biased and is more... approximates better the truth. So for example, if you know you will be excited about a given result, and not excited about another result, and you're scoring a phenotype you're gonna be more likely to score... to find and see the exciting phenotype than the phenotype that is less exciting. So a good way of deciding an experiment to control for that is to do blind scoring. So, if you do blind scoring, particularly double blind scoring, where even the person that is helping you to do the blind scoring doesn't know what the different experimental systems are, then you're able to, in an unbiased fashion, score the phenotypes... without influencing the outcome of the experiment.

- Blinding is when the experimenter doesn't know what they're scoring. They know how to score, but they don't know how that relates to the, the different experimental variables that they're testing.

- Imagine that you are studying how a particular mutation affects tumor growth. So you will have two population of cells, the cancer cells, and the cancer cells with that particular mutation that you have introduced in vitro in the lab. So then you go into mice and you introduce those two populations of cells. It's possible that that mutation increases, or decreases the tumor growth. You can have your evidence, suggesting that it is going to be one thing or the other. But the best way to not be biased toward one outcome, is to just be blind when you are monitoring the tumor growth in those mice. With this I mean to label those mice with numbers, not calling them "wild type" and "mutant" or this particular cell line and this mutation in this particular gene. Just a number, and often actually having someone else, not yourself measuring those tumors, will help you to be blind in this experiment. If that person doesn't know which mice were injected with which cells.

- In this case, you're extracting the bias because you cannot have a bias when you don't know what you're looking at and it's random. But that only comes if you recognize that that bias is there. So that's why it's important to recognize that those biases are inherent. And then, design experiments that account for those biases, rather than saying I'm gonna get to a point where I'm not biased about this. And then I'm gonna score it. That doesn't work.