Ep. 390: Occam’s Razor and the Problem with Probabilities

Pamela facepalm. Photo courtesy Chuck Tomasi

Pamela facepalm. Photo courtesy Chuck Tomasi


I’m not saying it’s aliens, but it’s aliens. Actually, it’s almost certainly not aliens, or a wormhole, or a multiverse. When scientists discover something unusual, they make guesses about what’s happening. But Occam’s Razor encourages us to consider the probabilities of different events before making any concrete predictions.

Download the show [MP3] | Jump to Shownotes | Jump to Transcript

This episode is sponsored by:  Casper, Swinburne Astronomy Online, 8th Light

Show Notes

Occam’s Razor
Occam’s Razor

Transcript

Transcription services provided by: GMR Transcription

Announcer: This episode of Astronomy Cast is brought to you by Swinburne Astronomy Online, the worlds’ longest running online Astronomy degree program. Visit astronomy.swin.edu.au for more information.

Fraser Cain: Astronomy Cast, Episode 390, Occam’s razor and the problem of probabilities. Welcome to Astronomy Cast, your weekly fact-based journey through the cosmos. We hope you understand not only what we know but how we know what we know.

My name is Fraser Cain. I’m the publisher of the Universe Today, and with me is Dr. Pamela Gay, a professor at Southern Illinois University Edwardsville and the director of CosmoQuest. Hey Pamela, how’re you doing?

Dr. Pamela Gay: I’m doing well, how’re you doing Fraser?

Fraser Cain: Good. So, the fans probably wanna know do you have any update on the fact that CosmoQuest has been renewed and funded for a few years?

Dr. Pamela Gay: No, not yet and in fact I’m waiting for a phone call from a contact at NASA headquarters so if my phone rings, I actually don’t have it off today –

Fraser Cain: You might have to take it.

Dr. Pamela Gay: I might have to take it.

Fraser Cain: Yeah, yeah and you won’t be able to record it so people won’t be able to hear.

Dr. Pamela Gay: Right, exactly.

Fraser Cain: But so this is the sad and terrible thing, right, is that we know that CosmoQuest is part of the group selected for education funding, but we don’t know for how much and for how long and any of that yet.

Dr. Pamela Gay: Well, so, no, I actually do know; 2.5 years, I can’t say the amount yet, but I have a target amount but that’s why I’m waiting for the phone call is to negotiate the target amount so yes, but at least five years.

Fraser Cain: Well, by the time you hear this, we may know.
Dr. Pamela Gay: Yes.

Fraser Cain: That’s exciting, awesome.

Announcer: This episode of Astronomy Cast is brought to you by 8th Light, Inc. 8th Light is an agile software development company. They craft beautiful applications that are durable and reliable. 8th Light provides disciplined software leadership on demand and shares its expertise to make your project better. For more information, visit them online at www.8thlight.com. Just remember, that’s www.8thlight.com. Drop them a note. 8th Light, software is their craft.

Fraser Cain: So, I’m not saying it is aliens, but it is aliens.

Dr. Pamela Gay: You always say it is aliens.

Fraser Cain: Actually, it’s almost certainly not aliens or a wormhole or a multiverse. When scientists discover something unusual, they make guesses about what’s happening, but Occam’s razor encourages us to consider the probabilities of different events before making any concrete predictions. Whatever has the least assumptions is probably right. All right Pamela, you threw this one onto the docket, and I guess this is sort of a reaction to the discovery of mega structures which we talked about in a previous episode or not discovery of mega structures.

Dr. Pamela Gay: I love the way you put it. We discussed the discovery as if it’s this set thing. So, it –

Fraser Cain: Incontrovertible evidence of mega structures.

Dr. Pamela Gay: Yeah, I know. So, this comes from two different things. First of all, I keep seeing people apply Occam’s razor inappropriately or at least changing what they think it means to something that isn’t actually what it means and then there’re so many awesome theories out there that we can neither prove nor disprove nor even put any probabilities on. And I just wanted to talk about how we have to try and deal with stuff like that.

So, on the one hand, we probably should start with what Occam’s razor actually is and this comes down to there was a dude, William of Ockham, and he lived from 1287 to 1347 and was a Franciscan friar and what he said was actually pretty simple. “Among competing hypotheses, the one with the fewest assumptions should be selected.” So, this means that if you have a whole bunch of different theories, theories usually involve having put in and I think this happens this much of the time or I have to assume that this thing happens; the one that has the least amount of guesswork involved is probably the one you should start with.

Fraser Cain: It’s kind of interesting that that a friar, a monk, someone who has dedicated himself to religion as opposed to necessarily science, as one of the ones who’s created one of the most longstanding, most useful concepts that’s been used in science and it really serves very well just in being a skeptic in general, right, Occam’s razor.

Dr. Pamela Gay: When it comes down to it, the fact that this came from a Franciscan monk actually isn’t all that surprising because in history, it was actually the monks, the Franciscans and the Jesuits who were often the scholars, often the scientists; were often the ones having the clearest heads in terms of trying to figure things out whether it be genetics or physics or any of the different sciences.

And I can just imagine this man of philosophy and science basically getting tired of people coming and going it’s witches, it’s leprechauns and having to respond. No, it’s germs. Well, they didn’t really have germs discovered yet, but they had dealt with the plague so they were starting to understand ideas of contagion and other things and even for a monk, it makes sense.

Fraser Cain: But I guess that’s a good example, right, that back in the day people were so, you know, they saw so many coincidences and they were so superstitious that they thought that I didn’t pay my tithes to the church and I didn’t do this and I didn’t do that and therefore my crops failed.

And so they’re making all these assumptions; they’re making three, four, five assumptions about what they did. I didn’t wear my lucky socks and therefore my crops failed and so I guess Occam is just saying you’re making a bunch of assumptions there that maybe you shouldn’t. So, how does this then apply to sort of science as it’s done in the modern day?

Dr. Pamela Gay: Well, today it means that when we’re trying to figure out what’s up with worlds like Titan where we see atmospheres that are out of chemical equilibrium. On one hand, we can say well, perhaps we don’t actually understand the chemistry that only requires one assumption, that we have a giant hole in our understanding. On the other hand, we can say well, maybe there are methane-eating bacteria and maybe these methane-eating bacteria produce the exact chemical difference that is what’s being observed in which case we’re now making two different assumptions.

Lots of things end up adding up like this so there is the idea that when we look at our universe and see expansion it’s because we’re precisely located in the very center of an over density of gravity and therefore everything around us is getting pulled out to that large ring or I guess sphere of material that is surrounding the visible universe.

Well, that requires the assumption that 1.) The rest of the universe is giant which isn’t too bad of an assumption, 2.) That there is a precise sphere of distributed matter outside of our visible universe, 3.)  We are specially located in the center of that sphere of over density in this very large universe or we can just assume we are at a random location in a universe that happens to be expanding, one assumption.

Fraser Cain: Right and that everybody is experiencing – wherever you are, you’re experiencing the same –

Dr. Pamela Gay: Effect.

Fraser Cain: – expansion, the same effect. For everyone’s perspective, it feels like they’re at the center of the universe.

Dr. Pamela Gay: And so when we actually deploy Occam’s razor, it’s a matter of how do we have the fewest assumptions embroiled in whatever it is that we’re discussing. This isn’t to say that whichever has the fewest assumptions has to be right; it’s just the starting point in where we should start our ideas, start our testing and once you’ve tested whatever’s the most simplistic, keep testing outwards until you find the truth.

Fraser Cain: Now, how do scientists make these assumptions and account for them and then sort of resolve them as part of their research paper? Like when you’re going in to do a piece of research on variable stars or extragalactic motions or whatever, are you aware of a series of assumptions that you’re making as part of the research?

Dr. Pamela Gay: With some research, you actually nowadays can pretty much get around most of the assumptions that you might need. So, for instance, with variable stars, I think the only assumption we use is one that you have to use for everything which is to say the way that you see something from one telescope can be standardized to match what you see on another telescope by using standard stars. So, if we can’t start with the assumption that everything looks the same no matter where you’re observing it from, science totally falls apart.

Fraser Cain: Right, right and so if you can’t make that one assumption then you can’t make any accurate predictions about what’s going on. Like we did a whole episode on homogenate and sort of that there’s nothing special about our spot in the universe because that has to be the baseline for you to make any kinds of predictions or any kind of real discoveries about what’s going on in the universe if you assume that everything is the same and that’s the one assumption that you’re making.

Dr. Pamela Gay: And this comes out in other places. For instance, the zero with law of thermodynamics is that if two different thermometers measure the same thing and they are calibrated then it’s actually the same temperature.

Fraser Cain: Okay and so, you know, you put this topic on the docket this week and I think part of it that you were doing sort of a reaction to this idea about the discovery KIC, phone number about a mega structure. So, what was going through your head as you sort of looked and reflected on the recent media storm and what’s probably going on?

Dr. Pamela Gay: What got me was the number of people that we talked to that when they heard that scientists were actually considering the argument that it’s aliens, we’re like okay, so, tell me the truth; tell me this isn’t true.

And I’m like I can’t because on one hand, you have to either assume that something very special happened that we happened to catch just precisely so 1.) There was a very rare phenomenon, 2.) We were looking in the right place at the exact right time to catch the very rare phenomena and it’s something that we can’t actually really justify in a lot of different ways in terms of like to get a comet storm you have to have had a nearby star pass through the exact right place, but the improbable can happen so that’s a bunch of really low probability things that we’re assuming happened.

Or single assumption, you have an advanced society that built stuff. Maybe you consider that two assumptions, but in either case we’re making assumptions about things that we don’t really understand what’s more likely and so there’s really absolutely no way for me to say that it’s more likely that there is a sudden flurry of comets falling into that inner solar system or that there’s an advanced society. I just can’t get there with science.

Fraser Cain: Right and so you can’t necessarily assign the probabilities to one or the other although, you know, is most likely not aliens, right, because comets are a physical phenomenon that we have observed; we know that there are comets, we know that there are comets surrounding the solar system, we know that there are comets surrounding other extra solar star or planets. We’ve seen other evidence of comets in other places and so comets are not impossible. We know they exist so that’s one possible assumption.

We know that there, you know, we know that things collide. We have seen evidence of things colliding and sometimes fairly recently. I think there’s like some number of stars we’re actually seeing evidence of collided planets, destroyed planets, things like that so, again, this is a thing that has happened and so as you look at all the assumptions that you’re being asked to make, there are ones that are more likely and ones that are less likely but you can’t rule it out.

Dr. Pamela Gay: And at the same time, we know that life can evolve because earth; we know that the concept of building mega structures can exist because we have the concept. We don’t know the probability of life existing in other worlds, just don’t know it. It could be 100 percent if you have the right chemistry. It could be one?thousandths of a percent; we have no concept for what the probability of life evolving a civilization is and without having any of that understanding, there’s actually no way for us to say what is more or less probable.

Fraser Cain: Right or zero, right, we don’t know. So, then what are some other kinds of conversations that happen in this world, in this sphere, that kinda get you a little on edge when a person quickly makes an assumption?

Dr. Pamela Gay: So, quickly making an assumption is something that can go both directions. On one hand, you have all of the people that don’t want there to be string theory, don’t want there to be relativity, don’t want dark matter or dark energy because these are complex ideas and the math isn’t particularly pretty when it comes to string theory and so they use Occam’s razor as the argument for why we need to get rid of string theory. Now, I admit I don’t like string theory. I don’t think there is enough scientific justification for it right now and so I’m perfectly willing to throw string theory out a window.

That said, Occam’s razor isn’t the reason you throw it out the window. The fact that it’s a theory that doesn’t have testable evidence might be a good reason, but you can’t just throw a theory out the window because it’s complicated.

Fraser Cain: Right. In fact some of the most accurate ones are like the –

[Crosstalk]

Dr. Pamela Gay: Quantum mechanics.

Fraser Cain: Quantum electrodynamics, right, is like the most accurate thing that’s ever been discovered. It is accurate to a level of decimal places that boggles the brain and yet incredibly complicated, very difficult for anyone to understand. Anyone who tells you they understand quantum mechanics doesn’t understand quantum mechanics, etc.

Dr. Pamela Gay: Exactly, so the idea that something isn’t a valid idea by invoking Occam’s razor simply because it’s complex doesn’t work as an argument. At the same time, you can’t use Occam’s razor to say that things that we don’t have a full understanding of thus aren’t allowed either.

Fraser Cain: So, for the lay person who’s looking at the – and we as sort of lay people, you know, we’re dealing with looking at all this stuff coming out in the media, I mean anyone who paid attention to any of both the space stuff and also just the non space stuff, Gizmodo, everyone was talking about alien mega structures. You look at the health issues; just today, it turns out bacon and sausages and all that is –

Dr. Pamela Gay: Carcinogenic.

Fraser Cain: – gives you cancer. Yeah, as carcinogenic as smoking and so on and so forth, right, so how do we as lay people who don’t necessarily have a deep knowledge in the field and know all the assumptions that are being made, how do we sort of examine this stuff as it comes out and try to sort of not panic or panic appropriately?
Dr. Pamela Gay: I think the example that you gave of the new processed meat discovery that was announced today is actually a really good one. One of the headlines I’ve seen several different places is that processed meat, and to be clear this is processed red meat, is as carcinogenic as cigarettes is a lie of simplification. The way that they classify things in terms of carcinogenic is has this proven to cause cancer? That puts it in the classification that cigarettes are in. Has this been shown to potentially be a source of cancer puts it into a different classification. So, in terms of identifying whether or not something is directly linked to cancer is how you classify what class of carcinogen it is.

Then the second thing that you have to look at is how likely is it to actually cause cancer so yes, this thing causes cancer but how likely is a given person to get cancer? Well, it turns out that with cancer-causing things, 86 percent of lung cancers can be tied directly to tobacco, but 21 percent of bowel cancers can be tied to consumption of processed red meat. So, there’s a difference of a factor of four, actually more than four, between your probability of getting lung cancer if you smoke versus your probability of getting a cancer of the bowels if you eat a lot of processed red meat. That’s a pretty big difference.

Now, admittedly, I’m looking at 21 percent and I’m thinking if you have a family proclivity towards cancer, just cut out processed red meat. But you do need to take into consideration 21 percent versus 86 percent with cigarettes.

Fraser Cain: And the dose, you know, for people who are smoking, they’re smoking many cigarettes a day while if you’re processing red meat or eating red meat, processed red meat, it’s only once a week, once every couple of weeks, depends on how often you eat it.

Dr. Pamela Gay: Yeah and this may mean stop eating a bologna sandwich every day and having a side of bacon all the time.

Fraser Cain: But this is an example that the World Health Organization has come down really strongly on this as a potential cause, but there’s a lot of other stuff which I’m sure people have seen where it’s like one, you know, oh, now we think that such and such causes cancer; oh, and now we think that it doesn’t cause cancer and it goes back and forth. Is a lot of that because people are making bad assumptions? Is a lot of it because the media is doing a really bad job of giving people context for where the assumptions are getting made?
Dr. Pamela Gay: Well, one of the things that we’re learning is true across pretty much all research is there’s a lot of lack of discussing bias in your sample that goes into research. So, for instance, in education research, if you have a group of test subjects that volunteered to take the special experimental class, you may see amazing learning gains because you’re dealing with a bunch of students who are actually really interested in the subject and doing great.

We see a lot of medical studies that are done on very small groups, especially selected people who are exactly deploying the protocol but real people may not be able to do that and so understanding the biases that go into research studies that involve humans is extraordinarily difficult. And even with science it’s often hard with the physical sciences to track down all of your biases. It can be things as simple as I’ve only looked at globular clusters so I’m only looking at stars that have a low metallicity or I’m only looking for galaxy clusters using this one set of filters and it turns out that I end up finding a completely different type of galaxy cluster if I instead look in the X?ray instead of in the optical.

We have to, with astronomy, examine things across the entire electromagnetic spectrum and across as much as we can the evolution of the universe to understand the biases in age, the bias in chemical composition available to construct something and the biases in what we’re able to see in different wave lengths and we don’t even sometimes know what biases go into our sample until suddenly we accidently look at something in a different wave length.

Fraser Cain: Yeah, yeah and a lot of cases what has to come up is that you – that’s what the whole peer review process is all about is to catch those biases that you may not realize that you have and someone else to say, you know, have you thought about this, what if you remove that, how does this relate to the research that you’re doing, etc., have you considered the implications of this potential bias and to include that.

And it’s tough, I mean for the regular, as I said, back to the lay people, I mean we have all these biases in our way of looking and appreciating the research; we cherry pick the stories even, the ones that just come to our eye and we’re like oh, you know, ah?ha, I knew it, you know, being a vegetarian was the right way to go because clearly red meat is really bad for you and we jump on those because of our biases.

Dr. Pamela Gay: And sometimes we just get excited. I think my favorite moment of wow, they screwed up but in a fascinating direction was there was a variety of different papers that have come out at different points in history where either an observer was smoking a cigar out on the dome floor and the chemicals from the cigarette appeared in the stellar spectra that were being taken.

There were cases of fluorescent lights being observed in a football stadium near a campus observatory that led to spectra being biased by the fluorescent lights and suddenly we go from thinking that we just discovered a new class of stars to realizing nope, we shouldn’t be smoking in the dome or nope, we shouldn’t be observing things when the football stadium is illuminated. And so our desire to discover the rare, our desire to discover the impossible and the new sometimes causes us to be blind to how we think about observations.

Fraser Cain: And so extraordinary claims require extraordinary evidence.

Dr. Pamela Gay: Exactly.

Fraser Cain: Awesome, all right, well, thank you very much Pamela. We’ll talk to you next week.

Dr. Pamela Gay: Sounds great Fraser.

Fraser Cain: Thanks for listening to Astronomy Cast, a nonprofit resource provided by Astrosphere New Media Association, Fraser Cain and Dr. Pamela Gay. You can find show notes and transcripts for every episode at astronomycast.com. You can email us at info@astronomycast.com, tweet us @astronomycast, like us on Facebook or circle us on Google+. We record our show live on Google+ every Monday at 12:00 p.m. Pacific, 3:00 p.m. Eastern or 20:00 GMT. If you miss the live event, you can always catch up over at cosmoquest.org.

If you enjoy Astronomy Cast, why not give us a donation? It helps us pay for bandwidth, transcripts and show notes. Just click the donate link on the website. All donations are tax deductible for U.S. residents.

You can support the show for free too. Write a review or recommend us to your friends. Every little bit helps. Click Support the Show on our website to see some suggestions. To subscribe to this show, point your podcasting software at astronomycast.com/podcast.xml or subscribe directly from iTunes. Our music is provided by Travis Searle and the show is edited by Preston Gibson.

Jump to Shownotes | Jump to Transcript

, , ,

No comments yet.

Leave a Reply

Powered by WordPress. Designed by WooThemes