The title seems like a stretch since they then go on to say,
"Audio samples were analysed objectively using acoustic analysis for measures of timing, vocal control, and quality. Subtle differences in speech timing, vocal effort, and voice quality may exist between cannabis and control groups, however data remain equivocal."
"After controlling for lifetime alcohol and tobacco use and applying a false discovery rate, only spectral tilt (vocal effort and intensity) differed between groups and appeared to change in line with duration of abstinence from cannabis use"
Basically no acoustical difference, except in perceived effort. You'd get the same intensity out of a group of tired people too, but that oddly doesn't seem to have been compared in the study
I feel like it’s more of a culture thing. You spend time with people who talk a certain way because of X reason and then you pick up their way of talking over time. This sounds like someone’s half assed attempt at a disertation for their graduate degree
Would be interesting to see a study comparing «lifestyle stoners» who surround themself with the culture and other smokers, to people who use cannabis, but does not identify with the stoner culture.
(Although I suspect this might be an oversimplified dichotomy)
I did not read the results, but this definitely depends on the effect size and the variance of whatever they are studying. If 31 people were raised from the dead, for example, it's plenty of sample size.
The issue is that the sample might be highly biased. If it's just 31 white males from Lubbock Texas then what does it even say? Stoners in that town speak different?
With such a small sample size, formal methods break down. The sample could be biased. In fact, it's almost guaranteed. The ability to p-hack (even inadvertently) is extremely high. Publication bias is also practically guaranteed. These kinds of issues void significance to a reasonably high degree.
At these kinds of sample sizes, the biggest issues are not going to be included or visible in the study. Unless the reported significance is extremely high, it's not meaningful.
I don't really participate in scientific literature or journals, but I see a lot of notes in HN around papers that appear to be overblowing the significance of their findings.
Is there a way for researchers to publish 'interesting observations' as just that, something that they measured that may or may not be noteworthy but felt worth sharing with the broader community?
The traditional solution to this problem is to insert a paragraph in the conclusion saying something along the lines of "this merits further study". However, that particular phrase has been overused in journal articles, to the point that it doesn't mean anything anymore.
It's not really something researchers can control. I was commenting on the news release, and it being posted on Hacker News, because that's more of an issue.
Yes agreed - the title doesn't seem to accurately reflect the details in the abstract.
Beyond a larger sample size, IMO it would have been also been useful to take repeat measurements of the cohort over time and also under the influence, especially given the observation of "change in line with duration of abstinence from cannabis use".
Perhaps those details are in the full paper, I haven't been able to find a copy.
> Basically no acoustical difference, except in perceived effort.
On the data points measured. It's easy to observe things like "heavy cannabis users tend to have flat affect," but very difficult to translate "flat affect" into something measurable.
Look at the way that Alex Grey or Terence McKenna talk. There is clearly something going on there, regardless of our ability to measure it.
It’s hard to separate the effect of culture. A lot of cannabis users have an affinity for the cannabis sub-culture, or with related ones like the hippie/hipster cultures that have many shared speech patterns. You might see similar results analyzing the speech of surfers, pilots, sorority members, etc.
I moved to Hollywood, CA from Rockford, IL to live with my father when I was 14 years old. When I started school at the end of Summer there my mid-western accent really stood out and I took quite a bit of ribbing over that. The kids there called me a "Hillbilly" because of my accent. But down in Texas, where I spent a summer a few years before that kids there called me a "Yankee" based on my accent.
But what stood out to me in Hollywood was how many "cliques" there were there. One of the first things kids there told me is I had to join one. They said I could choose from "BMX'ers, Surfers, Skateboarders, Freaks, Motocrossers, Bikers, Stoners" and probably a few others.
Each of those groups had their own phrases and very distinct accents and you had to talk and dress like them to be a part of that clique.
So, I think one could easily conclude the same thing about speech for skateboarders as pot smokers or any other social clique. But I don't think you could detect that influence with my speech, even though I've smoked weed since before I moved to Hollywood 48 years ago. You almost certainly could detect I'm from the mid-west though.
This specific speech pattern is pretty difficult to fake though. Anyone can talk like a surfer, but I think the average person would be pretty hard pressed to start imitating Alex Grey.
> You'd get the same intensity out of a group of tired people too
That would be interesting too, since AFAIK weed helps with sleep and gets you more deep sleep, but at the cost of REM sleep. There's a lot of theories tossed around as to what that can do to people, but I found it hard to find something solid in that area.
To point out the absurdity of this study using only 31 users with varying degrees of other types of drug use, as well as measuring some highly subjective metric, let me offer some purely anecdotal evidence that has a bigger sample size. I can name more than 31 rappers (try me) that have heavy, long-term cannabis use that have fast "speech timing", high "vocal quality", and low "vocal effort" compared to the general population.
It doesn't show that at all. It could mean that that profession selects people who are the best in the world at those things, and cannabis is entirely coincidental.
> You see, I think drugs have done some good things for us. I really do. And if you don't believe drugs have done good things for us, do me a favor. Go home tonight. Take all your albums, all your tapes and all your CDs and burn them. 'Cause you know what, the musicians that made all that great music that's enhanced your lives throughout the years were rrreal fucking high on drugs. The Beatles were so fucking high they let Ringo sing a few tunes.
"To point out the absurdity of this study using only 31 users with varying degrees of other types of drug use, as well as measuring some highly subjective metric, "
That is the reason they gave publicly, not the reason they don't like the study.
The study used participants without a history of other drug use.
And having the effect appear with only 31 randomly sampled users means the effect is easy to detect. This is a normal number of participants for this sort of study.
You see this sort of motivated reaction to a study every time there is one that triggers a population.
" You see this sort of motivated reaction to a study every time there is one that triggers a population. "
This is true, but in both directions.
Because unlike you said, they actually only tested 31 cannabis users against a controll group of 40. And found that maybe subtle differences exist.
"Speech samples were collected from a carefully described cohort of 31 adults with a history of cannabis use (but not use of illicit stimulant drugs) and 40 non-drug-using controls."
"Subtle differences in speech timing, vocal effort, and voice quality may exist between cannabis and control groups, however data remain equivocal"
> they actually only tested 31 cannabis users against a controll group of 40
This is exactly like I said. In fact nothing you say contradicts anything I've said, so I'm not sure what point you're trying to make.
Adding more participants would measure the same subtle difference with increasing precision, but it's unclear what the point of that would be.*
The next logical step is to look further into voice effects, since the voice effect (along with the similar motor effects) are just easily measurable proxies for other changes going on in the brain.
Generally, if one's criticism of a study uses only concepts you learn about in the first day of intro to statistics (e.g. "correlation does not imply causation" or "there are only N participants"), then chances are pretty low that one has discovered a serious flaw that was missed by the authors and reviewers.
*We can get into a wider discussion about power here, but that tends to be pretty subtle and nothing at all about the power of the study merits it being called absurd.
"nothing at all about the power of the study merits it being called absurd. "
Well, I have not read the paper in detail, because I do not want to buy it. So I would not claim it to be absurd in its whole.
But what I can read is that the title of it makes a factual statement, while in the text you find the word "maybe". So "absurd" might be fitting for that.
And 31 is a really, really low number, for the goal:
"The aim of this study was to investigate speech in individuals with a history of recreational cannabis use compared to non-drug-using healthy controls. "
Like others have said, there is a strong influence of subculture speech patterns.
Have 10 rastafarie influenced people in the group and sure you will notice difference in the speech pattern. And difference does not necessarily mean "worse". I would have to read the paper to judge what they exactly meassured. And of course, how much the "recreational use" of the person involved was. Because intuitivly: sure, a person smoking 10+ joints a day speaks different, if at all, compared to the normal person.
I think it's fair to criticize a study if you think the headline (or the popular writeup) oversells it. However, my reading here is that the abstract basically says
"It's possible that W, X, Y, and Z are different in our populations. However we can't conclude that because after doing the appropriate controls, we only find that Y and Z differ."
The existence of Y and Z differences is itself sufficient to justify the language in the headline. What the authors are pointing out is that there are some obvious other variables to check, and they checked them, but their measurements weren't precise enough to detect differences in all of them.
> Have 10 rastafarie influenced people in the group and sure you will notice difference in the speech pattern.
But obviously this is not what they did. Again, this fails the basic sniff test of whether your criticism uses concepts more difficult than day 1 of statistics 101.
Yes it is possible that the cannabis sample is biased in some way. But the point is it's not biased in any obvious way. The same is true of any scientific study. All of Newtonian physics was biased against objects moving near light speed. That doesn't make it absurd or bad, it just means you had to keep measuring things until you found out what the bias was.
> And difference does not necessarily mean "worse".
Sure, but in this case it does. Intensity was decreased and the speech was more effortful.
> And of course, how much the "recreational use" of the person involved was.
Given they only had 31 subjects, and the group with cannabis use without cannabis use differed drastically in alcohol and tobacco consumption, I think one should take these results with some degree of caution.
Yeah. There are other tricks to make a tiny n seem more relevant, like making the relevant (tiny) n a subset of a bigger sample, taking many separate samples from different populations and looking for correlations in them separately, and so on. Trust the science.*
*Whereas science is defined as getting a certain score in an abstract statistics game.
What I find most problematic is that there is clearly a mix of difference incentives at play. At the one hand, scientists are not getting tired of waving the scientific method about to the uninitiated. Claiming science has no agenda and is only interested in hard data. But on the other hand, it is rather clear that the career of individual scientists depends on how much they publish. Yes, there is peer review, but no, that doesnt help, because people form networks and have friends and allies. So, how do we ensure that the common people trust science more? I think we should start by ridiculing publications like this, if the scientific community by itself is not capable of filtering such noise out.
Not convinced a sample size of 31 is enough, but imagine the implications of a ML model that can detect if you’ve used cannabis recently. Seems like a privacy nightmare.
Well, the first thing I thought upon reading the headline is someone somewhere is going to build an AI based on this and there will be false positives and it'll suck.
Others have pointed out the flaws in this study. I want to add: at first blush, the basic premise seems plausible on its face. I bet I'd be able to pick out many long-time regular cannabis consumers (of whom I'm one) based on audio recordings of their conversational voice.
>Differences between groups may reflect longer term changes to the underlying neural control of speech.
A simpler hypothesis: Cannabis smoke is hot and messes with your throat. No need to invoke mysterious neural changes. See also: links between hot beverage consumption and throat cancer.
I've been a recreational Cannabis smoker for almost 20 years, since I was 14, and smoking it 100% has a noticeable effect on my speech. People, also stoners mind you, will make fun of me for it because of the level of effects it has on me. For instance, they would say I'm actually a model for why people shouldn't smoke pot (again these are users). Anyway, these effects go away with abstinence, but generally take up to a month to really go away (not that I've had many breaks that long in the last 20 years). Yes, weed effects my speech, but also other activities that involve hand-eye coordination or heavy thinking, such as sports, video games, programming, learning mathematics, etc.
It's not a huge stretch of the imagination (and a theory supported by research) that short term neurochemical changes will result in longer term structural, i.e. syntactical, changes in the brain. There is a lot understood about this type of mechanism (I did a PhD in computational neuroscience, so I've seen and understood the research in this area).
Is recreational cannabis somehow different than medical cannabis? Why specifically mention "recreational" at all instead of just cannabis itself? If you're taking it medically and in a comparable quantity/frequency, somehow it doesn't affect speech?
Do these folks study speech and know what to control for at all? Seems there are more critical confounding factors a study is likely to observe before psycho-active drug use is a significant factor https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2790192/
The sample size is too small to be definitive. But that could be fixed by doing the same tests in smartphone voice recognition programs and providing the results to advertisers. Facebook probably has enough information already to recognize cannabis users.
how’d i know every comment would reflexively try to find holes in the study. it’s funny i don’t see the same gusto for finding flaws in pro cannabis “studies” where it supposedly cures cancer and every other disease
Not sure why I got flagged for mentioning Scooby-Doo. The censors have no sense of humour. Shaggy spoke like a stoner and got munchies all the time.
"Older teenagers and adults have admitted to enjoying Scooby-Doo because of presumed subversive themes which involve theories of drug use and sexuality, in particular that Shaggy is assumed to be a user of cannabis"
The way America’s legalization effort has progressed has been really unfortunate. It seems as though no attempt at finding a middle ground was made, and consequently we wound up with complete legalization.
Marijuana is an addictive highly potent mood altering substance that’s become available on demand with no real discussion about the consequences. Will anyone argue that having 15% of the otherwise young and productive demographic zonked out 24hrs a day on “Mint Diesel x AK-47 DDH Live Resin (TM)” gives us an advantage over China? Hyperbolic discounting par excellence.
By that logic America should also ban video games because Chinese kids arn't "zonked out" on Fortnite. You're not having a real discussion either, you're employing overblown rhetoric in your request to bring the conversation to a more sane environment.
In most states there are limits on purchases for different products. While I totally agree that more regulation should be in place to inform consumers, I cannot at all agree with your rational or verbiage.
This is the sort of response that makes me wonder: dude, what have you been smoking?
Let's start with the whole notion of organizing our society based on some perceived need to compete with a country half the world away. That's the sort of paranoia combined with hazy connections that I'd expect from somebody who was a bad kind of high.
And then there's the concern about addiction. Marijuana is way down the list of addiction potential. Nicotine, alcohol, and caffeine are much stronger and the former two have widespread proven harm: between the two they cause ~600k deaths per year. [1][2] (People debate the latter more, but most of those arguing in favor of caffeine are addicts themselves, and nobody can deny that Starbucks is a plague upon the urban landscape.) And let's not forget sugar, the white powder that causes enormous medical harm and that whole industries work to hook our children on!!1!
Jokes aside, I'd note that at least here in California, we already have a "middle ground". Marijuana is only available from a relatively small number of places, with production and distribution heavily regulated. Alcohol and nicotine are much more widely available. If anything, I'd expect legalization has resulted in reduced availability to minors, as this has surely driven a lot of the previous weed dealers out of business.
I believe that we have a duty to prevent people from exploiting others, especially to prevent organized commercial exploitation. And I believe that duty is strongest where addiction is concerned, because that bypasses the rational choice that freedom depends on. But on balance I think we do a much better job with marijuana than drugs like alcohol, nicotine, and the legal opiates.
> And let's not forget sugar, the white powder that causes enormous medical harm and that whole industries work to hook our children on!!1!
> Jokes aside
You don't think sugar causes enormous medical harm, or that the industry is geared towards addiction of their users?
Walk into any supermarket, it's everywhere. It's even in wide display in the checkout area, their last chance before you leave, along with cigarettes and cute bottles of alcohol.
oh, I agree. But most people are so far from seeing it that I'm just going to joke about it. It's way easier for people to see the addictions of others than their own addictions.
I would argue that the injustice of mass incarceration resulting from inequitable and inconsistently applied drug laws was far more unfortunate and harmful. I say: more legalization, and now.
If you used a different strain like UK cheese it would provide an anti-anxiety effect to that group. Mint Diesel and AK-47 sounds like it would make you ampped and confused. Finding/using the proper strain for the effect you want is the missing context from your comment above.
The discussion of the consequences were raised 50 years ago and the accepted conclusion was the biggest harm was it acting as a gateway drug to crack specifically in the 80s. I don't think that belief is as accepted as it was before.
Your fears of a lost generation and losing a trade with China as a result is not based in reality. The economic loss to China is a result of large companies trying to open the market to their product while giving away too much and inbalances in currency and trade barriers. A super motivated extra 10-15% of the under 29 age group just doesn't matter when weighted against those bigger policies.
Federal legislation is far from finished. Who knows what we end up with.
I for one would prefer to pay rehabilitation therapy rather than supporting the ludicrous enforcement apparatus we have now. If someone wants to burn away their brain cells, let them do it. Why involve police and the justice system? Why throw them in jail, where they have even less chance of rehabilitating themselves?
I don't need to retread the old discussion "How on earth is alcohol legal?", since the effects of alcohol abuse are so much worse than the effects of marijuana abuse and it's rather hard to OD on the later. But in the end; what do I care?
There's the case to be made for protecting young adults, since there's solid evidence that smoking weed is harmful to the developing brain.
I don't use it, nor do I drink alcohol, smoke tobacco (or vape), and I don't use any other drugs at all. I haven't, for decades. I'm no fun at all.
That's because I burned through my quota by the time I was eighteen.
As far as I'm concerned, they should totally legalize it, to the same extent as booze (which, if discovered today, would easily be classified as a Schedule IV narcotic).
Here in New York, we had the Rockefeller sentencing laws in the 1970s. There's people still in jail, today, for offenses that would get a slap on the wrist, these days.
I doubt that it will suddenly make America look like the parking lot of a Grateful Dead show. People who would have problems with legal weed, are already having problems with it now, or are having problems with other drugs (like booze). Most folks won't have any real issues with it. Some will have problems that will cause things like traffic fatalities. Big Tobacco will take over the industry, destroying many of the mom-and-pothead dispensaries that we have now, and we'll probably see an "artisanal cannabis" industry, like the wine industry, today. I suspect that Napa Valley will become known for things other than wines.
Personally, I think it could be pretty cool. It will be another industry that would promote "back to the earth" values, like farming (except for the inevitable factory farming).
If we were really concerned about protecting people's health, we'd still be in prohibition, and tobacco would be 100% illegal, as would whole classes of junk food.
> Here in New York, we had the Rockefeller sentencing laws in the 1970s. There's people still in jail, today, for offenses that would get a slap on the wrist, these days.
That's so sad. Lives were ruined and society not only pays to to ruin their lives by keeping them in prison, but also basically destroys any means of rehabilitation. We really have to rethink our justice system when it comes to drug abuse.
Here's a good analysis by the CATO institute, and the conclusion is that legalization has not had much of an impact in either the positive or negative direction. Until there's good data to support the idea that cannabis legalization is creating harms it's probably best to remain neutral. Both pro and anti cannabis sides are going of "feels", rather than facts it seems.
Guinness World Record for biggest strawman. As if anywhere close to that number of people would be permanently "zonked out", or that the "zonked" ones wouldn't aready be as such on alcohol or illicit cannabis, or that the government should have the right to tell you exactly which mood-altering substances you can enjoy, or that cannabis isn't actually a beneficial mood-altering drug for many people.
In other words, I disagree with almost every premise you've established.
I think our ability to evolve, adapt, invent, and assimilate will keep us ahead of the game longer than others might, though eventually it will lose efficacy. The Chinese government is too genocidal, militaristic, and authoritarian to permit the same broad range of innovation as exists...basically everywhere else (including Russia). China will only continue to advance as long as the rest of the world permits their technology thievery and genocidal empire building to continue.
"Audio samples were analysed objectively using acoustic analysis for measures of timing, vocal control, and quality. Subtle differences in speech timing, vocal effort, and voice quality may exist between cannabis and control groups, however data remain equivocal."
"After controlling for lifetime alcohol and tobacco use and applying a false discovery rate, only spectral tilt (vocal effort and intensity) differed between groups and appeared to change in line with duration of abstinence from cannabis use"
Basically no acoustical difference, except in perceived effort. You'd get the same intensity out of a group of tired people too, but that oddly doesn't seem to have been compared in the study