Slashdot Mirror
Human Sense of Smell Rivals That of Dogs, Says Study (theguardian.com)
One scientific analysis is arguing that the human sense of smell has not only been underestimated over the years, but that it may rival that of dogs and rodents. John McGann, a neuroscientist at Rutgers University in New Jersey and the paper's author, said: "For so long people failed to stop and question this claim, even people who study the sense of smell for a living. The fact is the sense of smell is just as good in humans as in other mammals, like rodents and dogs." McGann has reached this unexpected conclusion after spending 14 years studying the olfactory system. The Guardian reports: McGann identifies a 19th century brain surgeon, Paul Broca, as the primary culprit for introducing the notion of inferior human olfaction into the scientific literature. Broca noted that the olfactory bulb -- the brain region that processes odor detection -- is smaller, relative to total brain volume, in people compared with dogs or rats. The discovery inspired Freud's belief that human sexual repression may be linked to our "usually atrophied" sense of smell. In the latest paper, published in Science, McGann points out that in absolute terms the human olfactory bulb is bigger than in many mammals and a literature search revealed that the absolute number olfactory neurons is remarkably consistent across mammals. McGann goes on to deconstruct other metrics that have been used to support the idea that human smelling abilities are limited. Humans have approximately 1,000 odor receptor genes, for instance, compared to 1,100 in mice, which some had taken as confirmation of mouse superiority. However, other work suggests there is not a tight relationship between the number of olfactory genes and smelling ability. One study found that cows have 2,000 such genes - far more than dogs.
I can't smell my own BO, let alone a nice stash of coke, weed, or vodka 2 feet from my nose.
Find me a human who can compete with a bloodhound or beagle in tracking a person, based on smelling a old shirt.
I know of drug and bomb sniffing dogs, I don't know of any drug or bomb sniffing humans.
Maybe we have as much processing hardware dedicated to smell as dogs, but we still don't do it as well as them. Maybe our sensors in the nose are worse, maybe our software running on that processing hardware is inferior. The end result ist: dogs do it better.
If our sense of smell is so good, can we tell whose urine is whose by smelling it?
------- Mark
In terms of absolute mass of brain dedicated to smell, yeah, maybe humans are the same as everything else. However, it's pretty well shown that dogs can distinguish individual scents massively better than humans. Their brain seems to divide the smells in to individual pieces whereas a human combines and b lends them all together. So a dog smells a specific bacteria, soap, cologne, and the each item in your lunch. Whereas a human just smells a college of mixed smells.
Talking about the number of genes is a bit silly, agreed. If you want to compare the two, compare them directly. There are humans and dogs trained in smelling things (in the fragrance industry, for example). Run a direct comparison test. Also of course you could directly test having untrained humans and dogs smell for food and other items.
DRUG dogs, specifically, have not fared well in blind in blind tests. While *some* dogs are probably quite good, in testing the typical police dog consistently "alerts" on wherever the handler thinks the drugs are. Tests have been done in which the drugs are in box #1, nothing is in box #2, and the police handler is *told* the drugs are in box #3. A police dog is more likely to alert on #3, where the cop thinks the drugs are, then box #1, where the drugs actually are.
From sniffing arse to kissing ass.
My ism, it's full of beliefs.
You can count the neurons and talk about the absolute size of the olfactory bulb versus relative fraction of the brain volume/weight. But proof is in the pudding. We don't have famed Scotland yard detectives sniffing their way from the murder scene. But the bloodhounds do. Till I see a human who can smell the difference between his own pee and his rival's... we need to give the prize to the dogs.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
My dog's got no nose
How does he smell ?
Just like you
Nullius in verba
I've studied olfaction, and this just doesn't seem right. This olfactory bulb argument seems like a straw man that no one in the field has been using since... the 19th century. Digging in to the article a bit, it seems the authors of the actual study agree with me, and are using different odors for humans to balance out some of our... differences. Their main point (which is right) is that the human sense of smell is much better than most people realize, and that you can be trained to follow a scent trail, distinguish similar odors, and notice the cognitive effects scent has on you. Anyone who has experimentally studied olfaction for a few years will notice themselves gaining these abilities (it goes away quickly when you're not smelling things professionally several hours a day).
So why is this summary so wrong?
First off, humans only have 400 different olfactory receptors, it doesn't matter if genetics say you should have 1000, you only get 400 (genotype =/= phenotype). Second, you have less "sensor" surface area than other mammals in real terms, not scaled for size. Third, you lack the ability to concentrate scent molecules by varying your rate of breathing like other mammals (this can be overcome by varying breathing through your mouth and nose, but other mammals don't have to do this).
Bullshit.
then why do they plant their faces into piles of dung? Can't they just get a whiff of the leavings of other animals from a distance?
That's not a bug but a feature?
A police dog is smarter than an FBI Director?
When we would pull up to my brother-in-law's farm, his dog would intently sniff all four tires of our car. Was the dog forming a mental image of the trip over hundred's of miles of highway that took us there?
A NOSE FOR ODORS
What do dogs have that we don't? For one thing, they possess up to 300 million olfactory receptors in their noses, compared to about six million in us. And the part of a dog's brain that is devoted to analyzing smells is, proportionally speaking, 40 times greater than ours.
from: http://www.pbs.org/wgbh/nova/n...
A dog's nose sure is longer than a humans...
If you look at a dog looking out of the window of a driving car - the activity is mainly smelling, seems to me dogs live in a world of smells rather visible objects.
The "sensitivity of the nose" is measured by odor detection thresholds.
Here are some values for humans:
http://www.tandfonline.com/doi...
And here are some values for dogs:
http://www.barksar.org/K-9_Det...
As you can see, both dogs and humans can detect some chemicals at below one part per billion. So, it's hard to say conclusively that dogs have "more sensitive noses" than humans. Humans and dogs are probably just sensitive to different compounds because we use smell differently. So, humans can't track prey by smell, but humans may be better at detecting dangerous chemical compounds and pathogens, something dogs often seem oblivious to.
I recall reading a book authored by Richard Feynman where he recounted a party where he used that as a party trick. He had one of his guests lay an arm somewhere inside a book, and Feynman was able to smell which two pages the person lay his arm between.
~Loyal
I aim to misbehave.
Which means they CAN smell better. GP is accurate.
While on our first rafting trip together on the Truckee River, my dog, a vizsla jumped out of the raft, cleared over 30' of water, ran up an embankment into some bushes and promptly returned with some old tennis ball that had been laying there -- that he clearly smelled from over 30' away and over moving water There is no way in hell our sense of smell rivals a dog with a nasal cavity larger than a pugs
"One scientific analysis is arguing that the human sense of smell has not only been underestimated over the years, but that it may rival that of dogs and rodents"
Utter bullshit.
If this was true they'd be using humans to sniff out drugs and bombs instead of dogs.
Just cruising through this digital world at 33 1/3 rpm...
I'll just go ahead and ship out this paper with a catchy headline and use the buzz for my next grant application.
This may be the Internet, but I CAN tell you're a dog.
Corruption is convincing someone that the selfless ideal is the same as their selfish ideal.
If the human sense of smell were as sophisticated as Dr. McGann believes, then we would never be in doubt about who "dealt it".
I've calculated my velocity with such exquisite precision that I have no idea where I am.
According to Jim Corbet, 1930s Tiger hunter with a national park named after him (seriously -- he really understood wild tigers).
In his book Corbet warned any readers that wish take up the sport of hunting tigers on foot through thick jungle that tigers do not realize that humans cannot smell. So if you are walking downwind you will be safe from an attack from behind. However, walking upwind can be extremely dangerous if there is a man eater nearby.
Information that I am sure Slash Dot readers will find very useful.
How about an experiment.
On a serious note, maybe nobody has really tried it. Dogs instinctively rely on their noses. People may have to train and practice. Being upright, we kind of have to go out of our way to sniff the ground a lot. It would take athletic training and conditioning to bend often, or use a shoveling device. Is that cheating?
A handful of deaf people have learned to use echo location to navigate by making clicking sounds. I tried it myself and indeed I got better over time, but not near as accurate as them. I found short "s" sounds more useful than clicks, more of a "ts, ts, ts...". I joke I accidentally discovered it while cussing: "shit! shit! shit!.....McAfee!"
Table-ized A.I.
Dry dogs smell OK to me, but wet dogs definitely smell worse than wet humans.
http://www.nature.com/news/hum...
A human nose has around 400 types of scent receptors. When the smell of coffee wafts through a room, for example, specific receptors in the nose detect molecular components of the odour, eliciting a series of neural responses that draw oneâ(TM)s attention to the coffee pot. But many details of that sequence are still unknown.
âoeThe relationship between the number of odorants that we can discriminate and the number of receptors that we have is unclear,â says Noam Sobel, a neuroscientist at the Weizmann Institute of Science in Rehovot, Israel. Some scientists assume that having more types of scent receptors indicates a more-sensitive sniffer.
2004 study
http://journals.plos.org/plosb...
However, some recent behavioral studies suggest that primates, including humans, have relatively good senses of smell. Resolution of this paradox may come from a larger perspective on the biology of smell. Here we begin by reassessing several overlooked factors: the structure of the nasal cavity, retronasal smell, olfactory brain areas, and language. In these arenas, humans may have advantages which outweigh their lower numbers of receptors. It appears that in the olfactory system, olfactory receptor genes do not map directly onto behavior; rather, behavior is the outcome of multiple factors. If human smell perception is better than we thought, it may have played a more important role in human evolution than is usually acknowledged.
Comparing the data on smell detection thresholds shows that humans not only perform as well or better than other primates, they also perform as well or better than other mammals. When tested for thresholds to the odors of a series of straight-chain (aliphatic) aldehydes, dogs do better on the short chain compounds, but humans perform as well or slightly better than dogs on the longer chain compounds, and humans perform significantly better than rats (Laska et al. 2000). Similar results have been obtained with other types of odors.
A third type of study demonstrating human olfactory abilities shows that in tests of odor detection, humans outperform the most sensitive measuring instruments such as the gas chromatograph.
2006 study
http://www.sciencemag.org/news...
A surprising new study suggests that people can track a scent across a grassy field--at least if they're willing to get down on their hands and knees and put their noses to the ground. The findings are unlikely to put hunting hounds and drug sniffing dogs out of work, but they may earn a little respect for the poorly regarded human sense of smell.
Humans are widely believed to be poor at tracking scents, especially when compared to other mammals such as dogs and rodents. But few had ever put that idea to the test. A research team led by Jess Porter and Noam Sobel at the University of California, Berkeley, dipped 10 meters of twine in chocolate essence and laid it in a field to form two straight lines connected at a 135Â angle. Then they blindfolded 32 undergraduate students and had them don earmuffs, thick gloves and kneepads to prevent them from using sensory cues other than smell. When set loose in the field, two-thirds of the subjects successfully followed the scent, zigzagging back and forth across the path like a dog tracking a pheasant, the researchers report online 17 December in Nature Neuroscience.
Nearly all the subjects reported that the task was challenging, Porter says, but four of them got a chance to improve with practice. Over the course of several days, they learned to follow the trail faster and deviate less. Even so, their performance remained well below what other researchers have reported in dogs.
And this ignores the not uncommon case of people who have more sensitive sense of smell than average.
She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
Irrespective of the size of the olfactory bulb, my dog can smell loads more than I can. We often walk in a forest, and for example I might see a deer ahead just over a rise, which my small dog does not (he'd go beserk he did). When we reach the spot where the deer was he does go beserk, pulling left and right (the direction the deer crossed the path), which can only be based on the deer's scent. I can't smell anything except grass and leaves, not even if I get down to his level.
Some of these scientists need to get out more.
The problem with humans and scent is that we have no standards to describe what we smell. Compared to sight, for instance, where children learn the names of popular colors. Consider that the ancient Greeks had only a handful of colors available, for the simple reason that there were no names for the rest of the spectrum. Our dependence on language means that a thing without a name does not exist.
If children were given standardized samples of each scent they are likely to encounter, along with a name for each one, their lives would be enriched forever. Who knows, symphonies of scent might evolve.
...omphaloskepsis often...
Simple field test. Take a large cloth object used by a person. Person walks to a point, unknown to anyone, about a mile away. Tear the cloth in half, approximately. Give one to a human, let the other be sniffed by a blood hound. Wait to see which one finds the human first, using only the cloth sample. The House will favor the dog.