Signal out of noise

I think I share with a lot of people a difficulty in deciphering what someone is saying when holding a conversation in a noisy room.  I can often pick out a few words, but understanding entire sentences is tricky.  A related phenomenon I've noticed is that if there is a song playing while there's noise going on -- in a bar, or on earphones at the gym -- I often have no idea what the song is, can't understand a single word or pick up the beat or figure out the music, until something clues me in to what the song is.  Then, all of a sudden, I find I'm able to hear it more clearly.

Some neuroscientists at the University of California - Berkeley have found out what's happening in the brain that causes this oddity in auditory perception.  In a paper in Nature: Communications, authors Christopher R. Holdgraf, Wendy de Heer, Brian Pasley, Jochem Rieger, Nathan Crone, Jack J. Lin, Robert T. Knight, and Frédéric E. Theunissen studied how the perception of garbled speech changes when subjects are told what's being said -- and found through a technique called spectrotemporal receptive field mapping that the brain is able to retune itself in less than a second.

The authors write:

Experience shapes our perception of the world on a moment-to-moment basis.  This robust perceptual effect of experience parallels a change in the neural representation of stimulus features, though the nature of this representation and its plasticity are not well-understood. Spectrotemporal receptive field (STRF) mapping describes the neural response to acoustic features, and has been used to study contextual effects on auditory receptive fields in animal models.  We performed a STRF plasticity analysis on electrophysiological data from recordings obtained directly from the human auditory cortex. Here, we report rapid, automatic plasticity of the spectrotemporal response of recorded neural ensembles, driven by previous experience with acoustic and linguistic information, and with a neurophysiological effect in the sub-second range.  This plasticity reflects increased sensitivity to spectrotemporal features, enhancing the extraction of more speech-like features from a degraded stimulus and providing the physiological basis for the observed ‘perceptual enhancement’ in understanding speech.

What astonishes me about this is how quickly the brain is able to accomplish this -- although that is certainly matched by my own experience of suddenly being able to hear lyrics of a song once I recognize what's playing.  As James Anderson put it, writing about the research in ReliaWire, "The findings... confirm hypotheses that neurons in the auditory cortex that pick out aspects of sound associated with language, the components of pitch, amplitude and timing that distinguish words or smaller sound bits called phonemes, continually tune themselves to pull meaning out of a noisy environment."

A related phenomenon is visual priming, which occurs when people are presented with a seemingly meaningless pattern of dots and blotches, such as the following:

Once you're told that the image is a cow, it's easy enough to find -- and after that, impossible to unsee.

"Something is changing in the auditory cortex to emphasize anything that might be speech-like, and increasing the gain for those features, so that I actually hear that sound in the noise," said study co-author Frédéric Theunissen.  "It’s not like I am generating those words in my head.  I really have the feeling of hearing the words in the noise with this pop-out phenomenon.  It is such a mystery."

Apparently, once the set of possibilities of what you're hearing (or seeing) is narrowed, your brain is much better at extracting meaning from noise.  "Your brain tries to get around the problem of too much information by making assumptions about the world," co-author Christopher Holdgraf said.  "It says, ‘I am going to restrict the many possible things I could pull out from an auditory stimulus so that I don’t have to do a lot of processing.’  By doing that, it is faster and expends less energy."

So there's another fascinating, and mind-boggling, piece of how our brains make sense of the world.  It's wonderful that evolution could shape such an amazingly adaptive device, although the survival advantage is obvious.  The faster you are at pulling a signal out of the noise, the more likely you are to make the right decisions about what it is that you're perceiving -- whether it's you talking to a friend in a crowded bar or a proto-hominid on the African savanna trying to figure out if that odd shape in the grass is a crouching lion.

****************************************

Not long ago I was discussing with a friend of mine the unfortunate tendency of North Americans and Western Europeans to judge everything based upon their own culture -- and to assume everyone else in the world sees things the same way.  (An attitude that, in my opinion, is far worse here in the United States than anywhere else, but since the majority of us here are the descendants of white Europeans, that attitude didn't come out of nowhere.)  

What that means is that people like me, who live somewhere WEIRD -- white, educated, industrialized, rich, and democratic -- automatically have blinders on.  And these blinders affect everything, up to and including things like supposedly variable-controlled psychological studies, which are usually conducted by WEIRDs on WEIRDs, and so interpret results as universal when they might well be culturally-dependent.

This is the topic of a wonderful new book by anthropologist Joseph Henrich called The WEIRDest People in the World: How the West Became Psychologically Peculiar and Particularly Prosperous.  It's a fascinating lens into a culture that has become so dominant on the world stage that many people within it staunchly believe it's quantifiably the best one -- and some act as if it's the only one.  It's an eye-opener, and will make you reconsider a lot of your baseline assumptions about what humans are and the ways we see the world -- of which science historian James Burke rightly said, "there are as many different versions of that as there are people."

[Note:  If you purchase this book using the image/link below, part of the proceeds goes to support Skeptophilia!]

Noise alert

The week after I retired I made the mistake of saying to my wife, "I don't know what I'm going to do with all of my free time!"

Two days later we found out we had to have major foundation work done on our house.  I do mean major; erosion and settling on one corner was causing the slab to twist, and if we didn't do something, we were going to have our slab -- and almost certainly our walls -- crack catastrophically.

So yeah.  Me and my big mouth.  It's times like this I have a hard time maintaining my status as Non-Superstitious Guy.

The foundation work required that we more or less gut our formerly-finished basement.  We were already planning on redoing it, just not this completely or this precipitously.  It could be a nice space -- it's got a walk-out (we're built on a hill, which is part of what caused the problem in the first place) and with some messing about it could be a den or even a rental apartment, now that we're empty nesters and it's just me and Carol in this big house.

Me and my son working on demolition.  You can probably see the amazing family resemblance between us.

In any case, this all comes up because of a paper that appeared last week in Nature Communications  about why we perceive some sounds as unpleasant (such as shop vacs, reciprocating saws, dehumidifiers, and air filters -- all of which we had going at once down there).  And it turns out that it's not just the volume (amplitude) of the sound waves.

In "The Rough Sound of Salience Enhances Aversion Through Neural Synchronisation," by Luc H. Arnal, Andreas Kleinschmidt, Laurent Spinelli, Anne-Lise Giraud, and Pierre Mégevand of the University of Geneva, we find that the degree of perceived unpleasantness of a sound has to do with repeated peaks in "fast repetitive modulations" in the sound.  Put simply, there are two kinds of frequency most sounds have: the fundamental frequency of the tone, which we perceive as its pitch; and the rise and fall of overall loudness.  And what the researchers discovered is when that second frequency is between 30 and 150 hertz, we find it really unpleasant.  (One hertz is one vibration per second; so even 30 hertz is fast enough that we're not consciously aware of it as a repetitive noise.)

Apparently sounds in that range cause our neurons to synchronize at that frequency, heightening awareness and making them difficult to ignore.  The researchers suspect that it may be an evolved response because those sorts of noises may signal danger, but that's speculation at this point.

The authors write:

Fast repetitive modulations produce “temporally salient” flickering percepts (e.g. strobe lights, vibrators, and alarm sounds), which efficiently capture attention, generally induce rough and unpleasant sensations, and elicit avoidance.  Despite the high ecological relevance of such flickering stimuli, there is to our knowledge no existing operational definition of temporal salience and only limited experimental work accounting for the intriguing aversive sensation such auditory textures produce and the reactions they trigger.  Here, we introduce and explore the notion of temporal salience and investigate its behavioural and neural underpinnings.  Of note, although salience may not systematically result in aversive percept, we argue that in this specific context, temporal salience—owing to the imperative effect of exogenously saturating perceptual systems in time—constitutes a valid proxy of aversion.  Therefore, we hypothesise that providing fast, but still discretisable and perceptible, temporally salient acoustic cues should enhance neural processing and ensuing aversive sensation.

This discovery led to some surprising connections.  "These sounds solicit the amygdala, hippocampus and insula in particular, all areas related to salience, aversion and pain.  This explains why participants experienced them as being unbearable," said Luc Arnal, who was the paper's lead author.   "This is the first time that sounds between 40 and 80 hertz have been shown to mobilise these neural networks, although the frequencies have been used for a long time in alarm systems...  We now understand at last why the brain can't ignore these sounds.  Something particular happens at these frequencies, and there are also many illnesses that show atypical brain responses to sounds at 40 Hz.  These include Alzheimer's, autism and schizophrenia."

Which is unexpected and startling.  What is happening in the brain at those frequencies -- and how does it connect with overall mental functioning?  Does schizophrenia (for example) involve some sort of "brain noise" that is at a frequency that the sufferer can't ignore?

In any case, it's a fascinating piece of research, and on a more banal level explains why I find that shop vac so damned annoying.  At least we've got the demolition done, so I won't have any more huge messes to clean up.

Unless the universe is listening and causes some catastrophic upheaval in another part of our house.  You never know.  Just because I'm not superstitious doesn't mean I can't jinx myself.

********************************

This week's Skeptophilia book recommendation is by the team of Mark Carwardine and the brilliant author of The Hitchhiker's Guide to the Galaxy, the late Douglas Adams.  Called Last Chance to See, it's about a round-the-world trip the two took to see the last populations of some of the world's most severely endangered animals, including the Rodrigues Fruit Bat, the Mountain Gorilla, the Aye-Aye, and the Komodo Dragon.  It's fascinating, entertaining, and sad, as Adams and Carwardine take an unflinching look at the devastation being wrought on the world's ecosystems by humans.

But it should be required reading for anyone interested in ecology, the environment, and the animal kingdom. Lucid, often funny, always eye-opening, Last Chance to See will give you a lens into the plight of some of the world's rarest species -- before they're gone forever.

[Note: if you purchase this book using the image/link below, part of the proceeds goes to support Skeptophilia!]

Silence is golden

Odd to say, for a veteran high school teacher, but I am not someone who thrives in a noisy, chaotic environment.

I can tolerate it for a while, and then I have a number of (often quite sudden) reactions.  First, I get a jolt of tension, sometimes manifesting as a physically painful clench in my stomach.  Second, I lose the ability to process voices individually -- something that is most striking in a crowded pub, where the voice of the person I'm talking to, sitting right next to, vanishes into a uniform, overwhelming Wall of Sound within which I can distinguish nothing at all.

Third, I feel like running away.

Turns out I'm not alone.  Some studies in the last couple of years have elucidated the restorative role of silence -- something all too few of us get to experience.  At least when I'm done at school, I can go home to my quiet house on a rural road, and immerse myself in something close to complete quiet.  But I can't imagine what it'd be like to live in a big city.

I think I'd go mad, honestly.

A 2013 study by Imke Kirste et al. of the Research Center for Regenerative Therapies in Dresden, Germany found a neurological underpinning to our desire for silence.  Baby mice exposed to various sorts of noise (music, white noise, the human voice, the vocalizations of other mice) showed, as you would expect, varying responses in terms of growth and interconnectedness of brain cells, levels of stress hormones, and so on.  What was surprising was that the group which was supposed to be the control -- a group of baby mice kept in total silence -- exhibited a statistically significant improvement in growth over any of the others in the number of new neurons in the hippocampus, a part of the brain associated with memory and spatial navigation.

Silence, by sculptor Alix Marquet (1921) [image courtesy of the Wikimedia Commons]

It doesn't just work with mice.  A study by Arline Bronzaft, published in the Journal of Environmental Psychology, found that installing noise-reduction technology in P. S. 93, a public school in the Bronx that was located very close to elevated railroad tracks, resulted in a significant uptick in student reading scores -- and, most tellingly, an equalization of scores in classes on the side of the building facing the tracks with the ones that were further away.

Then there's the study by Allyson Green et al. that appeared in the Journal of Residential and Environmental Public Health in 2015 that went even further, showing that noise was positively correlated with cortisol levels in a gold-mining community in Ghana.  Cortisol is not only a stress hormone, it's a natural anti-inflammatory; repeated long-term exposure to high blood levels of cortisol causes a receptor-weakening effect much like high sugar diet does in type-2 diabetes, with the result that inflammation all over the body increases.  Years of high cortisol levels have been associated with heart disease, ulcers, and arthritis -- so it's hard on you physically, not to mention the obvious psychological toll it takes.

What's most frightening about all of this is how much we've come simply to accept the amount of noise in our lives.  As much as I like listening to music on the radio, sometimes on my way home from work I have to switch it off -- I'm still too overwhelmed by the noise I experienced in the school to subject myself to more sounds (albeit pleasant enough ones) on my drive home.  But I'm struck then by how much noise the tires and engine make.  Even when we think we're in quiet, we seldom actually are -- and we view it as inevitable and unavoidable.

And we're so often unaware of it.  Look at how instantaneously you're aware of it when the power goes out, even in the daytime.  The sudden cessation of the noise of heaters, refrigerators, air conditioners, and so on can be as startling as a thunderclap.

It's getting worse, too.  Studies have concluded that the amount of ambient noise doubles roughly every thirty years -- outstripping population growth.  A 2015 study by Matthew Zawadski, Heather Costigan, and Joshua Smyth of Pennsylvania State University found that test subjects had a lower cortisol spike in their saliva (an indicator of stress) if they had a period of quiet and leisure prior to a high-stress activity (such as giving a speech to an unfriendly audience).  Being around turmoil and noise for extended periods of time leaves us less able to cope with the difficulties, large and small, that we face every day.

Me, I'm all about giving a try to having more silence in my life.  Maybe you'll give it a try, too -- turn off the radio, take out the earbuds, find a way to get out of the chaos of the city for a while.  Let me know what happens -- if the studies are correct, it should do you nothing but good.

Signal out of noise

I think I share with a lot of people a difficulty in deciphering what someone is saying when holding a conversation in a noisy room.  I can often pick out a few words, but understanding entire sentences is tricky.  A related phenomenon I've noticed is that if there is a song playing while there's noise going on -- in a bar, or on earphones at the gym -- I often have no idea what the song is, can't understand a single word or pick up the beat or figure out the music, until something clues me in to what the song is.  Then, all of a sudden, I find I'm able to hear it more clearly.

Some neuroscientists at the University of California - Berkeley have just found out what's happening in the brain that causes this oddity in auditory perception.  In a paper in Nature: Communications that came out earlier this week, authors Christopher R. Holdgraf, Wendy de Heer, Brian Pasley, Jochem Rieger, Nathan Crone, Jack J. Lin, Robert T. Knight, and Frédéric E. Theunissen studied how the perception of garbled speech changes when subjects are told what's being said -- and found through a technique called spectrotemporal receptive field mapping that the brain is able to retune itself in less than a second.

The authors write:

Experience shapes our perception of the world on a moment-to-moment basis.  This robust perceptual effect of experience parallels a change in the neural representation of stimulus features, though the nature of this representation and its plasticity are not well-understood.  Spectrotemporal receptive field (STRF) mapping describes the neural response to acoustic features, and has been used to study contextual effects on auditory receptive fields in animal models.  We performed a STRF plasticity analysis on electrophysiological data from recordings obtained directly from the human auditory cortex.  Here, we report rapid, automatic plasticity of the spectrotemporal response of recorded neural ensembles, driven by previous experience with acoustic and linguistic information, and with a neurophysiological effect in the sub-second range.  This plasticity reflects increased sensitivity to spectrotemporal features, enhancing the extraction of more speech-like features from a degraded stimulus and providing the physiological basis for the observed ‘perceptual enhancement’ in understanding speech.

What astonishes me about this is how quickly the brain is able to accomplish this -- although that is certainly matched by my own experience of suddenly being able to hear lyrics of a song once I recognize what's playing.  As James Anderson put it, writing about the research in ReliaWire, "The findings... confirm hypotheses that neurons in the auditory cortex that pick out aspects of sound associated with language, the components of pitch, amplitude and timing that distinguish words or smaller sound bits called phonemes, continually tune themselves to pull meaning out of a noisy environment."

A related phenomenon is visual priming, which occurs when people are presented with a seemingly meaningless pattern of dots and blotches, such as the following:

Once you're told that the image is a cow, it's easy enough to find -- and after that, impossible to unsee.

"Something is changing in the auditory cortex to emphasize anything that might be speech-like, and increasing the gain for those features, so that I actually hear that sound in the noise," said study co-author Frédéric Theunissen.  "It’s not like I am generating those words in my head.  I really have the feeling of hearing the words in the noise with this pop-out phenomenon.  It is such a mystery."

Apparently, once the set of possibilities of what you're hearing (or seeing) is narrowed, your brain is much better at extracting meaning from noise.  "Your brain tries to get around the problem of too much information by making assumptions about the world," co-author Christopher Holdgraf said.  "It says, ‘I am going to restrict the many possible things I could pull out from an auditory stimulus so that I don’t have to do a lot of processing.’ By doing that, it is faster and expends less energy."

So there's another fascinating, and mind-boggling, piece of how our brains make sense of the world.  It's wonderful that evolution could shape such an amazingly adaptive device, although the survival advantage is obvious.  The faster you are at pulling a signal out of the noise, the more likely you are to make the right decisions about what it is that you're perceiving -- whether it's you talking to a friend in a crowded bar or a proto-hominid on the African savanna trying to figure out if that odd shape in the grass is a crouching lion.

The source of the noise

A common theme in Skeptophilia is that people in general need to learn some scientific terminology.

Not only is science cool, and thus learning about it a worthy goal in and of itself; but knowing how science works, and some of the field's vocabulary, will keep you from being duped.  As we've seen over and over, the world is full of folks who either through ignorance or outright duplicity misrepresent what scientists are doing -- and without adequate mental firepower, you're gonna fall for their nonsense every time.

As an example, this weekend, a story started popping up all over that claimed that we'd finally received a good candidate for an alien signal from an extrasolar planet.  This immediately caught my eye -- it is one of my dearest wishes that we have incontrovertible evidence of alien intelligence before I die.  I'd be perfectly satisfied if it comes in the form of some kind of radio signal, but if an actual alien spacecraft landed in my back yard, that would also be acceptable, at least until they started vaporizing my dogs with laser pistols.

So my reaction was one of cautious enthusiasm.  Cautious, because I suspected that if this had actually happened, it would be all over the news, not just surfacing in the form of links on Twitter and Facebook.  But the stories all made the same claim.  Here's an excerpt from the version that appeared on UFO Blogger:

Astronomers have picked up a mystery "noise" which they believe could be coming from an Earth-like planet in the outer space [sic]. After analyzing the strange signals emitting from the object, scientiscientistssts [sic] are certain that a habitable planet exists some 22 light years away, a report said. 

In 2010, scientists had dismissed the mystery noise or signals as stellar bursts but after the latest research it was clear that an Earth-like planet, or Gliese 581d, has conditions which could support life, and is likely to be a rocky world, twice the size of Earth.

Okay, given the typos and grammar, it's not exactly the most credible of reports.  But all of the links I saw agreed; an extrasolar planet called Gliese 581d, 22 light years from Earth, had been reported as the source of a strange, unexplained noise.  The planet was bigger than Earth, but in the "Goldilocks Zone" -- the "just right" region around its star where water would be in liquid form, and therefore a place where life something like what we have here could evolve.

So immediately I started picturing Star Trek-style aliens, complete with fake rubber alien noses and bad accents.  Then I thought of the amazing final scene in Star Trek: First Contact, which is clearly the best movie the franchise ever produced, wherein Zefram Cochrane shakes hands with a Vulcan for the first time.  And before you know it, I had myself worked up into a lather about the possibilities.

Then I thought, "Calm down, dude.  Verify your sources."  So I typed "noise Gliese 581d" into a Google search to see if I could find out where this information had come from.  Clearly it was all similar enough that it had some kind of common origin, and it wasn't wacky enough to have come from The Weekly World News.  And after five minutes' search, I found the press release in Phys.org that had caused the stir -- the origin of the "noise."

And I put "noise" in quotation marks for a good reason, as you'll see.  The press release was a blurb summarizing a paper in Science by Anglada-Escudé and Tuomi called "Stellar activity masquerading as planets in the habitable zone of the M dwarf Gliese 581."  Here's the relevant passage:

A report published in Science has dismissed claims made last year that the first super-Earth planet discovered in the habitable zone of a distant star was 'stellar activity masquerading as planets.' The researchers are confident the planet named GJ 581d, identified in 2009 orbiting the star Gliese 581, does exist, and that last year's claim was triggered by inadequate analysis of the data.

The planet candidate was spotted using a spectrometer which measures the 'wobble', small changes in the wavelength of light emitted by a star, caused as a planet orbits it. In 2014 researchers revisiting the data said that the 'planet' was actually just noise in the data caused by starspots. The possible existence of the planet was widely dismissed without further questioning. 

But now researchers from Queen Mary University of London (QMUL) and University of Hertfordshire have questioned the methods used to challenge the planet's existence. The statistical technique used in the 2014 research to account for stellar activity is simply inadequate for identifying small planets like GJ 581d.

Note the use of the word "noise" twice.  This, apparently, is the source of the story of an alien noise coming from Gliese 581d.

Scientists use the word "noise" differently from the rest of us.  To a scientist, "noise" is scatter in the data, background junk, that might obscure something real and measurable (the "signal").  If you have enough noise, the signal becomes impossible to detect, so reducing the noise in a data set is critical.  A high "signal-to-noise ratio" is what you're after; lots of signal, little noise.  So when the astrophysicists re-analyzed the data from 2009 that had been rejected last year as supporting an Earth-like extrasolar planet around Gliese 581, they found a way of reducing the noise in the data, and were able to confirm that the planet did, actually, exist.

What they did not find was some kind of unexplained noise coming from Gliese 581d.  The noise was the scatter in the data, not some Klingon sending an insulting message at the Earth such as "Hab SoSlI' Quch" ("Your mother has a smooth forehead").

Not that I'm happy to report this, mind you.  No one would be more thrilled than me if we had received an alien communiqué, even if it contained an insult.  But unfortunately, the stories about mysterious and unexplained alien noises turn out to be unmysterious and completely explainable ignorance of bloggers regarding the use of scientific vocabulary.

So we keep waiting.  Given the number of extrasolar planets the astronomers are discovering, I'm still optimistic that one day, we'll find the aliens.  Or maybe they'll find us.  Either way, it'd be amazing, because, after all, Heghlu'meH QaQ jajvam ("Today is a good day to die").