by Christof Koch
December 19, 2013
Christof Koch is chief scientific officer at the Allen institute for Brain Science in Seattle. He serves on Scientific American Mind's board of advisers.
that consciousness is
universal,
offers some lessons in how
to think
about subjective
experience today
For every inside there is an outside, and for every
outside there is an inside; though they are
different, they go together.
- Alan
Watts
Man, Nature, and the
Nature of Man, 1991
I grew up in a devout and practicing Roman Catholic family with
Purzel, a fearless and high-energy dachshund.
He, as with all the other, much larger
dogs that subsequently accompanied me through life, showed plenty of
affection, curiosity, playfulness, aggression, anger, shame and
fear. Yet my church teaches that whereas animals, as God's
creatures, ought to be treated well, they do not possess an immortal
soul. Only humans do.
Even as a child, to me this belief felt
intuitively wrong. These gorgeous creatures had feelings, just like
I did.
Why deny them? Why would God
resurrect people but not dogs?
This core Christian belief in human
exceptionalism
did not make any sense to me. Whatever consciousness and
mind are and no matter how they relate to the brain and the rest of
the body, I felt that the same principle must hold for people and
dogs and, by extension, for other animals as well.
It was only later, at university, that I became acquainted with
Buddhism and its emphasis on the universal nature of mind.
Indeed, when I spent a week with His
Holiness the Dalai Lama earlier in 2013 [see "The
Brain of Buddha," Consciousness Redux; Scientific
American Mind, July/August 2013], I noted how often he talked about
the need to reduce the suffering of "all living beings" and not just
"all people."
My readings in philosophy brought me to
panpsychism, the view that mind (psyche) is found everywhere (pan).
Panpsychism is one of the oldest of all
philosophical doctrines extant and was put forth by the ancient
Greeks, in particular Thales of Miletus and Plato.
Philosopher Baruch Spinoza and
mathematician and universal genius Gottfried Wilhelm Leibniz,
who laid down the intellectual foundations for the Age of
Enlightenment, argued for panpsychism, as did philosopher
Arthur Schopenhauer, father of American psychology William
James, and Jesuit paleontologist Teilhard de Chardin.
It declined in popularity with the rise
of positivism in the 20th century.
As a natural scientist, I find a version of panpsychism modified for
the 21st century to be the single most elegant and
parsimonious explanation for the universe I find myself in.
There are three broad reasons why
panpsychism is appealing to the modern mind.
We Are All
Nature's Children
The past two centuries of scientific progress have made it difficult
to sustain a belief in human exceptionalism.
Consider my Bernese mountain dog, Ruby, when she yelps, whines,
gnaws at her paw, limps and then comes to me, seeking aid: I infer
that she is in pain because under similar conditions I behave in
similar ways (sans gnawing).
Physiological measures of pain confirm
this inference - injured dogs, just like people, experience an
elevated heart rate and blood pressure and release stress hormones
into their bloodstream.
I'm not saying that a dog's pain is
exactly like human pain, but dogs - as well as other animals - not
only react to noxious stimuli but also consciously experience pain.
All species - bees, octopuses, ravens, crows, magpies, parrots,
tuna, mice, whales, dogs, cats and monkeys - are capable of
sophisticated, learned, nonstereotyped behaviors that would be
associated with consciousness if a human were to carry out such
actions. Precursors of behaviors thought to be unique to people are
found in many species.
For instance,
bees are capable
of recognizing specific faces from photographs, can
communicate the location and quality of food sources to their
sisters via the waggle dance, and can navigate complex mazes with
the help of cues they store in short-term memory (for instance,
"after arriving at a fork, take the exit marked by the color at the
entrance").
Bees can fly several kilometers and
return to their hive, a remarkable navigational performance. And a
scent blown into the hive can trigger a return to the site where the
bees previously encountered this odor.
This type of associative memory was
famously described by Marcel Proust in
À la Recherche du Temps
Perdu. Other animals can recognize themselves, know
when their conspecifics observe them, and can lie and cheat.
Some people point to language and the associated benefits as being
the unique defining feature of consciousness. Conveniently, this
viewpoint rules out all but one species, Homo sapiens (which has an
ineradicable desire to come out on top), as having sentience.
Yet there is little reason to deny
consciousness to animals, preverbal infants [see "The
Conscious Infant," Consciousness Redux; Scientific
American Mind, September/October 2013] or patients with severe
aphasia, all of whom are mute.
None other than
Charles Darwin, in the last
book he published, in the year preceding his death, set out to learn
how far earthworms "acted consciously and how much mental power they
displayed."
Studying their feeding and sexual
behaviors for several decades - Darwin was after all a naturalist
with uncanny powers of observation - he concluded that there was no
absolute threshold between lower and higher animals, including
humans, that assigned higher mental powers to one but not to the
other.
The nervous systems of all these creatures are highly complex.
Their constitutive proteins, genes,
synapses, cells and neuronal circuits are as sophisticated,
variegated and specialized as anything seen in the human brain. It
is difficult to find anything exceptional about the human brain.
Even its size is not so special, because elephants, dolphins and
whales have bigger brains.
Only an expert neuroanatomist, armed
with a microscope, can tell a grain-size piece of cortex of a mouse
from that of a monkey or a human. Biologists emphasize this
structural and behavioral continuity by distinguishing between
nonhuman and human animals. We are all nature's children.
Given the lack of a clear and compelling
Rubicon separating simple
from complex animals and simple from complex behaviors, the belief
that only humans are capable of experiencing anything consciously
seems preposterous.
A much more reasonable assumption is
that until proved otherwise, many, if not all, multicellular
organisms experience pain and pleasure and can see and hear the
sights and sounds of life.
For brains that are smaller and less
complex, the creatures' conscious experience is very likely to be
less nuanced, less differentiated and more elemental.
Even a worm has perhaps the vaguest
sense of being alive. Of course, each species has its own unique
sensorium, matched to its ecological niche. Not every creature
has ears to hear and eyes to see.
Yet all are capable of having at least
some subjective feelings.
The Austere
Appeal of Panpsychism
Taken literally, panpsychism is the belief that everything is "enminded."
All of it. Whether it is a brain, a tree, a rock or an electron.
Everything that is physical also
possesses an interior mental aspect. One is objective - accessible
to everybody - and the other phenomenal - accessible only to the
subject.
That is the sense of the quotation by
British-born Buddhist scholar
Alan Watts with which I began
this essay.
I will defend a narrowed, more nuanced view:
namely that any complex system, as
defined below, has the basic attributes of mind and has a
minimal amount of consciousness in the sense that it feels like
something to be that system.
If the system falls apart, consciousness
ceases to be; it doesn't feel like anything to be a broken system.
And the more complex the system, the larger the repertoire of
conscious states it can experience.
My subjective experience (and yours, too, presumably), the Cartesian
"I think, therefore I am," is an undeniable certainty, one strong
enough to hold the weight of philosophy. But from whence does this
experience come? Materialists invoke something they call
emergentism to explain how
consciousness can be absent in simple nervous systems and emerge as
their complexity increases.
Consider the wetness of water, its
ability to maintain contact with surfaces. It is a consequence of
intermolecular interactions, notably hydrogen bonding among nearby
water molecules.
One or two molecules of H2O
are not wet, but put
gazillions together at the right temperature
and pressure, and wetness emerges. Or see how the laws of heredity
emerge from the molecular properties of DNA, RNA and proteins.
By the same process, mind is supposed to
arise out of sufficiently complex brains.
Yet the mental is too radically different for it to arise gradually
from the physical. This emergence of subjective feelings from
physical stuff appears inconceivable and is at odds with a basic
precept of physical thinking, the Ur-conservation law -
ex nihilo nihil fit.
So if there is nothing there in the
first place, adding a little bit more won't make something. If a
small brain won't be able to feel pain, why should a large brain be
able to feel the god-awfulness of a throbbing toothache? Why should
adding some neurons give rise to this ineffable feeling?
The phenomenal hails from a kingdom
other than the physical and is subject to different laws. I see no
way for the divide between unconscious and conscious states to be
bridged by bigger brains or more complex neurons.
A more principled solution is to assume that consciousness is a
basic feature of certain types of so-called complex systems (defined
in some universal, mathematical manner). And that complex systems
have sensation, whereas simple systems have none. This reasoning is
analogous to the arguments made by savants studying electrical
charge in the 18th century.
Charge is not an emergent property of
living things, as originally thought when electricity was discovered
in the twitching muscles of frogs.
There are no uncharged particles that in
the aggregate produce an electrical charge. Elementary particles
either have some charge, or they have none. Thus, an electron has
one negative charge, a proton has one positive charge and a photon,
the carrier of light, has zero charge. As far as chemistry and
biology are concerned, charge is an intrinsic property of these
particles.
Electrical charge does not emerge from
non-charged matter.
It is the same, goes the logic, with
consciousness. Consciousness comes with organized chunks of matter.
It is immanent in the organization of the system. It is a property
of complex entities and cannot be further reduced to the action of
more elementary properties. We have reached the ground floor of
reductionism.
Yet, as traditionally conceived, panpsychism suffers from two major
flaws. One is known as the problem of aggregates.
Philosopher John Searle of the
University of California, Berkeley, expressed it recently:
"Consciousness cannot spread over
the universe like a thin veneer of jam; there has to be a point
where my consciousness ends and yours begins."
Indeed, if consciousness is everywhere,
why should it not animate the iPhone, the Internet or the United
States of America? Furthermore, panpsychism does not explain why a
healthy brain is conscious, whereas the same brain, placed inside a
blender and reduced to goo, would not be.
That is, it does not explain how
aggregates combine to produce specific conscious experience.
Integrated
Panpsychism
These century-old arguments bring me to the conceptual framework of
the
integrated information theory (IIT)
of psychiatrist and neuroscientist Giulio Tononi of the
University of Wisconsin - Madison.
It postulates that conscious experience
is a fundamental aspect of reality and is identical to a particular
type of information - integrated information. Consciousness depends
on a physical substrate but is not reducible to it.
That is, my experience of seeing an
aquamarine blue is inexorably linked to my brain but is different
from my brain.
Any system that possesses some nonzero amount of integrated
information experiences something. Let me repeat: any system that
has even one bit of integrated information has a very minute
conscious experience.
IIT makes two principled assumptions.
-
First, conscious states are
highly differentiated; they are informationally very rich.
You can be conscious of an uncountable number of things.
Think of all the frames from all the movies that you have
ever seen or that have ever been filmed or that will be
filmed! Each frame, each view, is a specific conscious
percept.
-
Second, each such experience is
highly integrated. You cannot force yourself to see the
world in black and white; its color is an integrated part of
your view. Whatever information you are conscious of is
wholly and completely presented to your mind; it cannot be
subdivided. Underlying this unity of consciousness is a
multitude of causal interactions among the relevant parts of
your brain. If parts of the brain become fragmented and
balkanized, as occurs in deep sleep or in anesthesia,
consciousness fades.
To be conscious, then, you need to be a
single, integrated entity with a large repertoire of highly
differentiated states.
Even if the hard disk on my laptop
exceeds in capacity my lifetime memories, none of its information is
integrated. The family photos on my Mac are not linked to one
another.
The computer does not know that the boy
in those pictures is my son as he matures from a toddler to an
awkward teenager and then a graceful adult. To my computer, all
information is equally meaningless, just a vast, random tapestry of
0s and 1s. Yet I derive meaning from these images because my
memories are heavily cross-linked. And the more interconnected, the
more meaningful they become.
These ideas can be precisely expressed in the language of
mathematics using notions from information theory such as entropy.
Given a particular brain, with its neurons in a particular state -
these neurons are firing while those ones are quiet - one can
precisely compute the extent to which this network is integrated.
From this calculation, the theory
derives a single number, Φ (pronounced "fi") [see "A
Theory of Consciousness," Consciousness Redux; Scientific
American Mind, July/August 2009].
Measured in bits, Φ denotes the size of
the conscious repertoire associated with the network of causally
interacting parts being in one particular state.
Think of Φ as the synergy of the system.
The more integrated the system is, the more synergy it has and the
more conscious it is. If individual brain regions are too isolated
from one another or are interconnected at random, Φ will be low. If
the organism has many neurons and is richly endowed with synaptic
connections, Φ will be high.
Basically, Φ captures the quantity of
consciousness. The quality of any one experience - the way in which
red feels different from blue and a color is perceived differently
from a tone - is conveyed by the informational geometry associated
with Φ.
The theory assigns to any one brain
state a shape, a crystal, in a fantastically high-dimensional
qualia space. This crystal is the
system viewed from within. It is the voice in the head, the light
inside the skull. It is everything you will ever know of the world.
It is your only reality. It is the quiddity of experience.
The dream of the lotus eater, the
mindfulness of the meditating monk and the agony of the cancer
patient all feel the way they do because of the shape of the
distinct crystals in a space of a trillion dimensions - truly a
beatific vision.
The water of integrated
information (IIT) is turned into the wine of experience.
Integrated information makes very specific predictions about which
brain circuits are involved in consciousness and which ones are
peripheral players (even though they might contain many more
neurons, their anatomical wiring differs).
The theory has most recently been used
to build a consciousness meter to assess, in a quantitative manner,
the extent to which anesthetized subjects or severely brain-injured
patients, such as Terri Schiavo, who died in Florida in 2005,
are truly not conscious or do have some conscious experiences but
are unable to signal their pain and discomfort to their loved ones
[see "A
Consciousness Meter," Consciousness Redux; Scientific
American Mind, March/April 2013].
IIT addresses the problem of aggregates by postulating that only
"local maxima" of integrated information exist (over elements and
spatial and temporal scales):
my consciousness, your
consciousness, but nothing in between.
That is, every person living in the U.S.
is, self by self, conscious, but there is no superordinate
consciousness of the U.S. population as a whole.
Unlike classical panpsychism, not all physical objects have a Φ that
is different from zero. Only integrated systems do. A bunch of
disconnected neurons in a dish, a heap of sand, a galaxy of stars or
a black hole - none of them are integrated. They have no
consciousness. They do not have mental properties.
Last, IIT does not discriminate between squishy brains inside skulls
and silicon circuits encased in titanium.
Provided that the causal relations among
the circuit elements, transistors and other logic gates give rise to
integrated information, the system will feel like something.
Consider humankind's largest and most complex artifact, the
Internet. It consists of billions of computers linked together using
optical fibers and copper cables that rapidly instantiate specific
connections using ultrafast communication protocols.
Each of these processors in turn is made
out of a few billion transistors.
Taken as a whole, the Internet has
perhaps 1019 transistors, about the number of
synapses in the brains of 10,000 people. Thus, its sheer number of
components exceeds that of any one human brain. Whether or not the
Internet today feels like something to itself is completely
speculative. Still, it is certainly conceivable.
When I talk and write about panpsychism, I often encounter blank
stares of incomprehension.
Such a belief violates people's strongly
held intuition that sentience is something only humans and a few
closely related species possess. Yet our intuition also fails when
we are first told as kids that a whale is not a fish but a mammal or
that people on the other side of the planet do not fall off because
they are upside down.
Panpsychism is an elegant explanation
for the most basic of all brute facts I encounter every morning on
awakening: there is subjective experience. Tononi's theory offers a
scientific, constructive, predictive and mathematically precise form
of panpsychism for the 21st century.
It is a gigantic step in the final
resolution of the ancient mind-body problem.
