40,000 fMRI studies to throw away!

A research team recently revealed what they called a Cluster failure.

Most functional MRI studies conclusions may be erroneous because of their statistical basis. Widely used clustering inference techniques are actually pretty bad at properly inferring clusters due to “spatial autocorrelation functions that do not follow the assumed Gaussian shape”.

It means that “most common software packages for fMRI analysis (SPM, FSL, AFNI) can result in false-positive rates of up to 70%”… suggesting that some results were so inaccurate they could indicate brain activity that does not exist at all!. What is interesting is that neuroscientists are interpreting what they’re told by the statistical software rather than images.

These findings speak to the need of validating the statistical methods being used in the field of neuroimaging.

Therefore, 15 years of research on brain functioning could be invalidated!

But the issue is not limited to research but also extends to clinical use that is pretty worrying.

 

Image credits: fMRI by OpenStax from the Textbook OpenStax Anatomy and Physiology Published May 18, 2016




Your brain does process information but it is not a computer

I recently came across an article pretending that your brain “does not process information, retrieve knowledge or store memories” and cannot be viewed as a computer.

I personally think that this assertion is fundamentally wrong regarding the information processing.

The computer metaphor is still a valid one

First, I’d like to show you that the computer metaphor couldn’t be discarded so easily.

Of course our brain is not a computer. It is embodied and cannot be considered as an autonomous system.

Exercise Plays Vital Role Maintaining Brain Health, by a health blog

Credits: Exercise Plays Vital Role Maintaining Brain Health, by a health blog

An embodied system…

“Many features of cognition are embodied in that they are deeply dependent upon characteristics of the physical body of an agent” – RA Wilson and L Foglia, Embodied Cognition in The Stanford Encyclopedia of Philosophy

A downloaded brain, if it were possible, would probably not be able to function without a body having human-like sensors and effectors… since a lot of brain activity is dedicated to monitoring sensory inputs, regulating and interacting with the environment.

A system which is itself made up of of multiple systems

Rather than, the brain should be considered as multiples systems that interact together as a whole. Each of them is specialized and is able to communicate with others through different means. The basic elements of those systems are able to transmit signals at different speed, through different pathways: electrical signaling, chemical signaling (neurohormone)… even by using important diffuse systems (glia and immune system).

A limited metaphor…

To put in a nutshell,

“Humans, along with other organisms with brains, differ from computers because they are driven by emotions and motivations. The brain is much too hot and wet to be represented by a computer. The brain is electrical, but it is also driven by fluids (blood) and chemicals (hormones and neurotransmitters). Most importantly, the brain is part of a body which it drives to action, and research from an embodiment perspective also shows that the whole body (not just the brain) affects emotion, motivation, and other psychological processes.”

So what is the adequate metaphor?

Eddie Harmon-Jones, Ph.D., Professor of Psychology at Texas A&M University suggested that

“So, how can we replace the computer metaphor with a metaphor that more accurately represents the brain of an emotion-driven, motivated organism such as a human? I like the metaphor of a car.

A car may have a computer on board, and may be able to process information. But it is driven by fluids (gasoline, oil, etc.). It is both electrical and mechanical, and it can move.”

Renault Scénic Front Cut by Sovxx

Credits: Renault Scénic Front Cut by  Sovxx

Despite the fact that computers can be so complex using most recent advances (for instance VLSI or, simply supercomputers), the car metaphor is far more comprehensible.

We should also consider that computers have no significance if not considered through human perspective. On the contrary, all beings those have a brain exists by themselves.

But a useful metaphor

The metaphor of a brain as a computer helped scientists to gain a better understanding of our brain functioning. No more, no less. Just like both past metaphors of the brain and current ones.

Cognitivism (top down approach), connectionism (bottom up approach) and embodied cognition all succeed at explaining or modeling some aspects of our cognition. In neuroscience, several approaches contribute to explain the neural substrate functioning.

The working memory model , as well as the second revolution in linguistics, etc. were all useful for gaining a better knowledge. A knowledge that is helpful to evaluate individual’s cognitive aspects, model functions and, somehow, understand how the brain works.

These theories are still useful for natural language processing, cognitive remediation, and so on. We haven’t discarded Newtonian theory when Einstein published his general relativity. So is the computer metaphor.

The metaphor is helpful to explore the brain and to exploit its properties but it is not necessarily the ultimate metaphor. Nobody knows what the future has in store…

The brain process information

The brain is not able to recall a detailed picture you saw thousands times (a bank note for instance). That does not mean that it doesn’t store information.

As far as we know the brain stores information using synaptic plasticity (1, 2), i.e. connectivity changes, and brain network topology. The information is scattered through the brain and can be unlearned or mixed with new pieces of information.

Neurons network

What we exactly store is still on debate but it is synthetized and segmented information.

Actually, our brain does process information:

“Information is what is conveyed or represented by a particular arrangement or sequence of things” – Oxford dictionary

Which is exactly what the brain does.

To go further in this explanation:

“Information is an abstract concept. There is a temptation to think of information as fully representable by the bits used in a digital computer. However with signals, the timing of the signal makes a difference. When precisely a signal arrives at a neuron carries information about what the signal means, and there is increasing evidence that the relative timing among neural signals carries information as well. The brain accomplishes information processing using signal processing, but there is more going on than the phrase “information processing” alone would suggest. […]All signal processing is carried out by the spontaneous “random” interactions of molecular collisions, which are loosely guided by the continuously hierarchical structural form of the brain.” – Paul King, Computational Neuroscientist, Data Scientist, Technology Entrepreneur

What is your opinion on the question?

 

 

 




Do we live in a discrete world?

Today, I will share with you some recent thoughts I had.

Do you remember of analog chips (a.k.a linear integrated circuits)? These circuits are almost no longer in use and have been replaced by semi-conductors and discrete processing chips (digital ones). Some kind of analog signal processing still remains in use in vacuum tubes and may be soon integrated in some current devices thanks to recent improvements.

To sum up, we have the following kind of computers: analog, digital, quantum.

The ubiquitous type is digital computer (even your smartwatch is basically a computer). Analog almost disappeared (they are still in use in aircrafts) but may become more prominent due to interest in very-large-scale integration. Quantum ones are more theoretical than common.

I already mentioned that our universe behaviour is either mathematical or a quantum, perhaps both.

Nick Boström even suggested that we are most likely living in a computer simulation.  Some go further, hypothesizing in a thought experiment, that some of us may be p-zombies.

digital

The real nature of our universe is a prolific question in physics, philosophy, metaphysics, spirituality and even cognitive science since we extensively rely on our sensors and the perception of a physical world (a perception that is easily tricked, even at integration levels).

If the world is quantum, then it can be reduced to a quantum computational system. If it is analogous, it can be reduced to an analogous computational system. Both are not exclusive at all.

What if our universe were discrete?

The discrete option is really interesting in that it supports the metaphor of a computer and, ultimately, of a simulation. It is theorized in digital physics.

One can notice that if the universe processes information, then it can also generate and process knowledge. We know that information and knowledge spread in societies by well-known social processes and networks dynamics. Oriented and labelled graphs can organize knowledge just like ontologies in information science. I found the analogy quite interesting since both societies and ontology-based systems in computer science can generate knowledge.

Graph

In my humble opinion, it tends to extend the computer metaphor from the universe to societies: the same process underpinning the quantum level, the cognitive level and the social level. It explains why discrete graph models succeed to explain some levels’ information processing (1, 2). Information processing would be the essential nature of the universe and knowledge discovery (perhaps?) a goal for us. Enaction and embodiment taught us that this appropriation of knowledge is not necessarily academic but can be achieved from everyday life or manual work.

It also reminds me the Plato’s theory of Forms (and its limitations) and how to access the ultimate reality.

“We come here to a difficulty which has troubled many philosophic theologians. Only the contingent world, the world in space and time, can have been created; but this is the every-day world which has been condemned as illusory and also bad. Therefore the Creator, it would seem, created only illusion and evil. Some Gnostics were so consistent as to adopt this view; but in Plato the difficulty is still below the surface, and he seems, in the Republic, to have never become aware of it.”- Bertrand Russel, philosopher (sorry for the mention of a Creator, you can replace it by every concept that suits your beliefs).

social interaction

What is the intermediate level: human being. As you may already know, fractals are everywhere in our universe. The information processing is everywhere. So, it sounds interesting to explore the possibility that the same processes can be observed at all levels: from atoms to societies.

The point is that if our universe is discrete, then each level is discrete too.

Societies are discrete in the sense that information processing and knowledge spreading occurs temporarily by discrete steps. The processing speed is increased in our connected society.

Our brain is also discrete (a counterintuitive idea): from post synaptic potential triggering to time slices of perception as we recently discovered.

What about the quantum level?

Humanity might never be able to prove with certainty whether the universe is simulated, Chalmers said.

“You’re not going to get proof that we’re not in a simulation, because any evidence that we get could be simulated.” – David Chalmers, Professor of Philosophy and Director of the Centre for Consciousness at the Australian National University

But we have evidences for a simulation since universe has probably error correcting codes, just like in computer science.

“Error-correcting codes are what make browsers work, so why were they in the equations that I was studying about quarks, and leptons, and supersymmetry? […] That’s what brought me to this very stark realization that I could no longer say that people like Max [Tegmark] are crazy.” – James Gates, a physicist at the University of Maryland

We may also be able to prove that our universe is discrete by nature. The question is, ultimately: is there a smallest unit of length, beyond which you can’t divide any further?

It will probably be possible to confirm soon Giovanni Amelino-Camelia observations of Hubble’s quasar shift in high frequencies.

Pi

We should even not be required to deeply observe the universe since discrete and continuous may be two sides of the same!

“The most significant level of interaction is when one and the same phenomenon appears
in both the continuous and discrete setting. In such cases, intuition and insight gained from
considering one of these may be extremely useful in the other.” – László Lovász, Microsoft Research
If all the phenomenons we observe can be modelled (and perfectly predicted) in both discrete and continuous mathematics then we are justified to ask whether the universe is discrete rather than continuous (due to the low possibility that discrete succeed at modelling every essential phenomenons).



One of the best chips is bad at maths

I will share with you a promising innovation in chips. A MIT project that was funded by DARPA resulted in a chip …

capable of processing frames almost 100 times faster than a conventional processor restricted to doing correct math—while using less than 2 percent as much power

How?

chipx9991

The chip was actually doing imprecise summations but it was enough to perform very well on some hard tasks which do not required precise calculations. That is exactly what one does not expect from a chip’s functioning!

You can find more detail in this MIT review.




Do plants compute?

Plants are no longer considered as insensitive and passive life forms. We even found novel means they communicate. What is more interesting is that several recent news tend to suggest some ability to count in plants.

The case of Venus flytraps

venus flytrap

Venus flytrap by Jason

In this article, Jennifer Böhm et al. suggest that a carnivorous plant (Venus Flytrap Dionaea muscipula) is able to count the number of mechanical stimuli that trigger the production of digesting enzymes and sodium uptake modules.

The team asked the question as to “how many times trigger hairs (e.g. mechanosensors) have to be stimulated (e.g., how many APs are required) for the flytrap to recognize an encaged object as potential food, thus making it worthwhile activating the glands”.

What they found is the following:

One touch to a trigger hair sets the plant in a “ready” state. A second touch causes the trap to close around the prey. At five touches the plant begins to produce digestive enzymes and transporter molecules that take up nutrients.

It sounds like an algebraic summation that triggers different pathways.

Not yet impressed? Let me show you a more interesting computation in plants…

A more complex calculation…

Arabidopsis thaliana

A. thaliana by Alberto Salguero Quiles

Several years ago a team from the John Innes Centre found that Arabidopsis thaliana is able to perform some computations. In fact, this plant is able to have a perfect 5% of starch level after one night. Even when the night’s duration was artificially increased or decreased.

Actually, this well-known (but still unknown, in a way) plant can do arithmetic division to prevent starvation at night. The plant determines the remaining time before day then divide the starch level by this value to get the adequate velocity for consuming starch!

A novel perspective

Of course these calculations may not be conscious and are probably caused by a chemical imbalance. However, it opens really promising perspectives on the complexity of plants those are far more complex than initially thought.




Boundless consciousness?

I’m back with my second post of 2016 regarding consciousness. Hope it will be delightful.

Intelligences are multiple

Robot

Consciousness is not yet fully defined. As well as intelligence was until recently. The boundaries of intelligence were fluctuating across history and constantly redefined to ensure human superiority over other species. The tools to assess intelligence are often designed from our own abilities therefore fail at measuring other species intelligence.

However, we all now agree on various form of animal intelligences even in the human species. It was a long road before it became a consensus, harder than believe in multiple personalities traits which is actually intrinsically related.

What about consciousness?

Consciousness is hard to define due to lack of consensus. The definition I prefer is sentience, i.e. feeling as distinguished from perception or thought. One could also add self-awareness to the definition.

So what beings are conscious?

Coral_polyps_in_symbiosis_with_unicellular_dinoflagellates

Basically all animals (including humans) are sentient and sometimes self-aware. Plants may also be conscious beings. Although that they don’t have neurons or glia, they possess structures that allow them to process information and react to environment or even plan actions.

“They have ways of taking all the sensory data they gather in their everyday lives … integrate it and then behave in an appropriate way in response. And they do this without brains, which, in a way, is what’s incredible about it, because we automatically assume you need a brain to process information.”; “that the line between plants and animals might be a little softer than we traditionally think of it as.” – Michael Pollan

More interestingly, we may have created artificial beings that are conscious. The neuroscientist Christof Koch speculated that the Web might have achieved sentience. Pretty freaky, right?

I personally consider that at least cluster of networked machines (with their software) or intelligent machines may be sentient. Especially ones that auto-monitor. Some people argue that any system that is unpredictable may actually be sentient.

The fact is that such system may fail at the Turing test because they are so different than us and their consciousness is so different than what we expect. Their sensory modalities may even not map what we see as inputs and outputs.

Since they process information and act, even particles may be conscious, by flashes. Hard to believe…

Are you talking about panpsychism?

laniakea

Most of living scientists consider that considering non-human beings as conscious is panpsychism thus discarding it as pre-science. Panpsychism is the view that consciousness, mind or soul (psyche) is a universal and primordial feature of all things (to some degree).

This century marked a renewed interest in these hypothesis. Max Tegmark (MIT) argues that mater may be intrinsically conscious and a more restrictive scale have been invented to measure levels of consciousness of artificial agents.

An interesting conclusion

Depending of what frame of reference we consider, we may:

  • Be surrounded by above suspicion conscious beings: from artificial intelligences (not that terminator-like we all think about those are maybe not conscious of us at all) and even superintelligences (if they exist) to inert chromosomes.
  • Be living in a vast conscious system with a lot of subconscious entities.

The second option may be plausible since supercluster processes a lot of information as a complex system.

But the light speed limit tends to suggest a long scaled consciousness (if it exists) or consciousness at early stages of structures self-organisation only… although quantum entanglement or shared routines in a computational universe may also allow long-distance rapid information processing.




An affordable credit card-sized supercomputer by NVIDIA

Jetson TX1

NVIDIA announced yesterday the Jetson TX1, a small form-factor Linux system-on-module, credit card sized for various application ranging from autonomous navigation to deep learning-driven inference and analytics.

Jetson TX1

It will soon be available as development kit, e.g. a mini-ITX carrier board that includes the pre-mounted module and has low power consumption which provides an out of the box desktop user experience (it comes with a linux’s ubuntu custom distribution). Unfortunately, the development kit requires a USB hub to work with a keyboard and a mouse and the 16GB eMMC memory storage is probably too few.

Since I really enjoyed performing artificial intelligence at the university and during an experience as contractor in a public research center, I think I will ask the developer kit for christmas. I plan to use it as media center, intelligent home automation and for personal deep learning projets.

You may wonder why I chose this solution? Just because this card packs several interesting characteristics:

  • a Tegra X1 SoC : an ARM A57 CPU and a Maxwell-based GPU packing 256 CUDA cores (delivering 1 teraflop at 5.7W, i.e. the same peak speed as a small 15 years old supercomputer!)
  • 4GB of RAM shared between the CPU and GPU

It sounds interesting to me.




Deep learning for everyone!

TensorFlow

That’s great news! Google just open-sourced TensorFlow, its deep (machine) learning library.

The engine is widely used at Google: by speech recognition systems, in the new Google photo product, in Gmail, in search, etc.

TensorFlow

From now on startups will be able to develop systems as intelligent as a 4 year old children. More interestingly, code sharing in python between researchers or data scientists has never been easier.

The limitations of the previous system no longer exist:

[DistBelief] was narrowly targeted to [artificial] neural networks, it was difficult to configure, and it was tightly coupled to Google’s internal infrastructure — making it nearly impossible to share research code externally. […] TensorFlow has extensive built-in support for deep learning, but is far more general than that — any computation that you can express as a computational flow graph, you can compute with TensorFlow (see some examples). Any gradient-based machine learning algorithm will benefit from TensorFlow’s auto-differentiation and suite of first-rate optimizers. And it’s easy to express your new ideas in TensorFlow via the flexible Python interface.

Maybe the engine will soon get available for its cloud-based service on a clustered architecture…