While place and grid cells have been identified in the brain, we still have no idea how those functions are computed (people in my group and many others are working on this problem). We don't yet know how these representations are combined with our sensory experiences to form episodic memories (again, there are hypotheses, but no standard theory exists). There's no question that O'Keefe and the Mosers deserve the prize, but their work literally represents the mere beginning of this line of research.

The human brain is a wonder of engineering. While it might in principle be possible to construct a computing device with fewer of the flaws you mention, I strongly suspect that it will not be possible to do it without giving up either size, efficiency or latency (most likely all of those).

Your complaints regarding human memory demonstrate an ignorance of both engineering and neuroscience. Declarative memories are stored temporarily in the hippocampus, and some are over time consolidated into the neocortex. This long term storage of memory in the (sensory and association) cortices, where experiences and thoughts are processed and continuously compared (with zero latency) to a vast database of past experience, is precisely what allows these things to happen with the speed and effectiveness that they do. The fact that new memories must be integrated into existing networks is almost certainly what gives us both the aforementioned benefits as well as the drawbacks you mention. Making such a system less 'forgetful' or prone to false association would probably necessitate fundamental changes to its architecture.

To do what we can do with about 1 L of flesh that consumes just 20 watts of power is extraordinary. It's not the best tool for every job, but it's a far sight better than anything we've ever built for many important tasks. And we'd be well served to study it very closely, not just at a cognitive level, but at the network, cellular, and molecular level.

But they're wrong in a more important way: We've believed for years that the visual cortex is actually a visualization center! It just happens that when we're awake and looking with our eyes the visualization is constrained by sensory inputs. Sensory (and even association) cortices are basically simulators, that contain our best models of the world (what we expect the world to be like, based on prior experience), and the parts of those models that are active are dynamically constrained in real time by sensory data. When we have no sensory input at all, those models can run freely (this is essentially what is happening when we dream or have out of body experiences).

When individuals completely lose (or are born without) input for one modality, there's no reason to think they couldn't still use the corresponding cortical hierarchy for modelling (visualizing) that aspect of the world. The reported research is good evidence that this is exactly what happens.

Here's a crazy idea: try reading more than just the first sentence of the first article you find.

Also, the fact that you misread that very same sentence does not bode well for your reading comprehension. But seriously... while I would love to spend hours doing research for you and trying to convince you of the merits of this class of drugs, maybe you should try doing some reading beyond a single Wikipedia article.

https://en.wikipedia.org/wiki/Psychedelic_drug#General_psychological_effects
https://en.wikipedia.org/wiki/Psilocybin#Effects
https://en.wikipedia.org/wiki/Psilocybin#Medical_research

You've obviously never taken high doses of certain hallucinogens. Or if you had, you somehow completely missed one of the main benefits: insight into yourself and your relationship to the world around you. Ability to see problems for what they are, and often cope with them better.

And the fact that GHB is probably highly neurotoxic. There hasn't been a decent primate (non-human or otherwise) study completed yet, but the work done on rats is not encouraging (massive cell death in both PFC and hippocampus). If you're using GHB recreationally, I would seriously recommend reconsidering.

First of all, I'd like to say that I like Prof. Nutt and the work that he's done. In particular his lead authorship on Development of a rational scale to assess the harm of drugs of potential misuse and Drug harms in the UK: a multicriteria decision analysis represent a great public service.

But frankly this effort seems misleading. Either he's misrepresenting the potential of such a drug, or he's deluded himself. While I can easily imagine a drug that could be qualitatively similar to alcohol, though less harmful to the health of the user, I am aware of no popular recreational drug (outside of a small class of hallucinogens) that is completely non-neurotoxic. Prof. Nutt of course knows this. In particular, any drug that manipulates the excitation-inhibition balance as its primary action is almost certainly going to lead to habituation when used chronically and therefore will cause neuroexcitotoxicity when withdrawn.

Furthermore, any claims regarding lack of addiction potential should also be regarded sceptically. All widely used GABAergic drugs (ethanol, benzodiazepine, GHB) that I am aware of possess significant addiction potential. This may be avoidable, if the new drug were appropriately designed, though I cannot say that with certainty since I am not an expert on the binding properties of such molecules. But I should point out that the binding properties that give them their "desirable" effects also will cause them to bind to the interneurons of the VTA (Ventral tegmental area), part of the mesocorticolimbic dopamine system --- which is believed to be the initial locus of all drug addictions. For more information on this see Drug-Evoked Synaptic Plasticity Causing Addictive Behavior, Christian Luescher 2013 (JNeurosci).

Anyway, I think there are probably good reasons to prefer a tailor-made alcohol substitute, but I find it very strange that Prof Nutt is choosing to promote his efforts in this way. Given his history I would expect him to be a little more rational about how he promotes the work. On the other hand, given how irrational the public and the government are regarding drugs, maybe it's more reasonable to engage in a bit of propaganda. Unfortunately I think it goes against the spirit of his efforts in the last years.

Har har. Thanks genius.

This seems like a terrible idea. What could scientific computing with Ruby possibly offer that SciPy doesn't already? Way to split the potential work force guys. If you want to develop a scientific computing library for a rich dynamic language, then contribute to SciPy. What a wasted effort.

You're right, I overstated my case a bit, but the spirit of my point still stands: this is not a toy, but I presume a reasonably generic tool for exploring synaptic function. Just how generic it is comes down to the programmability of the various features.

I suppose one of us should go look at their paper if we actually want to settle it definitively. I did take a quick look at the PNAS website and looked for pre-prints on the author's website but couldn't find anything in the short time I looked.

No more time for this, back to my V1 model... :)

This is almost certainly wrong. It's unlikely they've modeled a single synapse type (I haven't read any of their papers, so I can't say with certainty). What's more likely is that they've modeled generic synaptic dynamics, including a standard set of ion species and programmable receptors (i.e. tune-able neurotransmitter affinities). There is a chance we'll discover a few more neurotransmitters, but we're probably good for 99% of the cases at this point.

It's not a toy: If you're interested in modeling complex synapses at high speed it's an extremely powerful tool.

Yeah, I don't understand why Scientific American considers this news. These are basic notions from machine learning and statistical modelling. It's not surprising, it's well established. If economists don't know this then they need to pick up a fucking textbook.

I just wanted to say thanks. I spend a lot of time developing in Python and after ten minutes of using PTVS, I'm seriously considering switching over from Eclipse+PyDev. Two problems though:

1) I have an unmodified package installed in site-packages, but sometimes I want to use a modified version contained in the project directory. We usually work around this by inserting the path to the modified package in the system path. Suddenly this doesn't work in PTVS.
2) For some reason I can't step into functions in said packages. Is there a problem with debugging 3rd party packages or something?

Anyway, please keep up the good work, it's looks fantastic. If I can work around these two issues, I'm pretty confident I'll make the switch.

Rather than argue the fine points of your Gedankenexperiment, I imagine it's sufficient to point out that were we to do so, we would be practicing precisely the thing you hoped to make obsolete: philosophy.

The only thing you've shown is that such a simulate function cannot exist, or at least cannot complete execution in time to affect the state of the world at T1. This follows from the fact that in order for the simulate() function to be a true simulation then it must also simulate itself. Put another way, if simulate() were to compute an accurate result, then it must include its own effect on the future state of the world. In essence it's a little paradox machine.

So much for your plan of replacing philosophy with computer science.