A brain and a robot walk into a bar.. (MEA2010, day 3)

Not surprisingly there are a few posters here on two-way connections between robots and neurons cultured on multielectrode arrays (MEAs). One of them offer an open source software package to do it, called Cult2Robot. The authors use the software to let a culture of neurons on an MEA move a robot in four directions and avoid obstacles, all through a Bluetooth connection. Spike rates in the culture are monitored and whenever they cross a threshold at one of the four edges of the square MEA the robot goes in that direction. If the robot's sensors detect an object in any of the four directions, electrical stimulation is applied to neurons on the opposite site of the array, making them more prone to fire and bring the robot away from the obstacle.

It's a simple principle but it illustrates a point I've been trying to formulate for months:

1. Activity in brains and neural networks can be understood as a fixed number of neurons with a spike rate 0-200 Hz (120.000 neurons in this particular culture) 
2. Some patterns of activity result in defined actions (here supra-threshold activity along an MEA edge results in ipsilateral movement) 
3. Some actions are adaptive, others maladaptive, depending on the circumstances 
4. Given adaptive sensory- and/or reward-feedback, neurons change their activity to produce more adaptive output (here objects are avoided by stimulation of neurons with contralateral output)
5. The number of adaptive activity patterns a network can reliably assume in the context of changing sensory- and/or reward-feedback is a measure of its operant control (in brains we call this creativity, intelligence, self-discipline etc; as an infant learns new words, its operant control increases)
6. By using sensory- and/or reward-feedback protocols and good electrophysiological or brain imaging techniques we can map the dynamic range of activity states a brain or network can assume and explore/model the network properties that determine its degree of operant control

A pressing question is how adaptive various networks can be. Could we for example program the MEA-culture-robot above to move not just in four directions but in the 360 directions of a circle? Could the network learn that certain spatial and/or temporal patterns of network output drive power-moves in the robot (like jumping, climbing or crawling) that scale difficult obstacles, the presence of which might be indicated by specific sensory feedback patterns? The link between specific obstacles and appropriate outputs could be strengthened by application of dopamine... you get the idea. Moreover, to make use of the rich and variable activity of neural networks, they should be given the ability to control the robot's actions along continuums like amplitude, duration, and correlation with other actions. And remember, its not just academic curiosity driving these explorations: a constant theme of this conference has been neural prostheses, and the ability of human brains to generate and respond to many arbitrary patterns of activity along continuums is exactly what gives them the ability to control and respond to brain computer interfaces that restore function and improve lives daily, all over the world, but which are still very immature and problematic.

MEA Meeting 2010, day two

I'm in Germany this week, in a sleepy little southern town called Reutlingen, attending the MEA2010 conference. I'm here to present our most recent work, talk to other MEA nerds and berate Multi Channel Systems employees until they implement basic spike sorting in MC_Rack. MEA stands for multielectrode array; the focus of the conference is multi-unit electrophysiology, mostly neural networks cultured on titanium arrays in glass dishes, but the whole spectrum of techniques and analysis methods is represented: yesterday we saw analysis of data from an electrode net the size of a small hand that goes right on the top of the brain, more on that in a bit.

German is frustratingly close to my native Swedish and yet impossible to understand. Almost. Spielplatz must mean playground, speil sounding like a variant of spel (game in Swedish) and platz like plats (ground in this context, though it really means location). Spielplatz = gameground. In Swedish we say lekplats, lek meaning play. It doesn't help, I'm still a tourist, and the heat is just as bad here as in England. People at the conference are mostly from Europe and East Asia, many languages spoken in the hallways and a lot of poor English (Globish, according to yesterday's Start The Week). There's maybe 200 people here in the big hall now, though it's still pretty early; leave it to the Germans to serve food and wine until 23:00 and then start talks at 8:30 the next morning. It's ok, my hotel is ten minutes away.

The opening lecture yesterday was by Pascal Fries and entitled 'Unravelling the brain-wide web of attention'. It was very good. Fries looks like he could be 25 years old but is a Prof., a P.I. and an M.D. He was showing us evidence that objects are held in visual attention by selective synchronization of distributed but functionally relevant brain regions in the gamma (feed forward, Granger = 0.02) and beta (feed back) bands (and a mystery band at 30 Hz). The data was recorded using a large net of a few hundred 1 mm diameter electrodes placed at 2 mm distance across almost the entire right hemisphere of two monkeys trained to respond to some visual stimuli and ignore others. The stimuli was known to hit specific regions in visual cortex, and activity in these region was subsequently compared with activity across the brain, using the net. The degree of synchrony in the gamma band predicted fast reaction time, so I'm wondering if synchrony in the molluskan buccal ganglia predict feeding rate. In question time I asked him what his thinking was on the mechanism by which the network associated with one stimulus becomes able to entrain others. The answer: given sufficient dopamine and noradrenaline tone, the network will establish coherence by activating interneurons in target regions. Signal-to-noise.

I actually asked him two questions, and having finished answering the first question he appeared ready to move to another questioner before catching himself and exclaiming "Oh, right, the second stimulus.. eh I mean second question". It's ok man, no need to speak humaneese, we're all neuroscientists here. Anyway, I should pay attention to the talks. Great wireless at a conference is both a blessing and a curse. Mostly blessing though.

In the news this morning

Former White House press secretary Scott McClellan releases a book highlighting the stench oozing off American politics. People who are outraged by his 'disloyalty' need a lecture on the concept of 'enabling' (or the Nuremberg trials).

A large review in Nature describe four groups that have spent the last eight years using deep brain stimulation to treat, successfully, obsessive compulsive disorder (ht: Laura). Their target, the ventral striatum, also called nucleus accumbens, is of course where rewarwding brain stimulation was first demonstrated over half a century ago in rats (Olds & Milner, 1954). Only the current differs.

The US Democratic party is getting ready to reconsider it's rules and rulings. One option: instead of stripping Florida and Michigan of their delegates' votes at the party's convention in August, let's "deny them passes to the convention for friends and spouses, and put them in sorry hotels"..

Finally, monkey uses brain implant controlled robotic arm (BICRA?) to feed itself (ht: Brandon). Poor monkey.