While it would certainly be nice for our minds to be, as they intuitively present to us, disembodied and perfect, they turn out to be more like machines, whose workings are amazingly intricate, but quite physical. A good deal has been learned about a few aspects of memory- how it gets recorded in synapses and in anatomical locations in the brain like the hippocampus, and how it is read back out.
The recent paper describes how spatial memory in rats relates to specific electrical/nerve signals called sharp wave ripples, or SWR for short. It was previously found that during sleep, rats (and by analogy, and more obviously, humans and dogs) replay scenes from the awake state in rapid fashion, so that researchers with sufficient electrodes in the subject's brain can even trace where the rat is in a dream, relative to a training maze it has been sent through during the day.
SWR's are common during these sleep+dreaming episodes, and are known to correlate with better learning. But SWR's take place in the awake state as well, and the current paper finds that they correlate with spatial memorization, and thus learning & performance in tests like mazes.
The experimental approach was to find anatomical locations where SWR were taking place in the hippocampus of rats while they were being trained to a new maze task, and then interfere electrically with those signals in a precise way that detects the beginning of a ripple and within 25 milliseconds cancels the rest of it, (total of about 50-100 milliseconds, typically). As a control, blasts were sent to the same locations, but at different times that didn't interfere with the SWR signal.
Rats with hippocampus damage eventually learn to go down the correct arms of the maze, but take far longer than normal rats. The job involves two quite different tasks- remembering which of two forks to take (the outbound decision). The rule is imposed by the experimenter, in an alternating sequence, using visually distinct mazes in successive trials, which requires remembering where one is and also what the prior trip was like. Second is the ability to remember how to get back to the start of the maze, (the center arm), which requires some degree of memory of where one is and where that was, which is, in a place-cell coordinate system, always the same. The experimenters claim that the inbound task is substantially easier, and it is the outbound test where they have in previous work established that rats use memory replays of past trips, (perhaps using more remote memories), not the inbound task.
The result was that only the outbound task was impaired by shorting the SWR signals. The inbound task was still learned at the normal speed. Whether the memory process is conscious or unconscious, the researchers were able to specifically interfere with the rat's thought process through a fine-grained electrical counter-stimulation; a fascinating development.
What are SWR's? They have been characterized over the last decade as rapid replays of navigation markers, (such as place cell firing), speeded up in time and replayed either forwards or backwards. They represent firing of place cells from throughout the hippocampus, as they would during a travelling sequence going from location to location, only the rat is at rest, and the sequence is speeded up twenty-fold! They are thus believed to constitute memory and simultaneously a way to convey this memory to other areas of the brain. It is truly a remarkable story.
Here are a few quotes from researchers doing this work:
"Sequences of neural activity occurring at the third time scale are observed during both sleep and awake but restful states, when animals are paused and generally inattentive, and are associated with sharp wave ripple complexes (SWRs) observed in the hippocampal local field potentials. During the awake state, these sequences have been shown to begin near the animal’s location and extend forward (forward replay) or backward (backward replay), and have been hypothesized to play a role in memory consolidation, path planning, and reinforcement learning." - thesis by Anoopum Gupta, 2011.
"During pauses in exploration, ensembles of place cells in the rat hippocampus re-express firing sequences corresponding to recent spatial experience. Such 'replay' co-occurs with ripple events: short-lasting (approximately 50-120 ms), high-frequency (approximately 200 Hz) oscillations that are associated with increased hippocampal-cortical communication. In previous studies, rats exploring small environments showed replay anchored to the rat's current location and compressed in time into a single ripple event. Here, we show, using a neural decoding approach, that firing sequences corresponding to long runs through a large environment are replayed with high fidelity and that such replay can begin at remote locations on the track. Extended replay proceeds at a characteristic virtual speed of approximately 8 m[eters]/s[econd] and remains coherent across trains of ripple events. These results suggest that extended replay is composed of chains of shorter subsequences, which may reflect a strategy for the storage and flexible expression of memories of prolonged experience." - abstract by Davidson, et al. 2009
"As we have noted, SWR-associated replay has been found to evolve approximately 20 times faster than behavior, and SWRs are on the order of 100 ms in duration. Given a running speed of 0.5 m/s, this means that the replay seen during a single SWR should recapitulate approximately 1 m of behavior."
"We make several novel contributions: we show that replay proceeds at a relatively constant 'virtual' velocity; that it can proceed over trajectories as long as the complete environment; that this extended replay spans trains of closely-spaced SWRs; and that replay can begin at locations remote from the animal." - thesis by Thomas Davidson, 2009Included in these findings is that SWRs can encompass not only replays of where rats have been, up to large areas and forward and reverse sequences, but also paths they have never taken, but could take, suggesting that planning may be taking place. So basically, (and however crudely and invasively), scientists are gaining the technology and knowledge to begin to eavesdrop on what rats are thinking- what they are remembering and what they are planning.
- How much must we destroy for oil?
- Atheism - out & proud.
- Salon's very funny New Yorker video drama.. especially the "shrink" episode.
- Religion, bad philosophy, and immorality seem to go together alot.
- Republican unworthiness continued.. fiscal edition. "There’s a reason why we can’t seem to make any progress on our fiscal mess: One of our two political parties has gone nuts."
- Cringely on IT outsourcing.. India's high school graduates man IBM's services, more or less.
- Europe.. is it nothing but class warfare, like it is in the US? "Unfortunately for the German population, while German business profited handsomely, and German Banks exported capital to the rest of the world, the costs were borne by German workers who faced wage pressure." (Capital which is, incidentally, going down the tubes. But no matter!)
- Open corruption continues in the US.
- MMT crows about its calls on the euro.
- But Margaret Thatcher, bless her, saw the euro crisis coming too.
- Bill Mitchell explains what he thinks is wrong with Alan Blinder's suggestion to stop interest support payments on bank reserves.
- Economics quote of the week, from Bill Mitchell, from an NGO report. Wealth doesn't trickle down, it washes out to sea.
"A global super-rich elite has exploited gaps in cross-border tax rules to hide an extraordinary £13 trillion ($21tn) of wealth offshore – as much as the American and Japanese GDPs put together … at least £13tn – perhaps up to £20tn – has leaked out of scores of countries into secretive jurisdictions such as Switzerland and the Cayman Islands with the help of private banks, which vie to attract the assets of so-called high net-worth individuals."
- Economics image of the week ... median income has been flat.