Neuroscience of Memory Part 1 DiGiovine Feb 20, 2013.ppt

Editor's Notes

• #4: Probably the most famous neuro patient, along with Phineas Gage!
• #5: Short term effects: Neurotransmitter effects. Long term effects: change in protein synthesis, which somehow leads to long term storage. We see this with our amnestic patients like HM who have b/l hippocampal damage who lose YEARS of memory (he lost about 3 yrs). 4 C’s is from Eichenbaum
• #6: e.g. pyramidal cell in the hippocampus has a long branching dendrite that extends upward from the cell body and receives inputs from other regions, as well as multiple dendrites that branch laterally from the cell to receive inputs from local neurons. The axon extends downward and branches. It may connect with other local cells or extend many millimeters to another brain region. Conduction: In the natural situation the resting potential is disturbed when chemical processes at the synapse result in a change in the permeability to a molecule, typically Na+  Action Potential:  Furthermore, when an action potential is generated in the initiation zone, the potential spreads electrotonically. This spread would decrement, but over a short distance would be more than sufficient to take the adjacent voltage–gated channels of axon membrane to threshold. This would regenerate the full magnitude of the action potential at that neighboring locus, and that potential would itself spread, reinitiating a full–blown action potential at its neighboring loci, and so on, continuing to reduplicate the action potential through the length of the axon. This simple regenerative mechanism, therefore, not only insures that the action potential achieves its full size at each locus but also insures its propagation for the full length of the axon regardless of the distance involved. The pumping mechanism sets this imbalance right, but during this recovery period a new action potential cannot be initiated, and so the cell is said to be in a refractory period. Synaptic Transmission: When the action potential reaches the end of the axon, at the presynaptic site, another mechanism takes over to mediate synaptic transmission.
• #7: Non-associative learning: does not involve association between stimuli; is a change in responsiveness to the same stimuli Following repetitions of siphon stimulation, the sensory neurons still produce action potentials, but the magnitude of the synaptic potential in the interneurons and motor neurons is reduced until eventually no AP’s are produced in the motor neuron. CONCLUSION: Loss of memory is PRESYNAPTIC. After habituation, the number of available vesicles did not decrease, but fewer of them became docked onto release sites in the habituated animals. Also, fewer sensory neurons had effective connections with motor neurons, and anatomical examination showed that the number of synaptic contacts between sensory neurons and interneurons and motor neurons was substantially reduced. CONCLUSION: Memory can be mediated by changes in synaptic efficacy, both through intracellular mechanisms that control transmitter release and through changes in anatomical connectivity. Sensitization: In the case of sensitization in Aplysia, the specific neurotransmitter of the modulatory interneurons is serotonin, which acts on the metabotropic receptors of the sensory neurons to increase their intracellular cAMP, (p. 46 ) which in turn activates PKA. This was found by showing that direct application of serotonin is sufficient to activate cAMP, and direct intracellular injection of cAMP is sufficient to enhance transmitter release by the sensory neuron, and to induce the facilitation of the reflex. 
• #8: In this form of learning two different stimuli are presented in close temporal proximity, such that typically a form of stimulation that does not ordinarily produce a response is presented before another stimulus that does produce the response. After multiple pairings the first stimulus acquires the ability to produce the response. In that sense, classical conditioning involves the acquisition of an association between the first, or conditioned stimulus, and the second, unconditioned stimulus. In Pavlov’s dogs the conditioned stimulus was a tone, that did not initially elicit salivation. It was presented repeatedly prior to injection of food into the mouth (the unconditioned stimulus), which did directly elicit salivation. After several pairings, the tone came to elicit the conditioned response of salivation.
• #9: Named by Lomo, who was studying cells in the hippocampus (specifically CA1) confirmed Hebb’s 1949 law that if a synapse is active when a postsynaptic neuron is active, the synapse will be strengthened Glutamate = primary excitatory transmitter in CA1 Found that tetanus-inducted changes (repetitive high-frequency stimuli of one pathway causing a greater population spike) lasted for several hours Called this “long-term potentiation” Evidence of hippocampal long-term depression, which enhances LTP at neighboring sites repetitive high-frequency electrical stimulation (called tetanus) of one pathway resulted in a steeper rise time (slope) of the excitatory synaptic potential to a subsequent single pulse. He also observed that following a tetanus there was recruitment of a greater number of cells reaching the threshold for an action potential, reflected in a greater “population spike,” the spike observed when many cells fire together (see examples in Fig.3–2B). Lomo found that the tetanus-induced changes in the synaptic and cellular responses to single pulses lasted for several hours, (p. 56 ) leading him to distinguish this phenomenon from short-lasting facilitations. And so, he called it “long-term potentiation.” EPSPs: excitatory postsynaptic potentials are increased by tetani that then lead to LTP  In particular, the NMDA receptors are selectively and competitively blocked by the antagonist D-2-amino-5-phosphono-valerate (AP5). A major discovery in revealing the mechanism of LTP was that AP5 has little effect on excitatory postsynaptic potentials (EPSPs) elicited by low-frequency stimulation, indicating that AMPA receptors, and not NMDA receptors, mediate normal synaptic transmission in the hippocampus. In contrast, AP5 completely blocks LTP following high-frequency stimulation trains, indicating that glutamate activation of NMDA receptors is critical to this form of synaptic plasticity.
• #11: Cooperative: more than one input must be active at the same time Associative: weak inputs are potentiated with co-occurring w/ stronger inputs Specificity: Only the stimulated synapse shows potentiation An intriguing 1997 study by Frey and Morris indicated that activity at hippocampal synapses that produces only a short-lived potentiation can nonetheless create a synaptic “tag” that lasts a few hours. Subsequent strong activation of a neighboring pathway within that period leads to lasting potentiation of both the “strong” pathway and the previously “tagged” synapses. Thus, LTP could serve to associate or integrate patterns of activity over a time window that has obvious behavioral significance. ________ Other studies suggest that the cascade of molecular events that is invoked by LTP may also mediate cortical plasticity that underlies memory. A particularly (p. 76 ) good example of this work involves a set of studies by Yadin Dudai and his colleagues focused on taste learning mediated by the gustatory cortex of rats. When rats are exposed to a novel taste and subsequently become ill, they develop a conditioned aversion specifically to that taste, and this learning is known to depend on the gustatory cortex. Blockade of NMDA receptors by infusion of the antagonist AP5 produces an impairment in taste aversion learning, whereas the same injections given prior to retention testing, or into an adjacent cortical area, had no effect. Thus, it is likely that modifications in cortical taste representations depend on LTP. Furthermore, blockade of protein synthesis in the gustatory cortex by infusion of an inhibitor prior to learning also prevents development of the conditioned taste aversion. By contrast, the same injection given into a neighboring cortical area or given to the gustatory cortex hours after learning has no effect. Consistent with this finding, MAP kinase as well as a downstream protein kinase were activated selectively in gustatory cortex within 10 minutes of exposure to a novel taste and activation peaked at 30 minutes, whereas exposure to a familiar taste had no effect. Conversely, a MAP kinase inhibitor retarded conditioned taste aversion. This combination of findings provides complementary lines of evidence that strongly implicate the NMDA mediated plasticity and subsequent specific protein synthesis as playing a critical role in cortical modifications that mediate this type of learning.
• #13: If rehearsal is prevented, lists of things are rapidly forgotten. Things don’t decay so much as we reach our limit – such as remembering numbers – we can only remember 7 or 8 in a row.
• #14: Regardless of the information in the items, the number of items retained is around 7* Dubbed this “span of immediate memory” Humans can recode information in packets that can be further broken down Sensory Memory is different! Observers can perceive 2-3x more information than they can report verbally without any cues, but this information is lost after 500ms In the auditory system, an “echoic store” can last up to 20 seconds How they figured out that we can remember more visually: A box of letters is flashed, it has 3 rows of 4 letters, for a total of 12 letters. This is over our 7 letter limit. But, when asked to recall a row (which is only 4 letters), we get it correct no matter which row asked. We don’t know beforehand which row will be asked, thus or visual memorial is better than our serial recall!
• #15: Craik and Lochkart (1972): level of processing matters 3 levels tested: Superficial (upper or lowercase letters) v. intermediate (rhyming) v. deep processing (judge the meaning of a word) Improved learning when given the deep processing task
• #16: Baddeley et al proposed the first variant of working memory – information that can be acted on and processed Phonological loop (left supramarginal gyrus, and L premotor region, which are only 2 of probably many areas) and visuospatial sketchpad (parieto-occipital region of both hemespheres) – independent of each other Subjects asked to recall a string of letters that were visually presented were found to be making acoustic mistakes (e.g. saying T for G). Giving a secondary visuospatial task during retention did not disrupt performance This was first discussed in the 1970’s
• #17: The Morris water maze task. A: A sketch of the apparatus and two trials of a typical training sequence. In the place navigation version of the task, the rat begins each trial at one of four randomly selected locations and must find a submerged platform (dashed circle) positioned at a constant location. In the cued navigation version the rat also starts at one of four locations, and the platform is visible (solid circle) and moved randomly across trials. B: Performance of rats with hippocampal lesions, with hippocampal lesions, and normal controls in acquiring the water maze task. Place navigation = hidden platform; cue navigation = visible platform. C: Performance on the transfer test. Left: Swim path of a control subject; dashed lines indicate quadrant of the maze in which the platform had been located. Right: Swim times of rats in different maze quadrants; black bar corresponds to the training quadrant (data from Morris et al., 1982).
• #18: R.B. is another pt like HM – he had an ischemic episode during bypass surgery. He developed anterograde amnesia – he couldn’t form new long-term memories. But, he also had retrograde amnesia back 1-2 years (so less severe than HM). When he died, autopsy showed he had a lesion specific to the Ca1 pyramidal cells of the hyppocampus b/l. Because damage to the medial temporal lobe doesn’t wipe out most of the episodic and semantic memories formed over a lifetime, we know the hippocampus is NOT where we store explicit knowledge. Somehow, it is important for that transition to the short to long-term memory. Lesions that damage lateral cortex of the anterior temporal lobe near the anterior temporal pole (so catching the entorhinal and perihippocampal cortex) causes severe retrograde amnesia (as in decades back). But, these patients may still be able to form ne long-term memoreies (this is isolated retrograde amnesia). It shows us that the anterior temporal lobe is important for memory storage, but not for acquiring new information/memories. Korsakoff – just like those w/ temp lobe amnesia, they lose explicit momory and preserve implicit (non-declarative) memory. Since pts w/ Korsakoff’s have no damage to the medial temporal lobe, this region cannot be the ONLY area responsible for forming explicit memory. Both areas must. Tulving (the one Dr. Rapin was referring to about the paper she sent out) found that their pt KC, who had severe antero- and retrograde amnesia, could indeed make semantic memories (albeit it took longer for him than control subjects), but wouldn’t remember HOW he learned this (he lost the episodic part of his memory acquisition), which is called source amnesia. Of note, he had damage to the medial temporal lobe, and frontal, parietal, and occipital cortices, L worse than R. Procedural memories - more implicit tasks….Studies have shown that pts like HM do better w/ procedural tasks w/ practice.