%%Reentrance Sources @article{hjorth71, title={Hippocampal efferents to the ipsilateral entorhinal area: An experimental study in the rat}, author={Hjorth-Simonsen, A}, journal={Journal of Comparative Neurology}, volume={142}, number={4}, pages={417--437}, year={1971}, publisher={Wiley Online Library} } @article{deadwyler75, desc={1st physio reentrance article}, title={Physiological studies of the reciprocal connections between the hippocampus and entorhinal cortex}, author={Deadwyler, Sam A and West, James R and Cotman, Carl W and Lynch, Gary}, journal={Experimental neurology}, volume={49}, number={1}, pages={35--57}, year={1975}, publisher={Elsevier}, notes={Method: 1. Stimd ipsi&contra CA3, IN-VIVO & recorded LFP in DG,EC,CA3 2. Made EC lesion to see if DG still activated afterwards Results:1. Saw that ipsi&contra CA3 stim will (1) activate EC (2) cause EC to reactivate DG! (reentrance) 2. Saw that lesioning EC will abolish 'late' (reentrant) DG potl from CA3com stim (see fig.7) 3. Saw that CA3c will also activate DG (Scharfman) 4. Saw that THETA commisural stim is ideal to potentiate EC & DG. **Note, CA3 not potentiated by this theta?? (pg 49) Discussion: 1. Avoids antidromic stim of MF (when studying CA3-->DG) by lowering current, since antidrom has higher threshold than orthodrom 2. Suggests CA3-->EC pathway is from direct connection, not via subiculum. Claims it probly isnt CA1 since stim @ CA1 doesnt give nearly as good EC readout! 3. Gives very good timing of delays in last paragraph (pg54): Commis.CA3 stim -> 0-3ms contra DG,CA3 -> +1-3ms CA1 -> +7-10ms EC +3 DG: 18-25ms tot loop time! Q1: which layer f EC did ghe use?} } @article{swanson77, title={An autoradiographic study of the organization of the efferet connections of the hippocampal formation in the rat}, author={Swanson, LW and Cowan, WM}, journal={Journal of Comparative Neurology}, volume={172}, number={1}, pages={49--84}, year={1977}, publisher={Wiley Online Library} } @article{swanson78, title={An autoradiographic study of the organization of intrahippocampal association pathways in the rat}, author={Swanson, LW and Wyss, JM and Cowan, WM}, journal={Journal of Comparative Neurology}, volume={181}, number={4}, pages={681--715}, year={1978}, publisher={Wiley Online Library} } @article{sorensen79, title={Projections from the subiculum to the deep layers of the lpsilateral presubicular and entorhinal cortices in the guinea pig}, author={S{\o}rensen, Knud Erik and Shipley, Michael T}, journal={Journal of Comparative Neurology}, volume={188}, number={2}, pages={313--333}, year={1979}, publisher={Wiley Online Library} } @article{finch86, desc={multiple reentrance paths}, title={Neurophysiology of limbic system pathways in the rat: projections from the subicular complex and hippocampus to the entorhinal cortex}, author={Finch, David M and Wong, Ernest E and Derian, Edie L and Babb, Thomas L}, journal={Brain research}, volume={397}, number={2}, pages={205--213}, year={1986}, publisher={Elsevier} } @article{buzsaki89, desc={topology preserved, reentrance modulated by behavior}, title={Two-stage model of memory trace formation: a role for “noisy” brain states}, author={Buzs{\'a}ki, Gy}, journal={Neuroscience}, volume={31}, number={3}, pages={551--570}, year={1989}, publisher={Elsevier}, notes={I. Stimulated perforant. path in-vivo & saw reentrance. Timed reentrance at 20-25ms see Fig6,7 -argues toppology preserved II. Talks about trisyn path parameters 1. EC>DG: each DG cell (1 million total) has ~4000 PP synapses 2. DG-->CA3: 250k CA3 cells. Each MF terminates on 10-15 CA3 cells. Each CA3 cell has 50-100 MF synapses 3. CA3 associational proj: each CA3 cell connected to 5% of all other CA3 cells (12.5k cells!!!!!!)each CA3 cell connected to 5% of all other CA3 cells (12.5k cells!!!!!!) III. CA3 trisyn responces modulated by behavior (lowered during TH activity & raised during sharp wave activity)} } @article{buzsaki91, desc={Lesioned subcortical connections & observed spontaneous oscill in hippo-EC loop}, title={Emergence and propagation of interictal spikes in the subcortically denervated hippocampus}, author={Buzs{\'a}ki, Gy{\"o}rgy and Hsu, Melissa and Slamka, Craig and Gage, Fred H and Horv{\'a}th, Zsolt}, journal={Hippocampus}, volume={1}, number={2}, pages={163--180}, year={1991}, publisher={Wiley Online Library} } @article{stringer92, title={Reverberatory seizure discharges in hippocampal-parahippocampal circuits}, author={Stringer, Janet L and Lothman, Eric W}, journal={Experimental neurology}, volume={116}, number={2}, pages={198--203}, year={1992}, publisher={Elsevier} } @article{pare92, title={Role of the hippocampal-entorhinal loop in temporal lobe epilepsy: extra-and intracellular study in the isolated guinea pig brain in vitro}, author={Pare, D and Llinas, R and others}, journal={The Journal of neuroscience}, volume={12}, number={5}, pages={1867--1881}, year={1992}, publisher={Soc Neuroscience}, notes={Method: 1. Used Guinea Pig slices, in-vitro w/ MDA epilepsy model (100Hz, while Stranger used 20Hz) 2. Stimulated EC. Recorded at EC, DG, CA3,CA1 Results:1. suggests seizures start in CA# and finish in EC (see Buszaki,1997 for counter). Only spanned ~2s, used visual evidence 2. divides seizure into stage1 (2-3Hz), stage2 (15-25Hz). In S1, EDs start in CA3-->CA1->EC and stop. In S2, they go through whole loop (inc. DG) and back. S2 marked by first ED in DG! 3. Shows show seizures synchronize along septotemporal axis 4. Records cells intracellularly and relates to extracellular potl: a. cell APs correspond to EDs b. CA3 cells exhibit bursting 100-150Hz c. CA3 & CA1 cells have PDS (ie elevated baseline after MDA) Discussion: 1. The first (as far as I know, see Buszaki 89,91) paper to suggest loop activity in seizures. Proof is: a. circumstntial evidence: CA1 EDs proceded by DG volume conductance b. sequintial unidirectional nature of EDs (see fig. 5ab,6) 2. Mentions connections w/ Strangers MDA (pg 1879). Says MDA==stage2 ED 3. talks about CA3-1 gap junctions as source of synchronization. pg 1879 4. Note S1 seems like IID while S2 seems like ID. See Avoli papers 5. emphasises DG gatekeeper role for epilepsy Q1: Interesting Sources: 1. Buzsaki, 1991- (see also Buszaki,1989) - shows loop cycles can occur spontaneously w/ fimbria/fornex lesion} } @article{sik94, title={Inhibitory CA1-CA3-hilar region feedback in the hippocampus}, author={Sik, Attila and Ylinen, Aarne and Penttonen, Markku and Buzsaki, Gyorgy}, journal={Science}, volume={265}, number={5179}, pages={1722--1724}, year={1994}, publisher={American Association for the Advancement of Science} } @article{tamamaki95, title={Preservation of topography in the connections between the subiculum, field CA1, and the entorhinal cortex in rats}, author={Tamamaki, Nobuaki and Nojyo, Yoshiaki}, journal={Journal of Comparative Neurology}, volume={353}, number={3}, pages={379--390}, year={1995}, publisher={Wiley Online Library} } @article{bartes95, desc={characterized loop w/ sigmoids of PSs}, title={Input-output relations in the entorhinal-hippocampal-entorhinal loop: entorhinal cortex and dentate gyrus}, author={Bartesaghi, Renata and Gessi, Tiziana and Migliore, Michele}, journal={Hippocampus}, volume={5}, number={5}, pages={440--451}, year={1995}, publisher={Wiley Online Library}, notes={Method: 1. Stimd comm. in guinea pigs Recorded EC, DG, CA3, CA1 2. Varied comm. stim amplitude and made IO sigmoids along hippo-EC loop! Results:1. Showed the hippo-EC loop can be activated w/ signle shock & not tetanus (see also Zosimov) 2. Found that comm>EC is roughly sigmoidal, but EC>DG, DG>CA extremely all-or-none. Thus there is linear/nonlinear part of loop. SEE FIG. 3. MOST IMPORTANT FIGGURE FOR US 3. spends rest of article talking abt comm>EC and EC>DG} } @article{iijima96, desc={optical imaging of hippo-EC w/ tetanus & bicuc. Saw RM of 1Hz}, title={Entorhinal-hippocampal interactions revealed by real-time imaging}, author={Iijima, Toshio and Witter, Menno P and Ichikawa, Michinori and Tominaga, Takashi and Kajiwara, Riichi and Matsumoto, Gen}, journal={Science}, volume={272}, number={5265}, pages={1176--1179}, year={1996}, publisher={American Association for the Advancement of Science}, notes={Method: 1. in-vitro stimd EC in rodent hippo-EC slice and OPTICALLY reocrded entire hippo-EC slice 2. applied bicuculline, 3. did 1Hz tetanus Results:1. Saw that activity would reverberate in hippo-EC and EC (!!!) when bicuculline applied 2. saw same thing after 1Hz tetanus applied (from 7th pulse onwards) Discussion: 1. Said can explain activity with 2 circuits, reverberating EC & reverberating hippo-EC 2. Suggests resonant mode of system on 1Hz. Note this confirms our CLPP paper RM in 1 Hz} } @article{nagao96, desc={pilocarpine model in hippo-EC slice. Similar to avoli97 w/o/ 1hz stim}, title={Epileptiform activity induced by pilocarpine in the rat hippocampal-entorhinal slice preparation}, author={Nagao, T and Alonso, A and Avoli, M}, journal={Neuroscience}, volume={72}, number={2}, pages={399--408}, year={1996}, publisher={Elsevier} } @article{avoli97, title={CA3-driven hippocampal-entorhinal loop controls rather than sustains in vitro limbic seizures}, author={Barbarosie, Michaela and Avoli, Massimo}, journal={The Journal of neuroscience}, volume={17}, number={23}, pages={9308--9314}, year={1997}, publisher={Soc Neuroscience} } @article{bragin97, title={Epileptic afterdischarge in the hippocampal--entorhinal system: current source density and unit studies}, author={Bragin, A and Csicsvari, J and Penttonen, M and Buzsaki, G}, journal={Neuroscience}, volume={76}, number={4}, pages={1187--1203}, year={1997}, publisher={Elsevier} } @article{wu98, title={Functional interconnections between CA3 and the dentate gyrus revealed by current source density analysis}, author={Wu, Kun and Canning, Kevin J and Leung, L Stan}, journal={Hippocampus}, volume={8}, number={3}, pages={217--230}, year={1998}, publisher={Wiley Online Library} } @article{ezrokhi01, desc={optical imaging of hippo-EC reentrance in both ambient & picrotoxin conditions}, title={Reverberation of Excitation in Living “Hippocampal Formation--Entorhinal Cortex” Slices from Rats. Optical Recording}, author={Ezrokhi, VL and Kas' yanov, AM and Markevich, VA and Balaban, PM}, journal={Neuroscience and behavioral physiology}, volume={31}, number={1}, pages={31--37}, year={2001}, publisher={Springer}, notes={Method: 1. Optically recorded in-vitro rodent Hippo slice (thus ALL areas of it, but not not EC) & stimd EC & DG 2. observed reentrance in both ambient (1 reverb) & picrotoxin (several reverb) conditions Results:1. so that activity reverberated w/ fixed latencies & periods of 60-150ms when picrotoxin was added but not w/o/ it, suggesting hippo-EC loop Discussion: 1. Said periodic activity is either from epileptoform activity OR hippo-EC reverberations. Argued for latter due to -CON: however latencies recorded here are much longer than all over papers (here 60-150ms, there 12-20ms) -fixed latenices, lack of spontaneous activity, proof of Iijima, Witter 96' paper} } @article{fell01, title={Human memory formation is accompanied by rhinal--hippocampal coupling and decoupling}, author={Fell, J{\"u}rgen and Klaver, Peter and Lehnertz, Klaus and Grunwald, Thomas and Schaller, Carlo and Elger, Christian E and Fern{\'a}ndez, Guill{\'e}n}, journal={Nature neuroscience}, volume={4}, number={12}, pages={1259--1264}, year={2001}, publisher={Nature Publishing Group} } @article{avoli02, title={Network and pharmacological mechanisms leading to epileptiform synchronization in the limbic system in vitro}, author={Avoli, Massimo and D’Antuono, Margherita and Louvel, Jacques and K{\"o}hling, R{\"u}diger and Biagini, Giuseppe and Pumain, Ren{\'e} and D’Arcangelo, Giovanna and Tancredi, Virginia}, journal={Progress in neurobiology}, volume={68}, number={3}, pages={167--207}, year={2002}, publisher={Elsevier} } @article{bartes03, title={Activation of perforant path neurons to field CA1 by hippocampal projections}, author={Bartesaghi, Renata and Gessi, Tiziana}, journal={Hippocampus}, volume={13}, number={2}, pages={235--249}, year={2003}, publisher={Wiley Online Library} } @article{kloosterman04, title={Two reentrant pathways in the hippocampal-entorhinal system}, author={Kloosterman, Fabian and van Haeften, Theo and Lopes da Silva, Fernando H}, journal={Hippocampus}, volume={14}, number={8}, pages={1026--1039}, year={2004}, publisher={Wiley Online Library}, notes={Presents evidence of Hippo-EC reentrance IN-VIVO Method: 1. uses uses 16 channel single-electrode to record at different depths along CA1-DG axis (fig.5b) 2. Uses current-source density analysis to identify source and sinks of LFPs. 3. stimulates Sub/SC and records in EC deep & superficial layers and along DG,CA1 axis in anethsized rats Results:1. Observes facilitation of Sub-->EC connection over several pulses of 5/10/20Hz (fig.2) 2. Observes LFP conection between EC5-->EC2 3. Observes 2 reentrant LFP pathways from EC-->DG (trisynaptic) and EC-->CA1 (temporoammonic)*** Discussion: 1. ECdeep>ECsuperficial connection 2. application of reentrance to memory, see Wang, 2001 3. says first evidence of reentrance was Deadwyler, 1975 4. pg.1036, says that temporoammonic reentrant LFP prob doesnt mean APs in CA1, just an EPSP. Thus LFP~=AP!!!! 5. pg.1038, says there was no evidence of oscillations, possibly due to subcortical inhibition. Says Buzsaki, 1989 showed oscillations in in-vivo denervated hippo Interesting Sources: 1. Wang, 2001 - relevance of oscillations to memory 2. Buszaki 1989 - in-vivo epilepsy model w/ oscilations by denervating hippo from subcortical inputs 3. Deadwyler, 1975 - first evidence of reentrance 4. Gloveli, 1997 - desribes 'pseudo-transfer-function' of EC neurons...} } @article{craig05, desc={LTP from CA1>EC >> CA1>Sub>EC}, title={Interaction between paired-pulse facilitation and long-term potentiation in the projection from hippocampal area CA1 to the entorhinal cortex}, author={Craig, Sarah and Commins, Sean}, journal={Neuroscience research}, volume={53}, number={2}, pages={140--146}, year={2005}, publisher={Elsevier}, notes={Method: 1. Stimulated CA1 in-vivo rate & recorded EC (layer ??? ) extracellularly. 2. Tested for paired-pulse facilitation (PPF) (2nd order nonlin) & LTP Results:1. Stimed CA1 w/ 20-480ms (2-50Hz) FIT trains and saw PPF in EC. SHOWS CA1->EC EXISTS IN-VIVO & THAT IT HAS 2ND ORDER NONLIN! 2. Induced LTP which lasted >30 mins Discussion: 1. Claims CA1>EC >> CA1>Sub (see Tamamaki,1995) 2. Discusses issue of wether LTP pre/post synaptic or both. Gives evidence that there is a presyn component! 3. discusses whether PPT & LTP connected/. Gives evidence they are! Q1: WHICH LAYER OF EC DID U RECORD FROM!!!! Interesting Sources: Tamamaki,1995- Claims CA1>EC >> CA1>Sub } } @article{bartes06, desc={characterized loop gain. BIG on multiple pathways: via DG & CA2 }, title={Input--output relations in the entorhinal cortex--dentate--hippocampal system: evidence for a non-linear transfer of signals}, author={Bartesaghi, R and Migliore, M and Gessi, T}, journal={Neuroscience}, volume={142}, number={1}, pages={247--265}, year={2006}, publisher={Elsevier}, notes={Method: 1. Same setup as all other papers (88,89,95,03). Also used sigmoid fitting Results:1. see Fig. 6-10 Discussion: 1. all-or-none behavior of PSs means output neuron population has (1) homogenous excitability & (2) # of input neurons 2. Is really big on story of 2 input paths to CA1: EC>DG>CA3>CA1, EC>CA2>CA1. this is really just a rehash of old story of temporoammonic path w/ a CA2 waystation.... 3. i think he did a good job in 95,06 paprs of characterizing loop gain as a whole...} } @article{zosimov08, title={Conditions required for the appearance of double responses in hippocampal field CA1 to application of single stimuli to Sh{\"a}ffer collaterals in freely moving rats}, author={Zosimovskii, VA and Korshunov, VA and Markevich, VA}, journal={Neuroscience and behavioral physiology}, volume={38}, number={3}, pages={313--321}, year={2008}, publisher={Springer} } @article{zosimov10, title={Return of Excitatory Waves from Field CA1 to the Hippocampal Formation Is Facilitated after Tetanization of Sch{\"a}ffer Collaterals during Sleep}, author={Zosimovskii, VA and Korshunov, VA}, journal={Neuroscience and behavioral physiology}, volume={40}, number={3}, pages={315--323}, year={2010}, publisher={Springer} } @article{zosimov12, title={Excitation Waves Returning to the Hippocampus via the Entorhinal Cortex Can Reactivate Populations of “Trained” Field CA1 Neurons during Deep Sleep}, author={Zosimovskii, VA and Korshunov, VA}, journal={Neuroscience and Behavioral Physiology}, volume={42}, number={2}, pages={133--143}, year={2012}, publisher={Springer}, notes={ Same as last article, but adds 2 very small things: 1. shows 3rd reverberation in some cases 2. adds theory that its going to CA1 via CA2 not EC. This is purely conjectural and is based largely on paper by R. Bartesaghi and T. Gessi, “Parallel activation of field CA2 and dentate gyrus by synaptically elicited perforant path volleys,” Hippocampus, 14, No. 8, 948–963 (2004) }} @article{uva14, title={Network Dynamics During the Progression of Seizure-Like Events in the Hippocampal--Parahippocampal Regions}, author={Boido, Davide and Jesuthasan, Nithiya and de Curtis, Marco and Uva, Laura}, journal={Cerebral Cortex}, volume={24}, number={1}, pages={163--173}, year={2014}, publisher={Oxford Univ Press}, notes={Method: 1. Used BMI Guinea Pig epiepsy model. Recorded extracellularly from slices in CA1,EC,DG,Sub,Parasub,Presub Results:1. Classified 2 types of seizures (based on HUMAN models): FA (fast activity) and HSA (hypersynchronous activity) a. FA - started in CA1/Sub/EC. Eventually DG got recuited @ B-early & made reentrant loop. Had fa & itts phase b. HSA - like FA, but w/o/ fa & itts phase (was Preictal --> bursting) 2. Identified 5 phases of seizures - preictal --> fa (20-30Hz) --> irregular spiking phase (itts) --> Bursting (B-early & B-late) 3. Noticed DG-->CA1 propogation times got shorter from b-early to b-late 4. Timed loop propogation delayes: EC-->DG: 12ms CA1-->EC: 13ms DG-->CA1 20ms Discussion: 1. talks about role of subiculum in starting / propogating seizures. Parasub & presub arent involved 2. talks about reentrant loop during bursting phase 3. See figure 9: it has alot of relavence to nested loop theory of hippocampus. Shows that info pathways evolve through seizure!} } %see also fell03, buzsaki09 %%Misc Hippocampal anatomy & physiology @article{jones93, desc={REVIEW: EC->Hippo connections}, title={Entorhinal-hippocampal connections: a speculative view of their function}, author={Jones, Roland SG}, journal={Trends in neurosciences}, volume={16}, number={2}, pages={58--64}, year={1993}, publisher={Elsevier}, notes={I. EC<->Hippo Anatomy 1. Discusses EC inputs from cortical areas (multimodal assoc. areas, some primary sensory areas, septum, thalamus, hypothal, etc...) 2. Rare connections: EC2->Sub, EC3->CA3, CA1->EC direct! II. Physiology 1. EC2 stellate cell responce has IPSP (dominated by interneuron release of GABA-A,B) followed by EPSP (AMPA/NDMA) -High freq potentiation gets rid of initial GABA IPSP 2. EC5 cells basically just have AMPA/NMDA excitatory EPSP III. Speculative Function: Access of info to hippi 1. Argues that EC2->DG route normally 'blocked' and info goes through EC->CA3. However, 5-10 Hz (theta) potentiation opens EC->DG route and allows info to flow from there! IV. Speculative Function: Learning & Memory V. Speculative Function: Epilepsy, Alzheimers 1. EC>CA1>Sub>>CA3 has most neurofibrillary tangles (NFT) in alzheimers! Interesting Sources: 1. Berger,Yeckel 1990 - says EC-->CA3 stronger than EC-->DG FOR US: III.1 suggests EC-->DG has PPTF which activated by theta. We need to see if his evidence can support this!! See potentiation in Strenger,Deadwyler,Berger articles:-) } } @article{yeckel90, title={Feedforward excitation of the hippocampus by afferents from the entorhinal cortex: redefinition of the role of the trisynaptic pathway.}, author={Yeckel, Mark F and Berger, Theodore W}, journal={Proceedings of the National Academy of Sciences}, volume={87}, number={15}, pages={5832--5836}, year={1990}, publisher={National Acad Sciences} } @article{scharfman07, title={The CA3 “backprojection” to the dentate gyrus}, author={Scharfman, Helen E}, journal={Progress in brain research}, volume={163}, pages={627--637}, year={2007}, publisher={Elsevier} } @article{mehta09, desc={hippo rate code & CA1 physiology}, title={The hippocampal rate code: anatomy, physiology and theory}, author={Ahmed, Omar J and Mehta, Mayank R}, journal={Trends in neurosciences}, volume={32}, number={6}, pages={329--338}, year={2009}, publisher={Elsevier} } @article{spruston07, desc={*The Hippocampus Book}, title={Structural and functional properties of hippocampal neurons}, author={Spruston, N and McBain, C}, journal={The hippocampus book}, pages={133--201}, year={2007}, publisher={Oxford University Press: New York} } @article{buzsaki94, desc={**CA3 recurrent network}, title={The hippocampal CA3 network: an in vivo intracellular labeling study}, author={Li, X-G and Somogyi, P and Ylinen, A and Buzsaki, G}, journal={Journal of Comparative Neurology}, volume={339}, number={2}, pages={181--208}, year={1994}, publisher={Wiley Online Library} } @article{fricker99, desc={shows extracell K+ increases baseline potl & after -54mV spont. cellular theta oscill emerge} title={Cell-attached measurements of the firing threshold of rat hippocampal neurones}, author={Fricker, Desdemona and Verheugen, Jos AH and Miles, Richard}, journal={The Journal of Physiology}, volume={517}, number={3}, pages={791--804}, year={1999}, publisher={Physiological Soc}, notes={Method: 1. Used wierd method of measuring extracellular K+ reversal to get cell thresh/baseline of CA1 pyr&IN cells in vitro. 2. Used NBQX,APV,Bicuc to remove Synaptic Glut/GABA events. 3. saw effects of increased K+ on threshold/spontinuity/baseline Results:1. saw baseline CA1 pyr potl was -84mV & -74mV for silent INs. This as 13mV more depol than traditional whole-cell recording method 2. Fig.3 GREAT! Shows that as [K+] increases, baseline potl increases. At 13mm K+ & -54mV, spontaneuous firing occurs! 3. Fig.5 subthreshoscillations proceeded spontaneous firing. APs were ussually on peak of oscil 4. increased Ca2+ raisses threshold but leaves baseline unaffected! 5. membrane potl in spont. cells was much hgiher than silent ones} } @article{west91, desc={*west neuron counting in CA3,CA1 of hippo} title={Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator}, author={West, MJ and Slomianka, LHJG and Gundersen, H Jẋ G}, journal={The Anatomical Record}, volume={231}, number={4}, pages={482--497}, year={1991}, publisher={Wiley Online Library} } %%Misc Epilepsy @article{westbrook00, title={Seizures and epilepsy}, author={Westbrook, GL}, journal={Principles of neural science}, volume={4}, pages={910--935}, year={2000}, publisher={McGraw-Hill New York} } @article{bertram13, title={Neuronal circuits in epilepsy: Do they matter?}, author={Bertram, Edward H}, journal={Experimental neurology}, volume={244}, pages={67--74}, year={2013}, publisher={Elsevier} } @article{fisch03, title={Interictal Epileptiform Activity: Diagnostic and Behavioral Implications: 2002 ACNS Presidential Address}, author={Fisch, Bruce J}, journal={Journal of clinical neurophysiology}, volume={20}, number={3}, pages={155--162}, year={2003}, publisher={LWW} } @incollection {jefferys12, Title = {Limbic Network Synchronization and Temporal Lobe Epilepsy}, Author = {Jefferys, JGR and Jiruska, P and de Curtis, M and Avoli, M}, Publisher = {National Center for Biotechnology Information (US), Bethesda (MD)}, Year = {2012}, Edition = {4th}, Booktitle = {Jasper's Basic Mechanisms of the Epilepsies}, URL = {http://europepmc.org/abstract/MED/22787650} } @article{lopantsev98, title={Participation of GABAA-mediated inhibition in ictallike discharges in the rat entorhinal cortex}, author={Lopantsev, Valeri and Avoli, Massimo}, journal={Journal of neurophysiology}, volume={79}, number={1}, pages={352--360}, year={1998}, publisher={Am Physiological Soc} } @article{pacia97, desc={*theta ECoG & epilepsy}, title={Intracranial EEG substrates of scalp ictal patterns from temporal lobe foci}, author={Pacia, Steven V and Ebersole, John S}, journal={Epilepsia}, volume={38}, number={6}, pages={642--654}, year={1997}, publisher={Wiley Online Library}, notes={As far as which frequencies are involved in hippocampal seizures, it depends whether you are looking at ECoG or EEG and whether the seizure began in the hippocampus or began in the neocortex and then spread to the hippocampus. For seizures which begin in the hippocampus, ECoG shows initial oscillations of 12-20Hz which progress to 5-9 Hz (theta). EEG can only pick up the 2nd stage of theta rhythms (this subtlety made this really confusing to research). Neocortical seizures usually have rhythms of 2-5 Hz. Once these seizures spread to the hippocampus, the 2-5 Hz rhythms can continue or they can turn into 5-9 Hz theta rhythms. } } @article{doose88, desc={**theta EEG & epilepsy}, title={Theta rhythms in the EEG: a genetic trait in childhood epilepsy}, author={Doose, Hermann and Baier, Wolfgang K}, journal={Brain and Development}, volume={10}, number={6}, pages={347--354}, year={1988}, publisher={Elsevier} } @article{medvedev01, desc={**gamma & epilepsy}, title={Temporal binding at gamma frequencies in the brain: paving the way to epilepsy?}, author={Medvedev, AV}, journal={Australasian Physics \& Engineering Sciences in Medicine}, volume={24}, number={1}, pages={37--48}, year={2001}, publisher={Springer} } @article{gonzalez11, desc={4-AP MEA model}, title={The 4-aminopyridine< i> in vitro epilepsy model analyzed with a perforated multi-electrode array}, author={Gonzalez-Sulser, Alfredo and Wang, Jing and Motamedi, Gholam K and Avoli, Massimo and Vicini, Stefano and Dzakpasu, Rhonda}, journal={Neuropharmacology}, volume={60}, number={7}, pages={1142--1153}, year={2011}, publisher={Elsevier}, notes={Method: 1. Used MEA to record LFP in CA3 & DG (not single spikes) under 4-AP model. Slices did NOT include EC 2. Saw effect of neuroactive agents of interictal initiation/propogation, including AMPA,NDMA,bicuculline,PB,THIP 3. Studied seperately effects of phasic & tonic GABA-A current using PB (Pentobarbitol-phasic&tonic GABA-A agonist) & THIP (tonic GABA-A agonist) Results:1. Fast CA3-driven IIE eradicated by iGluR antagonist (which also increases IIEs in DG) 2. Slow CA3-DG IIE is blocked from CA3-->DG propogation by bicuculline (thus GABA driven). 3. PB (GABA agonist) increased ii in CA3, suggesting GABA can be excitatory 4. THIP (GABA tonic agonist) decreased ii in both CA3 & DG Discussion: 1. Analyzed where exactly interictal events in Avoli,2002 (CA3-interictal & slow-interictal) come from w/ MEA 2. says fast CA3 interictal events caused by recurrent CA3 pyramidal cell connections 3. possiblity of excitatory GABA conductance (since PB increases ii in CA3) 4. Spread of ii from CA3-->DG &sync of CA3 & DG w/o/ EC contradicts idea that reentrance needed for ii spread & thus CLPP model... Interesting Sources: Dzhala & Staley, 2003 - says CA3 interictal events caused by recurrent CA3 connections} } @article{wendling02, desc={**recurrent connections > oscillations > epilepsy}, title={Epileptic fast activity can be explained by a model of impaired GABAergic dendritic inhibition}, author={Wendling, F and Bartolomei, F and Bellanger, JJ and Chauvel, P}, journal={European Journal of Neuroscience}, volume={15}, number={9}, pages={1499--1508}, year={2002}, publisher={Wiley Online Library} } @article{wendling08, desc={**REVIEW: computational models of epilepsy}, title={Computational models of epileptic activity: a bridge between observation and pathophysiological interpretation}, author={Wendling, Fabrice}, year={2008}, publisher={Expert Reviews Ltd London, UK} } @article{begley00, desc={**1-2\% of US population suffers epilepsy}, title={The Cost of Epilepsy in the United States: An Estimate from Population-Based Clinical and Survey Data}, author={Begley, Charles E and Famulari, Melissa and Annegers, John F and Lairson, David R and Reynolds, Thomas F and Coan, Sharon and Dubinsky, Stephanie and Newmark, Michael E and Leibson, Cynthia and So, EL and others}, journal={Epilepsia}, volume={41}, number={3}, pages={342--351}, year={2000}, publisher={Wiley Online Library} } @article{brodie96, desc={**30\% dont respond to medication}, author = {Brodie, Martin J. and Dichter, Marc A.}, title = {Antiepileptic Drugs}, journal = {New England Journal of Medicine}, volume = {334}, number = {3}, pages = {168-175}, year = {1996}, doi = {10.1056/NEJM199601183340308}, note ={PMID: 8531974}, URL = {http://www.nejm.org/doi/full/10.1056/NEJM199601183340308}, eprint = {http://www.nejm.org/doi/pdf/10.1056/NEJM199601183340308} } @article{engel03, desc={**epilepsy surgery remission rate}, title={Practice parameter: Temporal lobe and localized neocortical resections for epilepsy Report of the Quality Standards Subcommittee of the American Academy of Neurology, in Association with the American Epilepsy Society and the American Association of Neurological Surgeons}, author={Engel, Jerome and Wiebe, Samuel and French, Jacqueline and Sperling, Michael and Williamson, Peter and Spencer, Dennis and Gumnit, Robert and Zahn, Catherine and Westbrook, Edward and Enos, Bruce}, journal={Neurology}, volume={60}, number={4}, pages={538--547}, year={2003}, publisher={AAN Enterprises} } @article{naylor02, desc={saw how Wiener kernels of GWN-->DG decline w/ epilepsy}, title={Changes in Nonlinear Signal Processing in Rat Hippocampus Associated with Loss of Paired-Pulse Inhibition or Epileptogenesis}, author={Naylor, David}, journal={Epilepsia}, volume={43}, number={s5}, pages={188--193}, year={2002}, publisher={Wiley Online Library}, notes={Method: 1. Stimulated EC & recorded DG LFP. Didnt say if stim sub/supra thresh, but quotes source (Mazarati,1998)\ 2. Had epilepsy model where he induced SE by mix of 2/20Hz EC stim for an hour 3. stim w/ GWN & calculated Weiner kernals over time (before epil to 1 month after)! 4. Calc's Paired-pulse (PPI) responce Results:1. PPI lost after 1min 2Hz stim and stays lost for ~30 mins (figures not shown :-( ) 2. output LFP spike peak @ 7ms and memory~30ms ??? Discussion: 1. Result 1 shows how nonstationary hippo is, and why chronic stim can be theraputic 2. saud he compared w/ kernals of dead tissue which were much more linear 3. suggests large inhib input from EC-->DG due to neg kernals 4. Says thatoff-diagonal peak of 2nd kernal suggests a freq peak since that diff is prefered! (ask marm) Interesting Sources: Berger,1998- 3 papers using Weiner kernals of EC->DG (same as this but longer mem epoch) & compared w/ paired pulse} } %%Epilepsy DBS @article{penfield54, desc={**1st DBS report}, title={Epilepsy and the functional anatomy of the human brain.}, author={Penfield, Wilder and Jasper, Herbert}, year={1954}, publisher={Little, Brown \& Co.} } @article{morrell06, desc={REVIEW of DBS for epilepsy - clinical}, title={Brain stimulation for epilepsy: can scheduled or responsive neurostimulation stop seizures?}, author={Morrell, Martha}, journal={Current opinion in neurology}, volume={19}, number={2}, pages={164--168}, year={2006}, publisher={LWW}, notes={1. Reviewed several human DBS studies (see below) 2 had great section on DBS frequencies, biphasic effects, and kindling on pg 166} } @article{sun08, desc={technical description of neuropace} title={Responsive cortical stimulation for the treatment of epilepsy}, author={Sun, Felice T and Morrell, Martha J and Wharen Jr, Robert E}, journal={Neurotherapeutics}, volume={5}, number={1}, pages={68--74}, year={2008}, publisher={Elsevier}, notes={I. Reviews previous human dbs studies II. Detection- 3 tools: 1. half-wave tool 2. line-length tool - average sample-to-sample differences within a window. 3. area tool - average area-under-the-curve within a window short term window (<4s) compared to long window (4s-16mins). detection happens when threshold crossed III. Stimulation- 1-33hz, 1-12mA, 40-1000us pulse-widths. } } @article{sun14, desc={REVIEW of closed-loop neurostimulation in spinal pain management, epilepsy VNS, epilepsy DBS, parkinsons}, title={Closed-loop Neurostimulation: The Clinical Experience}, author={Sun, Felice T and Morrell, Martha J}, journal={Neurotherapeutics}, pages={1--11}, year={2014}, publisher={Springer} } @article{benmenachem14, desc={REVIEW of neuropace history & approval}, title={Epilepsy: Responsive neurostimulation [mdash] modulating the epileptic brain}, author={Ben-Menachem, Elinor and Krauss, Gregory L}, journal={Nature Reviews Neurology}, year={2014}, publisher={Nature Publishing Group}, notes={LETTER to editor reviewing Neuropace clinical trials. Particularly 2011 studsy & 2014 Christie Heck followup study 1. said neuropace was approved Nov. 2013 2. final 2 year efficacy was 53% of ppl had seizure reduction, median seizure reduction: 53%. 54% had >50% seizizure reduction. 9% seizure free for last 6 months. 3. 'implant' (lesion) effect of improvement w/ electrode placement was temporary: seizure rate eventually returned to baseline interesting source: Anderson, Kudela 2009 - computational model of DBS on large scale brain network } } @article{heck14, desc={**2-year neurpace followup study}, title={Two-year seizure reduction in adults with medically intractable partial onset epilepsy treated with responsive neurostimulation: Final results of the RNS System Pivotal trial}, author={Heck, Christianne N and King-Stephens, David and Massey, Andrew D and Nair, Dileep R and Jobst, Barbara C and Barkley, Gregory L and Salanova, Vicenta and Cole, Andrew J and Smith, Michael C and Gwinn, Ryder P and others}, journal={Epilepsia}, volume={55}, number={3}, pages={432--441}, year={2014}, publisher={Wiley Online Library} } @article{durand01, desc={REVIEW of DBS for epilepsy - mechanisms, technical}, title={Suppression and control of epileptiform activity by electrical stimulation: a review}, author={Durand, Dominique M and Bikson, Marom}, journal={Proceedings of the IEEE}, volume={89}, number={7}, pages={1065--1082}, year={2001}, publisher={IEEE}, notes={REVIEW of epilepsy DBS techniques & mechanisms. More technical, less clinical I. Intro - goal of epilepsy DBS is to create inhibition / desynchronization using currents. Also notes, DBS is the only soln. in 'elequent areas' (those which cant be resected) II. Uniform DC fields - uses 2 electrodes to create field between them. Effectiveness depends on relative orientation to cells (must be from basal to apical dendrites). Affective in high-K,low-Ca models. Works by hyperpolarizing cells. III. Localized DC fields - uses 1 electrode. Effective in penicillin,high-K,low-C,kindling models. Hyper/depol depends on whether electrode is by soma or dendrites. -works by DEPOLARIZIKNG the cell and causing 'depolarization block', which prevents APs when cell is too depold (~ -40mV). IV. LFS & Single Pulse stim- not orientation dept & only need to be applied during event, not chronically... Several methods exist A. Phase resetting & singularity- use single pulse right @ critical phase bifurcation ... ? (fig6). So far only used in slices. Assumes system has stable oscill & stable fixed point (nooscill) states. Says below technqiues have much animal studies but few clinical ones B. phase reseting & desynchronization- fig7 - assumes system has dysyncd stabvle modes C. chaos control - wtf??? D. LFS- gives some sources for 1-1.3hz , & TMS studies - suggests works by inducing LTD, but not sure V. HFS- says HFS have much clinical studies, but few experimental ones... Suggests it works by extracellular K+ rise which induces depoarization block. Other theories are NT buildup, loss of info transfer, specific activation of inhib pathways... -for depol block, see"On the mechanisms underlying the depolarization block in the spiking dynamics of CA1 pyramidal neurons"} } @article{kahane09, desc={REVIEW of DBS for epilepsy - clinical}, title={Manipulating the epileptic brain using stimulation: a review of experimental and clinical studies.}, author={Saillet, Sandrine and Langlois, M{\'e}lanie and Feddersen, Berend and Minotti, Lorella and Vercueil, Laurent and Chabard{\`e}s, Stephan and David, Olivier and Depaulis, Antoine and Deransart, Colin and Kahane, Philippe}, journal={Epileptic Disorders}, year={2009}, publisher={John Libbey Eurotext}, notes={I. About 30% of epileptic patients do not respond to antiepileptic drugs (Kwan and Brodie 2000), of which only a minority can benefit from resective surgery. II. VNS - Vagus Nerve Stimulation a. is FDA approved for chronic use to treat epilepsy. b. has other behavioral sideffects (less sleepiness, more alertness) c. reduces epilepsy by >50% in 50-60% of ppl d. CONS: no clear predicitve factors for responders, and precise mechanisms unknown III. DBS - used on corticol & subcortical structures A. Cerebellum - 1st DBS target in Cooper, 1973. Lately not that popular... 10Hz B. Thalamus - anterior thalamus & centromedian thalamus. more effective for GENERALIZED seizures. -rat study showed 130Hz thalamus stim led to reduction of focal hippocamp[al seizures C. Basal Ganglia- 1. Caudate nuclus (CN) - human CN: high,bad;low (4-6hz),good. 2. subthalamic nuclei (STN) - 130hz good 3. substania nigra (SNR)- 60hz good D. TMS @ seizure focus- ... E. Hippo Stim- 1. 1 Hz stim leads to quenching of seizures in rats (avoli) 2. 1hz & 50Hz of chronic stim good in humans (kinoshita, 2005) IV. Responsive/Adaptive/Closed-loop Stimulation 1. see osoria,2005, kossoff 2004, fountas 2005. Stim Freqs: cerebellum-10Hz; rath_thalamus-100Hz; 100hz-kainic acid rats. 6-8hz & 60Hz in human CM thalamus. human-CN: 4-6Hz. human-STN-130hz. Multiphasic effects of stimu: rat_thalamus-low,bad;high,good; human CN: high,bad;low,good. 6-8hz & 60Hz good in human-CM-thalamus. TNS stim of human motor cortex-low,good;high,bad. 1hz & 50hz in human hippo. LESIONAL/PLACEBO/CARRYOVER AFFECTS -: THALAMUS, } } @article{fisher14, desc={REVIEW of DBS for epilepsy - clinical}, title={Electrical brain stimulation for epilepsy}, author={Fisher, Robert S and Velasco, Ana Luisa}, journal={Nature Reviews Neurology}, year={2014}, publisher={Nature Publishing Group}, notes={REVIEW of DBS for epilepsy. Similar to Kahana,2009 Hippocampus most seizure prone structure in the brain Notes 1hz, 100hz, 185hz supressed seizures in rats Compares VNS,thalDBS & hippoDBS VNS- cheapest, safest*, but least effective. also, can affect other nerves.. Also controlled studies impossible since patients can feel stim.} } @article{anderson09, desc={computational DBS model of large scale neural netwrok & pulse reseting}, title={Phase-dependent stimulation effects on bursting activity in a neural network cortical simulation}, author={Anderson, William S and Kudela, Pawel and Weinberg, Seth and Bergey, Gregory K and Franaszczuk, Piotr J}, journal={Epilepsy research}, volume={84}, number={1}, pages={42--55}, year={2009}, publisher={Elsevier} } @article{wendling13, desc={computational DBS model of LFS, HFS on neural-mass thalamocortical network}, title={Modulation of epileptic activity by deep brain stimulation: a model-based study of frequency-dependent effects}, author={Mina, Faten and Benquet, Pascal and Pasnicu, Anca and Biraben, Arnaud and Wendling, Fabrice}, journal={Frontiers in computational neuroscience}, volume={7}, year={2013}, publisher={Frontiers Media SA} } @article{fisher97, desc={**+/- biphasic DBS: rat thalamic stim: 8hz bad, 100hz good}, title={Anticonvulsant effect of anterior thalamic high frequency electrical stimulation in the rat}, author={Mirski, Marek A and Rossell, Lisa Ann and Terry, John B and Fisher, Robert S}, journal={Epilepsy research}, volume={28}, number={2}, pages={89--100}, year={1997}, publisher={Elsevier} } @article{chkhenkeli97, desc={**+/- biphasic DBS: rat thalamic stim: 8hz bad, 100hz good}, title={Effects of therapeutic stimulation of nucleus caudatus on epileptic electrical activity of brain in patients with intractable epilepsy}, author={Chkhenkeli, SA and Chkhenkeli, IS}, journal={Stereotactic and functional neurosurgery}, volume={69}, number={1-4}, pages={221--224}, year={1997}, publisher={Karger Publishers} } @article{kinoshita05, desc={**+/+ biphasic DBS:}, title={Electric cortical stimulation suppresses epileptic and background activities in neocortical epilepsy and mesial temporal lobe epilepsy}, author={Kinoshita, Masako and Ikeda, Akio and Matsuhashi, Masao and Matsumoto, Riki and Hitomi, Takefumi and Begum, Tahamina and Usui, Keiko and Takayama, Motohiro and Mikuni, Nobuhiro and Miyamoto, Susumu and others}, journal={Clinical neurophysiology}, volume={116}, number={6}, pages={1291--1299}, year={2005}, publisher={Elsevier} } %Hippocampus & Memory @article{milner57, desc={**H.M. paper}, title={Loss of recent memory after bilateral hippocampal lesions}, author={Scoville, William Beecher and Milner, Brenda}, journal={Journal of neurology, neurosurgery, and psychiatry}, volume={20}, number={1}, pages={11}, year={1957}, publisher={BMJ Group} } @article{okeefe71, desc={**1st O'keefe paper on place cells}, title={The hippocampus as a spatial map. Preliminary evidence from unit activity in the freely-moving rat}, author={O'Keefe, John and Dostrovsky, Jonathan}, journal={Brain research}, volume={34}, number={1}, pages={171--175}, year={1971}, publisher={Elsevier} } @book{okeefe78, desc={**okeefe book on place cells}, title={The hippocampus as a cognitive map}, author={O'keefe, John and Nadel, Lynn}, volume={3}, year={1978}, publisher={Clarendon Press Oxford} } @article{hampson99, desc={**Hampson's 1st article on FCTs}, title={Distribution of spatial and nonspatial information in dorsal hippocampus}, author={Hampson, Robert E and Simeral, John D and Deadwyler, Sam A}, journal={Nature}, volume={402}, number={6762}, pages={610--614}, year={1999}, publisher={Nature Publishing Group} } @article{moser05, desc={**1st article on grid cells in EC}, title={Microstructure of a spatial map in the entorhinal cortex}, author={Hafting, Torkel and Fyhn, Marianne and Molden, Sturla and Moser, May-Britt and Moser, Edvard I}, journal={Nature}, volume={436}, number={7052}, pages={801--806}, year={2005}, publisher={Nature Publishing Group} } @article{moser06, desc={**Head direction cells in EC}, title={Conjunctive representation of position, direction, and velocity in entorhinal cortex}, author={Sargolini, Francesca and Fyhn, Marianne and Hafting, Torkel and McNaughton, Bruce L and Witter, Menno P and Moser, May-Britt and Moser, Edvard I}, journal={Science}, volume={312}, number={5774}, pages={758--762}, year={2006}, publisher={American Association for the Advancement of Science} } @article{buzsaki96, desc={hippocampal-neocortical dialogue}, title={The hippocampo-neocortical dialogue}, author={Buzs{\'a}ki, Gy{\"o}rgy}, journal={Cerebral Cortex}, volume={6}, number={2}, pages={81--92}, year={1996}, publisher={Oxford Univ Press} } %Reverberations, Hebb, & Learning @book{hebb49, desc={**Hebb's 2nd postulate:}, title={The organization of behavior: A neuropsychological theory}, author={Hebb, Donald Olding}, year={1949}, publisher={Psychology Press} } @article{amit95, desc={**hebbian assemblies make learning to maintaining reverberations}, title={The Hebbian paradigm reintegrated: Local reverberations as internal representations}, author={Amit, Daniel J}, journal={BEHAVIORAL AND BRAIN SCIENCES}, volume={18}, pages={617--657}, year={1995} } @article{johnson09, desc={**review of experiment theory of hebbian assembly in amygdala}, title={Hebbian reverberations in emotional memory micro circuits}, author={Johnson, Luke R and LeDoux, Joseph and Doyere, Valerie}, journal={Frontiers in neuroscience}, volume={3}, pages={27}, year={2009}, publisher={Frontiers} } @article{kaplan91, desc={**comp. model of hebbian assembly & good review of 2nd principle}, title={Tracing recurrent activity in cognitive elements (TRACE): A model of temporal dynamics in a cell assembly}, author={Kaplan, Stephen and Sonntag, Martin and Chown, Eric}, journal={Connection Science}, volume={3}, number={2}, pages={179--206}, year={1991}, publisher={Taylor \& Francis} } @article{wolters08, desc={**comp. model of hebbian assembly between PFC & posterior brain}, title={Coherence and recurrency: Maintenance, control and integration in working memory}, author={Wolters, Gezinus and Raffone, Antonino}, journal={Cognitive processing}, volume={9}, number={1}, pages={1--17}, year={2008}, publisher={Springer} } @article{lansner92, desc={**comp. model of hebbian assemblies of pyr. & IN cells}, title={Modelling Hebbian cell assemblies comprised of cortical neurons}, author={Lansner, A and Frans{\'e}n, E}, journal={Network: Computation in Neural Systems}, volume={3}, number={2}, pages={105--119}, year={1992}, publisher={Informa UK Ltd UK} } @article{fuster00, desc={presents theory of reverberatory activity for working memory}, title={Cortical dynamics of memory}, author={Fuster, Joaquin M}, journal={International Journal of Psychophysiology}, volume={35}, number={2}, pages={155--164}, year={2000}, publisher={Elsevier} } @article{wang01, desc={REVIEW of synaptic reverberation underlying persistant activity. mostly comp.}, title={Synaptic reverberation underlying mnemonic persistent activity}, author={Wang, Xiao-Jing}, journal={Trends in neurosciences}, volume={24}, number={8}, pages={455--463}, year={2001}, publisher={Elsevier} } %Misc @article{buzsaki12lfp, desc={REVIEW of EEG, ECoG, LFP relationship}, title={The origin of extracellular fields and currents—EEG, ECoG, LFP and spikes}, author={Buzs{\'a}ki, Gy{\"o}rgy and Anastassiou, Costas A and Koch, Christof}, journal={Nature Reviews Neuroscience}, volume={13}, number={6}, pages={407--420}, year={2012}, publisher={Nature Publishing Group}, notes={I. Contributors to Extracellular Fields 1. synaptic activity: Na+ going from extraell to intracell makes SINK. To maintain ELECTRONEUTRALITY, a passive SOURCE develops. This makes inverted extracell voltage (think Loeb final) -GABA makes little contrib. since its resting potl is so close to baseline 2. Action Potls- can only contribute a. to >100Hz activity (see fig.5) and ONLY if synchrony maintained. However AHP may make more lasting contrib. Thus >100Hz activity can be used as proxy for single spikes -suggests AHP behind delta waves 3. calcium spikes, gap junctions 4. instrinsic currents and resonances (I-h) - must be occuring synchronously, which is most common in interneurons 5. Ephatic effects- macro LFP feedback to cells. in normal comp. not much effects, but in hypersynchronous states, it can deliver as much current as TransCranialStim which is known to have effects! II. Determinants of LFP strength: architecture & synchrony 1. Neuronal geometry and architecture- pyr. cells are arranged in parr. & thus their dipoles can add up. -apparently, human hippo. has cell bodies vertically shifted to each other & thus human hippo. ECoG much weaker than rodent! 2. Temporal Synchrony- -good example of cerebellum, which has perfect geometry for LFPs, but since all computations are LOCAL, there is no synchrony and thus no LFP!!!!!!!!!!!!!!!! -1/f^n power law: power is inversely prop. to freq by 1/f^n law! This due to low-pass spatial filtering, slower freqs can recruit more neurons IV. Inverse problem - reconstructing micro events & dipole sources from macro LFP 1. due to VOLUME CONDUCTION, LFP ALWAYS requires verification of its local nature 2. CSD analysis can determine if source is close or far, kinda} } %Rhythms %Misc @inproceedings{crick90, desc={**apparently 1st to connect rhythms & consciousness}, title={Towards a neurobiological theory of consciousness}, author={Crick, Francis and Koch, Christof}, booktitle={Seminars in the Neurosciences}, volume={2}, pages={263--275}, year={1990}, organization={Saunders Scientific Publications} } %Resonance @article{yarom00, desc={REVIEW article of neural resonances}, title={Resonance, oscillation and the intrinsic frequency preferences of neurons}, author={Hutcheon, Bruce and Yarom, Yosef}, journal={Trends in neurosciences}, volume={23}, number={5}, pages={216--222}, year={2000}, publisher={Elsevier}, notes={I. INTRO: Q1. What causes brain rhythms A1a- connectivity between fundamentally nonoscillatory neurons - however this has never been found experimentally A1b- inherebt neuronal resonance - vast evidence for. Note, A1a&A1b can reinforce each other! II. Resonance as a probe of frequency preferance 1. gives sources (upto 2000) of studies using ZAP funcrtions on neurons (sources 5-14 on cardiac cells, inner ear, squid axon, neocortex) & some conductances used to generate them (Ca2+,K+,Na+,etc...) 2. BOX1: resonance is func. of impedance, z=1/wC=H(w)=Y(w)/X(w). Can be modeled w/ RLC circuit III. How to Make resonances: Rule of thumb 1. As in circuit to have resonance/pass-band you need low-pass component (here, membrane capacitance) and high-pass component (here,?) 2. High-pass component is 'current that oppose membrane voltage'- aka rectifying currents which have rev.potl. @ bottom of their protein gate activation curve (see Fig.2,3). They are negative-feedback. Their pulse (not impulse) response is thus high pass! Ex. are K+ IV. Amplifying currents, amplified resonance, and oscillations 1. To amplify resonance, you need positive feedback mechanism where reversal potl is at the top of its activation curve (like weak AP). Examples are I-Na, I-NDMA, I-CA2+ 2. BOX2: Uses simple neuron model & phase plane to show how diff. conductance (g-K,g-Na) parameters lead to either damped,damped-oscillatory,spontaneous oscillatory For us: 1. afterpotl cant be biological source of resonance since res. arises even subthreshold! 2. pg.218 - says in neocortex cells have 2 resonances (1-2Hz & 5-20Hz) which activate at diff. depol levels!} } @article{pike00, desc={CA1 pyr. + IN ZAP experiments}, title={Distinct frequency preferences of different types of rat hippocampal neurones in response to oscillatory input currents}, author={Pike, Fenella G and Goddard, Ruth S and Suckling, Jillian M and Ganter, Paul and Kasthuri, Narayanan and Paulsen, Ole}, journal={The Journal of Physiology}, volume={529}, number={1}, pages={205--213}, year={2000}, publisher={Wiley Online Library}, notes={Method: 1. stimulated CA1 pyr. cells & interneurons ('fast-spiking' in pyr. layer & horizontal in str. ori) w/ var freqs & ZAP, at depols just below threshold 2. Measured res. freq & 'spike-pref freq' (aka MFR like PPTF) for var freqs. 3. Did experiemnt w/ Na+ blocker TTX Results:1. Saw pyr. cells & horiz INs have theta peak (~4.5 Hz) & FS INs have gamma res. 2. Saw gamma res in INs blockd w/ TTX but not theta res in pyr.cells Discussion: 1. Issue w/ comparing to Leung,1998 due to him depoling neurons to right under thresh... 2. Saw 'suppressing freq' for horiz INs @ 20-30Hz !!!! 3. talks about how different neurons types in network transmit different streams of info @ their res. freqs 4. Says that optimal-spiking.freq > res.freq. Leung,1998 says res.freq>spont.freq. Difference? 5. Says gamma pref in FS INs caused by Na+ current! Interesting Sources: } } @article{jahnsen94, desc={CA1, thal responce to GWN}, title={A spectral analysis of the integration of artificial synaptic potentials in mammalian central neurons}, author={Jahnsen, Henrik and Karnup, Sergei}, journal={Brain research}, volume={666}, number={1}, pages={9--20}, year={1994}, publisher={Elsevier}, notes={Method: 1. stimulated CA1,thalamic neurons w/ GWN. Entirely subthreshold! Results:1. Showed CA1 neurons w/ GWN inputs DO NOT have resonance! Fig 3,5,7 2. CA1 neurons did have a res. peak for RECT inputs, due to I-h current (blcoked by Cs+), Fig6 . thalamic neurons did have GWN resonance @ 10-20Hz . shows I-V curve (aka slope=input resistance) of CA1 neurons. Some linear, some sigmoidal. . Shows (& discusses) that CA1 TF changes w/ depol, even when Rm linear. This surprising Discussion: 1. pg.18 good for explaining why MFR changes PPTF For Us: 1. As in simuls, OL CA3-->CA1 has no resonance! 2. I-h current suggests even subthreshold activity has 'feedback component' which needs feedback kernel (which is absent in Lu,2011,2012) Interesting Sources: 25- Sakai, 1992 - Review of Volterra/Kernel approaches in neuroscience/visual system! } } @article{leung98, desc={CA1 ZAP experiment}, title={Theta-frequency resonance in hippocampal CA1 neurons in vitro demonstrated by sinusoidal current injection}, author={Leung, L Stan and Yu, Hui-Wen}, journal={Journal of neurophysiology}, volume={79}, number={3}, pages={1592--1596}, year={1998}, publisher={[Bethesda, MD, etc.] American Physiological Society [etc.]}, notes={Method: 1. Stimulated CA1 cells w/ 'ZAP FUNCTION' (sine wave of increasing frequency from 2-20Hz) & calc'd FFT. Also stim'd w/ sine function 2. calc'd output power & coherence Results:1. Saw strong theta peaks in output power. 2. Saw harmonics in physiological data when stim power was high enough 3. Ran experiments w/ slight baseline depol. Saw that RMs shifted 4. Saw theta peak in 10/12 neurons when driving & only 5/12 spontaneously. Discussion: 1. fischer transform of coherence 2. interesting measure of sharpness, Q=Peak/HBW; HBW=half-peak power bandwidth which is -3dB from peak! Q1: Interesting Sources: Jahnsen, 1994 - stimd w/ GWN & saw little resonances. This makes sense cuz it was OL! In CL you may!!!! For us: 1. only went upto 20Hz 2. Saw harmonics!!!!!! 3. Saw input power (in our case MFR) had little input on pptf! 4 Ran experiments w/ slight baseline depol. Saw that RMs shifted. This confirms that TT can change RM power & location??? (must check) 5. saw theta peak in 10/12 neurons when driving & only 5/12 spontaneously. Thus we can argue we chose global PDMs of neurons which had theta activity!} } @article{white02, desc={**EC ZAP theta resonance}, title={Frequency selectivity of layer II stellate cells in the medial entorhinal cortex}, author={Haas, Julie S and White, John A}, journal={Journal of neurophysiology}, volume={88}, number={5}, pages={2422--2429}, year={2002}, publisher={[Bethesda, MD, etc.] American Physiological Society [etc.]} } @article{heinemann04, desc={**EC ZAP resonance}, title={Subthreshold resonance explains the frequency-dependent integration of periodic as well as random stimuli in the entorhinal cortex}, author={Schreiber, Susanne and Erchova, Irina and Heinemann, Uwe and Herz, Andreas V}, journal={J Neurophysiol}, volume={92}, number={1}, year={2004} } %NOTE: warren2011; hemond2009 do CA3 ZAP exp. Others do of EC & DG. see article summ %Theta %Physiology @article{green54, desc={**1st major article on theta}, title={Hippocampal electrical activity in arousal.}, author={Green, John D and Arduini, Arnaldo A}, journal={Journal of neurophysiology}, year={1954}, publisher={American Physiological Society} } @article{alonso87, desc={**1st article on EC theta}, title={Neuronal sources of theta rhythm in the entorhinal cortex of the rat}, author={Alonso, A and Garcia-Austt, E}, journal={Experimental brain research}, volume={67}, number={3}, pages={493--501}, year={1987}, publisher={Springer} } @article{gloveli97, desc={extracellular `ZAP` of EC}, title={Frequency-dependent information flow from the entorhinal cortex to the hippocampus}, author={Gloveli, Tengis and Schmitz, Dietmar and Empson, Ruth M and Heinemann, Uwe}, journal={Journal of neurophysiology}, volume={78}, number={6}, pages={3444--3449}, year={1997}, publisher={Am Physiological Soc}, notes={Method: 1. New Frequency Probing Method: (compare to intracellular ZAP injection): Extracellularly injected FITs of various freqs & measured cell responces intracellularly. -This method whole cell filter rather than just intracell filter: ie it includes AMPA/NMDA dynamics rather than just ion channels! Thus more realistic to in-vivo! 2. Used Sag & Rebound cell properties to characterize L2 (stellate) & L3a,b projection cells! Results:1. Saw that L2 stellate cells are high-pass filter (>6Hz), while L3 cells are low-pass filter (<10Hz). Theta activates both (fig2) 2. Gives Discussion: 1. Attempts physio reason for potentiation: regulated K currents lower MP & this removes Na channel inactivation, thus lowering threshold!!!! Thus advocartes dynamic threshold: Lu,2012 2. Shows AHP from .2-2s dep. on stim freq!!! 3. Mention EC-->CA1 (temporoamonic) may be inhibitory (see Q1: Interesting Sources: Staubli,1997 - Theta & LTP Soltesz, 1995 : EC-->CA1 inhibitory~!} } @article{kamondi98, desc={**theta in dendrites}, title={Theta oscillations in somata and dendrites of hippocampal pyramidal cells in vivo: Activity-dependent phase-precession of action potentials}, author={Kamondi, Anita and Acs{\'a}dy, L{\'a}szl{\'o} and Wang, Xiao-Jing and Buzs{\'a}ki, Gy{\"o}rgy}, journal={Hippocampus}, volume={8}, number={3}, pages={244--261}, year={1998}, publisher={Wiley Online Library} } @article{kocsis99, desc={shows CA3 is endogenous theta pacemaker}, title={Interdependence of multiple theta generators in the hippocampus: a partial coherence analysis}, author={Kocsis, Bernat and Bragin, Anatol and Buzs{\'a}ki, Gy{\"o}rgy}, journal={The Journal of neuroscience}, volume={19}, number={14}, pages={6200--6212}, year={1999}, publisher={Soc Neuroscience}, notes={Methods 1. recorded CA1-DG extracellularly using 16 electrode probe in-vivo rodents 2. Used partial coherence analysis (non directed!!!) to check for independent theta generators 3. Made EC lesion to confirm independnt theta generator Results 1. Found 2 indépendant TH generators: 1. EC-->(Str. Ori & Str. Lac). Notes EC-->IN-->Str. Ori pathway 2. CA3-->(DG,Str.Rad). Shows CA3 ind. TH generator (see Buzaki, 2002) 2. Used EC lesion to confirm CA3 is independent TH generator Discussion 1. Talks about CA3-->DG pathway (see Scharfman, 2002) 2. Suggests above pathway is normally suppressed by inhibitory DG-->CA3 pathway!!!! see Aksady,1998 3. Suggests nonlinear cross term between EC & CA3 TH generators to influence CA1 firing (see Mehta, 2004) Interesting Sources 1. Kamondi, 1998B - dendritic TH preferance 2. Acsady, 1998 - DG-->CA3 inhib! } } @article{buzsaki09, desc={Single Neuron theta around hippo-EC loop}, title={Theta oscillations provide temporal windows for local circuit computation in the entorhinal-hippocampal loop}, author={Mizuseki, Kenji and Sirota, Anton and Pastalkova, Eva and Buzs{\'a}ki, Gy{\"o}rgy}, journal={Neuron}, volume={64}, number={2}, pages={267--280}, year={2009}, publisher={Elsevier}, notes={Method: 1. Recorded SINGLE SPIKES from virtually all Hippo-EC sites (ECI-V,CA1,CA3,DG) IN-VIVO (in diff experiments). Analyzed both principal cells & interneurons 2. Studied theta phase procession & theta strength relationship of these different sites Results:1. Phase relationships (fig.3) a. Saw EC2&EC3 completely out of theta phase. b. Saw DG&CA3 in phase, but CA1 not w/ them. c. Saw interneurons are remarkably in phase globally (and thus out of phase w/ principal cells) 2. Did Cross-correlogram between several sites. Suggests 1st order dynamics? 3. Showed that cell theta-phase preferance varies w/ MFR, suggesting nonlinearites (similar to change in pTF w/ MFR) 4. showed that diff cells in EC repsond to theta differently: fast acting & slow acting. can tbe treated similarly} } @article{buzsaki02, desc={REVIEW: theta}, title={Theta oscillations in the hippocampus}, author={Buzs{\'a}ki, Gy{\"o}rgy}, journal={Neuron}, volume={33}, number={3}, pages={325--340}, year={2002}, publisher={Elsevier}, notes={Numba1: Brain systems involved in generation of theta rhythms 1. Hippocampal TH not. necc connected to TH in other parts of brain 2. distinction between 'current generator' (transmembrane currents resp. for magn. of LFP) & 'rhythm generator' - mechanisms resp. for emergence and contro, of oscillatory patterns 3. theta observed most strongly in CA1 (in terms of LFP) 4. rhythm generators can act either by 'permissive action' (allows emergence) or 'pacemaker action' (outputs theta) II. Hippocampal - Septum (or Septal Nuclei) connection 1. The minimum required condition for theta in HP is intact connection to MS-DBB (medial septum-diagonal band of Broca) 2. septum only projects to inhibitory interneurons in HP 3. HP interneurons are only HP cells which project to septum III. Classic Theta Model (fig.2) 1. Based on CSD (current source density) analysis of CA1. Shows excit in Lac-mol (from EC) and inhib. in str.oriens/pyr (from interneurons/septum) (fig.4) 2. Assumed EC \& septum are both pacemakers which can subthreshold potls across pyr. cells which (since pyr cells aligned) cause large extracell. potls! 3. This model has several weaknesses. see pg. 327 IV. 2 Types of Theta: 1. In aenesthized animals atropine (ACh blocker) comp. removes theta. In awake, it doesnt, suggesting 2 independant theta mechanisms: A. Atropine-Sensitive TH: facilitated by ACh; blocked by Atropine B. Atropine-Resistant TH: fac. by NMDA**, blocked by urethane, ketamine, EC lesion! Believed to be conveyed by EC-->CA V. Relationship of single cells to LFP 1. strong vs weak/threshold responders depol @ TH peak 2. strong resp. depol b4 peak and cause most of responce. Thus not all pyr. cells tuned to TH the same. This is spike-phase-procession / rate-coding VI. Relationship of TH to LTP Learning! -need to read closer VII. Theta in CA3 (pg330-331) 1. small TH sink in CA1 from CA3. Small since only small # of CA3 neurons active at any given time. There is about a 145deg TH phase shift between them (Buzsaki,2009) 2. CA3 doesnt have much TH LFP for logistical reasons (small dendritic arbor, etc...) Thus, it was largely ignored! 3. CA3 is able to generate inherent TH -if EC removed only TH in CA1 is in str. rad from CA3. -Probalby makes TH genreator from recurr. connections between itself \& maybe DG -is Cholinergic TH (atropine-sensitive), since w/o/ ACh it is abolished! Also, CA3 TH can be elicited in slices w/ ACh agonist!} } @article{williams09, desc={shows CA1 is endogenous theta pacemaker}, title={Self-generated theta oscillations in the hippocampus.}, author={Goutagny, Romain and Jackson, Jesse and Williams, Sylvain}, journal={Nature neuroscience}, volume={12}, number={12}, year={2009}, notes={Method: 1. Recorded in CA3,CA1 of removed intact whole hippocampus 2. isolated CA1 by (1)measuring coherence of CA3-CA1 (2)physically removing CA3 3. split septotemporal axis of CA1 in 2 by electrical stim? 4. Recorded pyr & IN intracellularly Results:1. Saw that CA1 can generate theta in isolation (5.1+/2Hz, range 3-10Hz) 2. Saw it was independant of CA3 (although CA3 also seemingly could?) 3. Saw it was AMPA/GABA/NMDA activated, but w.o. ACh (thus atropine resistant) 4. Saw that splitting CA1, seperate th. oscillators emerge in both regions, w/ slightly diff freq, suggesting entrainment 5. See FIG3c. Shows that theta peak synced w/ pyr. firing. INs syncd to offpeak & induce pyr firing. Thus a loop between CA1 Pyr & IN!!!!!!!!!!!!!!!! Discussion: 1. Talks about relation of intrinsic TH w/ MS-induced TH a. Says int. TH 10-20% of in-vivo values, thus strongly attenuated b. This study doesnt contradict MS lesion studies which abolish TH, since they measure TH power, not periodicity!!! (thus this power much less, but still periodic). Also, all those studies report slight residual TH ascribed to bad lesion Q1: Can CA3 independantly generate theta? Interesting Sources: 1. Glovellu,2005 - says that gamma & theta are synced to traverse & longitudonal axis, respectively. Thus they form grid!, Relvence to theta-gamma comodulation (TGC) FOLLOWUP NEWS REPORT IN NATURE: [Colgin, Laura Lee, and Edvard I. Moser. "Hippocampal theta rhythms follow the beat of their own drum." Nature Neuroscience 12.12 (2009): 1483-1484.] 1. SAYS IN axons extend micrometers in septotemporal axis, thus can lead to coherence across CA1!} } @article{colgin13, desc={REVIEW: theta}, title={Mechanisms and functions of theta rhythms}, author={Colgin, Laura Lee}, journal={Annual review of neuroscience}, volume={36}, pages={295--312}, year={2013}, publisher={Annual Reviews}, notes={I. pg.297 gives good review of previous 2TH hypothesis: AS-TH from MS & AR-TH from EC. -Reviews MS role in TH: (1) HP TH starts 500ms after MS TH. (2) HP-TH 80ms syc lag behind MS TH II. Intrinsic CA1 TH- 1. suggests from horizontal INs since they have HCN channels (& Ih) 2. thus presents, 'weakly coupled oscillator theory'. something I thought of 5` months ago! III. Theta Functions 1. Arise during movement & REM 2. Discrete processing entity theory 3. Intrinsic theta in dendrites 4. Hippo-PFC theta coupling: stronger in correct taks. May be reduced in diseases like schitzo Interesting Sources Giocomo,2007 - shows TH freq proportional to HCN freq! Kamondi,1998 - TH in CA! dendrites inherent!} } @article{colom06, desc={REVIEW: septal theta}, title={Septal networks: relevance to theta rhythm, epilepsy and Alzheimer's disease}, author={Colom, Luis V}, journal={Journal of neurochemistry}, volume={96}, number={3}, pages={609--623}, year={2006}, publisher={Wiley Online Library}, notes={notes={I. Septal Anatomy & Connectivity: S has 3 parts: lateral (LS), medial (MS), and posterior (PS). Hippo inputs to all 3. Only the MS inputs to the hippo, via the FF complex. LS & MS connected to hypothalamus II. 3 neuron types: GABAergic, Cholinergic, Glutamatergic 1. GABA- evenly distributed in S 2. ACh- mostly in MS-DBB (where they are 2x GABAergic). Thought to be ~9000 in mice. 3. Glut- Also cnc in the MS-DBB (16,000) in rat. III. Local in-vivo Theta - unclear. GABAn's involved. Chol? 1. pg615 septum the place where broadband hypothal input gets turned into theta input IV. In vitro theta- seems to be generated by AHP & Ih V. Septo-Hippo System - All 3 types of N's project to hippo. But, CholN's-->pyr&int, while GABAN's-->ints only!!!! Glut-->? VI. Septum & HippoTheta 1. Distinction 1: TH prev during exploration & REM, while LIA (large amplitude activity)/Bursting is prev. during immobility & SWS 2. Distinction 2: TH#1, atropine-resistant TH prev during exploration, while TH#2, atropine-sensitive prev. during immobility and SWS + urethane anesthesia 3. reasons why TH important a. LTP sensitive to TH phase b. human recordings c. place cells d. -- e. septal lesions decrease memory performance 4. Both chol & GABA Ns important to hippo TH. Lesioning either will effect it 5. See fig. 3 for SCHEMATIC~!!!!!!! VII. Epilepsy & (AD) 1. AD lnked to loss of ACh neurons (all FDA drugs increase this). Epilepsy kills these neurons due to neurodegeneration through ischemia/hypoxia. Thus E-->ND-->ACh loss -->AD 2. Suggests loss of septal ACh input leads to lack of ACh in hippo & lack of atr-sensitive theta & thus AD -proof1: injection of B-plaque in septum kills ACh cells & lowers task performance -proof2: in AD humans, ChAT (ACh making enzyme) reduced by 60 in MS-DBB!!!! } } } @article{lubenov09, desc={theta are waves along septotemporal axis}, title={Hippocampal theta oscillations are travelling waves}, author={Lubenov, Evgueniy V and Siapas, Athanassios G}, journal={Nature}, volume={459}, number={7246}, pages={534--539}, year={2009}, publisher={Nature Publishing Group}, notes={Method: 1. used 'microdrive array' which is depth MEA? to record along septotemporal axis of CA1 str.ori. (since there theta constant across depth) Results:1. Saw CA1 theta not in phase across CA1 but is more like traveling wave which spreads from CA3-->sub Discussion: 1. Used coherence analysis to show that although theta across CA1 out of phase it was still the same theta (since coherence across CA1=1) 2. See discussion on last page about possiblities of theta emergence. Gives 4: 1.MSpacemaker 2.? 3. weakly coupled oscillators 4. FF&FB networks 3. EC2 cells have inherent theta bursting (see sources 45,46) 4. Colgin,2013 says this paper strongly supports weakly coupled nested oscillator theory} } @article{traub03, desc={*IN/pyr interactions to make gamma <& some TH> + gap junctions}, title={< i> Interneuron Diversity series: Inhibitory interneurons and network oscillations< i> in vitro}, author={Whittington, Miles A and Traub, Roger D}, journal={Trends in neurosciences}, volume={26}, number={12}, pages={676--682}, year={2003}, publisher={Elsevier} } @article{kopell05, desc={**computational model of how pyr./IN interactions make theta}, title={Slow and Fast Inhibition and an H-Current Interact to Create a Theta Rhythm in a Model of CA1 Interneuron Network}, author={Rotstein, Horacio G and Pervouchine, Dmitri D and Acker, Corey D and Gillies, Martin J and White, John A and Buhl, Eberhardt H and Whittington, Miles A and Kopell, Nancy}, journal={J Neurophysiol}, volume={94}, pages={1509--1518}, year={2005} } %Behavioral @article{winson78, desc={1st study showing loss of theta impairs behavior}, title={Loss of hippocampal theta rhythm results in spatial memory deficit in the rat}, author={Winson, Jonathan}, journal={Science}, volume={201}, number={4351}, pages={160--163}, year={1978}, publisher={American Association for the Advancement of Science}, notes={Method: 1. recorded (w/ depth electrodes) abilibty of rats on spatial maze task before and after septal lesion Results:1. showed that those spetal lesions which abolished theta (~.5) decreased performance, while others didnt 2. showed that theta-abolished animals were still able to relearn task leter, probably using different mechanisms Discussion: 1. 1st article on animal theta memory deficit. only hint earlier was H.M.} } @article{berry78, desc={2nd theta-behavior study. Showed that TH power b4 trl was pos. corr. w/ training rate}, title={Prediction of learning rate from the hippocampal electroencephalogram}, author={Berry, Stephen D and Thompson, Richard F}, journal={Science}, volume={200}, number={4347}, pages={1298--1300}, year={1978}, publisher={American Association for the Advancement of Science}, notes={Method: 1. recorded RABBIT CA1 w/ depth electrode. Had rabbit do behavioral task Results:1. saw that theta power b4 trl positively corr. w/ performance (R2=.72) Discussion: 1. 2nd theta-behavior trl after Winson} } @article{wiebe01, desc={**theta correlates w/ performance in (odor) DNMS task}, title={Recognition memory correlates of hippocampal theta cells}, author={Wiebe, Sherman P and St{\"a}ubli, Ursula V}, journal={The Journal of Neuroscience}, volume={21}, number={11}, pages={3955--3967}, year={2001}, publisher={Soc Neuroscience} } @article{olvera02, desc={**theta & performance in morris water maze task}, title={Place-learning, but not cue-learning training, modifies the hippocampal theta rhythm in rats}, author={Olvera-Cort{\'e}s, E and Cervantes, M and Gonzalez-Burgos, I}, journal={Brain research bulletin}, volume={58}, number={3}, pages={261--270}, year={2002}, publisher={Elsevier} } @article{chrobak89, desc={**chemical septal inhibition reduces theta & rodent performance in maze task}, title={Intraseptal administration of muscimol produces dose-dependent memory impairments in the rat}, author={Chrobak, James J and Stackman, Robert W and Walsh, Thomas J}, journal={Behavioral and neural biology}, volume={52}, number={3}, pages={357--369}, year={1989}, publisher={Elsevier} } @article{staubli95, desc={**augmenting of theta IMPROVES performance}, title={Effects of 5-HT3 receptor antagonism on hippocampal theta rhythm, memory, and LTP induction in the freely moving rat}, author={Staubli, U and Xu, Fang Bo}, journal={The Journal of neuroscience}, volume={15}, number={3}, pages={2445--2452}, year={1995}, publisher={Soc Neuroscience} } @article{kahana01, desc={REVIEW: theta in humans}, title={Theta returns}, author={Kahana, Michael J and Seelig, David and Madsen, Joseph R}, journal={Current opinion in neurobiology}, volume={11}, number={6}, pages={739--744}, year={2001}, publisher={Elsevier}, notes={I. Rodent Theta a. Behavioral correlates - locomotion, orientiing, voluntary behaviors b. LTP & Phase Reset - 'windows' LTP via phase reset phenom whereby theta phase reset locks TH in with stimulus presentation, as seen in event related potl (ERP) - P300 c. Role in spatial memory & coding place - theta phase procession d. Lesioning/removing TH reduces memory performance II. Human Theta a. ECoG allowed researchers in 2000's to confirm TH activity increases in humans in complex tasks such as 3D maze navigation, word recall (nonspatial task) III. Good Point- TH detection is human tasks doesnt proove its corr. w/ memory. It could just mean its corr. w/ attention }, } @article{axmacher06, desc={REVIEW: theta & gamma in memory formation}, title={Memory formation by neuronal synchronization}, author={Axmacher, Nikolai and Mormann, Florian and Fern{\'a}ndez, Guillen and Elger, Christian E and Fell, Juergen}, journal={Brain research reviews}, volume={52}, number={1}, pages={170--182}, year={2006}, publisher={Elsevier}, notes={I. Intro: synchronization by itself reduces amount of info in neuron ensambles, so whats the benefit? 2 things: strengthens syncd input onto target output cell 2) syncd inputs raise LTP in target output cell. II. Gamma Frequency Hypothesis a. gamma suggested to be critical because it has resonance with LTP, which is hypothesized to occur at 10-30ms (33-100Hz) 2.1 a. "gamma-interneuron hypothesis"- gamma caused by pyr-IN interactions & INs inhibit pyr cells and thus carve out precise time windows during which pyrs may fire b. "gamma phase procession"- more depolarized pyr cells fire earlier in gamma cycle. Thus pyr cells encode the strength of their synaptic inputs in their gamma phase! NOTE: this is still a hypothesis 2.2 a. only subset of cells syncd to th & gma. thus it has proved difficult to detect th & gama changes w/ behavioral task 2.3 Theory of DG/EC-->CA3 Gamma encoding: 1. increased gamma coherence in responce to stimulus (how this happens is open question) 2. Makes input (DG/EC) spikes fire at same gamma phase as output (CA3) spikes, which induces LTP 3. memorys encoded as I/O synapse strengts in LTP -NOTES: involves CA# recurrent conn. works w/ weak inputs; 'hebbian' (I & O dept) plasticity III. Theta-Gamma Hypothesis 3.2 THETA RESET- there is a reset to a fixed theta phase across hippo-EC after stimulus presentation, which leads to TH sync & learning in following steps 1. inputs arrive to output at random TH phases & thus cant elicit APs/LTP 2. stimulus occurs & TH reset happens, now all inputs arrive in same peak TH phase to output, which leads to LTP -proof: it has been shown that inputs arriving at TH peak lead to LTP, while those at TH trough lead to LTD 3. after learning, the same inputs can elicit spikes in earlier in TH cycle to to the LTP -this can result in place cell TH phase procession -NOTES: thus TH dept synaptic plasticity is LTP between input & TH phase, not input & output as w/ gamma 3.3 Theta thus responds to stimulus, while gamma does not. Their comodulation/interaction can thus lead to answering our Q in 2.3.1 (ie how gamma coh results from stimuli pres) IV. Sharp Waves & HFOs -sharp waves: "During immobility and consummatory behavior as well as during slow-wave sleep, the recurrent collaterals of the CA3 region of the hippocampus are released from subcortical inhibition so that activation of a small subset of CA3 cells gives rise to synchronous population discharges" -physiological sharp waves are 100-200Hz, epileptic HFOs are 200-500Hz these are replayed at sleep to consolidate memory....} } @article{fell03, desc={saw increased theta in human memory task}, title={Rhinal--hippocampal theta coherence during declarative memory formation: interaction with gamma synchronization?}, author={Fell, Juergen and Klaver, Peter and Elfadil, Hakim and Schaller, Carlo and Elger, Christian E and Fern{\'a}ndez, Guill{\'e}n}, journal={European Journal of Neuroscience}, volume={17}, number={5}, pages={1082--1088}, year={2003}, publisher={Wiley Online Library}, notes={Method: 1. Recorded hippo & EC in HUMANS w/ multicontact depth electrodes (see Fried paper). recorded from both lobes. 2. performed word recall task 3. Had method to check if coh. increase coming from amplitude or phase (see pg. 1085). Dont really get this 3. review ANOVA, phase syncronization Results:1. Found TH coh between hippo & EC increased during successful recall. NOTE: ANOVA showed all freqs-coh increased but not any specific one (ie theta) was signif. However t-test showed only theta coh signif. Discussion: 1. Dep. on stat test they either found coh. increased across all freqs (ANOVA) or only in TH (t-test)} } @article{fried12, desc={stim of EC can increase performance. TH reset}, title={Memory enhancement and deep-brain stimulation of the entorhinal area}, author={Suthana, Nanthia and Haneef, Zulfi and Stern, John and Mukamel, Roy and Behnke, Eric and Knowlton, Barbara and Fried, Itzhak}, journal={New England Journal of Medicine}, volume={366}, number={6}, pages={502--510}, year={2012}, publisher={Mass Medical Soc}, notes={Method: 1. stimulated EC & hippo w/ ECoG electrodes in patients w/ epilepsy during spatial learning task 2. Stimulated at 50 Hz for 5s on off Results:1. Saw that EC (but NOT hippo) stim produced memory improvement, on trials after (but less so during) stimulation 2. Saw that improved memory correlated well w/ theta 'phase resetting' aka difference in theta power in EPS wave before & after task 3. stim had no effects on 2 other behavioral tasks (pg507) Discussion: 1. speaks about possiblity of DBS device for alzheimers 2. suggests increased EC activity corr. w/ theta phase reseting which is critical to memory tasks (see Axmacher,06)} } @article{curran10, desc={REVIEW: theta & Gamma in episodic memory}, title={Functional role of gamma and theta oscillations in episodic memory}, author={Nyhus, Erika and Curran, Tim}, journal={Neuroscience \& Biobehavioral Reviews}, volume={34}, number={7}, pages={1023--1035}, year={2010}, publisher={Elsevier}, notes={REVIEW: theta & gamma in episodic (experiential) memory 3.1 - gives sources for idea that theta emerges from pyr. cell & SLOW IN interactions 4. SEE TABLE 1: GIVES SUMMARY OF EVERY HUMAN THETA EEG/ECoG STUDY DONE TO DATE!!!!!!! 4.1 - Hippocampal memory indexing theory - hippocampus binds diverse sensory inputs into unified episodic memory using LTP 4.1.1 - BBS (binding-by-synchrony) theory in visual system suggests visual objects encoded by gamma oscillations in visual cortex 4.1.2 - gamma & LTP - says its ~20ms time res. due to absorbtion of NMDA taking 20ms 4.1.3 - Theta & LTP - positive peak of theta leads to opening of NMDA channels which is critical to LTP -quotes a bunch of studies showing theta imporoves LTP in-vivo, in-vitro, in DG, CA1 -quotes a bunch of animal studies correlating higher theta w/ improved performance THE REST OF ARTICLE TALKED ABOUT COGNITIVE PSYCH THEORYU OF HOW THETA-GAMMA BINDING MAKE MEMORY} } @article{givens96, desc={1st description of theta-reset}, title={Stimulus-evoked resetting of the dentate theta rhythm: relation to working memory}, author={Givens, Bennet}, journal={Neuroreport}, volume={8}, number={1}, pages={159--163}, year={1996}, publisher={LWW}, notes={Method: 1. recorded rodent DG LFP during working/episodic (sim. to DNMS) memory task & referance memory (similar to long-term memory) task in rodents Results:1. saw that DG LFP averaged across trls (ie ERP) had consistent theta reset phase in WM task, but not ref. memory task (see fig2) Discussion: 1. suggests that theta reset ERP ideal to induce LTP since info comes in at trough. See givens,2004} } @article{givens04, desc={theta reset & LTP}, title={Theta reset produces optimal conditions for long-term potentiation}, author={McCartney, Holly and Johnson, Andrew D and Weil, Zachary M and Givens, Bennet}, journal={Hippocampus}, volume={14}, number={6}, pages={684--687}, year={2004}, publisher={Wiley Online Library}, notes={Method: 1. same as givens,96 but this time also HFS stimd DG at various points in theta ERP. Measured LTP as amplitude of stim EPSP Results:1. Saw that stim at ERP theta peak did induce LTP, but at trough, actually depotentiated EC>DG. See fig.2 & last paragraph} } @article{tesche00, desc={1st human theta reset study}, title={Theta oscillations index human hippocampal activation during a working memory task}, author={Tesche, CD and Karhu, J}, journal={Proceedings of the National Academy of Sciences}, volume={97}, number={2}, pages={919--924}, year={2000}, publisher={National Acad Sciences}, notes={Method: 1. recorded MEG from human patients doing sternberg (number recognition) task Results:1. observed theta-reset / ERP in L,R hippocampus -saw that L,R were different -saw that TH reset was longer when task was more difficult (ie more numbers had to be recognized) Discussion: 1. 1st human theta reset study} } @article{lisman95, desc={**theta & LTP}, title={Bidirectional synaptic plasticity induced by a single burst during cholinergic theta oscillation in CA1 in vitro}, author={Huerta, Patricio T and Lisman, John E}, journal={Neuron}, volume={15}, number={5}, pages={1053--1063}, year={1995}, publisher={Elsevier} } @article{moser13, desc={REVIEW: connection of spatial, episodic, & semantic memory w/ theta}, title={Memory, navigation and theta rhythm in the hippocampal-entorhinal system}, author={Buzs{\'a}ki, Gy{\"o}rgy and Moser, Edvard I}, journal={Nature neuroscience}, volume={16}, number={2}, pages={130--138}, year={2013}, publisher={Nature Publishing Group}, notes={ I. 2 types of navigation: path integration (based on calculating past positions and velocities & map-based/allocentric navigation based on landmarks (fig.1) A. Conn. of allocentric navigation & semantic memory 1. GRID CELLS- EC layer2 (& subiculum) cells which have periodic lattice-triangle firing feilds (fig2a) 2. unline place cells, grid cells active in ALL most environments, and their remapping only involves phase/offset changes. Thus they form a universal map -suggests that EC grid cells form ensemble input to downstream target of place cells. Thus, they allow each place cell to fire only at very specifc combos (ie places). SEE FIG. 3!!!!!!!!!!!!!!!!!! 3. suggests grid/place cell / hippo/EC network cant just we for nav. since (1) insects can do nav. w/ much less cells (2) only a dozen grid cells needed to define environment... a. thus this system can be used not only for relationships to map but also events/objects in SEMANTIC MEMORY b. further proof is (1) assymetry of free recall mirrors assymetry of place cells (ie reverse nav. doesnt have same fields). (2) temporal contiguity of recollection (ie if i jump your memory to event you can proceed from there much easier B. path integration & episodic memory 1. episodic memory is path integration through experiance/time 2. theta phase segregation of gamma cycles- suggests theta has function of seperating gamma bound neural ensembles from mixing w/ each other. See fig.6d NOTE: i didnt get the last p[art of the article. the main thing i got was connection of spatial & declaritive memory....} } %Gamma @article{buzsaki12gama, desc={REVIEW: gamma}, title={Mechanisms of gamma oscillations}, author={Buzs{\'a}ki, Gy{\"o}rgy and Wang, Xiao-Jing}, journal={Annual review of neuroscience}, volume={35}, pages={203--225}, year={2012}, publisher={Annual Reviews}, notes={I. Neural Assemblies- groups of pyr. neurons which together coform a gamma oscill. Within this timespan (10-30ms, also GABA,AMPA time constant & time for LTP induction), all inputs are like 1 event to downstream neuron. Outside it, the are seperate. See Fig1 & Buzsaki,2010 II. Gamma oscill. Models A. I-I models - synchronization via interneurons (w/o/ pyr) B. E-I models - sync. between pyr & INs. Explains gamma phase shift between pyr & IN in-vivo II. Cellular Mechanisms A. "Extracellularly recorded gamma waves brought about by synchronous IPSPs in pyr. cells, brought about by fast spiking [PV-basket] INs" -PV cell spiking corr. to gamma rhythm & they have gamma peak in their FFT (see fig 3a,b) -subthreshold gamma oscill. has reversal potl of Cl- (GABA-A) in pyr. cells -have inherent gamma resonance (Pike,2000) III. Gamma coherence in distal networks - evidence of gamma coh. in L/R hippo! See fig 6B. IV. Theta gamma synchrony correlated w/ behavioral task (diff. from TG-comodulation!) Interesting Sources: 1. Buzsaki,2010 idea of cell assemblies! 2. Jarvis & Mitra- spike trains FFT - Jarvis, M. R., and P. P. Mitra. "Sampling properties of the spectrum and coherency of sequences of action potentials." Neural Computation 13.4 (2001): 717-749.} } @article{fell01, desc={saw increased gamma in human memory task}, title={Human memory formation is accompanied by rhinal--hippocampal coupling and decoupling}, author={Fell, J{\"u}rgen and Klaver, Peter and Lehnertz, Klaus and Grunwald, Thomas and Schaller, Carlo and Elger, Christian E and Fern{\'a}ndez, Guill{\'e}n}, journal={Nature neuroscience}, volume={4}, number={12}, pages={1259--1264}, year={2001}, publisher={Nature Publishing Group}, notes={Method: 1. Same as Fell, 2003 Results:1. Saw decreased hippo gamma power & increased hippo-EC gamma coh in recalled words tasks} } @article{csicsvari03, desc={**gamma mechanisms in hippo}, title={Mechanisms of gamma oscillations in the hippocampus of the behaving rat}, author={Csicsvari, Jozsef and Jamieson, Brian and Wise, Kensall D and Buzs{\'a}ki, Gy{\"o}rgy}, journal={Neuron}, volume={37}, number={2}, pages={311--322}, year={2003}, publisher={Elsevier} } @article{mann05, desc={REVIEW: gamma mechanisms}, title={Mechanisms underlying gamma (‘40Hz’) network oscillations in the hippocampus—a mini-review}, author={Mann, Edward O and Paulsen, Ole}, journal={Progress in biophysics and molecular biology}, volume={87}, number={1}, pages={67--76}, year={2005}, publisher={Elsevier}, notes={I. Gamma Oscillations in-vivo 1. 2 Gamma generators: CA3 & DG A. DG gamma is driven by EC & abolished w/ EC lesions 2. CA3 gamma generation is inherent & entrains CA1 A. probably network activity between pyr cells & basket/axo-axonic INs B. Presumably goes CA3 --> CA1int -->CA1pyr. This is why CA1 pyr. earlier in gamma cycle than CA3 pyr. 3. LFP vs. individual neurons relationship A. interneurons do fire at gamma freq, however ussually pyr. cells only fire <5sps, HOWEVER pyr. ENSEMBLES fire together at gamma!!!!! B. CA3 recurrent conn. make pyr. cells recruit other pyr. cells w/ EPSP & then inhib by IPSP, thus making gamma. Thus its a multicell process (pg. 71 last comment) II. Gamma oscill. In-Vitro Induced by musc. ACh receptors (mAChR) which activate I-h & I-CAN (nonspecific cation current) III. Mechanisms: 1. INs do jave gamma resonance (Pike,2000). Also maybe coupled via gap junctions 2. caused by recurrent IN & pyr interactions} } @article{bibbig07, desc={*missed beat beta}, title={Beta rhythms (15--20 Hz) generated by nonreciprocal communication in hippocampus}, author={Bibbig, Andrea and Middleton, Steven and Racca, Claudia and Gillies, Martin J and Garner, Helen and LeBeau, Fiona EN and Davies, Ceri H and Whittington, Miles A}, journal={Journal of neurophysiology}, volume={97}, number={4}, pages={2812--2823}, year={2007}, publisher={Am Physiological Soc} } @article{colgin09, desc={**low CA1 gamma from CA3, high CA1 gamma from EC}, title={Frequency of gamma oscillations routes flow of information in the hippocampus}, author={Colgin, Laura Lee and Denninger, Tobias and Fyhn, Marianne and Hafting, Torkel and Bonnevie, Tora and Jensen, Ole and Moser, May-Britt and Moser, Edvard I}, journal={Nature}, volume={462}, number={7271}, pages={353--357}, year={2009}, publisher={Nature Publishing Group} } %Marmarelis/Berger Modeling @article{marm86, desc={minimum-order wiener model}, title={Minimum-order Wiener modelling of spike-output systems}, author={Marmarelis, VZ and Citron, MC and Vivo, CP}, journal={Biological cybernetics}, volume={54}, number={2}, pages={115--123}, year={1986}, publisher={Springer} } @article{marm93, desc={PDMs in neurosceince}, title={Modeling of neural systems by use of neuronal modes}, author={Marmarelis, VZ and Orme, ME}, journal={Biomedical Engineering, IEEE Transactions on}, volume={40}, number={11}, pages={1149--1158}, year={1993}, publisher={IEEE} } @book{marm04, desc={2004 BOOK}, title={Nonlinear dynamic modeling of physiological systems}, author={Marmarelis, Vasilis Z}, year={2004}, publisher={Wiley-Interscience} } @article{berger05, desc={}, title={Restoring lost cognitive function}, author={Berger, Theodore W and Ahuja, Ashish and Courellis, Spiros H and Deadwyler, Samuel A and Erinjippurath, Gopal and Gerhardt, Gregory A and Gholmieh, Ghassan and Granacki, John J and Hampson, Robert and Hsaio, Min Chi and others}, journal={Engineering in Medicine and Biology Magazine, IEEE}, volume={24}, number={5}, pages={30--44}, year={2005}, publisher={IEEE} } @article{song07, desc={Dong's 1st article on his MIMO model}, title={Nonlinear dynamic modeling of spike train transformations for hippocampal-cortical prostheses}, author={Song, Dong and Chan, Rosa HM and Marmarelis, Vasilis Z and Hampson, Robert E and Deadwyler, Sam A and Berger, Theodore W}, journal={Biomedical Engineering, IEEE Transactions on}, volume={54}, number={6}, pages={1053--1066}, year={2007}, publisher={IEEE} } @article{zanos08, desc={Zanos MIMO article}, title={Nonlinear modeling of causal interrelationships in neuronal ensembles}, author={Zanos, Theodoros P and Courellis, Spiros H and Berger, Theodore W and Hampson, Robert E and Deadwyler, Sam A and Marmarelis, Vasilis Z}, journal={Neural Systems and Rehabilitation Engineering, IEEE Transactions on}, volume={16}, number={4}, pages={336--352}, year={2008}, publisher={IEEE} } @article{song09par1, desc={}, title={Parametric and non-parametric modeling of short-term synaptic plasticity. Part I: Computational study}, author={Song, Dong and Marmarelis, Vasilis Z and Berger, Theodore W}, journal={Journal of computational neuroscience}, volume={26}, number={1}, pages={1--19}, year={2009}, publisher={Springer} } @article{song09par2, desc={}, title={Parametric and non-parametric modeling of short-term synaptic plasticity. Part II: Experimental study}, author={Song, Dong and Wang, Zhuo and Marmarelis, Vasilis Z and Berger, Theodore W}, journal={Journal of computational neuroscience}, volume={26}, number={1}, pages={21--37}, year={2009}, publisher={Springer} } @article{zanos09bool1, desc={Boolean modeling part 1: computational}, title={Boolean modeling of neural systems with point-process inputs and outputs. Part I: Theory and simulations}, author={Marmarelis, Vasilis Z and Zanos, Theodoros P and Berger, Theodore W}, journal={Annals of biomedical engineering}, volume={37}, number={8}, pages={1654--1667}, year={2009}, publisher={Springer} } @article{zanos09bool2, desc={Boolean modeling part 2: application}, title={Boolean modeling of neural systems with point-process inputs and outputs. Part II: Application to the rat hippocampus}, author={Zanos, Theodoros P and Hampson, Robert E and Deadwyler, Samuel E and Berger, Theodore W and Marmarelis, Vasilis Z}, journal={Annals of biomedical engineering}, volume={37}, number={8}, pages={1668--1682}, year={2009}, publisher={Springer} } @article{goonawar10, desc={**Goonawardena 2010 Cannabinoid article}, title={Cannabinoid and cholinergic systems interact during performance of a short-term memory task in the rat}, author={Goonawardena, Anushka V and Robinson, Lianne and Hampson, Robert E and Riedel, Gernot}, journal={Learning \& Memory}, volume={17}, number={10}, pages={502--511}, year={2010}, publisher={Cold Spring Harbor Lab} } @article{lu11, desc={volterra model of RIT(onto CA3-SC)->CA1. NO dynamic threshold}, title={Nonlinear dynamic modeling of synaptically driven single hippocampal neuron intracellular activity}, author={Lu, Ude and Song, Dong and Berger, Theodore W}, journal={Biomedical Engineering, IEEE Transactions on}, volume={58}, number={5}, pages={1303--1313}, year={2011}, publisher={IEEE}, notes={Method: 1. made volterra model of RIT(onto CA3-SC)->CA1. Half of RIT pulses generated AP. 2. Volterra model was 2rd order forward & 1st order back. Had trigger to superimpose stereotyped AP. Thus modeled sub&supra threshold 3. Metrics: NMSE (for cont9inous sig. compare), SPER (for ST compare). SPER=(FP+FN/total) 4. does global search for alpha,L on train/test set 5. M=300ms, bin=? L=3; a_forward=.972; a_back=.91; Results:1. K1CA1 biology. CA1 neuron length is 1mm. Says CA3 also synapse onto basal dendrites in str.ori 2. LSE vs MLE Q1: what is BW???? this is important for ROC results, alpha, etc... Interesting Sources: [55] - Sekerli,2004 - dynamic neruon threshold} } @article{lu12, desc={same as Lu,11 w/ dynamic nonlinear threshold}, title={Nonlinear Dynamic Modeling of Neuron Action Potential Threshold During Synaptically Driven Broadband Intracellular Activity}, author={Lu, Ude and Roach, Shane M and Song, Dong and Berger, Theodore W}, journal={Biomedical Engineering, IEEE Transactions on}, volume={59}, number={3}, pages={706--716}, year={2012}, publisher={IEEE}, notes={Method: 1. Same experiment as Lu,2011: made model of RIT(onto CA3-SC)->CA1. Half of RIT pulses generated AP. 2. Estimated RIT-->Threshold 2nd order Volterra model (M=2000ms), which he then combined w/ Lu,2011 model 3. Had method to estimate threshold trigger, both phenominologiclly (using 3rd derivative) and empirically (thresh is voltage at which sometimes spike, sometimes not) 4. interesting way of using k0=k0+f; f=offset to evaluate ROC plots (pg712) Results:1. resting CA1 threshold (k0) = 9mV? 2. dynamic threshold increased SPUR by 33% (on average)!!! Also greatly increased ROC area! 3. Saw threshold increases w/ recent AP. Discussion: 1. He has 2 feedback kernels: h, for afterpotential, and k for threshold dynamics. h is for potential output and k is for ST output. In ST-->ST models, h & k comined into 1 feedback kernel. That having been said, this shows that in ST-->ST models, h has 1)2nd order dynamics 2)fast (afterpotential) and slow (threshold) dynamics, thus 2 alphas! ***Q1: Why voltage set at 0 and why threshold so small! Q2. Did they try 1st order threhold nonlin!} } @article{marm12, desc={optimal stimulation & TLF article}, title={Design of optimal stimulation patterns for neuronal ensembles based on Volterra-type hierarchical modeling}, author={Marmarelis, VZ and Shin, DC and Hampson, RE and Deadwyler, SA and Song, D and Berger, TW}, journal={Journal of neural engineering}, volume={9}, number={6}, pages={066003}, year={2012}, publisher={IOP Publishing} } @article{berger12, desc={Berger neuroprosthetic results}, title={A hippocampal cognitive prosthesis: multi-input, multi-output nonlinear modeling and VLSI implementation}, author={Berger, Theodore W and Song, Dong and Chan, Rosa HM and Marmarelis, Vasilis Z and LaCoss, Jeff and Wills, Jack and Hampson, Robert E and Deadwyler, Sam A and Granacki, John J}, journal={Neural Systems and Rehabilitation Engineering, IEEE Transactions on}, volume={20}, number={2}, pages={198--211}, year={2012}, publisher={IEEE} } @article{hampson12, desc={Hampson neuroprosthetic results}, title={Facilitation and restoration of cognitive function in primate prefrontal cortex by a neuroprosthesis that utilizes minicolumn-specific neural firing}, author={Hampson, Robert E and Gerhardt, Greg A and Marmarelis, Vasilis and Song, Dong and Opris, Ioan and Santos, Lucas and Berger, Theodore W and Deadwyler, Sam A}, journal={Journal of neural engineering}, volume={9}, number={5}, pages={056012}, year={2012}, publisher={IOP Publishing} } @article{hampson12b, desc={FCTs in MIMO article}, title={Closing the loop for memory prosthesis: Detecting the role of hippocampal neural ensembles using nonlinear models}, author={Hampson, Robert E and Song, Dong and Chan, Rosa HM and Sweatt, Andrew J and Riley, Mitchell R and Goonawardena, Anushka V and Marmarelis, Vasilis Z and Gerhardt, Greg A and Berger, Theodore W and Deadwyler, Sam A}, journal={Neural Systems and Rehabilitation Engineering, IEEE Transactions on}, volume={20}, number={4}, pages={510--525}, year={2012}, publisher={IEEE} } @article{song09, desc={}, title={Nonlinear modeling of neural population dynamics for hippocampal prostheses}, author={Song, Dong and Chan, Rosa HM and Marmarelis, Vasilis Z and Hampson, Robert E and Deadwyler, Sam A and Berger, Theodore W}, journal={Neural Networks}, volume={22}, number={9}, pages={1340--1351}, year={2009}, publisher={Elsevier} } @article{marm13, desc={marm PDM article}, title={Nonlinear modeling of dynamic interactions within neuronal ensembles using Principal Dynamic Modes}, author={Marmarelis, Vasilis Z and Shin, Dae C and Song, Dong and Hampson, Robert E and Deadwyler, Sam A and Berger, Theodore W}, journal={Journal of computational neuroscience}, volume={34}, number={1}, pages={73--87}, year={2013}, publisher={Springer} } @article{marm13cl, desc={closed-loop cerebral hemodynamics model}, title={Closed-loop dynamic modeling of cerebral hemodynamics}, author={Marmarelis, VZ and Shin, DC and Orme, ME and Zhang, R}, journal={Annals of biomedical engineering}, volume={41}, number={5}, pages={1029--1048}, year={2013}, publisher={Springer}, notes={Method: 1. recorded MABP, CBFV (cer. blood vel), EDCO2 in 15 control subjects & got 2 global PDM systems 2. constructed 3 VARIABLE CL model!!!! EDCO2 was the '3rd variable' which was NOT predicted by model but a sort of 'predefined' 2nd input disturbance (see Fig2) 3. has some basic math showing nature of CL. see eq. 1-4 4. has good math of SiCL vs OL 5. good math on comparing w/ kinear system in laplace domain Results:1. in fig10 compared OL & SiCL reposnce to a pulse . showed that cL results are different! 2. compared reosnances only in CL mode. (fig11) 3. compared w/ 2 linear models (linear-PDM & traditional-linear) fig12 Discussion: 1. pg. 1034: suggests that you can 'measure' nonlin of system by checking coherence across frequencies (table1). Also checked residuals of each input in freq domain (fig8) 2. assessed influence of each input by explaatory power (%NMSE) they covered in output! 3. pg. 1030, talks about if CL true, whjat it would mean for OL results: "These previous studies have yielded useful results, but they do not take into account the “closed-loop” nature of the subject system—i.e., the fact that changes in blood flow may also affect the blood pressure of cerebral perfusion, and not only the other way around—which is expressed herein as our “working hypothesis.” If this working hypothesis becomes validated, the “open-loop” studies and the previously obtained results are not invalidated. However, they must be interpreted in the open-loop context, which is distinct from the “closed-loop” context discussed in this paper. We maintain that there is considerable scientific value (and potential clinical utility) in studying and understanding the closed-loop characteristics of the CFA-CVMR system, as elaborated below. This paper presents a general methodology with which physiological closed-loop systems can be studied in general and, in particular, it demonstrates its application to the closed-loop analysis (CLA) of the CFA-CVMR system of cerebral hemodynamics." 4. very open about lack of evidence either theoretical or validation: " The relative evaluation of the closed-loop results of these methods is not possible in the absence of a meaningful criterion, since the prediction errors based on the experimental data are the same in closed-loop and open-loop analysis. Possible differences between open-loop and CLA can be evaluated through correlation with clinical data in the case of a specific disease (not relevant to this study) or through model predictions of data collected in specialized experiments where specific external disturbances are imposed as stimuli (not available in this study)." " It is posited herein as a working hypothesis that the system of cerebral hemodynamics operates in “closed-loop” (i.e., changes in cerebral perfusion pressure affect cerebral blood flow and vice versa, and CO2 tension affects both blood flow and blood pressure). Since direct physiological evidence of such a closed-loop operation in cerebral hemodynamics is heretofore limited (e.g., the modulation of blood pressure by flow-mediated changes in cerebrovascular impedance and/or a possible “central baroreflex” in response to oscillations in blood flow), the presented CLA is intended to generate, rather than to test, our key working hypothesis of closed-loop operation in cerebral hemodynamics." 5. residual language: "whereby the residuals of each model prediction attain the role of “systemic disturbances” that essentially drive the closed-loop system in its spontaneous operation." 6. OL v CL: " However, the open-loop analysis may be misleading when a question is addressed that is affected by the closed-loop operation of this system (provided, of course, our working hypothesis is correct). The prediction of the internal (in-the-loop) variables for an externally imposed change in arterial pressure, blood flow or CO2 tension is significantly affected under the closed-loop arrangement, since the latter has effects that are “recycled” through the closed-loop system. For example, an imposed pulse change in the unobserved pressure disturbance signal, P d(t), of Fig. 2 may transform to a different waveform by the time it “shows up” as a change in the measured arterial pressure signal, P(t), that is the input to the forward model A" " We take the view that the effect of external or “imposed” changes must be studied in the context of closed-loop modeling to retain a realistic understanding of physiological states (e.g., for simulating the effect of a pulse change in arterial pressure or CO2 tension upon the cerebral flow velocity" " The closed-loop effects are also likely to affect the assessment of the impact of possible hemodynamic alterations caused by pathologies, clinical conditions or the administration of drugs. "} } @article{eikenberry13, desc={**stefens NARV model for HH dynamics}, title={A nonlinear autoregressive Volterra model of the Hodgkin--Huxley equations}, author={Eikenberry, Steffen E and Marmarelis, Vasilis Z}, journal={Journal of computational neuroscience}, volume={34}, number={1}, pages={163--183}, year={2013}, publisher={Springer} } @inproceedings{sandler13, desc={ME!!!!!!}, title={Closed-loop modeling of the hippocampus and design of neurostimulation patterns for suppressing seizures}, author={Sandler, R and Shin, DC and Song, D and Hampson, RE and Deadwyler, SA and Berger, TW and Marmarelis, VZ}, booktitle={Neural Engineering (NER), 2013 6th International IEEE/EMBS Conference on}, pages={1143--1146}, year={2013}, organization={IEEE} } @article{marm14, desc={2014 Primate global PDM article}, title={On parsing the neural code in the prefrontal cortex of primates using principal dynamic modes}, author={Marmarelis, Vasilis Z and Shin, Dae C and Song, Dong and Hampson, Robert E and Deadwyler, Sam A and Berger, Theodore W}, journal={Journal of computational neuroscience}, volume={36}, number={3}, pages={321--337}, year={2014}, publisher={Springer} } @article{sandler14pbv, desc={ME!!!!!!!!}, title={Probability-based nonlinear modeling of neural dynamical systems with point-process inputs and outputs}, author={Sandler, Roman and Song, Dong and Hampson, Robert E and Deadwyler, Sam A and Berger, Theodore and Marmarelis, Vasilis}, journal={BMC Neuroscience}, volume={15}, number={Suppl 1}, pages={P102}, year={2014}, publisher={BioMed Central Ltd} } %misc mathematical @book{higgins93, desc={nonparametric statistics book}, title={Intoruction to Modern Nonparametric Statistics}, author={Higgins, James J.}, year={2003} } @article{hanely1982, desc={ROC,MW equivalency}, title={The meaning and use of the area under a receiver operating characteristic (ROC) curve}, author={Hanley, James A and McNeil, Barbara J}, journal={Radiology}, pages={29-36}, year={1982} } @article{victor97, desc={Victor-Purpura distance}, title={Metric-space analysis of spike trains: theory, algorithms and application}, author={Victor, Jonathan D and Purpura, Keith P}, journal={Network: computation in neural systems}, volume={8}, number={2}, pages={127--164}, year={1997}, publisher={Informa UK Ltd UK} } @article{vanrossum01, desc={van-rossum distance}, title={A novel spike distance}, author={van Rossum, Mark CW}, journal={Neural Computation}, volume={13}, number={4}, pages={751--763}, year={2001}, publisher={MIT Press} } @article{brown11, desc={GLM Point-Process granger causality model}, title={A Granger causality measure for point process models of ensemble neural spiking activity}, author={Kim, Sanggyun and Putrino, David and Ghosh, Soumya and Brown, Emery N}, journal={PLoS computational biology}, volume={7}, number={3}, pages={e1001110}, year={2011}, publisher={Public Library of Science} } @article{barbieri13, desc={Barbieri's nonlinear volterra purely AR model of point-processes}, title={Point-process nonlinear models with Laguerre and Volterra expansions: instantaneous assessment of heartbeat dynamics}, author={Valenza, Gaetano and Citi, Luca and Scilingo, E and Barbieri, Riccardo}, year={2013}, publisher={IEEE}, notes={Method: 1. Made completely autoregressive model for point processes. Ie w/ no covariates a. Model composed of inhomogenous inverse gaussian (IIG) renewal process, where mean paramter g is determined by 2nd order volterra model exapnded on laguerre basis. Only covariates are signal history 2. Uses adaptive tracking algorithm to track paramters through time 3. He converts his nonlinear autoregressive (NAR) model to an equivalent inf order volterra model (think autoregessive filter can be inf FIR filter...) 4. uses bispectrum & trispectrum to probe higher order spectra...? } }