Index of /Neural nets/Grossberg/captions html

• Parent Directory
• 0_pCaptionL all.txt
• 0_pLog dir_pHtmlL_refc_to_pPngL.txt
• 0_pLog dir_pImgCapL_refc_to_pPngL.txt
• 2_pCaptionL problems.txt
• 3_pCaptionL test 1.01-1.07.txt
• 3_pCaptionL test 1.01-1.25.txt
• 3_pCaptionL test 1.26-17.06.txt
• 4_pImgCapL fix cover image 231019.txt
• 4_pImgCapL fixes 231019.txt
• cover image.html
• p002fig01.01 Seeing an object vs knowing what it is.html
• p002fig01.02 Dalmation in snow.html
• p003fig01.03 Amodal completion.html
• p004fig01.04 Kanizsa stratification: transparency images.html
• p008fig01.05 Noise-saturation dilemma: cell activity; current activity.html
• p009fig01.06 Primacy gradient of activity in a recurrent shunting OC-OS network.html
• p011fig01.07 signal function determines how initial activity pattern is transformed.html
• p012fig01.08 Sigmoidal signal: a hybrid of [same, slower, faster]-than-linear.html
• p013fig01.09 sigmoid signal: quenching threshold; contrast enhancement.html
• p016fig01.10 Minimal adaptive prediction: blocking and unblocking.html
• p016fig01.11 BU-TD mismatch -> orienting system -> nonspecific arousal.html
• p018fig01.12 Peak shift and behavioural contrast: prefer new experiences.html
• p019fig01.13 Affective circuits are organized into opponent channels.html
• p023fig01.14 gated dipole opponent process: sustained on-response; transient off-response.html
• p024fig01.15 READ circuit: REcurrent Associative Dipole.html
• p025fig01.16 Cognitive-Emotional-Motor (CogEM) model: sensory cortex, amygdala, PFC.html
• p025fig01.17 Sensory-drive heterarchy vs drive hierarchy.html
• p026fig01.18 Inverted-U behaviour vs arousal.html
• p027fig01.19 ventral What [percept, class], dorsal Where [spatial represent, action].html
• p029tbl01.01 complementary streams [visual boundary, what-where, perception & recognition, object tracking, motor target].html
• p030fig01.20 neo-cortex 6 layers: same canonical laminar design cart [vision, speech, cognition].html
• p030tbl01.02 complementary streams: What- [rapid, stable] learn invariant object categories, Where- [labile spatial, action] actions.html
• p032fig01.21 [Retina, LGNs, V[1,2,3,4], MT] to What & Where areas.html
• p035fig01.22 Presentation [normal, silence, noise replaced].html
• p036fig01.23 working memory [longer list, bigger chunk]s.html
• p037fig01.24 sentence [learn, store, class] via 3 streams.html
• p038fig01.25 ART Matching Rule stabilizes learning: [real time learn, object attention].html
• p039tbl01.03 [consciousness, movement] links: visual, auditory, emotional.html
• p042tbl01.04 six main resonances which support different kinds of conscious awareness.html
• p051fig02.01 laterial inhibition: darker appears darker; lighter appears lighter.html
• p052fig02.02 Adaptive Resonance reactivation: features bottom-up; categories top-down.html
• p057fig02.03 neuron basic [anatomy, physiology].html
• p058fig02.04 Learning a global arrow in time.html
• p059fig02.05 Effects of intertrial and intratrial intervals.html
• p059fig02.06 Bow due to backward effect in time.html
• p060fig02.07 Error gradients depend on list position.html
• p061fig02.08 neural networks can learn forward and backward associations.html
• p063fig02.09 Short Term Memory (STM): Additive Model.html
• p064fig02.10 STM Shunting Model, mass action in membrane equations.html
• p064fig02.11 MTM habituative transmitter gate; LTM gated steepest descent learning.html
• p065fig02.12 Three sources of neural network research: [binary, linear, nonlinear].html
• p068fig02.13 Hartline: lateral inhibition in limulus retina of horseshoe crab.html
• p068fig02.14 Hodgkin and Huxley: spike potentials in squid giant axon.html
• p071fig02.15 Noise-Saturation Dilemma: functional unit is a spatial activity pattern.html
• p071fig02.16 Noise-Saturation Dilemma:sensitivity to ratios of inputs.html
• p072fig02.17 Vision: brightness constancy, contrast normalization.html
• p072fig02.18 Vision: brightness contrast, conserve a total quantity, total activity normalization.html
• p073fig02.19 Computing in a bounded activity domain, Gedanken experiment.html
• p073fig02.20 Shunting saturation occurs when inputs get larger to non-interacting cells.html
• p073fig02.21 Shunting saturation: how shunting saturation turns on all of a cells excitable sites as input intensity increases.html
• p073fig02.22 Computing with patterns: how to compute the pattern-sensitive variable.html
• p074fig02.23 Shunting on-center off-surround network: no saturation! infinite dynamical range, conserve total activity.html
• p075fig02.24 Membrane equations of physiology: shunting equation, not additive.html
• p076fig02.25 Weber law, adaptation, and shift property, convert to logarithmic coordinates.html
• p076fig02.26 Mudpuppy retina neurophysiology, adaptation- sensitivity shifts for different backgrounds.html
• p077fig02.27 Mechanism: cooperative-competitive dynamics, subtractive lateral inhibition.html
• p077fig02.28 Weber Law and adaptation level: hyperpolarization vs silent inhibition.html
• p078fig02.29 Weber Law and adaptation level: adaptation level theory.html
• p078fig02.30 Noise suppression: attenuate zero spatial frequency patterns- no information.html
• p078fig02.31 Noise suppression -> pattern matching: mismatch (out of phase) suppressed, match (in phase) amplifies pattern.html
• p079fig02.32 Substrate of resonance: match (in phase) of BU and TD input patterns amplifies matched pattern due to automatic gain control by shunting terms.html
• p080fig02.33 How do noise suppression signals arise: symmetry-breaking during morphogenesis, opposites attract rule.html
• p080fig02.34 Symmetry-breaking: dynamics and anatomy.html
• p081fig02.35 Ratio contrast detector: reflectance processing, contrast normalization, discount illuminant.html
• p081fig02.36 [Noise suppression, contour detection]: uniform patterns are suppressed, contrasts are selectively enhanced, contours are detected.html
• p082fig02.37 Modelling method and cycle (brain): proper level of abstraction; cannot derive a brain in one step.html
• p085fig02.38 Modelling method and cycle, technological applications: at each stage [behavioural data, design principles, neural data, math model and analysis].html
• p087fig03.01 Emerging unified theory of visual intelligence: BU-TD interactions overcome complementary processing deficiencies.html
• p089fig03.02 What do you think lies under the two grey disks (on a checkers board).html
• p090fig03.03 Kanizsa square and reverse-contrast Kanizsa square precepts.html
• p091fig03.04 blind spot and veins can occlude light to the retina.html
• p092fig03.05 A cross-section of the retinal layers: light stimuli need to go through all retinal layers.html
• p093fig03.06 Every line is an illusion!: boundary completion, surface filling-in.html
• p094fig03.07 Complementary properties of boundaries and surfaces.html
• p095fig03.08 Computer simulation of a Kanizsa square percept.html
• p095fig03.09 Simulation of a reverse-contrast Kanizsa square percept.html
• p096fig03.10 The visual illusion of eon color spreading.html
• p096fig03.11 Another example of neon color spreading.html
• p098fig03.12 Einstein's face: [edges, texture, shading] are overlaid.html
• p100fig03.13 Ehrenstein percept weakened as lines deviate from perpendicular.html
• p100fig03.14 Perpendicular induction at line ends: [locally [,un], globally] preferred.html
• p100fig03.15 orientations: [transient before, equilibrium after] choice.html
• p102fig03.16 Ts and Ls group together based on shared orientations, not identities.html
• p102fig03.17 Positions of squares give rise to a percept of three regions.html
• p103fig03.18 different spatial arrangements of inducers: emergent [horizontal, diagonal] groupings, but inducers have vertical orientations.html
• p103fig03.19 [diagonal, perpendicular, parallel]: thats how multiple orientations can induce boundary completion of an object.html
• p104fig03.20 Sean Williams: how boundaries can form.html
• p104fig03.21 Four examples of how emergent boundaries can form in response to different kinds of images.html
• p105fig03.22 3D vision and figure-ground separation: [multiple-scale, depth-selective] boundary webs.html
• p105fig03.23 pointillist painting: Georges Seurat, A Sunday on la Grande Jatte.html
• p106fig03.24 Do these ideas work on hard problems: Synthetic Aperture Radar [discount illuminant, filling-in, boundaries].html
• p107fig03.25 Matisse, The Roofs of Collioure.html
• p107fig03.26 drawing directly in color leads to colored surface representations.html
• p108fig03.27 Matisse: Open Window, Collioure, [continuously, sparsely] indiced surfaces.html
• p108fig03.28 Baingio Pinna, Watercolor illusion filled-in regions bulge in depth, [multiple-scale, depth-selective] boundary web.html
• p109fig03.29 Chiaroscuro- Rembrandt self-portrait; Trompe l oeil- Graham Rust.html
• p109fig03.30 Jo Baer triptych: Primary Light Group [red, green, blue].html
• p110fig03.31 Henry Hensche painting: The Bather, is suffused with light.html
• p110fig03.32 Claude Monet painting: Poppies Near Argenteuil.html
• p112fig03.33 Boundary web gradient can cause self-luminosity, similar to watercolor illusion.html
• p112fig03.34 Examples of Ross Bleckner's self-luminous paintings.html
• p113fig03.35 Highest Luminance As White (HLAW) rule, Hans Wallach.html
• p113fig03.36 Blurred Highest Luminance As White (BHLAW) rule.html
• p114fig03.37 Perceived reflectance vs cross-section of visual field: anchored brightness, self-luminous.html
• p114fig03.38 Color field painting: Jules Olitski, spray paintings of ambiguous depth.html
• p115fig03.39 Gene Davis paintings [full color, monochromatic]: percepts of grouping and relative depth.html
• p116fig03.40 Mona Lisa by Leonardo da Vinci: T-junctions and perspective cues give strong percept of depth.html
• p117fig03.41 Boundary contours and feature contours- no inhibition, feature signals survive and spread.html
• p117fig03.42 Two paintings by Frank Stella.html
• p120fig03.43 Four paintings by Monet of the Rouen cathedral under different lighting conditions.html
• p120fig03.44 Rouen Cathedral at sunset (Monet 1892-1894): equiluminant, obscured and less depth.html
• p121fig03.45 Rouen Cathedral full sunlight (Monet 1892-1894): non-uniform lighting, more detail and depth.html
• p121fig03.46 Rouen Cathedral full sunlight (Monet 1892-1894): T-junctions greater depth.html
• p123fig04.01 Combining stabilized images with filling-in.html
• p124fig04.02 closed boundaries prevent brightness from flowing.html
• p126fig04.03 Color constancy: compute ratios, discount the illuminant, compute lightness.html
• p128fig04.04 reflectance changes at contours: fill-in illuminant-discounted colors.html
• p129fig04.05 reflectance changes at contours: color contours.html
• p129fig04.06 reflectance changes at contours: fill-in color; resolve uncertainty.html
• p130fig04.07 brightness constancy: boundary peaks spatially narrower than feature peaks.html
• p131fig04.08 brightness constancy: discount illuminant, ratio-sensitive feature contours.html
• p131fig04.09 Simulation of brightness contrast.html
• p132fig04.10 Simulation of brightness assimilation.html
• p132fig04.11 Simulation of double step and COCE.html
• p133fig04.12 Simulation of the 2D COCE.html
• p133fig04.13 Contrast constancy, relative luminances can be reversed, discounting illuminant.html
• p134fig04.14 Experiments on filling-in: in-the-act; simulation.html
• p138fig04.15 oriented filtering to grouping and boundary completion.html
• p139fig04.16 Simplest simple cell model: threshold linear, half-wave rectification.html
• p140fig04.17 Complex cells: pool like-oriented simple cells of opposite polarity.html
• p141fig04.18 Binocular Disparity to reconstruct depth from 2D retinal inputs.html
• p141fig04.19 Laminar cortical circuit for complex cells.html
• p142fig04.20 [, reverse-]Glass patterns give rise to different boundary groupings.html
• p143fig04.21 Hierarchical resolution of uncertainty for a given field size.html
• p144fig04.22 End Gap and End Cut simulation.html
• p145fig04.23 A perceptual disaster in the feature contour system.html
• p145fig04.24 Hierarchical resolution of uncertainty- End Cuts.html
• p146fig04.25 How are end cuts created: two stages of short-range competition.html
• p148fig04.26 End cut during neon color spreading via 2 stages.html
• p149fig04.27 Bipole cells boundary completion: long cooperation & short inhibition.html
• p150fig04.28 Bipole property: boundary completion via long-range cooperation.html
• p151fig04.29 bipole cells in cortical area V2: first neurophysiological evidence.html
• p151fig04.30 anatomy: horizontal connections in V1.html
• p152fig04.31 Bipoles through the ages.html
• p153fig04.32 Double filter and grouping network.html
• p156fig04.33 emergent boundary groupings can segregate textured regions.html
• p157fig04.34 texture: Boundary Contour System resolves errors of complex channels model.html
• p159fig04.35 Spatial impenetrability prevents grouping.html
• p159fig04.36 Graffiti art by Banksy: amodal boundary completion; spatial impenetrability.html
• p161fig04.37 Boundary Contour System model: analog-sensitive boundary completion Kanizsas.html
• p162fig04.38 Cooperation and competition during grouping.html
• p163fig04.39 LAMINART model explains key aspects of visual cortical anatomy and dynamics.html
• p164fig04.40 Koffka-Benussi ring.html
• p165fig04.41 Kanizsa-Minguzzi ring.html
• p166fig04.42 Computer simulation of Kanizsa-Minguzzi ring percept.html
• p167fig04.43 T-junction sensitivity: image, Bipole cells, boundary.html
• p168fig04.44 main [boundary, surface] formation stages: LGN-> V1-> V2-> V4.html
• p168fig04.45 ON and OFF feature contours: filled-in regions when adjacent to boundary.html
• p170fig04.46 regions can fill-in feature contour inputs when [adjacent to, collinear with] boundary contour inputs.html
• p170fig04.47 A double-opponent network processes output signals from FIDOs.html
• p171fig04.48 closed boundaries -> filling-in; open boundaries -> color spread.html
• p171fig04.49 DaVinci stereopsis and occlusion.html
• p173fig04.50 closed boundary at prescribed depth: addition of [bi, mon]ocular boundaries.html
• p174fig04.51 figure-ground separation, complementary consistency [boundaries, surfaces].html
• p174fig04.52 Stereogram surface percepts: surface lightnesses are segregated in depth.html
• p176fig04.53 OC-OS [within position, across depth]: brighter Kanizsas look closer.html
• p178fig04.54 figure-ground separation: bipole cooperation and competition.html
• p178fig04.55 Amodal completion of boundaries and surfaces in V2.html
• p179fig04.56 Visible surface 3D perception: boundary enrichment, surface filling-in.html
• p181fig04.57 relative contrasts induce: unimodal and bistable transparency; or flat 2D surface.html
• p182fig04.58 LAMINART explains many percepts of transparency.html
• p186fig05.01 Learn many-to-one (compression, naming), one-to-many (expert knowledge) maps.html
• p186fig05.02 Many-to-one map, two stage compression: [visual, auditory] categories.html
• p186fig05.03 Many-to-one map: IF-THEN rules: [symptom, test, treatment]s; length of stay.html
• p189fig05.04 hippocampus & several brain regions [learn, remember] throughout life.html
• p192fig05.05 LGN [ON, OFF] cells respond differently to [side, end]s of lines.html
• p192fig05.06 BU-TD circuits between the LGN and cortical area V1, ART Matching Rule.html
• p193fig05.07 detailed connections between [retinal ganglion cells, LGN, V1].html
• p193fig05.08 LGN [activation, inhibition], with[, out] top-down feedback.html
• p194fig05.09 [feature, boundary] contours from Ehrenstein disk stimulus.html
• p198fig05.10 Competitive learning and Self-Organized Maps (SOMs).html
• p199fig05.11 Instar learning: bottom-up adaptive filter for feature patterns.html
• p200fig05.12 Duality of [outstar, instar] networks.html
• p200fig05.13 Expectations focus attention: instar BU filters, outstar TD expectations.html
• p200fig05.14 Outstar learning, both [in, de]creases for LTM to learn STM pattern.html
• p201fig05.15 Spatial learning pattern, outstar learning.html
• p202fig05.16 Geometry of choice and learning, classifying vector.html
• p202fig05.17 Geometry of choice and learning, trains the closest LTM vector.html
• p205fig05.18 catastrophic forgetting due to [competition, associative] learning.html
• p207fig05.19 ART: [attentional, orienting] systems learn novel categories, no catastophic forgetting.html
• p211fig05.20 [PN match, N200 mismatch] computationally complementary potentials.html
• p211fig05.21 ART predicted correlated P120-N200-P300 ERPs during oddball learning.html
• p213fig05.22 If inputs incorrectly activate a category, how to correct the error.html
• p213fig05.23 A [category, symbol, other] cannot determine whether an error has occurred.html
• p214fig05.24 Learning top-down expectations occurs during bottom-up learning.html
• p214fig05.25 Error correction: [learn, compare] TD-BU inputs, Processing Negativity ERP.html
• p214fig05.26 Mismatch triggers nonspecific arousal, N200 ERP from orienting system.html
• p215fig05.27 Every event has [specific attentional cue, nonspecific orienting arousal].html
• p215fig05.28 BU+TD mismatch arousal and reset if degree of match < ART vigilance.html
• p220fig05.29 Vigilance [excitation: search better match, inhibition: resonance & learning].html
• p221fig05.30 predictive error -> vigilance increase just enough -> minimax learning.html
• p221fig05.31 Fuzzy ARTMAP can associate categories between ART networks, minimax learn.html
• p224fig05.32 Learning the alphabet with two different levels of vigilance.html
• p225fig05.33 Some early ARTMAP benchmark studies (no image - link instead).html
• p225fig05.34 ARTMAP learned maps of natural terrains better than AI expert systems.html
• p226fig05.35 Code instability sequences: [competitive learning, self-organizing map].html
• p226fig05.36 catastrophic forgetting without ART Matching Rule due to superset recoding.html
• p228fig05.37 neurotrophic Spectrally Timed ART (nSTART) model.html
• p230fig05.38 Synchronous Matching ART (SMART) spiking neurons in laminar cortical hierarchy.html
• p231fig05.39 SMART: vigilance increase via nucleus basalis of Meynert acetylcholine.html
• p232fig05.40 SMART generates γ oscillations for good match; β oscillations for bad match.html
• p232fig05.41 mismatch reset interlaminar events sequence [data, SMART predictions].html
• p233fig05.42 Evidence for the [gamma, beta] prediction in 3 parts of the brain.html
• p236fig05.43 nucleus basalis of Meynert releases ACh, reduces AHP, increases vigilance.html
• p240fig05.44 The 5-4 category structure example: ART learns the same kinds of categories as human learners.html
• p240fig05.44 models using only local computations look like an ART prototype model.html
• p242fig05.46 Distributed ARTMAP variants learn the 5-4 category structure.html
• p245fig05.47 [long-range excitatory, short-range disynaptic inhibitory] connections realize the bipole grouping law.html
• p246fig05.48 LAMINART model: BU adaptive filtering, horizontal bipole grouping, TD attentional matching.html
• p248fig05.49 LAMINART explains Up and Down states during slow wave sleep, ACh dynamics.html
• p252fig06.01 surface-shroud resonance forms as objects bid for spatial attention.html
• p253fig06.02 Surface-shroud resonance BU-TD OC-OS: perceptual surfaces -> competition -> spatial attention.html
• p254fig06.03 ARTSCAN Search model learns to recognize and name invariant object categories.html
• p255fig06.04 The ARTSCAN Search for a desired target object in a scene: Wheres Waldo.html
• p257fig06.05 Spatial attention flows along object boundaries: Macaque V1.html
• p258fig06.06 Neurophysiological data & simulation: attention can flow along a curve.html
• p258fig06.07 Top-down attentional spotlight becomes a shroud.html
• p259fig06.08 dARTSCN spatial attention hierarchy [Fast Where, Slow What] stream.html
• p260fig06.09 Crowding: visible objects & confused recognition, increased flanker spacing at higher eccentricity.html
• p260fig06.10 cortical magnification transforms coordinates: artesian (retina) to log polar (V1).html
• p261fig06.11 Crowding: visible objects and confused recognition.html
• p261fig06.12 A more serial search is needed due to overlapping conjunctions of features.html
• p265fig06.13 basal ganglia gate perceptual, cognitive, emotional, etc through parallel loops.html
• p267fig06.14 Perceptual consistency and figure-ground separation.html
• p268fig06.15 saccades within an object: figure-ground outputs control eye movements via V3AA.html
• p270fig06.16 Predictive remapping of eye movements, from V3A to LIP.html
• p271fig06.17 Persistent activity in IT to [view, position, size]-invariant category learning by positional ARTSCAN.html
• p272fig06.18 pARTSCAN: positionally-invariant object learning.html
• p272fig06.19 persistent activity needed to learn positionally-invariant object categories.html
• p273fig06.20 pARTSCAN simulation of Li & DiCarlo IT cell swapping data.html
• p274fig06.21 pARTSCAN [position invariance, selectivity] trade-off of Zoccolan etal 2007.html
• p274fig06.22 pARTSCAN: IT cortex processes image morphs with high vigilance.html
• p275fig06.23 IT responses to image morphs, data vs model.html
• p275fig06.24 Sterogram surface percepts: surface lightnesses are segregated in depth.html
• p276fig06.25 saccades: predictive gain fields [binocular fusion, filling-in of surfaces].html
• p277fig06.26 Predictive remapping maintains binocular boundary fusion as eyes move.html
• p278fig06.27 knowing vs seeing resonances: What [knowing, feature-prototype], Where [seeing, surface-shroud].html
• p278fig06.28 knowing vs seeing resonances: visual agnosia- reaching without knowing.html
• p283fig07.01 Boundary competition: spatial habituative gates, orientation gated dipole, bipole grouping.html
• p284fig07.02 Persistence decreases with flash illuminance & duration [data, simulations].html
• p285fig07.03 Persistence decrease: rebound to input offset inhibits bipole cells.html
• p286fig07.04 Illusory contours persist longer than real contours.html
• p286fig07.05 Illusory contours inhibited by OFF cell rebounds, propagate to center.html
• p287fig07.06 Persistence: [less, more] as adaptation orientation [same, orthogonal].html
• p287fig07.07 Persistence increases with distance, due to weaker spatial competition in hypercomplex cells.html
• p290fig08.01 Motion pools contrast-sensitive information moving in the same direction.html
• p291fig08.02 Complex cells respond to motion: opposite [direction, contrast polarities].html
• p292fig08.03 Visual aftereffects: [form- MacKay 90 degree, motion- waterfall 180].html
• p293fig08.04 Local vs overall motion: aperture problem of EVERY neurons receptive field.html
• p295fig08.05 sparse feature tracking signals [capture ambiguous, determine perceived] motion direction.html
• p296fig08.06 Simplest example of apparent motion: two dots turning on and off.html
• p296fig08.07 continuous motion illusions: [Beta with, Phi without] percept.html
• p297fig08.08 Delta motion when [luminance, contrast] of flash 2 is larger than flash 1.html
• p297fig08.09 motion in opposite directions perceived when 2 later flashes on either side of 1st flash.html
• p298fig08.10 motion speed-up perceived when flash duration decreases.html
• p298fig08.11 illusory contours: double illusion in V1-V2, motion V2-MT interaction.html
• p300fig08.12 Single flash: Gaussian receptive fields, recurrent OC-OS winner-take-all.html
• p300fig08.13 Nothing moves: [single flash, exponential decay], Gaussian peak fixed.html
• p300fig08.14 Visual inertia: flash decay after the flash shuts off.html
• p301fig08.15 two flashes: cell activation by first waning while second one is waxing.html
• p301fig08.16 sum Gaussian flash activity profiles: [waning 1st, waxing 2nd] -> travelling wave.html
• p302fig08.17 maximum long-rang apparent motion: Gaussian kernel spans successive flashes.html
• p302fig08.18 G-wave theorem 1: wave moves continuously IFF L <= 2*K.html
• p303fig08.19 No motion vs motion at multiple scales: flash distance L, Gaussian width K.html
• p303fig08.20 G-wave theorem 2: [speed-up, scale] independent of [distance, scale size].html
• p304fig08.21 Equal half-time property: multiple scales generate motion percept.html
• p304fig08.22 Korte Laws: ISIs in the hundreds of milliseconds can cause apparent motion.html
• p305fig08.23 Ternus motion: ISI [small- stationary, intermediate- element, larger- group].html
• p305fig08.24 Reverse-contrast Ternus motion: ISI [small- stationarity, intermediate- group (not element!), larger- group] motion.html
• p306fig08.25 Motion BCS model [explain, simulate]s long-range motion percepts.html
• p306fig08.26 3D FORMOTION model: track objects moving in depth.html
• p307fig08.27 Ternus motion: [element- weak, group- strong] transients, element [visual persistence, perceived stationarity].html
• p308fig08.28 Ternus group motion: Gaussian filter of 3 flashes forms one global maximum.html
• p310fig08.29 when individual component motions combine, their perceived direction & speed changes.html
• p311fig08.30 3D FORMOTION model: feature tracking [get directional, inhibit inconsistent] signals.html
• p311fig08.31 Motion BCS stages: locally ambiguous motion signals -> globally coherent percept, solving the aperture problem.html
• p312fig08.32 Schematic of motion filtering circuits.html
• p312fig08.33 Processing motion signals by a population of speed-tuned neurons.html
• p314fig08.34 VISTARS navigation model: FORMOTION front end for navigational circuits.html
• p315fig08.35 How to select correct direction and preserve speed estimates.html
• p316fig08.36 Motion capture by directional grouping feedback.html
• p317fig08.37 Motion capture by directional grouping feedback: [short, long]-range filters, transient cells.html
• p319fig08.38 Solving the aperture problem takes time.html
• p320fig08.39 Simulation of the barberpole illusion direction field at two times.html
• p321fig08.40 [, in]visible occluders [do, not] capture boundaries they share with moving edges.html
• p322fig08.41 motion transparency: asymmetry [near, far], competing opposite directions.html
• p323fig08.42 Chopsticks: motion separation in depth via [, in]visible occluders [display, percept].html
• p324fig08.43 ambiguous X-junction motion: MT-MST directional grouping bridges the ambiguous position.html
• p325fig08.44 The role of MT-V1 feedback: [motion-form feedback, bipole boundary completion.html
• p325fig08.45 Closing formotion feedback loop [MT, MST]-to-V1-to-V2-to-[MT, MST].html
• p326fig08.46 How do we perceive relative motion of object parts.html
• p327fig08.47 Two classical examples of part motion: Symmetrically moving inducers; Duncker wheel.html
• p328fig08.48 vector decomposition: (retinal - common = part) motion.html
• p328fig08.49 What is the mechanism of vector decomposition, prediction: directional peak shift.html
• p329fig08.50 How is common motion direction computed? retinal motion-> bipole grouping (form stream)-> V2-MT formotion.html
• p329fig08.51 Large and small scale boundaries differentially form illusory contours.html
• p330fig08.52 Correct motion directions after the peak shift top-down expectation acts.html
• p330fig08.53 Simulation of the various directional signals of the left dot through time.html
• p331fig08.54 Motion directions of a single dot moving slowly along a cycloid curve through time.html
• p331fig08.55 Duncker Wheel, large: stable rightward motion at the center captures motion at the rim.html
• p332fig08.56 Duncker Wheel, small: wheel motion as seen when directions are collapsed.html
• p332fig08.57 MODE (MOtion DEcision) model: Motion BCS -> saccadic target selection -> basal ganglia.html
• p333fig08.58 LIP responses during RT task correct trials: coherence and [activation, inhibition].html
• p334fig08.59 LIP responses for FD task: predictiveness decreases with increasing coherence.html
• p334fig08.60 [RT, FD] task behavioral data: more coherence in the motion causes more accurate decisions.html
• p335fig08.61 RT task behavioural data: reach time (ms) vs % coherence.html
• p335fig08.62 LIP encodes not only where, but also when, to move the eyes - No Bayes.html
• p338fig09.01 optic flow through brain regions: moving observer [navigate, track] moving object.html
• p338fig09.02 Heading (focus of velocity field) from optic flow: humans accurate +- 1 to 2 degrees.html
• p339fig09.03 Heading with [body move, eye rotate, combined] -> optic flow [expand, translate, rotate].html
• p339fig09.04 How can translation flow (eye rotation) be subtracted from spiral flow to recover the expansion flow.html
• p340fig09.05 efference copy command: may use outflow movement commands to eye muscles.html
• p340fig09.06 Corollary discharges from outflow movement commands that move muscles.html
• p340fig09.07 Log polar remapping of optic flow: [expansion, circular] motion maps to single direction.html
• p341fig09.08 optic flows [retina, V1, MT, MSTd, parietal cortex], V1 log polar mapping.html
• p341fig09.09 MSTd cells are sensitive to [spiral, rotation, expansion] motion.html
• p342fig09.10 Retina -> log polar -> MSTd cell, heading eccentricity.html
• p342fig09.11 importance of efference copy in real movements.html
• p343fig09.12 two retinal views of the Simpsons: [separate, recognize] overlapping figures.html
• p343fig09.13 How do our brains figure out which views belong to which pear.html
• p344fig09.14 Heading sensitivity unimpaired: MT tuning width 38°, MSTd spiral tuning 61°.html
• p345fig09.15 MT double opponent directional fields: relative motions [objects, backgrounds].html
• p346fig09.16 macrocircuit of 13 brain regions used to move the eyes.html
• p347fig09.17 leftward eye movement model: retina-> MT-> MST[v,d]-> pursuit.html
• p347fig09.18 MST[v,d] circuits enable predictive target tracking by the pursuit system.html
• p348fig09.19 MSTv cells: target speed on retina, background speed on retina, pursuit speed command.html
• p349fig09.20 Steering from optic flow: goals are attractors, obstacles are repellers.html
• p349fig09.21 Steering dynamics goal approach: [obstacle, goal, heading] -> steering.html
• p350fig09.22 negative Gaussian of an obstacle: avoid obstacle without losing sight of goal.html
• p350fig09.23 Unidirectional transient cells: [lead, trail]ing boundaries, driving video.html
• p351fig09.24 Directional transient cells respond most to motion in their preferred directions.html
• p351fig09.25 M+ computes global motion estimate from noisy local motion estimates.html
• p352fig09.26 heading direction final stage: beautiful optic flow, accuracy matches humans.html
• p354fig10.01 [Top-down attention, folded feedback] supports predicted ART Matching Rule.html
• p355fig10.02 seeing vs knowing distinction is difficult because they interact so strongly.html
• p356fig10.03 Laminar computing: [self-stabilize learning, fuse [BU pre-,TD]attentive processing, perceptual grouping no analog sensitivity].html
• p357fig10.04 Laminar Computing: combines feed[forward, back], [analog, digital], [pre,]attentive learning.html
• p359fig10.05 Activation of V1 by direct excitatory signals from LGN to layer 4 of V1.html
• p359fig10.06 Why another layer 6-to-4 signal: on-center off-surround.html
• p359fig10.07 Together [LGN-to-4 path, 6-to-4 OC-OS] do contrast normalization if cells obey shunting or membrane equation dynamics.html
• p360fig10.08 [IC 6-to-4, BU-OS LGN-to-6-to-4] excitations BOTH needed to activate layer 4, ART Matching Rule.html
• p360fig10.09 Grouping starts in layer 2-3: long-range horizontal excitation, short-range inhibition of target pyramidal.html
• p361fig10.10 Bipole property controls perceptual grouping: inputs [excitatory sum, inhibitory normalize].html
• p362fig10.11 Final grouping: folded feedback, strongest enhanced on-center, weaker suppressed off-surround, interlaminar functional columns.html
• p363fig10.12 V2 repeats V1 circuitry at larger spatial scale.html
• p364fig10.13 6-to-4 decision circuit common to [BU adaptive filter, intracortical grouping, top-down intercortical attention].html
• p364fig10.14 Explanation: grouping and attention share the same modulatory decision circuit.html
• p367fig10.15 Attention protects target from masking stimulus.html
• p367fig10.16 Flankers can enhance or suppress targets.html
• p368fig10.17 Attention has greater effect on low contrast targets.html
• p368fig10.18 Texture reduces response to a bar: [iso-orientation, perpendicular] suppression.html
• p369fig10.19 Unconscious learning of motion direction, without [extra-foveal attention, awareness] of stimuli.html
• p371fig11.01 FACADE theory explains how the 3D boundaries and surfaces are formed to see the world in depth.html
• p372fig11.02 3D surface filling-in of [lightness, color, depth] by a single process: FACADE.html
• p373fig11.03 Both [contrast-specific binocular fusion, contrast-invariant boundary perception] are needed to see the world in depth.html
• p374fig11.04 Three processing stages of [monocular simple, complex] cells.html
• p374fig11.05 Contrast constraint on binocular fusion: only contrasts which are derived from the same objects in space are binoculary matched.html
• p375fig11.06 Binocular fusion by obligate cells in V1-3B when =[left,right] contrasts.html
• p375fig11.07 3D LAMINART: [mo, bi]nocular simple cells binocularly fuse like image contrasts.html
• p376fig11.08 Correspondance problem: How does the brain inhibit false matches? contrast constraint not enough.html
• p376fig11.09 V2 disparity filter solves correspondence problem: false matches suppressed by line-of-sight inhibition.html
• p376fig11.10 3D LAMINART with disparity filter: 3D boundary representations via bipole grouping cells.html
• p377fig11.11 DaVinci stereopsis: monocular information and depth percept.html
• p378fig11.12 3D LAMINART: V2 monocular+binocular line of sight inputs -> depth perception.html
• p379fig11.13 3D LAMINART, DaVinci stereopsis (occlusion): emergent from simple mechanisms working together.html
• p380fig11.14 3D LAMINART, DaVinci stereopsis (polarity): same explanation as occlusion.html
• p381fig11.15 DaVinci stereopsis variant of (Gillam, Blackburn, Nakayama 1999): same mechanisms.html
• p382fig11.16 DaVinci stereopsis of [3 narrow, one thick] rectangles: same explanation.html
• p383fig11.17 Venetian blind effect: [left, right] eye matching bars.html
• p384fig11.18 Venetian blind effect: Surface[, -to-boundary] surface contour signals.html
• p385fig11.19 Dichoptic masking: [left, right] images have sufficiently different contrasts.html
• p385fig11.20 Dichoptic masking, Panum's limiting case: simplified version of Venetian blind effect.html
• p386fig11.21 Craik-O'Brien-Cornsweet Effect: 2D surface at a very near depth.html
• p387fig11.22 Julesz stereogram: boundaries with[out, ] surface contour feedback.html
• p388fig11.23 Sparse stereogram, large regions of ambiguous white: correct surface in depth.html
• p388fig11.24 depth-ambiguous feature contours: boundary groups lift to correct surface in depth.html
• p389fig11.25 Boundaries: not just edge detectors, or a shaded ellipse would look [flat, uniformly gray].html
• p390fig11.26 Multiple-scale depth-selective groupings determine perceived depth.html
• p391fig11.27 Multiple-scale grouping and size-disparity correlation.html
• p391fig11.28 Ocular dominance columns, LGN mappings into layer 4C of V1.html
• p392fig11.29 3D vision figure-ground separation: multiple-scale, depth-selective boundary webs.html
• p392fig11.30 How multiple scales vote for multiple depths, scale-to-depth and depth-to-scale maps.html
• p393fig11.31 LIGHTSHAFT model: determining depth-from-texture percept.html
• p393fig11.32 Kulikowski stereograms: binocular matching of out-of-phase [Gaussians, rectangles].html
• p394fig11.33 Kaufman stereogram: simultaneous fusion and rivalry.html
• p395fig11.34 3D LAMINART vs 7 other rivalry models: stable vision and rivalry.html
• p396fig11.35 Three properties of bipole boundary grouping in V2: boundaries oscillate with rivalry-inducing stimuli.html
• p397fig11.36 temporal dynamics of [rivalrous, coherent] boundary switching.html
• p398fig11.37 Simulation of the no swap baseline condition (Logothetis, Leopold, Sheinberg 1996).html
• p399fig11.38 Simulation of the swap condition of (Logothetis, Leopold, Sheinberg 1996).html
• p399fig11.39 Simulation of the eye rivalry data of (Lee, Blake 1999).html
• p400fig11.40 How do ambiguous 2D shapes contextually define a 3D object form.html
• p401fig11.41 3D LAMINART: [angle, disparity-gradient] cells learn 3D representations.html
• p401fig11.42 hypothetical cortical hypercolumn: how [angle, disparity-gradient] cells may self-organize during development.html
• p402fig11.43 A pair of disparate images of a scene from the University of Tsukuba.html
• p402fig11.44 3D LAMINART disparities [5, 6, 8, 10, 11, 14]: images of objects in common depth planes.html
• p403fig11.45 SAR processing by multiple scales: reconstruction of a SAR image.html
• p405fig12.01 [What ventral, Where-How dorsal] cortical streams for [audition, vision].html
• p406fig12.02 Three S's of movement: Synergy formation, muscle Synchrony, volitional Speed.html
• p407fig12.03 Motor cortical cells: vectors for [direction, length] of commanded movement.html
• p409fig12.04 VITE simulations: difference vector emergent from network interactions.html
• p410fig12.05 VITE: velocity profile invariance [short, long] movements for same GO signal.html
• p410fig12.06 Monkeys transform movement: 2 -> 10 o'clock target, 50 or 100 msec after activation of 2 o'clock target.html
• p411fig12.07 VITE: higher peak velocity due to target switching.html
• p411fig12.08 GO signals gate agonist-antagonist [difference, present position] vector processing stages.html
• p412fig12.09 Vector Associative Map: difference vector mismatch learning calibrates [target, present] position vectors.html
• p413fig12.10 VITE: cortical area [4,5] combine [trajectory, inflow] signals from [spinal cord, cerebellum] for [variable loads, obstacles].html
• p414fig12.11 [data, simulation]s from cortical areas 4 and 5 during a reach.html
• p415fig12.12 [VITE, FLETE, cerebellar, opponent muscle] model for trajectory formation.html
• p416fig12.13 DIRECT model: Endogenous Random Generator learns volitional reaches.html
• p416fig12.14 DIRECT reaches [unconstrained, with TOOL, elbow@140°, blind].html
• p417fig12.15 From Seeing & Reaching (DIRECT) to Hearing & Speaking (DIVA): homologous circular reactions, [tool use, coarticulation].html
• p418fig12.16 Anatomy of DIVA model processing stages.html
• p419fig12.17 Auditory continuity illusion: backwards in time through noise, ART Matching Rule.html
• p420fig12.18 ARTSTREAM: auditory continuity illusion, stream as a spectral-pitch resonance.html
• p422fig12.19 ARTSTREAM: derive streams from [pitch, source direction].html
• p423fig12.20 SPINET: log polar spatial sound frequency spectrum to distinct auditory streams.html
• p424fig12.21 Pitch shifts with component shifts, pitch vs lowest harmonic number.html
• p424fig12.22 Decomposition of a sound in terms of three of its harmonics.html
• p425fig12.23 ARTSTREAM: auditory continuity illusion- continuity does not occur without noise.html
• p426fig12.24 Spectrograms of -ba- and -pa- show the transient and sustained parts of their spectrograms.html
• p428fig12.25 ARTSPEECH: auditory-articulatory feedback loop & imitative map, [auditory, motor] dimensionally consistent, motor theory of speech.html
• p430fig12.26 NormNet: speaker normalization via specializations of mechanisms for auditory streams.html
• p431fig12.27 ARTSTREAM & NormNet strip maps: variants of occular dominance columns in visual cortex.html
• p432fig12.28 SpaN: spatial representations of numerical quantities in the parietal cortex.html
• p433fig12.29 What stream: place-value [number map, language category]s; to Where stream: numerical strip maps.html
• p436fig12.30 cARTWORD: laminar speech model- future disambiguates past, resonanct wave propagates through time.html
• p436fig12.31 Working memory: temporal order STM is often imperfect, then stored in LTM.html
• p437fig12.32 Free recall bowed serial position curve.html
• p437fig12.33 Working memory models: item and order, or competitive queuing.html
• p438fig12.34 LTM Invariance Principle: [STM, LTM] new words must not cause catastrophic forgetting of subwords.html
• p439fig12.35 Normalization Rule: total activity of working memory has upper bound independent of number of items.html
• p439fig12.36 [Item, Order] working memories: [content-addressable categories, temporal order, [excitatory, inhibitory] recurrence, rehearsal wave.html
• p440fig12.37 Normalization Rule: primacy bow as more items stored.html
• p441fig12.38 LTM Invariance Principle: new events do not change the relative activities of past event sequences.html
• p442fig12.39 [LTM invariance, Normalization Rule] Shunt normalization -> STM bow.html
• p442fig12.40 [LTM Invariance, normalization, STM steady attention]: only [primacy, bowed] gradients of activity can be stored.html
• p443fig12.41 Neurophysiology of sequential copying: [primacy gradient, self-inhibition].html
• p444fig12.42 LIST PARSE: Laminar cortical model of working memory and list chunking.html
• p445fig12.43 LIST PARSE laminar Cognitive Working Memory in VPC, is homologous to visual LAMINART circuit.html
• p446fig12.44 LIST PARSE: immediate free recall experiments transposition errors, list length.html
• p447fig12.45 LIST PARSE: order errors vs serial position with extended pauses.html
• p448fig12.46 Masking Field working memory is a multiple-scale self-similar recurrent shunting on-center off-surround network.html
• p449fig12.47 Masking Field self-similar [recurrent inhibitory, top-down excitatory] signals to the item chunk working memory.html
• p452fig12.48 Perceptual integration of acoustic cues: [silence vs noise] durations.html
• p453fig12.49 ARTWORD: acoustic cues, phonetic [features, WM], Masking Field unitized lists, gain control.html
• p453fig12.50 ARTWORD perception cycle: sequences-> chunks-> compete-> top-down expectations-> item working memory-> develops item-list resonance.html
• p454fig12.51 Resonant transfer: as silence interval increases, a delayed additional item can facilitate perception of a longer list.html
• p455fig12.52 cARTWORD dynamics 1-2-3: resonant activity in item and feature layers corresponds to conscious speech percept.html
• p456fig12.53 cARTWORD dynamics 1-silence-3: Gap in resonant activity of 1-silence-3 in [item, feature] layers corresponds to perceived silence.html
• p456fig12.54 cARTWORD dynamics: 1-noise-3: Resonance of 1-2-3 in [item, feature] layers restores item 2.html
• p457fig12.55 cARTWORD dynamics 1-noise-5: Figures 12.[54, 55] future context can disambiguate past noisy sequences that are otherwise identical.html
• p459fig12.56 Rank information on the position of an item in a list using numerical hypercolumns in the prefrontal cortex.html
• p460fig12.57 lisTELOS for saccades: prototype to [store, recall] other [cognitive, spatial, motor] information.html
• p461fig12.58 lisTELOS shows [BG nigro-[thalamic, collicular], FEF, ITa, PFC, PNR-THAL, PPC, SEF, SC, V1, V4-ITp, Visual Cortex input] and [GABA].html
• p462fig12.59 TELOS: balancing reactive vs. planned movements.html
• p463fig12.60 Rank-related activity in PFC and SEF from two different experiments.html
• p464fig12.61 SEF saccades microstimulating electrode: spatial gradient of habituation alters order, but not which, saccades are performed.html
• p464fig12.62 The most habituated position is foveated last: because stimulation spreads in all directions, saccade trajectories tend to converge.html
• p465fig12.63 lisTELOS and data: microstimulation biases selection so saccade trajectories converge toward a single location in space.html
• p467fig12.64 Some of the auditory cortical regions that respond to sustained or transient sounds.html
• p468fig12.65 [PHONET, ARTPHONE] linguistic properties: creates rate-invariant representations for variable-rate speech, paradoxical VC-CV category boundaries.html
• p469fig12.66 PHONET: relative duration of [consonant, vowel] pairs can [preserve, change] a percept.html
• p469fig12.67 PHONET [transient, sustained] cells that respond to certain [consonant transient, sustained vowel] sounds.html
• p471fig12.68 Mismatch vs resonant fusion: effect of silence interval length.html
• p473fig12.69 ART Matching Rule properties explain error rate and mean reaction time (RT) data from lexical decision experiments.html
• p474fig12.70 macrocircuit model to explain lexical decision task data.html
• p476fig12.71 Word frequency data model.html
• p481fig13.01 Cognitive-Emotional-Motor (CogEM): macrocircuit of [function, anatomy].html
• p483fig13.02 CogEM: motivated attention [closes cognitive-emotional feedback loop, focuses on relevant cues, blocks irrelevant cues].html
• p483fig13.03 CogEM: supported by anatomical connections [[sensory, orbitofrontal] cortices, amygdala].html
• p484fig13.04 Cognitive-Emotional resonance: top-down feedback from the orbitofrontal cortex closes a feedback loop.html
• p484fig13.05 Classical conditioning: perhaps simplest kind of associative learning.html
• p485fig13.06 Classical conditioning: inverted-U vs InterStimulus Interval (ISI).html
• p485fig13.07 Paradigm of secondary conditioning.html
• p486fig13.08 Blocking paradigm: cues lacking different consequences may fail to be attended.html
• p486fig13.09 Equally salient cues can be conditioned in parallel to an emotional consequence.html
• p486fig13.10 Blocking: both [secondary, attenuation of] conditioning at zero ISI.html
• p487fig13.11 CogEM : three main properties to explain how attentional blocking occurs.html
• p488fig13.12 Motivational feedback and blocking.html
• p489fig13.13 CogEM and conditioning: positive ISI; inverted-U vs ISI.html
• p490fig13.14 Cognitive-Emotional circuit: for proper conditioning, sensory needs >= 2 processing stages.html
• p490fig13.15 CogEM is an ancient design that is found even in mollusks like Aplysia.html
• p492fig13.16 Polyvalent CS sampling and US-activated nonspecific arousal.html
• p493fig13.17 Learning nonspecific arousal and CR read-out.html
• p494fig13.18 Learning to control nonspecific arousal and read-out of the CR: two stages of CS.html
• p494fig13.19 CogEM: secondary conditioning of [arousal, response], multiple [drive, input]s, motivational sets.html
• p496fig13.20 A single avalanche sampling cell can learn an arbitrary space-time pattern.html
• p497fig13.21 nonspecific arousal: primitive crayfish swimmerets, songbird pattern generator avalanche.html
• p498fig13.22 Adaptive filtering and Conditioned arousal: Towards Cognition, Towards Emotion.html
• p499fig13.23 Self-organizing avalanches [instars filter, serial learning, outstars read-out], Serial list learning.html
• p500fig13.24 Primary [excitatory, inhibitory] conditioning using opponent processes and their antagonistic rebounds.html
• p501fig13.25 Unbiased transducer in finite rate physical process: mass action by a chemical transmitter is the result.html
• p501fig13.26 Transmitter y [accumulation, release]: y restored < infinite rate, evolution has exploited this.html
• p502fig13.27 Transmitter minor mathematical miracle [accumulation, release]: S*y = S*A*B div (A + S) (gate, mass action).html
• p502fig13.28 Habituative transmitter gate: fast [increment, decrement]s of input lead to [overshoot, habituation, undershoot]s, Weber Law.html
• p503fig13.29 ON response to phasic ON input has Weber Law properties due to the habituative transmitter.html
• p504fig13.30 OFF-rebound transient due to phasic input offset: arousal level sets ratio ON vs OFF rebounds, Weber Law.html
• p504fig13.31 Behavioral contrast rebounds: decrease [food-> negative Frustration, shock-> positive Relief] reinforcers.html
• p505fig13.32 Behavioral contrast: [response suppression, antagonist rebound] both calibrated by shock levels.html
• p505fig13.33 Novelty reset- rebound to arousal onset: equilibrate to [I, J]; keep phasic input J fixed; interpret this equation.html
• p506fig13.34 Novelty reset: rebound to arousal onset, reset of dipole field by unexpected event.html
• p506fig13.35 Shock [cognitive, emotional] effects: [reinforcer, sensory cue, expectancy].html
• p509fig13.36 Life-long learning: selective without [passive forgetting, associative saturation].html
• p510fig13.37 A disconfirmed expectation inhibits prior incentive, but is insufficient to prevent associative saturation.html
• p510fig13.38 Dissociation of LTM read-[out, in]: dendritic action potentials as teaching signals, early predictions.html
• p510fig13.39 Learn net dipole output pattern: [shunting competition, informational noise suppression] in affective gated dipoles, back-propagation.html
• p512fig13.40 Conditioned excitor extinguishes: [learning, forgetting] phases, shock expectation disconfirmed.html
• p513fig13.41 Conditioned inhibitor does not extinguish: [learn, forget] phases, same [CS, teacher] can be used.html
• p513fig13.42 Conditioned excitor extinguishes when expectation of shock is disconfirmed.html
• p513fig13.43 Conditioned excitor extinguishes: expectation that -no shock- follows CS2 is NOT disconfirmed.html
• p514fig13.44 Analog of the COgEM model maps of [object X, proto-self], assembly of second-order map.html
• p519fig14.01 Coronal sections of prefrontal cortex.html
• p520fig14.02 pART [cognitive-emotional, working memory] dynamics: main brain [regions, connections].html
• p523fig14.03 MOTIVATOR model generalizes CogEM by including the basal ganglia: supports motivated attention for [, un]conditioned stimuli.html
• p524fig14.04 Basal ganglia circuit for dopaminergic Now Print signals from the substantia nigra pars compacta in response to unexpected rewards.html
• p530fig14.05 Visual [pop-out, search]-> reaction time experiments.html
• p531fig14.06 ARTSCENE: classification of scenic properties as texture categories.html
• p531fig14.07 ARTSCENE voting achieves even better prediction of scene type.html
• p532fig14.08 ARTSCENE: using [sequence, location]s of already experienced objects to predict [what, where] the desired object is.html
• p533fig14.09 ARTSCENE search [data, simulation]s for 6 pairs of images.html
• p540fig15.01 [Delay, trace conditioning] paradigms: require a CS memory trace over the ISI.html
• p541fig15.02 nSTART hippocampal Cognitive-Emotional resonance: feeling of what happens, knowing causative event.html
• p541fig15.03 Timed responses from adaptively timed conditioning: Weber laws, inverted U as a function of ISI.html
• p542fig15.04 blinks of [nictitating membrane, eyelid] are adaptively timed: closure occurs at arrival of the US following the CS, obeys Weber Law.html
• p543fig15.05 Learning with two ISIs: each peak obeys Weber Law, strong evidence for spectral learning.html
• p543fig15.06 Circuit between [dentate granule, CA1 hippocampal pyramid] cells seems to compute spectrally timed responses.html
• p544fig15.07 Spectral timing: STM sensory representation-> Spectral activation.html
• p544fig15.08 Habituative transmitter gate: spectral activities-> sigmoid signals-> gated by habituative transmitters.html
• p544fig15.09 Habituative transmitter gate: increases with accumulation, decreases from gated inactivation.html
• p545fig15.10 A timed spectrum of gated sampling intervals.html
• p545fig15.11 Associative learning, gated steepest descent learning: output from each population is a doubly gated signal.html
• p546fig15.12 Computer simulation of spectral learning: fastest with large sampling signals when the US occurs.html
• p546fig15.13 Adaptive timing is a population property, random spectrum of rates achieves good collective timing.html
• p547fig15.14 [Un, ]expected non-occurences of goal: a predictive failure leads to: Orienting Reactions, Emotional- Frustration, Motor- Explorator.html
• p547fig15.15 Expected non-occurrence of goal: some rewards are reliable but delayed in time, do not lead to orienting reactions.html
• p548fig15.16 Homolog between ART and CogEM model: complementary systems.html
• p548fig15.17 The timing paradox: want [accurate timing, to inhibit exploratory behaviour throught ISI].html
• p549fig15.18 Weber Law: reconciling accurate and distributed timing, different ISIs- standard deviation = peak time, Weber law rule.html
• p549fig15.19 Conditioning, Attention, and Timing circuit: Hippocampus spectrum-> Amgdala orienting system-> neocortex motivational attention.html
• p550fig15.20 Adaptively timed Long Term Depression between parallel fibres and Purkinje cells-> movement gains within learned time interval.html
• p551fig15.21 Cerebellum: important cells types and circuitry.html
• p551fig15.22 Responses of a turtle retinal cone to brief flashes of light of increasing intensity.html
• p552fig15.23 Cerebellar biochemistry: mGluR supports adaptively timed conditioning at cerebellar Purkinje cells.html
• p556fig15.24 Cerebellar cortex responses: [data, model] short latency responses after lesioning.html
• p557fig15.25 Computer simulations of adaptively timed [LTD at Purkinje cells, activation of cereballar nuclear cells].html
• p557fig15.26 Brain [region, process]s that contribute to autistic behavioral symptoms.html
• p559fig15.27 Spectrally timed SNc learning: brain [region, process]s release of dopaminergic signals, unexpected reinforcing.html
• p559fig15.28 Neurophysiological data and simulations of SNc responses.html
• p560fig15.29 Excitatory pathways that support activation of the SNc by a US and the conditioning of a CS to the US.html
• p560fig15.30 Inhibitory pathway: striosomal cells predict [timing, magnitude] of reward signal to cancel it.html
• p561fig15.31 Expectation timing: timing spectrum, striosomal cells delayed transient signals, gate [learning, read-out].html
• p561fig15.32 Inhibitory pathway expectation magnitude: is a negative feedback control system for learning.html
• p563fig15.33 MOTIVATOR: thalamocortical loops through basal ganglia.html
• p563fig15.34 Distinct basal ganglia zones for each loop.html
• p564fig15.35 GO signal to recurrent shunting OC-OS networks: control of the [fore, hind] limbs.html
• p565fig15.36 (a) FOVEATE: control of saccadic eye movements within the peri-pontine reticular formation.html
• p566fig15.37 FOVEATE: steps in generation of a saccade.html
• p567fig15.38 Gated Pacemaker of [diurnal, nocturnal] circadian rythms: whether phasic light turns the pacemaker on or off.html
• p568fig15.39 MOTIVATOR hypothalamic gated dipoles: inputs, [object, value, object-value] categories, reward expectation filter.html
• p569fig15.40 GO and STOP movement signals: control by [direct, indirect] basal ganglia circuits.html
• p573fig16.01 Hippocampal place cells: discovery from rat [experimental chamber, neurophysiological recordings].html
• p574fig16.02 Neurophysiological recordings of 18 different place cell receptive fields.html
• p575fig16.03 Rat navigation: firing patterns of [hippocampal place, entrorhinal grid] cells.html
• p578fig16.04 Cross-sections of the hippocampal regions and the inputs to them.html
• p580fig16.05 GridPlaceMap hierarchy of SOMs with identical equations: learns 2D [grid, place] cells.html
• p581fig16.06 Trigonometry of spatial navigation: coactivation of stripe cells.html
• p582fig16.07 Stripe cells multiple [orientation, phase, scale]s: directionally-sensitive ring attractors, velocity, distance.html
• p582fig16.08 Evidence for stripe-like cells: entorhinal cortex data, Band Cells position from grid cell oscillatory interference.html
• p583fig16.09 GRIDSmap: stripe cells for rat trajectories, self-organizing map learned hexagonal grid cell receptive fields.html
• p583fig16.10 GRIDSmap embedded into hierarchy of SOMs: [angular head velocity, linear velocity] signals to place cells.html
• p584fig16.11 GRIDSmap learning of hexagonal grid fields, multiple phases per scale.html
• p584fig16.12 Temporal development of grid fields: orientations rotate to align with each other.html
• p585fig16.13 Hexagonal grid cell receptive fields: somewhat insensitive to [number, directional selectivities] of stripe cells.html
• p585fig16.14 GRIDSmap: Superimposed firing of stripe cells supports learning hexagonal grid.html
• p586fig16.15 Why is a hexagonal grid favored: stripe cells at intervals of 45 degrees, GRIDSmap does not learn, oscillatory interference does.html
• p586fig16.16 Grid-to-place SOM: formation of place cell fields via grid-to-place cell learning.html
• p587fig16.17 A refined analysis: SOM amplifies most frequent and energetic coactivations, stripe fields separated by [90°, 60°].html
• p588fig16.18 GridPlaceMap hierarchy of SOMs: coordinated learning of [grid, place, inomodal] cell receptive fields.html
• p589fig16.19 How does the spatial scale increase along the MEC dorsoventral axis.html
• p590fig16.20 Dorsoventral gradient in the rate of synaptic integration of MEC layer II stellate cells.html
• p590fig16.21 Frequency of membrane potential oscillations in grid cells decreases along the dorsoventral gradient of the MEC.html
• p591fig16.22 Dorsoventral [time constant, duration] gradients in AHP kinetics of MEC layer II stellate cells.html
• p591fig16.23 Spectral spacing model: map cells respond to stripe cell inputs of multiple scales, How do entorhinal cells solve the scale selection problem.html
• p592fig16.24 Parameter settings in the Spectral Spacing Model that were used in simulations.html
• p593fig16.25 Spectral Spacing Model equations for [STM, MTM, LTM].html
• p593fig16.26 Gradient of grid spacing along dorsoventral axis of MEC.html
• p594fig16.27 Gradient of field width along dorsoventral axis of MEC.html
• p595fig16.28 Peak and mean rates at different locations along DV axis of MEC.html
• p596fig16.29 Subthreshold membrane mV oscillations: decreasing Hz at different locations along DV axis of MEC.html
• p596fig16.30 Spatial phases of learned grid and place cells.html
• p597fig16.31 Multimodal place cell firing in large spaces.html
• p597fig16.32 Model fits data about grid cell development in juvenile rats: grid [score increases, spacing flat].html
• p598fig16.33 Model fits [place, grid, directional] cell data about grid cell development in juvenile rats: [spatial information, inter-trial stability] vs postnatal day.html
• p598fig16.34 spiking GridPlaceMap: generates theta-modulated place and grid cell firing, unlike the rate-based model.html
• p599fig16.35 anatomically overlapping grid cell modules: effects of [different modules in one animal, DV location, response rate].html
• p600fig16.36 entorhinal-hipppocampal system: ART spatial category learning system, place cells as spatial categories.html
• p602fig16.37 Hippocampal inactivation by muscimol disrupts grid cells.html
• p603fig16.38 Role of hippocampal feedback in maintaining grid fields, muscimol inhibition.html
• p605fig16.39 Disruptive effects of MS inactivation in MEC.html
• p607fig16.40 Effects of medial septum (MS) inactivation on grid cells: data, simulations, gridness.html
• p611fig16.41 back-propagating action potentials, recurrent inhibitory interneurons: control learning, regulate rythm- read-out is dissociated from read-in.html
• p612fig16.42 Macrocircuit of the main SOVEREIGN subsystems: visual, motor.html
• p613fig16.43 SOVEREIGN [visual form, motion processing] stream mechanisms.html
• p613fig16.44 SOVEREIGN[target position, difference] vectors, volitional GO computations] to control decision-making and action.html
• p614fig16.45 [distance, angle] computations learn dimensionally-consistent [visual, motor] information for [decision, action]s.html
• p615fig16.46 SOVEREIGN uses homologous processing stages to model the [What, Where] cortical streams, motivational mechanisms.html
• p615fig16.47 SOVEREIGN: multiple parallel READ circuits, sensory-drive heterarchy amplifies motivationally favored option.html
• p616fig16.48 SOVEREIGN tests using virtual reality 3D rendering of a cross maze.html
• p616fig16.49 SOVEREIGN animat converted inefficient exploration into an efficient direct learned path to the goal.html
• p617fig16.50 Spectral Spacing models of [perirhinal what, parahippocampal where] inputs, fused in the hippocampus.html
• p627tbl17.01 Homologs between [reaction-diffusion, recurrent shunting cellular network] models of development.html
• p628fig17.01 A hydra.html
• p628fig17.02 how different [cut, graft]s of the normal Hydra may [, not] lead to the growth of a new head.html
• p629fig17.03 How an initial morphogenetic gradient may be contrast enhanced to exceed the threshold for head formation.html
• p630fig17.04 Morphogenesis: use cellular models vs [chemical, fluid] reaction-diffusion models.html
• p631fig17.05 How a blastula develops into a gastrula.html
• p634fig17.06 How binary cells with a Gaussian distribution of output thresholds generates a sigmoidal population signal.html
• pxvifig00.01 Macrocircuit of the visual system.html