Brain & Cognition 🧠Flashcards
the retina prre-processes the rod and cone signals
via bipolar cells to ganglion cells
ganglion cells
pass the preprocessed signals to the brain
long wavelength cone
responds well to red or yellow
medium wavelength cone
responds best to green less to yellow
short wavelength cone
responds best to blue
rhodopsin
translaties light into the closing of Na+ channels so that the membrane hyperpolarizes > neural signal that is sent to bipolar >? ganglion cell
retinal color blindness
absence of a particular cone type
fovea
cup shaped highest density photoreceptors, mainly cones; sharpest vision, color vision
age related macular degeneration
older age, smoking diet, genetic , loss of central vision, acqity loss, pigment epithelium (receptors) are lost due to accumulation of toxic products,
dry macular degeneration
damage to the fovea, yellow deposits (drum) accumulate in macula
why are rods and cones not switched with the photoreceptors placement?
the pigment epithelium prevents light scatter, so that sharper vision is possible (also provides nutrients )
RGC fibers lying on top causes the blind spot
place where all retinal ganglion cell fibers pass through the eye (optic disk)), and no receptors are present
glaucoma
increase of pressure inside of the eye, damage of nerve fibers of the RGC’s : optic nerve, loss of peripheral vision first (but may vary), treatment : eyedrops, surgery
the need for data compression
optic nerve only contains 1 mil nerve fibers and data compression gebeurt in het oog van 130 mil.
how is retinal information compressed?
the photoreceptor responds to light by hyperpolarization (closing of Na+ channels) to dark by depolarisation (opening of Na+ channels) : a graded potential signal
the one and off systems originate at the level of the bipolar cells
the receptors make sign conserving synapse with the off bipolar cells and sign inverting synapses with on bipolar cells that have an unique neurotransmitter receptor site (mGluR6)
horizontal cells
receive signals from widespread region of receptors, . they provide negative feedback on the receptors
the retinal network
from luminance (receptors) to contrast (bipolar and ganglion cells). from graded potentials (hyper/depolarization) in cones, bipolars, horizontals, to action potentials (in ganglion cells, because of long axons)
retinal ganglion cells
encode contrast, luminance is discarded
Hermann grid illusion
‘side effect’ of the data compression by the retina (and higher visual areas), comparable to artefacts caused bt JPEG compression
how the retina solves the data compression problem
- contrast coding (ON and OFF center surround Retinal Ganglion cells)
- rod signals pass through the same RGC’s as cone signals
- colours are coded as R/G, G/R or B/Y contrasts
- parallel pathways (M/P>P/M) for color //bw, detail/global
how do rods work? because they are mainly working in the dark
- rods mostly connect to rod bipolar cells
- they do not connect directly to RGC’s
- they connect to cone driven bipolar cells via the amacrine cells
- so bipolar and RGC’s have (overlapping) receptive fields from cone and rod iputs
- horizontal cells also give the rods a suppressive surround > Mexican hat RF profile
diagram of primary rod-driven signal pathways via four synapses
rods -> rod bipolar cell (RBC) -> all amacrine cell (AII-AC) -> OFF or ON (cone) bipolar cell (OFF-BC, ON-BC) -> OFF or ON ganglion cell (OFF-GC, ON-GC).
rod cell
sensitive enough to respond to a single photon of light and is about 100 times more sensitive to a single photon than cones.
- sensative to single wavelength (hence are useless for color vision)
- rod bipolar receive input from multiple rods, hence have larger receptive fields
- therefore, vision in the dark is less sharp (also because rods sit mainly peripheral)
dark adaptation
- pupil dilation
- cone> rod transition
- ‘bleaching’ (depletion) of pigment (opsin) in photoreceptors that become undone
- less receptor signal > less negative feedback from horizontal cells
retinis pigmentosa
genetic disorder (>50 genes involved)
- progressive degeneration of receptors: rods first, followed by cones
- pigments deposits at affected parts of retina, depigmentation at vulnerable sites
- night blindness* > loss of peripheral vision > tunnel vision > full blindness
two types of ganglion cells
midget(X, small receptive fields) and parasol cells (Y, larger receptive fields)
midget cells
small receptive field, single cone center and surround: color contrast selective, slow sustained responses, tuned to high spatial frequencies// parvocellular layers of the LGN
parasol cells
large receptive field, many cones input to center and surround; not color sensitive, fast transient response, tuned to low spatial frequencies// Magnocellular layers of the LGN
continuous spectrum, yet color opponent of retinal ganglion cells:
sampling of red/green, green/red, blue vs yellow (not yellow blue because yellow is made up off of two cones and that cannot be put in the center)
the concept of spatial frequency decomposition
every image/ contour can be decomposed into the spatial frequencies it contains
- sharp edges (square waves) contain both low and high spatial frequencies
- so do small spots of light (impulse function)
spatial frequency sensitivity depends on
- contrast & brightness
- receptive field size
center component of RGC is sensitive from
low to high SF’s, surround from low to intermediate SF’s, combined, this gives the characteristic SF running of RGC’s
parvocellular pathway
(from midget cells) has sustained response sees collor low contrast agin higher spatial resolution slower
magnocellular pathway
(from parasol cells) transient réponse monochrome high contrast gain lower spatial resolution faster
Y-type (parasol) RGC axons have …. conduction velocities than X-type (midget)
faster
magnocellular fibers of LGN … than parvocellular fibers
faster
global precedence:navon task (hierarchical letter stimuli)
global targets are detected faster than local targets. also, congruent stimuli are faster than incongruent
seems to be a hemispheric asymmetry in the processing of global versus local information
global information is faster in the right hemisphere (stimuli in left visual field)
local information is faster in the left hemisphere (stimuli in right visual field)
this corresponds to the finding that patients with right hemisphere damage have trouble copying the global shape, while patients with left hemisphere damage have trouble copying the local shapes
what is the purpose of the brain interpreting the incoming wavelengths (color is not wavelength)
color constancy as a brain code for ‘tasty’ or ‘nasty’
V4
color constant emerges at the level of V4. V4 receptive field is large enough to integrate numerous color opponent signals and discount the illuminant
achromatopsia
lesions to V4/V8 results in cortical color blindness// lesion in V4 also had deficits in discriminating complex shapes. he could distinguish luminance, orientation and motion but not illusory contours and complex shapes
simplified Reinhardt detector model for direction selectivity
the cell receives input from two cells that have spatially separate receptive fields. one of the cells has a delayed input. only when the stimulus moves in the right direction, the cell receives simultaneous input from both cells and will fire
computational power of the neuron
adding up all inputs - in a weighted manner- to generate an action potential or not
the neuron is a coincidence detector
only when enough inputs combine at the same time, an action potential is generated
direction of motion selectivity
V1, V3, MT
the aperture problem
detecting motion through an aperture is ambiguous, many motion vectors can yield the same motion through the aperture. V1 cells suffer from the aperture problem
pattern cell
MT 50%, the cell is ‘seeing’ the pattern motion, no longer the movement of the individual compontnetns
component cell
V1, the cell is not seeing the pattern motion but the movement of the individual components// so basically it is seeing things that you are not seeing, you perceive it as going left but the component cell perceives the two individual directions that it is made out of
MST
neurons répond selectively to particular motion flow fields. often caused by self-motion
biological motion
recognising species, sex, mood etc from movement
superior temporal sulcus
selectively activated by biological compared to non-biological (scrambled) motion
akinetopsia
bilateral MT,MST,STS lesion. zihl’s motion blind patient: sees only stationary scenes, with objects flashing from one position to the next
motion (visual area
component V1, V3, MT
optic flow MST
biological STS
color visual area
wavelength V1., V2
colour constancy V4
dorsal pathway
magno dominated
ventral pathway
parvo dominated
hypothalamus
regulation of circadian rhythms
pretectum
reflexive control of pupil and lens
superior colliculus
converts a retinatopic map into a saccade map too that stimuli are foveated //orienting movements of head and eyes
layers of LGN
magnocellular, parvocellular, koniocellular
magnocellular
y-type (parasol) input
parvocellular
X-type (midget) input
layers 1,4,6 of LGN
(depends on which LGN) contralateral eye
layers 2,3,5 of LGN
ipsilateral eye
retino-cortical expansion
there is much more area in your visual cortex looking at the central part of your visual field than the surrounding peripheral visual field
V1
starting point of projection to the other visual areas and the rest of the brain
definition of visual area
each area contains a separate, retinotopic, map of the visual field, need not to be a complete map, sometimes only central visual field or peripheral or only lower or upper.
oculair dominance columns
strong separation in layer 4, weaker in layers 2/3 & 5/6
amblyopia
pathological dominance fo one eye , if this happens during the critical period in development (2-6 years) input from that eye occupies less space in V1, the OD column of that eye becomes smaller. after that critical period, this reduced representation remains, and amblyopia (lazy eye) remains
receptive field tuning
a neuron will only respond to a stimulus within its receptive field if that stimulus has certain characteristics.
feature selectivity
orientation and direction of motion tuning in V1
orthogonal organisational components of primary visual cortecx
orientation columns
ocular dominance columns
CO blobs
hypercolumn
basic processing unit in V1 : Cytochome oxidase activity, Ocular dominance column, orientation column
Gabor filter
sinusoid combined with Gaussian envelope
simple cells
Gabor filters tuned to orientation and spatial frequency, with on and off zones
quadrature pairs
of simple cells combine to form complex cells
colour constancy
we perceive the same color despite differences in illumination. the wavelength of the light from the banana is very different in the morning versus the evening, yet it is perceived as yellow the whole day
the simplified Reinhardt detector model for direction selectivity
The cell receives input from two cells that have spatially separate receptive fields. One of the cells has a delayed input*. Only when the stimulus moves in the right direction, the cell receives simultaneous input from both cells and will fire
monocular depth cues
respectieve size constancy occlusion cast shadowns shading & contour areal/athmospheric perspective texture motion parallax
disparity
when you fixate a point P, you converge the two eyes. P will project on the fovea (F) in each eye. every other stimulus (Q) that falls within the same plane of depth (horopter) will be projected at equal distance form the fovea in the two eyes. //A stimulus that is more nearin depth will produce unequal distance projections (red). This is called disparity (near disparity in this case). Similar for far disparity.
the brain calculates depth from disparity
correspondence problem
Each eye/camera views three image primitives (dots). The problem then is, which dots in the left eye correspond to which dots in the right eye ? The 9 dots represent all the possible matches that could be made, the black dots are the correct matches and the rest are incorrect, (referred to as either false targets'' orghosts‘’).Confronted by these 9 possible matches, we ourselves are capable (in this instance) of making the 3 correct matches. The interesting thing about using the dot example is that no high level information or cues are presented to help the viewer in matching. This led researchers to believe that stereo matching is performed early in the human visual process, it is assumed to be a low level operation. Only after the two images are matched is any attempt made to understand what is actually being viewed.
panui’s area
Only objects that are not too far or too near relative to the horopter have small enough disparity to result in fusion (and hence depth perception). This region in 3D space is called Panum’sarea. Objects that are farther or closer than Planum’s area result either in diplopia orsuppression of their image in one of the eyes (usually the non-dominant eye, remember the eye-dominance test)
Strabismus
misalignment of the eyes , due to congenital eye-muscle disorders, or due to cranial oculomotor nerve disorders (neurological disease)
amblyopia
Particularly during childhood (~0-4 years) the result will be the permanent suppression of the input from one eye:
binocular rivalry
ccurs when the two images are completely different, which is typically only achieved experimentally using devices to present different images to the two eyes. While the images in the two eyes remain constant, the (conscious) percept spontaneously switches between the one and the other image, with different durations of dominance for each stimulus
visual features : shape
orientation (V1, V2, V4)
complex shapes (V4, LOC)
real world shapes (IT, faces hands)
invariant object coding
visual features : position
Position-retinal (V1,V3, MT)-head centered (VIP)-body centered (area 5, PRA)
the binding problem
Perceptual organization Combining distant elements and features into a coherent ‘whole’, objects and background
perceptual organization
/ A powerful, yet automatic mechanism, that always tries to make sense of visual information, even with minimal input/making sense of visual information, even when there is minimal input
biological motion
High level perceptual organization with minimal informationAlso the effect of ‘knowledge’ or memory: we are very often exposed to moving bodie
gestalts law of perceptual organization
proximity, similarity (shape) connectedness good continuation symmetry closedness pregnant(good form) common fate figure-ground
Apperceptive Agnosia
Right Hemisphere occipital / temporal / parietal lesions usually quite diffuse (e.g. CO damage)
Patients see individual features, such as color, orientation or motion, but cannot integrate this into a whole
Failure of perceptual organization•Gollinpicture task is bad, Copying is bad
•Unusual views / shading task is bad
Integrative Agnosia
- Somewhat similar to Apperceptive Agnosia
- Can copy, but piece by piece
- Unusual views variable
- Overlapping objects are not seen individually
THE FAST FEEDFORWARD SWEEP
Speeds through the visual system within 80 msby means of feedforward connections•Provides the neurons with their receptive field tuning properties•Provides fast detection of ‘hardwired’ features and feature constellations•In that sense provides a set of grouping operations•Enables ‘intelligent’ sensorimotor reflexes
face selective cells in inferotemporal cortex
responds selective to faces, and. not to other objects or scrambled faces. tuning to face properties: direction of gaze, emotional expression, identity
more than one face area in human brain
occipital face area, fusiform face area, superior temporal sulcus
PPA parahippocampal place area
responds to stimuli with a location or place element, such as landscapes houses building. not to other man-made objects
prosopagnosia
face selective neuropsychological deficit after extensive bilateral or right sided lesions in temporal and occipital lobes
wholistic analysis
a face is not so much recognised by its parts, it is recognised by the configuration of the parts. the relation the different parts have to each other
composite effect
Separate faces that are aligned are analyzed as ‘wholes’ instead of by their parts, making it more difficult to recognize the top half when aligned with different lower halves. This effect disappears when the faces are no longer aligned. > faces are recognized as ‘wholes’.
capgras delusion
intact face recognition, but thinking they are imposters
horizontal connections
interconnect cells with distance receptive fields
horizontal connections in V1 reflect gestalt laws of perceptual organization
proximity, similarity (orientation), good continuation
feedforward connections
carrying orientation information
horizontal connections
lateral inhibition between neurons tuned to the same orientation. this creates bumps. present at every level in the hierarchy
feedback connections
these connections do a retinatopic feedback, so that the region between the bumps get progressively filled in
hierarchical model (theoretical view of how the representation of objects is implemented in the brain)
here is feedforward convergence of hierarchical RF properties: from cells detecting low level features to cells detecting increasingly complex feature constellations, up to cells detecting highly specific objects: pontifical cells (also called ‘grandmother cells’ or ‘yellow Volkswagen cells’, to indicate that for every possible object you would need a specific cell, which would yield a ‘combinatorial explosion’)
dynamical assembly formation model (theoretical view of how the representation of objects is implemented in the brain)
cells each encode specific low and higher level features, but objects are encoded by the formation of dynamic assemblies (groups of cells) via horizontal and feedback connections, that together encode a particular object
combinatorial explosion
you cannot have a cell for every possible object
ensemble coding theory
- objects are represented by sets of neurons each representing the individual features
- individual nodes may participate in different ensembles
dorsal pathway
vision for action. uses position and shape information for visually guided action. motion& position ,magno dominated
ventral pathway
vision for perception. uses shape (and position) information for visual perception, memory. color & shape, parvo dominated
object (shape) discrimination
impaired by lesion of temporal lobes
landmark (position) discrimination
impaired by lesion of parietal lobes
associative agnosia
mostly left hemisphere occipital / temporal/parietal lesions. They cannot match objects by functionPatients cannot name objects that they see (but can recognize what they feel, so they still know the name). Perceptual organization is fine Also fine at copying drawings.
optic ataxia patient
posterior parietal lesion, can recognise orientation, but cannot post
final common pathway of all m0tor output
the ventral horn alpha motor neuron and the stretch reflex
dorsolateral part spine
distal muscles, fine movements
ventromedial part spine
proximal muscles, posture
stretch reflex
keeping posture. . Muscle Spindle senses stretching-Activates Alpha Motor Neuron > muscle contracts-Gamma motor neuron contracts muscle spindle during voluntary movement so that they stay ‘short enough’ to sense stretching when muscles are short. Input from Pons
reciprocal inhibition of antagonistic muscles
when extensor (quadriceps) contacts, flexor relaxes
crossed extensor reflex
as one limb flexes, the other extends
Golgi tenden reflex
protecting from overload
extrapyramidal systems
subrospinal tract, tectospinal tract, vestibulospinal tract, reticulospinal tract
rubrospinawl tract
upper motor neurons in red nuclei•control muscle tone and distal limb muscles that perform more precise movements
tectospinal tract
upper motor neurons in superior and inferior colliculi•Receive visual (superior) and auditory (inferior) info•Reflex-like Orienting Response: head, neck, upper limbs move towards visual and auditory stimuli
vestibulospinal tract
Info from vestibulococlearnerve (VIII) (inner ear)•Monitor position and movement of the head to maintain Posture and Balance
reticulospinal tract
Reticular Formation, input from many pathways•Controls many reflexes (excitability)•State of arousal
pyramidal system
•Voluntary (conscious) control of skeletal muscles:•begins at upper motor neurons of primary motor cortex and other cortical areas (SMA, PMC)•axons descend into brain stem and spinal cord to synapse on lower (alpha) motor neurons
corticulobulbar tract
Towards cranial nerve nuclei that move eye, jaw, face, and some muscles of neck and pharynx (throat)
corticospineal tract
Corticospinal tracts visible along ventral surface of medulla oblongata as pair of thick bands, the pyramids•Control of all non-facial somatic muscles•Lateral CS tract crosses at pyramidal decussation at high level•Anterior CS tract cross to opposite side of spinal cord at lower level in anterior white commissure
damage to corticospineal tract
paralysis/paresis, spasticity/flaccidity, changed reflexes (babinski sign)
cerebellum
provides a subcortical-cortical loop. finetuning of movements (lesions or alcohol result in ataxia). timing of automated movement sequences, motor memory, maybe also timing in general
spinocerebellum
balance, walking, affected by alcohol use
neocerebellum
control of fine movements, finger to nose test, speech
vestibulocerebellum
coordination of eye movements with body movements , (vestibule-ocular-reflex)
cerebellar ataxia
endpoint tremor, slurred speech
basal ganglia, anatomy
Striatum (Caudate nucleus, Putamen)
•Globus Pallidus (Externa•Interna) •Substantia Nigra( Dopamine)• Subthalamic nucleus
bg function
direct pathway, indirect pathway, dopamine release by SNc, D1 D2 receptors in striatum
higher motor control
•Action planning•Action selection•Affordances•Mirror Neurons•Reference frames
primary motor control
direct motor control
supplementary motor area & PFC
internally guided action (selecting which object to pick up)
premotor cortex & posterior parietal cortex
externally guided, stimulus driven action
hemiplegia
•Half sidedparalysisduetolesionsof uppermotor neurons comingfromM1
apraxia
Loss of motor skill, not due to muscular, upper (M1) or lower motor (Spinal Cord) neuron deficit
•Lesions to SMA, PMC, PPC
ideomotor apraxia
rough idea of movement can be executed (SMA, PMC)
ideational apraxia
no idea what to do, use of wrong tools (PPC)
M1 (and PMC)
neurons encode movement direction. Individual motor neurons encode vector of movement, i.e. are tuned for the directionof limb movemen
afforance competition hypothesis
Sensory inputs create many potential motor responses (affordances). Depending on needs and potential payoffs, one of these has to be selected
premotor cortex
encodes population vectors of multiple potential actions
mirror neurons
Neurons that encode an action, yet also are activated by seeing (or hearing) the same action performed by others// are widespread in motor cortex and parietal cortex
posterior parietal cortex
translating movement from retinal (eye-centred) to hand- head- or body-centred reference frames
proximity
parts are grouped when close together
similarity
parts are grouped when similar
connectedness
parts are grouped when connected
common fate
parts that move together// in MT most neurons are direction selective. Patchy horizontal connections connect cells with similar direction preference
closure
convex shapes are preferred over concave
good continuation, collinearity
lines that continue in the most fluent way/logical way
symmetry
contours that form a symmetrical shape are grouped over contours that form an asymmetrical shape
pregnanz
good form, reduction to the simples forms
gestalt law of figure ground
some region is figure other is background that continues behind figure
similarity grouping features
orientation, direction of motion, disparity (Depth), spatial frequency (size), luminance and color
horizontal connections in v1 reflect gestalt laws of perceptual organization
proximity, similarity (orientation) and good continuation
pigment epithelium
prevents light to scatter, so that sharper vision is possible (also provides nutrients)
hyperpolarization
closing of NA+ channels
depolarization
opening of Na+ channels
a graded potential signal
the photoreceptor responds to light by hyperpolarization to dark by depolarisation
sign conserving synapse
When the photoreceptor cell depolarizes, the bipolar cell also depolarizes. Accordingly, this synapse is sign-conserving: when light falls on the rod or cone, the bipolar cell hyperpolarizes (because the photoreceptor hyperpolarizes
sign inverting synapse
between photoreceptors and ON bipolar cells is the foundation for the ON pathway in vision. The synapse is sign-inverting because glutamate, released in darkness by rod and cone photoreceptors, hyperpolarizes the membrane of ON bipolar cells.
indirect pathway
inhibits GPe (which inhibits GPi/SNr), so that GPi/SNr activity increases > more inhibition of thalamus, cortex> less movements
direct pathway
inhibits GPi/SNr so that their inhibitory effect is diminished > more activation of cortex > more movements
off bipolar cells
preserve the input from the cone and are hyperpolarized by light
on bipolar cells
reverse the sign of the cone and are depolarised by light
full perceptual organisation
feedback connections link features processed across areas
hierarchical coding
what does the existence of face selective cells actually mean, is object recognition implemented in the brain by hierarchical coding?
gain modulation
some VIP neurons change their response as a function of eye position.// RF Is the same respective to the fixation point// some neurons are stronger depending on where the receptive field is relative to the fixation point
scotopoic vision
rods are primary source of visual information at night
how do we perceive depth?
monocular depth cues: perspective, size constancy, occlusion, cast shadows, shading & contour, areal/atmospheric perspective, texture, motion parallax. Binocular depth cues: disparity
near disparity
a stimulus that is more near in depth will reduce unequal distance projections (red)
Inferior temporal cortex neurons
encode perceived depth not just contrast disparity.. they are really about the depth that you perceive
a common theme in the computations performed in the visual cortex
(component cells) neurons responding to visual features that are preceding perception (component motion, disparity) vs. (pattern cells) neurons responding to visual features that are perceived (pattern motion, depth)
IT
perceived depth
V2
Neurons tuned for the orientation of illusory contours (and real contours of course)
LOC
visual area responds more strongly to shapes (regardless of whether they are familiar or not) than to scrambled objects// responds to shapes even when defined by second order cues, like contour from motion stimuli
object constancy
we can recognise objects from any position, angel and under all sorts of different shading and lighting conditions. in all these cases the image on the retina (the distribution of light and contrast) is highly different -> object recognition is therefore viewpoint invariant
repetition priming
show the same object twice, and the visual cortex will respond weaker the second time around
viewpoint invariant coding
left anterior and posterior fusiform gyro Repetition priming for identifcal objects at different viewpoints
size invariant coding no viewpoint invariant coding
right anterior and posterior fusiform gyro (and parietal) show repetition priming for identical objects of different size, but not for different viewpoints
VIP, LIP, area 7a (in monkeys)
cortical areas where position is transformed from retinal to other reference frames such as head or the body. // vip neurons show all the intermediate steps of transforming retinatopic (eye centred) RF’s into head centred RF’s
PRR : parietal reach region (area 5 in monkeys)
how the coordinate frames of motor outputs.
feedforward convergence
cells get increasingly larger receptive fields. and get progressively more complex tuning properties. there are multiple feedforward sweeps, for example those coming from M/P going to dorsal/ventral
the type of collinear spread of the tracing of horizontal connections in V1 seems to be substrate for the gestalt law of
good continuation or collinearity
contextual modulation
reflect figure-ground organization. through feedback connection the stimulus will react differently when the background is different around the receptive field.
