Systems Exam 2

Lower motor neurons

Directly innervate skeletal muscles

move limbs, voluntary body

Not anatomically correlated, can have LMN in brainstem innervate face

Local circuit neurons

Interconnected with LMN to form local circuits (involved in reflexes and simple pattern generators)

Lower motor system

LMN + local circuit neurons

Upper motor neurons

Indirect connections to muscles, only rarely projecting to lower motor neurons diretcly

Not anatomically correlated, can have LMN in brainstem innervate face

Command, coordiinate LMN

Not anatomically correlated, can have LMN in brainstem innervate facial musclkes or form local reflex circuit, even though located physically higehr ip than some UMN

Smooth Muscle

Spindle shaped, nonstriated, uninucleated fibers

Occurs in walls of internal organs

Is involuntary

(visceral, digestion)

Cardiac muscle

Has striated, branched, generally uninucleated fibers

Occurs in walls of heart

Is involuntary

Skeletal muscle

Has striated, tubular, multinucleated fibers

usually attached to skeleton

Is voluntary

Also involved in reflexes

Skeletal muscle organization

Skeletal muscles has multiple fascicles

Fascicle has multiple muscle fibers (cells)

sarcolemma is the fine, specialized plasma membrane that envelops each individual muscle fiber

Sarcoplasm is the specialized cytoplasm of a muscle fiber (muscle cell), enclosed by a membrane called the sarcolemma

Muscle cell has many myofibrils (specialized cell organelle, engines to move muscle)

Muscle movement organization

Muscles move when actin and myosin in myofibrils slide along each other

In striated muscle, these alternating filaments have specific length, arranged in parallel, give distinctive striped appearance

One stripe called sarcomere

Relaxed configuration

• 2 Actins connected to z disks on either end , myosin in middle of actins (vertically and horizontally

• Actin extended on either side of myosin

Contracted configuration

• Add ATP, Ca2+

  • Actin closes gap between 2, z disks closer together

Spinal cord lower motor neurons located in ventral horn

Ventral horn (bigger, thicker part of X, usually on bottom)

• Changes size based on how many cells

Retrograde tracers injected into 2 calf muscles show anatomical distribution of motor units innervating them

• Motor neurons form narrow columns in ventral horn, clustered at specific distance from midline, stay this distance from midline all the way up

• Topographical organization of how neurons located relative to what they innervate

• Retained all the way up the spine

Topographic mapping of ventral horn muscles

Proximal muscles have their lower motor neurons innervating them in medial part of VH

Distal muscles have LMNs innervating them in lateral part of ventral horn T

Types of lower motor neurons

Muscles contain 2 different types of fibers, innervated by 2 different types of motor neurons

Alpha motor neurons- contract extrafusal muscle fibers, responsible for actual movement

Gamma motor neurons- contract intramural muscle fibers (spindles), part of sensory feedback systems (proprioception, set muscle tone)

alpha motor neurons

Innervate extrafusal skeletal muscle cells (generate movement)

Receive input from

• Upper motor neurons

• Muscle spindles

• Local circuit neurons

Located in Ventral Horn (DH has sensory neurons)

Most extrafusal skeletal muscle fibers (cells) are innervated by single alpha motor neuron

• each alpha MN typically innervates multiple muscle fibers

These innervated fibers distributed fairly widely in the muscles

• Damage to one motor neuron doesn’t carry as great an effect on whole muscle

  • Single muscle has many alpha motor neurons innervating

Neuromuscular junction

Synapse between alpha motor neuron and muscle fiber

Large synapse, 1 presynaptic AP = 1 postsynaptic AP (muscle twitch/contraction)

One of most studied synapse

Single NMJ has many active zones (contains neurotransmitter vesicles on pre-synaptic side, ACh and voltage gated Na channels on postsynaptic side)

Motor unit

Single AMN and the muscle fibers it innervates

Motor( neuron) pool

Group of AMNs that together innervate all fibers of a given muscle

Two biochemical pathways for generating energy (both yield ATP)

Glycolysis

• Breakdown of glycogen, does not use O2

Oxidative metabolism

• Requires O2

Different types of muscle fibers (motor units)

Slow oxidative (Type 1, slow, dark meat)

• Contract slowly, slow twitch muscles

• Adapted for slow sustained postural contractions (neck muscles)

• Slow fatigue

• Relies on blood flow, oxygen (dark)

  • Small fibers

Fast glycolytic (Type 2B, fast fatigable, white meat)

• Contract quickly, fast twitch

• Fast transient bursts, fine skills (finger muscles)

• Rapid fatigue

• Doesn’t need oxygen

• Large diameter

Fast oxidative (Type 2A, fast fatigue resistant)

• Hybrid of 1 and 2, oxidative (dark), but fast twitch (ex. leg muscles)

• Medium contraction speed and fatigue

• Relies on oxygen, needs lot of blood flow

• small fibers

Different fiber types can co-occur within given muscle, can train these

Different skeletal muscle tissue generates different force

Single AP resulting force

• Fast fatigable > Fast fatigue resistant > Slow

Repeated stimulation

• Fast fatigable

  • Force decreases rapidly over time

• Fast fatigue resistant

  • Force decreases at slower rate, consistent but gradual decreases

• Slow

  • Consistent, doesn’t fatigue, always 100% force

Progressive recruitment of motor units during muscle contraction

Gradual progression of recuruitment of motor units

• Start gently, then need more force

• Slow → Fast fatigue resistant → fast fatigable

Size principle

• Recruitment of smaller S motor units first (sustained oxidative fibers), don’t innovate as many fibers

• Then larger motor units (innervate many fibers), with powerful fatigable, glycolytic fibers

An input to a motor pool (like from UM system) ramps up

• Smaller motor units recruited first, generating small amount first

  • With more force, progressively larger motor units (in same pool) are recruited, firing rate increases

Local interneuron circuits

Most movements involve coordination of activity in multiple muscles (synergist and antagonist)

• When contract muscle, have to relax antagonist (flex vs extend)

• Coordinate AMN between bicep flexing, tricep relaxing

Some of this coordination accomplished through local interneuron circuits in the spinal cord

• Medial local circuit neurons

  • Tend to be longer distance, terminate bilaterally

  • Postural control, coordinated limb movements

• Lateral local circuit neurons

  • Shorter distance, unilateral

    • Fine movements in fingers

Muscle spindle is a sensory organ that signals stretch

• Muscles spindles are bundles of intrafusal fibers, innervated by gamma motorneurons

• Intrafusal fibers consist of 2 components, bag fibers and chain fibers

• Bag fibers

  • Innervated by Group 1a afferent (sensory axons)

  • Signal stretch velocity, (changes in stretch: initiates stretch. reflex, maintain muscle tone)

  • Muscle spindles are bundles of bag fibers

  • Initiates stretch reflex

• Chain fibers

  • Innervated by Group 2 afferent axons

  • Signal stretch amount

    • Important for proprioception, sense of limb and body position

  • Even if no change in stretch, still fire

  • Not source of stretch reflex

Gamma motor neurons enable muscle spindles to signal changes in stretch across wide range of muscle lengths

Muscle spindles need to adjust their length to match that of muscle itself

• Have to ensure some tension when muscle contracts, other just slack

Otherwise spindles will not be be able to detect stretch, because spindle is slack fro contracted muscle

By activating alpha and gamma motor neurons together, spindle afferent stay in correct dynamic range (range of stretch levels in which changes can be detected)

Pathway:

• Without gamma

  • Stimulate AMN, extrafusal muscle fibers contract, increase in muscle force

  • IF muscle fibers don’t contract, slack, can’t detect EF fiber change

    • Gap in afferent activity

• With gamma

  • Stimulate AMN, muscle EF fiber contracts

  • Stimulation of GMN, muscle IF fiber contracts to match length, can signal afferent activity Ia response when muscle contracts

GMN set muscle tone

Muslces always under some degree of stretch, so GMN involved in maintenance

• Steadu baseline level of tension referred to as muscle tone

• If y not active, slack muscle spindles, not responsive to change in stretch

Hypotonia: state of low muscle tone (floppy baby syndrome) involves various disruptions to control and signaling in muscle spindles

• Causes: genetic (Down syndrome), early neonatal (muscle dystrophy), autoimmune

Muscle afferents are fastest axons in human body

Thick = fast

Muscle spindle receptor (proprioception)> Touch (Ia, II) > A beta, A delta (Pain, temp itch)

Fast signaling crucial for feedback control

Feedforward vs feedback control

• Cannon

  • No sensors

  • Once launched, no opportunity for control

• Saturn V rocket

  • Equipped with sensors (gyroscope) and controls

  • Can course correct

  • “feedback”, muscle feedback only works if fast enough

3 spinal cord reflex circuits that rely on inhibitory feedback control1

• Stretch reflex (myotonic, knee jerk)

  • Negative FB loop to maintain muscle length

  • Passive stretch, muscle spindle longe than expected

  • 1a afferents signal stretch, contract homonymous muscle

• Polysynaptic inhibitory reflex (reverse myotonic, clasp knife)

  • Negative FB loop to maintain muscle force

  • Active contraction: GTO detects muscle force

  • 1b afferents signal froce

  • Relax homonymous muscle through inhibitory interneuron

• Flexion reflex (withdrawal)

  • Neg FB loop to maintain pain free state

  • AD and other nociceptive fibers signal threat

  • Contract ipsilateral flexor muscle, relax extensor muscle

  • Local circuit coordinates with contralateral side to maintain balance

Stretch reflex

Muscle spindles detect stretch

• Soda poured into cup held by arm, bicep EF fiber passively stretched, IF stretches, reports via 1a afferent axon

• Gamma motor neuron activity sets muscle tone

  • More GMN and IF activity → shorter muscle spindle→ faster and larger response to stretch

  • Sets gain of feedback loop, how much you correct given perturbation away from set point

1a sensory afferents signal change, synapse onto homonymous AMN (innervating same muscle 1a afferent came from)

• 1a goes from bicep to AMN that innervates bicep

makes negative feedback loop, synergistic muscle now contracts (antagonist relaxes), counteracting stretch and restoring to original position

• AMN contracts bicep more, interneuron inhibits antagonist, relaxes it

• Undoes whatever external force acted upon it

Knee jerk reflex

• Tap on muscle tendon under knee

• Passively pulls on/stretches quadriceps EF muscle fibers

• Reported to AMN via 1a axon from muscle spindle

• AMN fires and contracts quad, inhibits antagonist, leg lengthens

Muscle spindles (bag fibers) initiate stretch reflex, not chain fibers

Golgi Tendon Organ

Between muscle fibers and Tendon

Muscle spindles signal passive stretch (because if active, gamma motor neurons compensate)

• Feedback system for maintaining length

Passive stretch

• IF muscle fibers report muscle lengthening/ EF stretch via spindle afferent, feedback to maintain muscle length

• GTO doesn’t report passive stretching of muscle, afferent activity stays the same

GTO signals active contraction (because if passive, muscle but not GTO stretches first)

• Feedback system for maintaining force

• Active contraction

  • Stimulate AMN, muscle EF contracts

  • Muscle spindle goes slack

  • GTO signals shortening of muscle via increasing in spikes

Information about muscle contraction from GTO carried by 1b afferents, initiates reflex loop to reduce muscle tension

• Pathway

  • GTO senses tension in muscle, sends signal along 1b afferent

  • 1b inhibitory interneuron inhibits homonymous muscle (relax), excites antagonist

• Prevents muscles from generating excessive tension (preventing damage)

• Also known as polysynaptic inhibitory reflex, clasp knife, or reverse myotatic reflex

Can also damage is passive movement too large, can’t absorb all force

Flexion reflex

Pain withdrawal reflex

Mediated by pain fibers (A delta) originating from nociceptors

Involves local circuit neurons to coordinate withdrawal on ipsilateral side with extension on contralateral side to maintain balance

Can be modulated by descending pathways

Pathway

• Step on nail, cutaneous receptor activated, sends signal via AD

• Excitatory interneuron excites antagonist (flexor), inhibitory interneuron inhibits extensor, flex leg away from painful stimulus

• Opposite in other leg, extends to support (excite extensor muscle, inhibits flexor muscle)

GMN enable muscle spindles to signal changes in stretch across wide range of muscle lengths

GMNs need to adjust length to match that of muscle

Otherwise spindles will not be able to detect stretch, because spindle is lack fro contracted muscle

By activating A and G MNs together, the spindle afferents stay in correct dynamic range (range of stretch levels in which changes can be detected)

Passive movement

• No parallel motor command to IF muscle fibers

• Longer than expected

• Increase Ia afferent firing

Active movement

• parallel motor command to IF fibers

• As expected

• No change in Ia afferent firing rate

Motor Handout

• 1a afferents respond to changes in muscle stretch, signal provided by muscle spindles

• Signal encoded by muscle spindle afferents describe by tuning curve

  • X and Y is IF fiber length and firing rate

  • Above certain length, no more changes in firing rate (response saturates), under certain length, firing rate is zero (spindle slack)

  • Between these two, linear increase from fiber length and firing rate

• Small change in muscle length cannot be detected from 1a afferent activity when in saturation or slack phase

• Gamma motor neurons make sure IF Fibers at length at which they can respond to changes

Stretch reflex

• Correct for passive stretch, sense stretch occurring, trigger motor response to correct for it

• Other explanation is that it attempts to correct for stretch prediction error

  • Error = EF length - IF length

• Voluntary movement

  • ALpha and gamma fire at same time

  • IF and EF fiber length shorten at same time, in tandem

  • No error, Ia afferent fibers don’t fire, no difference between EF and IF

• Obstruction preventing EF fibers from contracting even when motor command issued

  • Alpha and gamma fire at same time

  • IF fibers shorten in prediction of EF shortening, but EF don’t shorten/contract

  • Error goes up when disconnected, EF more stretched than expected

  • 1a afferent fibers increased fire, signal muscle that IF contracted, but not EF contracted, now difference between stretches

    • Spindle stretched relative to muscle

  • IF stretched too much, 1a fires, AMN fires, contracts EF even harder

• Knee jerk reflex

  • No motor commands from either A or G

    • EF lengthen via passive force

    • Tension on IF because no change in length

      • Increases 1a firing, mismatch in length

      • Same error, EF longer than IF

      • Increases 1a firing

• Signals prediction error

  • Predicts something contracts, but doesn’t contract, 1a fires

  • Predicts something stays, but contracts, 1a fires

Lower motor circuits can self generate patterns

Rhythmic behaviors (breathing, walking, swimming) consist of repeated stereotyped patterns of muscle activation

After transecting spinal cord, walking pattern for cat hind legs remains intact, showing the pattern is generated locally, not in motor cortex

• COuldn’t voluntarily walk, but could walk as reflex on treadmill

• Speed and initiation of pattern controlled by mesencephalic locator region

Pacemaker neuron

• Single neuron that generates persistent activity in absence of external input *(may or may not be rhythmic)

Central pattern generator

• Small network of neurons, including one or more pacemaker neurons, generating repeating pattern/rhythm

• Build networks of neurons, even if not pacemaker neurons, phenomenon is generation of pattern

Simple patterns

• Two pacemaker neurons (one rhythmic, one tonically active)

  • Oscillate out of phase with each other once coupled

• two tonically active neurons

  • Still can generate rhythmic activity once coupled

    • Reciprocal inhibition when coupled, out of phase

Pattern generators for swimming

Alternate contraction of dorsal longitudinal muscle and ventral longitudinal muscle

• Out of phase propagating waves between dorsal and ventral cell, alternate contract and relax

Reciprocal inhibition between dorsal and ventral pacemaker motor neurons, establishes basic pattern swim in leech

• How basic motor actions founded, then cortex added on top

Other examples of simple motor CPGs

Swimming in lamprey

• Small neuron networks generate rhythmic movement

Stomach movements in crab crayfish

Human spinal cord CPGs

Can’t give voluntary motor commands to legs

But if stimulate spine, can start CPG

• Electrical stimulation of lumbar spinal cord in humans with complete long standing spinal cord injury can induce patterned locomotor like activity

Selecting between different lower motor patterns

Which pattern to execute

Swimming and crawling in leech controlled by different motor patterns

• Shown in graph by recording from leech ganglion roots, as well as by voltage sensitive dye imaging

How are patterns selected

• Different stimuli causes leech to walk or swim (IC electrode)

• Bottom of pool

Exoperimt

• Stimulate sensory neuron (DP) intracellularly such that sometimes response is swim, other times crawl

• Activity of a specific neuron (cell 208) predicted with high accuracy which response elicited

  • If active, swim, inactive, crawl

Overall organization of motor systems is hierarchy

• Local circuits in lower motor system can do refelexes, basic motor programs

• Commanded, modulated, gated, sequenced, done by higher circuits

• Sensory feedback from muscles (proprtiocpetion) is key, not just top down

• Add high levels to modulated lower simple parts

Cell 280 example of higher control

Mike headless chicken

Decapitated by ax, lost head, lived

• Retained brainstem, jugular

  • Locomotion and balance

Separation of descending motor tracts

Direct to Lower motor tracts vs Indirect, goes to other upper motor tracts

Lateral descending tracts

• Lateral corticospinal

  • pyramidal - Descends in pyramidal medulla tract

    • Direct to Lower motor neurons

  • Crossed

• Corticobulbar

  • pyramidal

  • Crossed

• Rubrospinal

  • Extrapyramidal (doesn’t descend in pyramidal tract)

    • More indirect, to lower an upper motor centers

  • Crossed

Ventral-medial

• Ventral corticospinal

  • Uncrossed

  • P

• Vestibulospinal

  • Uncrossed

  • EP

• Reticulospinal

  • Uncrossed

  • EP

• Tectospinal

  • Crossed

  • EP

EP = Vestibulo, rubro, tecto, retículo

Cortex vs Brainstem

Upper motor control results from two distinct but cooperating system:

• Motor cortex (M1, premotor areas)

• Brainstem centers vestibular, reticular, collicular, rural)

Locations of descending tracts from motor cortex (lateral white matter) and brainstem (medial white matter) gives clues to respective roles (proximal muscles = medial, distal muscles = lateral)

Lateral vs Ventral

Lateral division

• Involved in fine control of distal musculature, flexor muscles

ventral division

• Involved in control of muscles to trunk

Lateral vs ventral corticospinal tract

Corticobulbar tract: a primary descending motor pathway that carries voluntary motor commands from the cerebral cortex to the brainstem. It connects the motor cortex to the cranial nerve nuclei

Motor cortex

Anterior to central sulcus

Primary motor cortex (M1, Area 4) and associated premotor and supplementary motor areas contain upper motor neurons, which project either directly to LMNs in brainstem or spinal cord or indirectly onto local interneuron circuits

Neurons direct to LMN, or indirect to other UMN

Upper motor neurons in motor cortex reside in Layer 5

• In M1, 10% of these are Betz cell, largest cell bodies in human brain

  • Part of descending path directly to LMN and circuits

  • Big b/c need axons to travel far distance, need lot of machinery

Projections from M1

• Layer 5

  • Contains most long range projection neurons

  • Target local circuits and motor neurons in brainstem and spinal cord

    • Also striatum and other cortical areas

Start of direct paths

• M1 to BS is corticobulbar tract

  • Upper motor neurons in motor cortex control face and tongue muscles (by contacting local circuit neurons in brainstem nuclei)

  • Also cingulate motor area, indirectly innervates eyebrow muscles

• M1 to spinal cord is Corticospinal tract (motor control of rest of body)

  • Splits

    • ventral

      • Uncrossed but some bilateral

      • Contacts local circuits for axial and proximal muscles

    • lateral

      • Crossed

      • Contacts local circuits for distal muscles, some UMN contact AMN directly

        • Privileged access for fast fine movements of distal muscles

Both tracts also project to dorsal spinal cord regions (sensory, modulation)

Also projects thalamus, contralateral cortex 2ndary MA and Premotor Area, ipsilateral cortex all 3 areas, SS cortex, striatum

Coarse topographical organization in M1

Origin of corticobulbar (lateral M1) and corticospinal (dorsal/medial M1) tracts includes topographical organization in M1

• Spine/body separated from face

  • Upper body of CS tract next to face/CB tract, shows topo organization

Homunculus not only in primary motor cortex, also in Premotor MA, Supplementary MA

Not nearly as precise as in somatosensory cortex because M1 encodes movements not muscles

Also seen in spine

M1 encodes movements, not muscles

Put electrode in M1, prolonged stimulation, see what effect is

• results in purposeful movements

  • Bring hand to face as in eating

  • bring face out into space, as if to inspect object

• Independent of starting position

  • All different starting positions (+), all end up in same general place in middle, or all end up outside of body

  • In each instance, common end point reached independent of starting position

  • Fits with notion of motor hierarchy

    • Lower levels figure out details to accomplish movements, don’t know about overall goal

    • upper levels know overall goal, not minute movement details

• Also works with facial expressions

Mosaic map of motor movements

• Idea of smooth topographic motor map oversimplfiied

• Microstimulation studies reveal mosaic organization with only rough topography

  • Rough topography, fragmented/mosaic

    • Several portions of M1 for same muscle, spread out

  • For many endpoints, need to use same muscles

    • Map of target movements, not individual muscles

Single M1 neuron excites many individual muscles

EMG activity in single muscle following single M1 neuron spike

• See large spike in many muscles, coordinated movement

• movements, not muscles

Spike triggered averaging

• he Spike-Triggered Average is a neurophysiology technique that works backward from a neuron's electrical spikes to find the average stimulus feature that caused them to fire.

Corticospinal tract neurons contact lower motor neurons that innervate multiple muscles

CS tract from M1, connects to lower motor neurons

• Each lower motor neuron synapses multiple times with same muscle, motor unit

  • Connected to different motor units, coordinated movements

• All motor neurons don’t go onto same muscle

What do neurons in M1 encode

Monkey cued to make movement to light

• Record M1 neurons as monkey moves hand from center position to illuminated position

Raster plot generated

• Each row is trial, each dot is spike

• Recording for same neuron when monkey moves in different directions

  • Increased spikes = increase firing = increased preference for that direction

Population coding in M1

Population coding in M1

• Primary motor cortex cells broadly turned to motion direction

• Given firing rate consistent with multiple directions

  • Cosine tuning curve for single cell

    • Similar firing for bunch of directions

    • Can’t just take firing and guess movements

    • Have to pool info from different neurons, summate signals

• Precise representation of movement direction requires summation of signals of a population of motor cortex cells

• Simultaneous activity over array of cells, referred to as vector or population coding

Each cell has preferred direction

• Cell 1 likes 135, 2 likes 30, 3 likes 270 degrees

• Population vector means give each a vote, make vector of all of them

• If 135 neuron active, pulls population vector towards it, movement towards it

  • 135 vector larger than 30 and 270 vector

All directions in 3D equally represented

Motor cortex cells tuned to motion in 3D not just 2

Cell surveys show that preferred direction of cell varies, but all directions of motion equally represented

Brian control of prosthetic limb

Understanding coding of movement in motor cortex made it possible to decode this activity and use it to control a robotic arm using brain activity alone

Robotic arm reads population vector of M1 neurons of monkey

problems: infection, M1 changes with learning, readouts not consistent

Premotor areasd

Premotor cortex

• Supplies 30% of axons in CS tract

Lateral premotor areas

• Important for preparing and initiating movements in response to cues (also mirror neurons)

broca’s area

• Rostro-lateral premotor

• Speech production

medial pre-motor/supplementary motor areas

• Important for initiating spontaneous, self initiated movements

• Not response to cue

Cingulate motor areas

• Facial expressions

Example of preparatory activity

During delay task in which monkey must withhold response/movement to cue, pending go cue

• Population vector points in correct direction throughout delay

Muscle potentail EMG recordings from arm showed no muscle activity during delay

• Suggests premotor cortex involved in motor planning, not just execution

Experiment 2

• monkey has to make movement 90 degrees from cued direction

  • Cue 1 way then change cue other way

• under conditions, movement vector initially points in direction of cue, then rotates to direction of actual movement

• Like mental rotation

  • if monkey messes up, see if error in premotor area, or downstream

Mirror neurons in ventral premotor cortex

Compare neuron between conditions of subject making movement vs observing someone else make movememnt

• See some firing when see someone else grasp button with hand, then more firing when movement itself made

• See no firing of mirror neurons when someone else uses pliers, then monkey grasps with hand

Mirror neurons respond not just when making movement, but also when obvsebing same movement

• Can be specific (no response to pliers)

  • Role in imitation leanring, empathy?

Indirect paths

Vestibule, retiruclo, colliculo/tecto, rubro

Descending systems, some are crossed

Spinal cord ventral horn terminatino

• Colliculospinal terminates in cervical portion, for head and neck movements

Lateral and medial vestibulospinal tracts

Use info from vestibular system

Involved in muscle movements for postural adjustments

• Mainly involving axial muscles

• Excitatory to extensor muscles

• Balance, upright posture, head stabilization

Lateral tract

• Helps maintain upright and balanced posture

• Inneravtion of leg extensor muscles

medial tract

• Stabilizes head position by innervating neck muscles

vestibulospinal tracts active in cat right reflex

Cat righting reflex augmented by flexible backbone, absence of collarbone in skeleton of cat

tremendous flexibility, upper body rotation

By turning the head and forefeet while falling, rest of the body naturally follows and cat is re-able to orient itself

Reticular formation: Loose (net like) collection of neuronal clusters

Extends through Brainstem, well placed to coordinate muscle groups, spread out net

Rostral part:

• Modulates forebrain activity

• Part of reticular formation that modulates corticothalamic activity and responsiveness, control of arousal, awakening, consciousness (reticular activating system)

Caudal part

• Controls muscles involved in balance

• Motor reflex like sneezing, hiccuping, yawning, swallowing, gagging, vominting, laughing, crying

• Cardio and resp control

Reticular formation also involved in coordinating anticipatory muscle activity

• Experiment

  • Bell sounds, have to pull lever

  • In advance of bicep pulling lever, see contraction of leg muscle to adjust balance

  • B4 bicep contracts, anticipate change in posture from bicep contraction, see leg contract BEFORE biceps contract

• Motor hierarchy

  • LMN in leg doesn’t know why contracting

  • Higher levels know need to keep balance

• Also vestibulospinal helps out for unanticipated postural instability

  • EF/IF fibers correct

Direct vs indirect control of movement by motor cortex

• Similar reaching experiment in cats

  • Lifting and extending forepaw requires postural adjustments of other legs

• Inactivated caudal reticular formation

  • Forepaw movement (direct pathway from cortex to spinal cord) intact

  • Feedforward postural adjustments (indirect pathway, reticular formation) impaired

Collicoluspinal/tectospinal tract

Descending projections from superior colliculus coordinate head and trunk movements (orienting) with eye movements

• Orient to something in periphery

Important mech in many movements we make

• Shift in gaze (saccade) precedes reaching and orienting movements

Most descending projections from colliculus are actually indirect, synapsing in reticular formation

Rubrospinal tract

Controls distal musculature of the upper body, not fine motor control in fingers (corticospinal)

• Hand withdrawal reflex

Named because of origin in red nucleus in midbrain

lower and upper motor systems interact

Stretch and withdrawal reflex mediated through spinal cord circuits (no cortex)

• But can be modulated and coordinated by upper motor control

Cortical motor area generally do not innervate motor neurons directly (exception is fine motor control in fingers)

• But do so through brainstem centers (reticular formation and Local circuit neurons)

• Motor systems can be thought of as a hierarchy

brain architecture as superposition of loops

• Small loop is just spinal cord, large loop is full cortex

Lower motor disorders

Duchenne muscular dystrophy, ALS, Polio, Spinal cord injury, NMJ autoimmune syndromes

Involves muscle weakness, decreased tone, decreased reflexes, muscle atrophy

• Muscles gradually waste away, or lose ability to control them

Clinical signs different from damage to upper motor centers and pathways

• Spasticity, rigidity, loss of fine motor control

Amyotrophic lateral sclerosis ALS

Progressive death of AMNs initially in lateral ventral horn, progresses to other areas, eventually causes death

Duchenne muscular dystrophy

Characterized by progressive muscle weakness and wasting

genetic component

• X linked

• Recessive mutation in gene for dystrophin

  • Structural protein that binds actin in sarcolemma

• Causes muscle fibers to tear and die

No cure

Upper motor system disorders

nothing wrong with muscles themselves

When pyramidal tracts severed

• Initially

  • Paralysis and spasticity of distal muscles on contralateral side

  • Posture Ok

• Some recovery (not all fibers cross over, compensation through other tracts)

  • Persistent loss of fine motor control and some loss of strength

    • Damage to motor cortex results in similar symptoms (if partial, less severe/more localized)

Population vector handout

Tuning curve

• X axis = direction of reaching

• Y axis = firing rate

• Can’t decode reaching direction from one neuron because same spike level/threshold can correspond to different directions (cosine curve)

  • Activity of individual neuron varies a lot second to second as well

Record spikes from 5 M1 neurons simultaneously, each with different preferred direction

• Now arrange their tuning curves according to preferred direction, 0 degrees up, 90 degrees right

Question: If subject moves 45 degrees draw vector for each neuron

• Mark 45 degrees on x axis, note firing rate

• Draw arrow length proportional to firing rate

• Direction of movement is vector sum of arrows

  • Each neuron votes for preferred direction, strength proportional to firing rate

  • Population vector, direction corresponds to direction encoded by population

See population vector pointing in 90 degree direction, decode neural activity, see they made movement in 90 degree direction

length of population vector

• Smal arrow

  • Low firing rate

  • vectors cancel out, going in opposite directions

• When neurons don;t agree, no movement, maybe thinking

• Long population vector

  • Confident decision, make activity, could correlate to strength/speed

Basal ganglia

Do not control motor neurons or local circuits directly

• influence movement by regulating activity of upper motor neurons, mostly in cortex

Collection of deep brain structures

• (Neo)striatum

  • consists of caudate and putamen

  • Also nucleus accumbent, olfactory tubercle

  • Input stage of BG

• GPe (Global pallidus external segment)

• GPi (Global pallidus internal segment)

• STN (sub thalamic nucleus)

• SNr (substantial nigra, pars reticulata)

• SNc (substantia nigra, pars compacta)

Deep brain structures bound together by parallel loops in anatomy

• Forms parallel loops with cortex

• Striatum in input stage of basal ganglia (corticostriatal pathway)

• Basal ganglia are topographically organized, forming parallel loops that project back (through thalamus) to same cortical area the input came from

  • Every part in cortex has own BG looop

• 2 paths

  • Direct path from cortex (excitatory) to striatum (inhibitory) to GPi (inhibitory), then to thalamus(excitatory) back to cortex

  • Indirect path: NS to SN back to NS

    • GPE to STN to GPI

    • Both paths back to origin are excitatory

  • Excitatory = Glutamate, layer 5 pyramidal neurons, inhibitory = GABA (atypical)

    • SN has dopaminergic projections

Inputs to basal ganglia

Input to striatum from cortex (corticostriatal)

• Nearly all regions of cortex project striatum, except V1

Caudate:

• Inputs from Frontal eye areas (FEF), multimodal association area (frontal, motor)

Putamen

• Inputs from sensory, visual (not V1), (pre)motor, auditory cortical areas

Striatum also gets dopaminergic input from SNc and VTA

BG Inputs: parallel, topographical loops

• Projections from cortex to striatum are excitatory and topographically organized

  • Adjacent areas of cortex project to adjacent areas of striatum

• Projection neurons of striatum are medium spiny neurons

  • Large spiny dendritic arbors

  • Receive glutamatergic inputs from up to 10,000 cortical neurons

  • Spines allow gathering of info from BG

  • Gabaergic, inhibitory

3 motor circuits in Basal ganglia (after striatum)

• Collicular pathway

  • Caudate/putamen to SNr to Superior Colliculus

  • Atypical, like motor circuit, eye movements

• Direct pathway

  • Caudate/putamen to GPi to thalamus to cortex

• Indirect pathway

  • caudate/putamen to GPe to STN to GPi to thalamys to cortex

Collicular pathway

• Caudate/putamen to SNr to Superior Colliculus

Caudate MSNs and SNr projection neurons both inhibitory

SNr cells tonically active, always inhibiting SC

Caudate activity leads to disinhibition of colliculus neurons

• Activation of Caudate, leads to inhibition of tonically active SNr, leads to loss of inhibition of colliculus, leads to eye movement

BG direct pathway

• Caudate/putamen to GPi to thalamus to cortex

• Same principle as collicular pathway, different structure

  • inhibitory neurons in GPi are tonically active

  • Activation of Caudate/Putamen MSNs leads to inhibition of tonically active inhibitory GPI neurons, leads to disinhibition/release in inhibition of thalamus, excitation of motor cortex

  • Transient excitatory inputs from Cortex to Cpu

Indirect pathway

CPu → GPe→ GPi→ thalamus

• opposite sign to direct pathway

• Inhibits thalamus/cortex

• - - - = -

Also though surround inhibition of STN (- - + - = -)

Basal ganglia as action selection system

Direct “Go” pathway = pros of action, facilitates specific, selected action

Indirect “NoGO” pathway = cons of action, diffuse inhibition to non selected actions

Need to suppress or promote actions

• Actions depend on distance, muscles activated, motivation

Different populations of neurons within each area represent different possible actions to be selected between

Can’t go by individual neuron

Like center surround

• Intended action is like center

• Suppress other actions like surround, shut down alternate/similar actions

• Cortex provides many possible actions, then BG loop until one action dominates via most active responbse

Dopaminergic inputs to basal ganglia

Dopamine = neuromodulator, modify activity (plasticity, learning) rather than direct/fast transmission

But does affect BG directly

Dopamine releasing neuron in SNc project to Cpu MSNs in striatum

CPU MSNs have 2 types

• D1R expressing MSNs

  • Direct go pathway

  • High tonic dopamine, D1R excited, increased activity in neuron

• D2R Expressing MSNs

  • High tonic dopamine, D2R is less excitable, decreased activity

    • Different G protein, decreased CAMP

Parkinsons disease

Progressive DAergic neuron degeneration

• Less dopamine released in striatum from SNc

D1 neurons less active, less GPi inhibition → “Go” GPi neurons fire more → thalamus Go signal more inhibited, less likely to select correct action

• Weaker +

D2 neurons more active, NoGO GPi neurons fire more, thalamus NoGo more inhibited, less likely to select incorrect actions

• D2 inhibition increased, stronger -

Explains clinical observations of difficulty in initiating actions, slow actions

Sx

• Slowness of voluntary movements

• Difficulty initiating and stopping movements

• Tremors at rest, but not when moving

• Rigidity

  • Increased resistance to passive displacement

Less dark neurons on photo of SN

Deep brain stimulation in STN, improves tremors

• Site of action unclear, can include local circuits, off target effects, downstream action sites via axon projections

Why does DBS work

• EEG electrodes record a signal that reflects summed contributions of many sources (mostly synaptic inputs, not spikes)

  • Synced neurons result in stronger summed EEG, desynced has less net effect

• Intracranial EEG is referred to as LFP, local field potential

  • Like stadium noise

Pathological oscillations in BG

• in BG patients, LFPs can be recorded from DBS electrodes, often show elevated, pathological oscillations in Beta range (12-30 Hz)

• DBS reduces power of beta oscillations, reduces peak in Beta range in Subthalamic Nucleus

  • Hypothesis that pathological oscillations prevent normal operation of BG network

PD treatments

Pharmacological

• Systemically administer L-Dopa, dopamine precursor, to restore normal dopamine levels

• Pro: non invasive

• Con: DA degeneration is not uniform, causes excess dopamine in not yet degenerated areas, only works in early stages

DBS

• Chronically implant electrode in BG, break pathological rhythms

• Pro: can be effective after L-DOPA fails

• Con: Does not stop degeneration, invasive, mechanisms unclear

Stem cell therapy

• Can potentially stop/reverse DA neuron degenration

• Con: UInproven in humans, invasive

Huntington’s Disease

D2R MSN degeneration

Degeneration of indirect pathway

Uncontrollable movements, jerky, rapid

Movements with no clear purpose

Sometimes occur in facial musculature

Onset in 30s, hereditary, chromosome 4

Pathway

• Fewer D2 inputs to GPe, GPe more active/less inhibited, NoGo GPi neurons fire less, thalamus NoGo less inhibited (more likely to select incorrect actions)

• Explains clinical observations of undesired ballistic movements

Complexity of BG network requires computational modeling

To understand better, many pathological/path predictions come from these models

BG involvement in motivational, affective, cognitive processes

• Parallel loops, most cortices + area, interact w/ BG

• Topographucally organized

Motor loops: body movement, oculomotor

Non-motor loops: prefrontal loops, limbic loop

Beyond action selection: habits and the basal ganglia

Many everyday decision involve more than just one action at a time

Rather they involve sequences of actions, which can be highly stereotyped

• Syntactic chains in rat grooming

  • Always have same grooming sequence repeated over and over again

• Walking to familiar room or office in building

Habitual sequence, can do other actions during habit, don’t need to think about it

make life easier, but also can be problematic

Defining habits

Defining habits

• Watson and Carr, early rat experiment

  • Train them on same complex maze again and again until they can run it super quickly

  • Change a wall, rats run headfirst right into it

  • Don’t look, just do it on autopilot even though environment changed

• 2nd experiment

  • Place rats in maze that looks like a cross

  • Overtrain them, always turn left to get food

    • When placed in new starting location, execute fixed response, turn left even though won’t get them food

    • Strategy depends of dorsal striatum

    • Same response after many trials requires dStr

    • Lesion dStr, no habits

  • If inactivated, rats revert to place strategy

    • Dependent on hippocampus

    • Go to same place after change of location after few trials requires hippocampus

    • Lesion HC, no place strategy

The place strategy is a hippocampus-dependent navigation method where the brain relies on an internal "cognitive map" of the environment to locate a destination. Rather than following rigid sequences of turns, it provides flexible, goal-oriented navigation using allocentric spatial cues (e.g., landmarks and room-coordinates

Insensitivity to devaluation

• training phase

  • Lever 1 results in water, lever 2 in food

  • Equal preference for both

• Devaluation phase

  • Specific satiation

  • Devalue one lever, stuff rat with endless food

Experiment

• Testing phase (extinction) after few trials

  • Rat will prefer water lever

  • Lower preference for food since been stuffed with it

  • Sensitive to devaluation, goal directed

    • Requires dorsomedial striatum

• Testing phase (extinction) after many trials

  • No preference, each lever pressed equally

  • Overtrained

  • Rat will press each lever out of habit even though stuffed with food

  • Insensitive to devaluation, habitual

    • Requires dorsolateral striatum

Working definition of habits

Habits are actions (action sequences) that are insensitive to changes in goal value

A kind of stimulus response strategy

• If in box press lever

• If in elevator turn left

Learning habits

All about dopamine (neuromodulator, not fast response)

Midbrain dopamine pathways

• Nigrostriatal (SNc)

  • main dopamine pathway to caudate and dorsal putamen

• Mesolimbic (VTA)

  • Not cortical

  • Dopaminergic input to ventral putamen, nucleus accumbent, prefrontal cortex, other limbic areas

• Mesocrotuical (VTA)

  • Diffuse input to other cortical areas

Dopamineergic neuron synapses localize near cortical pyramidal/MSN synapses, neurons

VTA and SNc adjacent and continuous to each other, similar inputs, firing properties

Midbrain dopamine neurons signal reward prediction error

Pavlovian conditioning

• initial neutral Conditioned stimulus repeatedly paired with unconditioned stimulus

• Animals learn that conditioned stimulus predicts reward delivery (US)

Recording VTA dopamine neuron

• No prediction, reward occurs

  • Positive prediction error

  • neuron increased fires when reward

  • Reward when not predicted

• Reward predicted, reward occurs

  • Neutral prediction error

  • After conditioning, increase in firing to prediction, not to reward

• Reward predicted, no reward occurs

  • Increased firing when reward predicted

  • Less firing when no reward occurs

  • negative prediction error

    • No reward when predicted

Model of Basal Ganglia as action selector

• Striatal MSNs with largest input will win and implement the choice/action

• Have dopamine neuron inputs that can signal prediction error, drive learning

Deciding between steak (MSN 1) and fish (MSN 2),

• More dopamine to MSN 2, fish because expect that fish will be better, neuron fires more at ordering of fish, expectation

  • However, when fish is awful, Negative RPE,

  • dopamine neurons pause, cause long term depression (LTD), on direct pathway “fish” neuron inputs)

  • Inputs to fish stratal neurons are weakened, so next time, steak neurons will have larger input, win competition, choose steak

• If fish better than expected, Positive RPE, neuron fires more at taste of fish

  • Dopamine neurons fire, Long term potentiation on direct pathway fish neurons

  • Input to fish striatal neurons are strengthened via LTP, next time fish neurons have even larger input, wins competition, fish is chosen

Demonstration of D1 and D2 pathways in learning

• If increased Dopamine way to learn from feedback, influences future actions

• Experiment

  • Instead of delivering actual reward, stimulate D1 (direct, better than expected) or D2 (indirect, worse than expected) MSNs directly via optogenetics

    • Increased DA (better than expected): Upregulated D1, downregulate D2

    • Decreased DA (worse than expected: Upregulated D2, downregulate D1

  • Touch left sensor, stimulate D1, Right sensor inactive

    • Increased D1 lever pressing,

  • OR Touch left sensor inactive, right sensor stimulate D2

    • Avoid D2 lever, negative feedback

  • Forms long term memory

    • Day 2 and Day 3 have higher dMSN and lower iMSN presses

Learning from feedback in PD patients

Subjects don’t know Japanese, learn which characters would predict reward

Compared PD Pt off meds vs PD pt on meds

Found that learning from reward (positive feedback, choose A) relies on potentiation of direct pathway (impaired in PD patients, better in PD on meds)

Found that learning from punishment (negative feedback, Avoid B) relates on potentiation of indirect pathway (improved in PD patients)

Breaking habits

Same experiment as before:

• rats in T maze, cue to go a direction, better with training, overtrained until insensitive to devaluation

• Do devaluation procedure

• Inactivate infra limbic cortex (ventromedial prefrontal cortex in humans) during devaluation

  • Rats sensitive to devaluation again, no longer habitual

    • precise mechanism is subject of research

basal ganglia: the limbic loop

Ventral part of striatum is called nucleus accumbent (has core and shell subdivisions)

• Involved in addiction

• Shares parallel loops and DAergic input architecture (from VTA) with other striatal subdivisions

• Uniquely, also receives inputs from amygdala and hippocampus

Basal ganglia on drugs

Every known drug of abuse has a site of action in basal ganglia dopamine circuit

• Within BG, NAcc has high conc. of opiate,cannabinoid receptors

Cocaine is dopamine transporter/reuptake inhibitor

• increases EC dopamine

• Hijack learning from feedback system, brain thinks something better than actually is

• Taking cocaine results in artificially large DA burst that cannot be compensated for by RPS,

  • Conseuqences for learning and subsequent choice: outcomes always better than expected

Adiction: a pathology of motivation and choice

More generally, exposure to drugs of abuse causes synaptic, molecular/cellular, circuit level changes in BG-dopamine circuit

Reinforcers hijack the existing circuitry to favor drug seeking rather than natural reward, which become less effective in engaging these circuits

Dopamine/RPE handout

Play slot game, 3 symbols line up, win some money

• Question 1: Positive and Negative RPE in a Slot Machine

  Positive RPE (δ(t) > 0): You win money when you didn't expect to (or win more than expected). For example, if you expected to lose $1 but the reels line up and you win $5, R(t) > E(t), so δ(t) is positive. Dopamine neurons fire more than baseline.

  Negative RPE (δ(t) < 0): You lose when you expected to win (or win less than expected). If you expected to win $2 but got nothing, R(t) < E(t), so δ(t) is negative. Dopamine neurons dip below baseline firing.

• When 2/3 slots line up, updated reward predictions

RPE = Received reward - expected reward

• Easy to quantify

Can use RPEs to learn from feedback

• If action better than expected, revise expectation upwards

• E (t+1) = E(t)+ Aδ(t), where A is learning rate, small number that signals change in prediction based on one event

• Expected value is previous expected value + error

L-DOPA

• Incrases dopamine on brain

• Artificially inflates RPE

• δ(𝑡) = 𝑅(𝑡) − 𝐸(𝑡) + 𝑐

  which implies that even if you

  receive what you expect, a positive prediction error is signalled (for positive c)

• Do more action than data warrants

• Since δ(t) is always artificially elevated, the update rule E(t+1) = E(t) + α·δ(t) means E(t) will continuously drift upward over time. The brain keeps revising expectations higher and higher, even if the slot machine keeps paying out the same amount. Eventually E(t) overshoots reality — the patient develops inflated expectations that can never quite be met, which has implications for compulsive/addictive behavior sometimes seen with L-DOPA treatment.

• Increased risk for gambling addiction, linked with dopamine

Cerebellum

Does not control motor neurons directly

Output influences movement by regulating UMN activity

• Primarily but not exclusively in brainstem

Contains Largest cell in human brain: cerebellar Purkinje cell (200,000 synapses from granule cells)

• Has many dendrites to combine inputs from many granule cells

Contains most numerous cell type in human brain: cerebellar granule cell (up to 50 billion, more than in entire cerebrum

Cerebellum 2

Pattern generator with ability to learn

Thought to primarily motor part of brain, important for proper coordination of muscle movements

• Synchronizes timing of activating of different muscles

Reaches largest size in humans, need for synergy of muscle in learned activities that require precision

Many connections with motor systems of brainstem, cortex, basal ganglia, thalamus

Cerebellum 3

Cerebellar hemisphere have 3 main anatomical lobes: anterior, posterior, flocculonodular

Also primary fissure, superior part of cerebellum

Each hemisphere concerned with coordinating movements on same side of body (ipsilateral)

3 main functional divisions:

Cerebrocerebellum,

• Lateral zone

• Receives info primarily from cerebral cortex

  • Particularly well developed in primates and humans

• Involved in overall planning and initiation of skilled, sequential, motor movements

• Concerned with what is going to be happening next

• Finely skilled movements

spinocerebellum,

• Somatosnesory topographic mapping in cerebellum

  • Contains multiple maps

    • Vermis vs intermediate zone

• Vermis + intermediate zone

  • Vermis

    • Control of muscle movements of core/axial body, neck, head, shoulders, eye movements

  • Intermediate zone

    • Control of muscle contractions in distal portions of upper and lower limbs

• Receives most information from spinal cord

• Regulates muscle tone and adjusting ongoing movements

• Posture/gait

vestibulocerebellum

• Flocculus + nodules + flocculonodular zone

• oldest part of cerebellum

• Receives input primarily from vestibular system

• body equilibrium, balance, posture, reflex eye movement

Cerebellar peduncles

3 pathways that connect cerebellum to rest of brain

Needs one input (Middle), one output (sup)

Superior CP

• Serves cerebrocerebellum

• OUTPUT pathway from deep cerebellar nuclei (dentate nucleus, interposed and fastigial nuclei) to VL/VA thalamus, red nucleus, superior colliculus, cortex

Middle CP

• Serves cerebrocerebellum

• INPUT pathway from pons

Inferior CP

• Spinocerebellum and vestibulocerebellum

• Mixed input and output, vestibular nuclei, reticular formation

Cerebellar inputs and outputs

Cortex, red nucleus, pons, inferior olive, spinal cord, vestibular nuclei

• Many mediated by middle CP

• different parts of cerebellum have different inputs + outputs

Some inputs and outputs cross midline

• Confusing but principle is the cerebellum concerned with ipsilateral side of the body

• Most but not all inputs and outputs synapse in cerebellar nuclei (4 per hemisphere: dentate, interposed (2) and fastigial)