Reflexes and pattern generation

reflexes and the generation of simple, repetitive patterns are among the
    simplest behaviors

reflex -- a simple, relatively stereotyped action caused by a specific stimulus

reflex arc -- the neural pathway that underlies a reflex
    usually consists at least three elements:
        sensory neuron, interneuron, and motoneuron
    interneuron sometimes called association neuron because it
        associates the sensory input with the motor output

reflexes can vary in the complexity of the pathway
    monosynaptic reflex -- just sensory and motor neurons, no interneuron
    polysynaptic reflex -- sensory, motor, and many association neurons

knee-jerk reflex: familiar test on physical exam
    also know as the monosynaptic stretch reflex
    main reflex pathway consists of a sensory neuron from a stretch
        receptor in a muscle to the motoneurons innervating the muscle
        (Ia sensory neurons and a motoneurons, to be precise)
    action -- tap on the tendon, stretch the muscle, activate the stretch
        receptors and sensory neurons, fire the motoneurons, contract
        the muscle -- knee-jerk!
    the neurophysiological role of the monosynaptic stretch reflex is to
        maintain the tension of a muscle at some particular value that is
        needed for proper motor control

flexor/crossed-extensor reflex -- example of a more complex reflex
    described by Sir Charles Sherrington in early 1900s
    its action is to withdraw a limb from a noxious stimulus
    it involves both excitation of synergistic muscles and inhibition of
        antagonist muscles, say, in your legs
    action -- noxious stimulus, activate pain sensitive sensory neurons,
        excite motoneurons to flexors on pain side and extensors on
        opposite side, but also inhibit extensor on pain side and flexors
        on opposite side -- leg is withdrawn without falling down!

scratch reflex -- example of a very complex and sophisticated reflex
    put a noxious stimulus on a turtle or frog and the scratch reflex will
        coordinate many muscles to raise a leg and wipe it away!

reflexes are specific:
for example, a locust will move its leg reflexively in response to
    stimulation of its leg sensory hairs
the direction of leg movement is appropriate to the location of the
    stimulated hair -- stimulate a hair on the front of the leg and the
    locust will move its leg back
the different sensory neurons involved contact different sets of
    interneurons, depending upon their location on the leg, and
    those interneurons contact a particular set of motoneurons that
    cause movement in the appropriate direction
the brain may use these sets of interneurons for non-reflexive
    movements also

chain of reflexes?? -- inspired by the work of Sherrington, many researchers
    thought that repetitive behaviors (like walking) were nothing but a
    chain of reflexes, where one motion excites receptors that reflexively
    cause another motion, which then excite receptors that reflexively
    cause the first motion, and so on in cycles

later research disproved the chain-of-reflexes hypothesis

innovative research proved that simple repetitive movements are not
    produced by a chain of reflexes

central pattern generator (CPG) -- a neural circuit capable of producing repetitive activity in the absence of any sensory input
    although a CPG doesn't need sensory input in order to produce
        oscillatory output, its activity can be affected by sensory input
    CPGs were discovered by Donald Wilson in 1961, when he
        demonstrated that the oscillatory output of the locust wing beat
        CPG was maintained in the complete absence of sensory input
    Wilson observed that the frequency of oscillation was 50% slower
        following the removal of the sensory input, but the oscillatory
        pattern was maintained intact
    CPGs are important for almost all rhythmic behaviors that animals
        engage in; although these CPGs can oscillate in the absence of
        sensory input, such input plays an important regulatory role

fictive pattern -- a motor pattern recorded in the absence of any actual
    movement of the animal
    Wilson recorded the activity of the motoneurons from the cut wing
        stumps of the locust -- a fictive pattern
    this pattern consisted of alternating bursts in the elevator and
        depressor motoneurons, where those of the (cut) hindwing were
        phase advanced relative to the (cut) forewing

modeling CPGs -- involves reciprocal inhibition
    computer modeling can provide insight into how neural systems work
    reciprocal inhibition -- the connectivity pattern in which two neurons
        inhibit each other
    reciprocal inhibition is a feature of many neural structures, and it lies
        at the heart of CPG models

in Wilson's model of the locust wing beat CPG, two neurons
    reciprocally inhibit each other, also they both receive the same
    constant input but one neuron is more sensitive than the other,
    and both neurons fatigue
    Wilson's CPG produces an oscillatory pattern as follows: the constant
        input comes on and excites the more sensitive neuron first; that
        neuron fires and inhibits the second neuron; after a short while
        the first neuron fatigues and releases the second neuron from
        inhibition; the second neuron is now free to respond to the
        constant input; the second neuron fires and inhibits the first;
        then the second neuron fatigues and releases the first from
        inhibition; again the first neuron fires and inhibits the second,
        and the cycle repeats

post-inhibitory rebound -- the tendency of a neuron to fire a burst
        following release from inhibition
    not all neurons show this property, but some do, and it works like this:
    hyperpolarization reduces the number of inactivated sodium channels
    in a nerve cell membrane; this reduces its firing threshold; if the
    nerve cell repolarizes abruptly (as when it's disinhibited), then
    the membrane potential crosses the threshold and the cell fires
the cell won't fire if the membrane repolarizes gradually, allowing the
    sodium channels to inactivate and thus raise the threshold

a simpler CPG model can be constructed using post-inhibitory rebound
    consider two reciprocally inhibitory neurons that fatigue and show
    post-inhibitory rebound -- they oscillate as follows:
    constant inhibitory input to the first neuron is released; the first
    neuron rebounds and starts to fire, inhibiting the second neuron;
    then the first neuron fatigues, and its inhibition of the second
    neuron is released; the second neuron rebounds and starts to
    fire, inhibiting the first neuron; then the second neuron fatigues
    and its inhibition of the first neuron is released; the first neuron
    again rebounds and the cycle repeats until it is broken by
    constant inhibitory input

real CPGs have features in common with the models
    both have neurons connected with reciprocal inhibition
    both are interposed between the brain and the motoneurons

real CPGs differ from the models is several important respects
    real CPGs are much more complex than the models:
        real CPGs have many more neurons
        real CPGs have more complex interconnections
        real CPGs appear to be composed of several sub-circuits
 
a well-studied example of a real CPG is the lobster stomatogastric ganglion
    the stomatogastric ganglion controls the oscillatory movements of the
        esophagus, pylorus (stomach), and gastric mill (including teeth)
        of the lobster
    the lobster stomatogastric ganglion contains several distinct CPGs
        (pyloric, gastric, and esophageal networks), each of which is
        composed of many sub-oscillators
    each CPG oscillates with its own rhythm, determined by the coupled
        interaction of its sub-oscillators

CPGs are found in vertebrates as well as in invertebrates
    a well-studied example of vertebrate CPGs are those that underlie
        undulatory swimming in the lamprey

individual muscle groups each have their own CPGs; synchronization
    between CPGs is brought about by coordinating interneurons
    a good example is supplied by the crawfish, each of whose
        swimmerets is controlled by its own CPG
    swimmeret beating is synchronized along the tail so that the rear ones
        beat first and the forward ones follow in sequence
    this synchronization is produced by coordinating interneurons that run
        between individual CPGs

inputs from neurons external to CPGs can modulate their activity
    these modulatory inputs can cause CPG sub-circuits to act as
        independent oscillators, or couple sub-circuits and even whole
        CPGs into super-circuits that express different rhythms
    an example of this is the coupling of the pyloric, gastric, and
        esophageal CPGs in the lobster stomatogastric ganglion by the
        PS neurons
    after they are coupled, the CPGs oscillate at a common frequency that
        is intermediate between the frequencies of the uncoupled CPGs
    the modulating inputs are associated with various transmitters, and
        effect certain critical cells in CPGs, turning some on others off