NROC34H3 Lecture Notes - Bistability, Krypto, Mollusca

42 views10 pages
5 Apr 2012
School
Department
Course
Professor
Lec 11: Decisions and Control of Behavior
Slide 1: Decision and Control of Behavior
Forming link btn sensory and motor systems
Decision-making from the neuoroethological point of view
Slide 2: Decisions
Diff approaches to study decisions
Psychological/cognitive science approach:
oLargely involves formal models of psychophysical data (= data from
experiments where you carefully control the stimulus conditions and
connect what qualities of the stimulus are associated w/ a particular beh or
discrimination (deciding what’s the same or diff)
oLow-resolution neural data (brain-imaging/EEG) = low spatial/temporal
resolution
oHumans & monkeys
Neuroethological/behavioural approach
oEmphasizes natural behavior
oFocused on circuit-level description
Leech paper is aiming at the possibility that there could be some unification of
these 2 approaches
Slide 3: A decision model
Architecture that represents a set of inputs which can be thought of as sensory
inputs that are sensitive to diff possibilities:
oSo, these could be motion-sensitive visual neurons in the brain of a monkey
that’s staring at cpu screen watching dots move in random ways; get reward
for picking right direction
oSo, have set of sensory inputs that are accumulating info in favor of diff
alternatives; may be watching diff axes of mvt; when one of these reaches
some critical level, then its activity is integrated by these integrators
oThe first one of the inputs to reach a critical level makes the decision; cross
cxns btn the integrators represent inhibition; competition among diff
sensory inputs representing diff alternatives
oThis is a standard way of thinking about how decisions may be
implemented in NS
oThis kind of approach is designed to describe response time of tasks such as
the one about the monkey staring at a screen, watching random dot pattern,
there’s some overall direction of mvt monkey indicates which direction
it thinks the dot is moving; can set this up so that there’s more or less time
to make the right decision and there’s a tradeoff between the speed of
making the choice and the accuracy of the choice
oCharacteristic relationship btn speed and accuracy
Slide 5: Typical neuroethology approach to decisions
Unlock document

This preview shows pages 1-3 of the document.
Unlock all 10 pages and 3 million more documents.

Already have an account? Log in
Idea of command neurons
Sensory info coming in; have pathways that are dedicated to particular behaviors
The process sensory info does or does not activate some critical element of the
nervous sys, often referred to as command neuron
And if it does activate command neuron, then beh occurs; one of the effects of
activating a command neuron could be to inhibit other command neurons
Some similarities to the abstract model from above; winner takes all mechanism
where the 1st decision making element to be activated shuts off its competitors
structure of processing circuit
Slide 6:
This kind of view: circuit-level description of the action of the NS and its relation to
behavior comes from a tradition that wasn’t focused on behavior at all but rather an
engineering approach to NS
oEx: gastronemius prep of a frog, nerve stimulated and see muscle contract
oPieces of animals used to look at little parts of the circuits; focused on low-
level properties of the NS (cell and synapse level)
But Sherrington wrote book: Integrative Fxn of the NS; where he imagined how
this small scale mechanism could be used to describe the NS as a whole and how it
controls behavior
oBased on idea of reflexes as fxn’l unit; neuron when activated has hard-
wired cxn to some effector (muscle), will always result in a certain beh
oIf many of these elaborately connected together automaton of animal;
chain of deterministic reflexes
oHigher mental fxns and emotions could never be addressed in this
mechanistic way
Earthworm:
oTouch it, quickly w/draws into burrow; quick and robust w/drawal reflex
oCircuit diagram of w/drawal reflex pathway; w/drawal reflex carried thru by
nerve cord from segment to segment
oIs this enough to describe the fxn of the NS?
oIs NS capable of endogenously generated activity?
Slide 8: Spontaneous/Endogenous Behavior
Is NS capable of endogenously generated activity? Whether or not everything was
just a stimulus-response set-up? NS is simply driven by stimuli that impinge on the
animal and are processed in some way OR whether you can have activity
spontaneously, independently generated by the NS itself?
This lead to the idea of central pattern generators, rhythmic activity
Rhythmic activity were key to demonstrating that endogenous beh is possible
Slide 9: Oscillation in Neural Networks
Breakthrough was that you can have self-sustained cyclical activity in NS that IS
NOT driven by sensory input
oRhythmic motor activity
Unlock document

This preview shows pages 1-3 of the document.
Unlock all 10 pages and 3 million more documents.

Already have an account? Log in
oSensory process (olfaction; cycles of synchronized activity as the basis of
coding mechanism)
oDynamic interplay btn central and sensory processes
There is nevertheless a role for sensory input for these processes
Slide 11: What does it take to create a pattern generator?
Simulated APs based on this circuit; 2 neurons, each w/ reciprocal inhibitory cxns,
each of which has the property of inhibitory rebound (= when its membrane is
hyperpolarized and that inhibition stops, it overshoots on its return to resting level
and fires an AP)
oWhich means when you inhibit one of these neurons, when it rebounds, it
has burst of APs that inhibits the other neuron; and when 2nd neuron
rebounds, it has burst of APs that inhibit 1st neuron alternating bursts of
activity btn these 2 neurons
oJust need something to get started, then have self-sustained osciallation of
activity
Can be modified in diff ways:
oAdditional neurons may interfere; inhibits one of them, results in changes in
timing reset phase of the cycle (stop and start one of them, will change
phase)
2 neurons with these properties: (reciprocal inhibition, inhibitory rebound) and
you’ll have a simple pattern generator
Slide 12: Molluscan Swim
2 fins, flips back and forth to swim
the neural circuits that controls those fin mvts is basically equivalent to simple
simulation from above
pair of neurons in the ganglion of fin of each side that are each connected to motor
neurons; one makes fins go down, the others make it go up; down and up neurons
reciprocally inhibitory relationship and alternating spikes in those 2 neurons in each
ganglion of each fin that controls the swimming mvts
Slide 14: Central Pattern Generators
1) this works in combination of BOTH: NETWORK AND INTRINSIC
PROPERTIES;
odepends on if the neurons are wired up; neurons have to be reciprocally
connected to each other
oand have certain intrinsic properties this inhibitory rebound is an
intrinsic property of membrane
based on varying those kinds of characteristics : connectivity of neurons involved
and their intrinsic properties; there are diff variations in complexity of pattern
generators that you might find
2) another important property is modulation
oso if you have network which depend on certain properties of neuron wiring
and membrane properties (intrinsic) of the neurons involved, if you have
Unlock document

This preview shows pages 1-3 of the document.
Unlock all 10 pages and 3 million more documents.

Already have an account? Log in

Get OneClass Notes+

Unlimited access to class notes and textbook notes.

YearlyBest Value
75% OFF
$8 USD/m
Monthly
$30 USD/m
You will be charged $96 USD upfront and auto renewed at the end of each cycle. You may cancel anytime under Payment Settings. For more information, see our Terms and Privacy.
Payments are encrypted using 256-bit SSL. Powered by Stripe.