1/45
Looks like no tags are added yet.
classical trace conditioning
it involves a thirsty head-fixed mice that are exposed to a brief auditory tone (0.25s, 12kHz) paired with a sucrose reward. They manipulated the inter-trial interval (ITI) across 1-hour day sessions.
How was the inter-trial interval (ITI) broken up?
a 60-second ITI group (50 trials per session) = 50 mins vs 600-second ITI group (6 trials) = 60 mins
What was hypothesis 1?
learning in fixed number of trials
What was the result of hypothesis 1?
600-s ITI group 10X longer to reach the learning threshold compared to the 60-s ITI group.
What was hypothesis 2?
trial spacing effect
What was the result of hypothesis 2?
longer trial spacing moderately increases the per-trial learning rate
What was hypothesis 3?
learning in a fixed total time
What was the result of hypothesis 3?
10X increase in ITI yields 10X more learning per trial
What was found about the anticipatory licking?
it emerges in significantly fewer trials for a mouse in the 600-s ITI group than one in the 60-S ITI group
dLight1
human DA receptor coupled to circularly permuted green flourescent protein (cpGFP)
dLight1.3b
genetically encoded fluorescent dopamine sensor that is injected into the NAcC of head-fixed trace conditioning mice
What occurred as a result of the emergence of cue-evoked mesolimbic dopamine release in the nucleus accumbens core?
it scaled in exact proportion to trial spacing.
scale invariance
multiplying x by a constant multiplies y by a constant, whatever x is, so the pattern looks the same at every scale.
What was found while looking at an equivalent learning over a given time period regardless of number of trials?
despite experiencing vastly different numbers of trials within a fixed duration, animals accumulate equivalent overall learning.
Retrospective casual learning
cause-and-effect relationships are inferred backward in time after a significant event occurs
Retrospective casual learning: reward arrives
A meaningful event triggers learning. The brain asks what might have caused it.
Retrospective casual learning: look back in memory
Recent cues are still held in memory, so the brain checks which ones came just before.
Retrospective casual learning: compare with chance
Cues that come before reward more often than by chance are learned as its likely causes.
Partial reinforcement
identifies the exact trial at which dopaminerfic cue responses and behavioral licking emerge.
What is learning rate controlled by?
the time between rewards rather than the time between cues.
What did non-learnings show?
they did not exhibit anticipatory licking but they developed intact cue-evoked dopamine responses identical to behavioral learners.
What does the emergence of anticipatory behavior require?
crossing an internal evidence threshold, whereas cue-evoked dopamine release develops directly as associative evidence accumulates.
classical conditioning
it forms an association between a condition stimulus (CS) and unconditioned stimulus (US).
backward conditioning
when the US follows the CS, conditioning is generally more effective than when the US precedes the CS
trace conditioning
it includes a gap between the CS and US. This involves the hippocampus because the animal must bridge a temporal interval.
fear conditioning
A form of classical conditioning using a tone paired with a shock. The amygdala is whats assocated with it.
what was the behavioral task and design?
The animals learned an association between an auditory cue and a sucrose reward. The task used thirsty, head-fixed mice. The cue was 0.25 secs at 12 kHz, followed by sucrose after a delay.
anticipatory licking
licking after the cue before the reward which served as a behavioral measure of learning.
how did the behavioral graphs come out?
when learning was plotted against the trial number, the curves differed, when scaledd by trial spacing, their trajectories were similar.
What did the paper say about “practice makes perfect”
It challenges the idea that more trials alone determine learning. Repetition is still necessary, but how practice is spaced matters greatly.
what was the main finding of dopamine?
it is involved in regulating learning but is not by itself a direct measure of whether behavioral learning has occured.
50% partial reinforcement
The cue was rewarded on only half of its presentations. The interval between rewards was therefore longer than the nominal ITI.
10% partial reinforcement
Only one in ten cue presentations was rewarded, producing a much longer average interval between rewards.
What did the partial reinforcement reveal?
there is a dissociation between dopamine dynamics and behavioral expression
retrospective causal learning
An outcome to look backward and identify events that may have caused it.
what happened at the 3600 seconds?
Learning was greatly reduced or absent under the tested conditions
if one studies the behavior or physiology of a “head-fixed” animal, what type of sensory information might be significantly altered and affect results?
vestibular and prorioceptive
in this study, as animals undergo trace conditioning they exhibit what behavior?
anticipatory licking
dLight1 is a combination of GFP and a segment of a dopamine receptor. How does binding of dopamine to dLight1 result in a change in GFP fluorescence?
GFP hydrolyzes dopamine
A change in dopamine signal in the nucleus accumbens implies what?
a change in dopamine neuron firing
what is the significance of the “power law”?
for every n-fold increase in reward spacing, n-fold fewer trial experiences are need for learning
aversive learning
a type of leaning where the animal associates a specific behavior with an unpleasent outcome. in the paper, the animals would freeze in response to a tone cue paired with a mild foot shock (cue-shock conditioning)
why is repetition needed for cue-reward learning?
sensory or motor discrimination.
what model was updated at every reward?
ANCCR
what might happen if cue-reward are replayed during sleep?
one trial learning
What is the function of dopamine in cue-reward learning proposed in the paper?
it influences learning rate