Final Deck Flashcards
Define stimuli
Anything that can be perceived by the animal
Animals are always learning- info is provided by everything in environment
Learned information then causes some kind of behavioral response
Types of stimuli
Unconditioned or primary stimuli- environmental information
Conditioned stimuli- cues/signals
What can be a cue
Anything the animal can perceive and trainer can use consistently
Define latency
time to respond to a cue
part of criteria- slow responses may have been shaped
How to condition the cue
Cue is associated with history of reinforcement:
Presence of cue- performance of a particular behavior is likely to be reinforced
Absence of cue- performance of a particular behavior is not likely to be reinforced
How to establish a cue
Shape behavior first
Teach cue after behavior is established
When you can predict animal will respond, start associating cue
Tips for training a cue
Give cue just as behavior is starting and gradually back up cue in time
Discrimination of cue and no cue
Work in blocks
Know cue is established once you have stimulus control
How do you know the animal understands the cue
Behavior always occurs immediately when cue is given
Behavior never occurs in absence of cue
Behavior never occurs in response to another cue
No other behavior occurs in response to the cue
How to transfer a cue
Pair second cue with presentation of first.
Commonly use a physical and verbal cue
Creates flexibility in how worked
Pros and cons of fading a cue
Cue can be made smaller in magnitude to make it less obvious to the audience. It can increase the animal’s attentiveness but is limited by the animal’s perceptual abilities
How to correct anticipation: animal offers behavior before it is asked
Correct by using set time frames
Set time goal for animal to NOT offer behavior
Reset time if animal offers behavior before goal is reached
Importance of feeding for attention/waiting
Define limited hold
Technique used to reduce latency when animal is not responding quickly enough to the cue
Reinforce for average response time, reduce over time with variability (like bell curve)
How to chain behaviors
Train last to first
Completion of one behavior cues start of next
Con- Animal may anticipate and skip steps in chain
Tips for trouble shooting behaviors
Discuss use of LRS
Think about application to ABCs of training
If something is not working, look at Antecedent and Consequence. Is there something in the cue/set up that is not allowing the animal to succeed? Is there something about the consequence not allowing the animal to succeed?
What is the goal of any session
To set the animal up to succeed
Why is the LRS beneficial
By using the LRS, we don’t reinforce incorrect behavior but allow animal to keep working
Pacing of session and handling LRS encourages good behavioral momentum
What determines whether something is reinforcing or not
Natural history
Subject’s state:
hungry, fearful, wanting attention, liking novelty
The arrival of reinforcement communicates what to the subject
The arrival of the reinforcer tells the subject what is being reinforced
(Baiting can reinforce a non-response)
(Negative reinforcer needs to cease as soon as desired behavior occurs)
Why is it beneficial to use the smallest size of reinforcement possible
Reduces time animal spends eating food
Spends more time learning
Takes longer to be full
Gives trainer more flexbility
Define jackpots
Reinforcement that is much bigger than normal
Surprise to the subject
May be used to reinforce a breakthrough
Define a continuous interval reinforcement schedule
AKA Continuous reinforcement (CRF)
With each behavior, a reinforcement is delivered. Every selected behavior is reinforced
Define a fixed schedule reinforcement schedule
When the responses are not reinforced every time they occur, but are reinforced at set times
Define Fixed Interval (FI)
the first behavior after a set period of time is reinforced. Ex: rat presses lever. This is limited to once a minute. So first press after that minute time frame is reinforced. Rat may keep pressing lever within the time window, but these are not reinforced
Define Fixed Ratio (FR)
Animal is reinforced after a set number of behaviors
So rat has to press lever 5 times to get reinforcement
Define Variable Interval (VI)
Time interval between reinforcements varies
Rat may be reinforced for touching lever once after a minute, then after 30 seconds, then after 5 minutes. Rat may keep pressing lever within the time window, but these are not reinforced.
Set a variance: average time and standard deviation time schedule moves around
Define Variable Ratio (VR)
Behavior reinforced after varying number of reiterations
Again need set variance with average and SD to vary around
Define differential reinforcement
AKA selective reinforcement
Adjusting reinforcement based on animals performance
Shaping tool
T or F: All behavior fall on a bell curve as to how “well” they are done
True. Using different schedules will change how much variability you have, and can be used to change how well an animal does a behavior
Define Random and Interrupted Reinforcement (RIR)
Reinforcers delivered intermittently in a randomized order.
So if a behavior occurred correctly, may or may not bridge, may or may not reinforce, and could deliver an unconditioned or conditioned reinforcement. All at random chance
Define Variable Ratio with Reinforcement Variety (VRRV)
Variable ratio schedule with the addition of secondary reinforcers in addition to primary ones
Define Continuous Reinforcement with Variable Ratio Primary Reinforcement Follow up (CRVRPRF)
With an animal that is well worked with, with a new behavior you always want to reward (ie continuous reinforcement), but you can deliver a variable primary reinforcer. Like using food, water, ice, etc. in varying amounts
As a new trainer, what type of reinforcement schedule should you use
Continuous schedule, fixed ratio. Also use when training a new behavior. With more experience (both you and your animal), introduce fixed ratio mixed with continuous, then can move to mix with variable ratio with differential reinforcement for established behaviors.
Use approximations to train reinforcement schedule
What type of reinforcement schedule will strengthen your behaviors
Variable schedules.
Why should you be careful about the timing of your reinforcement
You can accidently reinforce a negative behavior. Example: Polar bear pacing. Boomer ball given. Polar bear stops pacing. Ends up pacing more often because that was when reinforcement was given. Can also be positive- reinforcing while he is in water can encourage to spend more time in water and less time pacing
Define Least Reinforcing Stimuli (LRS)
Incorrect response leads to no change
A 3-5 second no response by the trainer for an incorrect response given by the animal following a signal.
Continue doing what you were doing- it is not changing the situation
What are the ABC’s of training
Antecedent
Behavior
Consequence
T or F: Every interaction you have with your animal is a ‘training’ session
True.
Define shaping
Technique for training new behaviors- small shifts of behavior in right direction toward ultimate goal.
(If behavior already occurs for animal, can capture, if doesn’t, can shape)
How does shaping affect the bell curve of behavior
Shaping increases the behavior in a specific direction on curve. Uses reinforcement of successive approximations.
Define successive approximations
Baby steps. Increasing or altering a behavior incrementally by repeatedly changing the environment to amplify or extend the behavior
What was the 1st use of the term “shaping” and “conditioned reinforcer”
Skinner 1951 “How to teach animals”
What is the 1st Law of Shaping
Raise criteria in increments small enough that the subject always has a realistic chance for reinforcement. Don’t raise criteria beyond what the animal has offered. Subject has to understand that the rules have changed but reinforcement is still available. Ex: Sugar glider glide
What is the 2nd Law of Shaping
Train one aspect of any particular behavior at a time; don’t try to shape for two criteria simultaneously
Each reinforcement conveys only one piece of information
Trying to train multiple criteria will be confusing and frustrating
Make sure you are clear about what criteria is, and isolate each piece. Ex: Nova
What is the 3rd Law of Shaping
During shaping, put the current level of response onto a variable schedule of reinforcement before adding or raising the criteria.
Allows differential reinforcement. Makes behavior more resistant to extinction.
Ex: Val
What is the 4th Law of Shaping
When introducing a new criterion, or aspect of the behavioral skill, temporarily relax the old ones
Already learned behavior may get sloppy when you start working on a diff part of the behavior. Ex: Puppy on barrel, generalized to a new location
What is the 5th Law of Shaping
Stay ahead of your subject: Plan your shaping program completely so that if the subject makes sudden progress, you are aware of what to reinforce next
You can’t predict what parts of training the animal will find difficult and what parts it will find easy
Be prepared to jackpot/continue progressing. Ex: Miki
What is the 6th Law of Shaping
Don’t change trainers in midstream; you can have several trainers per animal, but stick to one shaper per behavior
Consistency: for standards from animal, how trainers cue/reinforce
Ex. Nova
What is the 7th Law of Shaping
If one shaping procedure is not eliciting progress, find another; there are as many ways to get behavior as there are trainers to think them up
Go back to operant principles, examine your assumptions about what is reinforcing, and think about other approximations
What is the 8th Law of Shaping
Don’t interrupt a training session gratuitously; that constitutes a punishment. Ending a session:
-Plan your session: follow training plans and set animals up for success
-Think about how much reinforcement you have/how you will use it
-Always save plenty for end-plan for emergencies, end on a good note
Pros and Cons of an “End of Session” signal
(Cue that signals a training session has ended)
Can indicate when an animal is in session versus not. May cause aggression or punish an animal who does not want to end session. May cause “rushing” or animal physically trying to stop end of session. Not having an end of session cue prevents anticipating end of session, and can produce more variety for a session
What is the 9th Law of Shaping
If behavior deteriorates, “go back to kindergarten”, quickly review the whole shaping process with a series of easily earned reinforcers
Many factors impact animals performance
Learning is contextual
T or F: Learning is context dependent
True.
What is the 10th Law of Shaping
End each session on a high note. Plan to end positively. Set up for success.
Jackpot- high magnitude/high value reinforcement to mark great behavior
Memory will be strongest for start and end of session
What do good training plans include
Background information (trainer, animal, relevant previous training information)
Description of what final behavior will look like/how it is cued
Successive approximations that realistically take animal from where performance is now to end goal
Each step should have clear criteria for when to move onto next step
What do good training plans include
Background information (trainer, animal, relevant previous training information)
Description of what final behavior will look like/how it is cued
Successive approximations that realistically take animal from where performance is now to end goal
Each step should have clear criteria for when to move onto next step
Define anthropomorphism
Assigning human qualities, characteristics, or motivations to animals
Can lead to poor training or care decisions. Affects how we approach training and problem solve. Changes relationship
How can a trainer’s emotions affect the animal
Attachment to the animal can cloud judgement
Outside influences on emotions can too
If emotional state is not normal: Set emotions aside, postpone training session, don’t make decisions
Describe animal intelligence
Animals are as smart as they have needed to be to survive
Individual variation
Perceptual abilities
Varied environments and pressures
You can shape the behavior of the animal regardless of its intelligence level
T or F: Any interaction is training
True
Aspects of developing a relationship with an animal
Trust account
Knowing what to expect
Developed through interactions
Non-formal
Reinforcement/ formal training history
What is animal training
Training is a tool for changing behavior
Increasing the occurrence of desired behaviors
Decreasing the occurrence of undesired behaviors
Two types: Classical and Operant
Training is a language we establish with the animals. It is a game with set rules. The animal follow the rules. Change behavior to change the outcomes.
What is classical conditioning AKA pavlovian conditioning
Reflex training
Unconditioned stimulus and response is
Associated to conditioned stimulus and response
What is Operant AKA Skinnerian Conditioning
An animal’s behavior is determined by past experience
After a behavior occurs, the possible outcomes are referred to as:
Reinforcement
Punishment
Positive
Negative
Define reinforcement
Something which increases the occurrence of a behavior
Define punishment
Something which decreases the occurrence of a behavior
Define Positive
Something which is given to (aka added) to the animal
Define Negative
Something which is taken away (aka subtracted) from the animal
Is reinforcement or punishment more effective for learning
Learning is faster, has better latency and better retention with reinforcement. For punishment to be comparably effective significant physical harm must occur
Define primary reinforcement
A biological need that must be reinforcing to the animal
Basics needed for survival: food, water
This is good for the animal regardless of previous experience
Define secondary reinforcer
Some argue same as conditioned reinforcer
Some argue inherently valuable to animal
Ex: Grooming, sunning, enrichment
Define conditioned reinforcer
Another reinforcement that has been paired with the primary reinforcement
Learned as good to the animal based on experience with it
What factors may influence the value of a reinforcement
Individual preference
Motivation
Frequency of use
Every interaction with an animal can be reinforcing (or punishing)
Dropping enrichment, social attention, etc. Every interaction with an animal matters
What is a bridge
The bridge is a conditioned reinforcer
Tells the animal the exact moment they have done a desired behavior
How do you establish a bridge
Pair a conditioned stimuli (CS) with an unconditioned stimuli (US). CS should be reinforcing, and highly relevant to animal (primary)
What can be a bridge
Anything an animal can perceive
Typically clicker or whistle
Verbal’s like “good”
If verbal, or human whistle, think about consistency
How long do you have to pair US/CS
Immediately (but up to 7 seconds). There are many factors that influence.
T or F: Unconditioned Stimulus should be novel- learning occurs more quickly if animal has not had past experience with US
True. Due to US pre-exposure effect. Initial experience with a stimulus presented by itself slows subsequent conditioning of that stimulus.
When should you bridge
Immediately as the desired behavior occurs/ Bridge delays >.5 seconds significantly reduces learning
How much time between giving bridge and delivering primary reinforcement
Can reinforce 60 sec+ after bridge if needed. Bridge timing is most important.
Name some different training techniques
Capturing/scanning
Shaping
Selective reinforcement of approximations
What was the relationship between early humans and animals and how did it evolve
Early humans cooperated to drive predators off kills to take food
Moved to working with predators to kill prey and share food
Domestication- live and work with animals
Hunting, riding, messengers, herders, hauling, farm animals, etc.
Menageries- maintained and cared for wildlife as hobby. Sign of wealth/power
Who was Ivan Pavlov
Russian, late 1800s to early 1900s
Varied background, advanced studies in medicine and physiology
Noble Prize 1904 studying digestion
Demonstrated a conditioned reflex
Dogs salivated at the sound of a bell after associated with food
Classical conditioning
Define the ethologist
Studied animal learning and behavior through genetics (heredity) and instinctual behaviors.
Developed in Europe
Animal learning is the result of animal’s evolutionary biology and its genetic predispositions
Konrad Lorenz, Niko Tinbergen
Define the behaviorist
Studied animal learning and behavior through observation and manipulation of cues/consequences
Developed in the US. All animal learning is due to consequences that occur after behavior
B.F. Skinner, J.B Watson.
Environment shapes behavior
Operant conditioning
Define instinctual drift
Animal is operantly trained for one behavior and drifts back to instinctual behavior
Need to understand natural history when using behavioral techniques to shape behavior
Important things to understand about the species and individual animal in order to be an effective teacher (trainer)
Natural history
Anatomy and physiology
Natural habitat
Social structure
Feeding habits
Natural behavior
Communication
Acquisition
Captive or wild born
Parent or hand reared
Source, rehab, past facilities, pet?
Experiences at past/current facilities, previous human interactions, housing and care
Training, medical, enrichment, behavior, diet
Motivation, enclosure parameters, social changes, structure, seasonal changes, records, etc.
Why are records important
Track progress
Identify trends objectively
Help identify where changes in behavior occur and why they may have occurred
Communication with co-trainers/staff
Useful for future trainers
Documentation for other facilities if animal moved
Required for USDA/accreditation
Important features of good record keeping
Easy retrieval of information
Easy entry of information
Standard formatting
Legends for rating scales
Complete and consistent
Need whole animal team to enter notes
Important to have other records to understand behavior within training sessions
Behavior observations
When should you write a training record and what is the format
After every training session.
Format: Date, time, duration, people involved, location, behavioral rating, comments, stage goal of session, objective summary of what happened, suggested steps for next session
Why do we train animals
Husbandry
Education
Medical behaviors
Enrichment
Research
Conservation
Habitat use
Work
