BLP Exam 4

negative reinforcement (avoidance learning)

Response → No aversive stimulation
(negative contingency)
Responding increases
Example: I clean the litter box to get rid of the smell

positive punishment

Response → produces aversive stimulation
(positive contingency)
Responding decreases
Example: wearing high heels makes my feet hurt

what do negative reinforcement and positive punishment have in common?

Subjects are changing how they respond to minimize exposure to the aversive stimulus

Two Factor Theory

There are two processes that underlie avoidance learning
1. Classical Conditioning of fear to signal
- Tone paired with Shock; Tone →Fear
2. Operant Response to escape the signal (and stop the fear)
- Tone → Fear → Press Lever → Fear reduction
Important Note: According to two-factor theory, subjects do NOT make the response to avoid shock.
Subjects respond to turn off (escape) the stimulus that has become associated with shock

Evidence for Two Factor Theory (Miller Experiment)

if there are two factors, then we should be able to manipulate them independently.
Phase I: Grp 1: White box → shock; Grp 2: no training
Phase II: Place each subject in white box (they can turn wheel to escape)
Results: Only Grp 1 learns to turn wheel to escape (only group motivated to escape)

Evidence Against Two Factor Theory (Kamin Experiment)

Four groups:
Grp 1: Normal avoidance learning (tone → press lever → stops tone and no shock occurs)
Grp 2: Signal termination (tone → press lever → stops tone, but shock still occurs)
Grp 3: US avoidance (tone → press lever → tone continues, but no shock occurs)
Grp 4: Pure classical conditioning (tone → press lever → tone continues and shock occurs)
Two-factor Theory predicts: any condition in which
responding terminates the SIGNAL should result in an
increase in responses
Grps 1 and 2 should learn to press the lever
Grps 3 and 4 should not learn to press the lever
Actual Results:
Grp 1 Learns to press lever consistently when exposed to tone
Grps 2 and 3 learn to press lever some times
Grp 4 does not learn to press lever

Evidence Against Two Factor Theory (Extinction Problem)

After avoidance learning is obtained, Subjects are making the response reliably and avoiding aversive stimulation.
Two-factor theory would predict that the fear to the signal would eventually extinguish, however the avoidance response never extinguishes even after hundreds of trials.

Seligman, Maier, Overmier Dogs - Learned Helplessness

Used the shuttle box task to learn more about how
prior experience affects future performance.
Found that there are deleterious consequences of
long-term, uncontrollable, aversive events.

Characteristics of Learned Helplessness

Associative Deficit
Motivational Deficit
Seem depressed, sickly, anxious
-LH rats show higher levels of anxiety on elevated plus maze and social interaction tests
-Benzodiazepines (anti-anxiety medications) reduce LH
-Give anxiety-producing drugs to rats – they show LH (even without any prior exposure to shock)
-LH rats are more susceptible to developing tumors and have trouble fighting off disease.

LH Therapy and Immunization Experiments

LH subjects can learn to escape shock if they are shown how to escape
Prior exposure to escapable shock protects subjects from developing LH later on

Species-Specific Defensive Reaction (SSDR) Theory

Premise 1: Aversive stimuli elicit innate, species-specific defensive responses (SSDRs)
Premise 2: Which SSDR is elicited depends on the situation and configuration of the environment.
Evidence: SSDRs are more easily learned in an avoidance paradigm than other behavioral responses.

Predatory Imminence Continuum Theory

Premise 3: Which SSDR is elicited depends on the level of danger faced by the animal.
Premise 4: A signal (CS) associated with an aversive event (US) will elicit an SSDR
Prediction: The delay (time) between the CS and US will determine which SSDR will occur.
Example: If CS and US are presented close together during training, then the CS will elicit an SSDR
typically shown when the animal is in imminent danger. If there is a longer delay between the CS and
US during training, then the CS will elicit an SSDR typically shown when the animal is preparing for
danger (danger not imminent).

Predatory imminence and the Brain

More danger perceived – more activation of limbic brain areas (e.g., amygdala)
PTSD patients – amygdala activation even when no imminent threat

Factors that affect punishment

1. Punishment results in a decrease in responding – therefore, you can only observe punishment’s effects
by studying a behavior that is, initially, likely to occur.
2. Initial exposure to punishment determines responses to later punishment:
Start strong - If you severely punish a response on trial one, then the response will not be performed again.
Start weak - If you slightly punish a response on trial one, then you will likely see recurrence, even if you increase punishment severity later on.
3. If the aversive stimulus is/is not presented contingent on the target response. Contingency is necessary (only receive the aversive stimulus when making the response).
4. Interval between target response and aversive stimulus.
If you wait too long between the response and application of the aversive stimulus, no learning will occur.
5. Differing schedules of punishment
Some schedules lead to better reduction in responding than others.
6. Whether “bad” behavior is otherwise reinforced in a positive fashion – it will reverse the effects.
Example: speeding (don’t always get caught and you get to your destination faster)

comparative cognition

the study of animal behavior that focuses on the mechanisms by which animals acquire, process, store, and act on information from the environment

General Process Approach vs Comparative Cognition

General Process Approach: Study learning in animals because of what it tells us about learning in general.
Comparative Cognition: Focus on the differences in cognitive mechanisms between humans and animals.

Cognitive Etiology vs Comparative Cognition

Cognitive Ethology: Presumption that animals are capable of conscious thought and intentionality.
Problems with Cognitive Ethology
Comparative Cognition: employ the simplest possible explanations that explain as much of the data as possible.

Problems with anthropomorphism seen in cognitive ethology

Biases research and hampers knowledge
Overemphasis on the human experience

Memory

the retention of information or experiences over time

Stages of Memory

acquisition (encoding), retention interval (storage), retrieval

Learning Studies vs Memory Studies

Learning: manipulate acquisition conditions; test in same conditions
Memory: same acquisition conditions; manipulate testing conditions

procedural memory

memory for how to do something

perceptual memory

memory for how things look (percieved)

episodic memory

form of reference memory, recall episodes and experiences from the past

semantic memory

form of reference memory; recall facts and meanings of words

working memory

short-term memory, the retention of information just long enough to complete a task

ex: which spices you have already put into the stew

reference memory

stored memory of information that can be recalled to help use new information

ex: recall what spices look like, how they taste, where they are kept when making a stew

Working Memory in Animals - Delated Matching to Sample

Procedure: Sample → Choice task
Variations
Length of exposure to the sample stimulus
Duration of the retention interval
Appearance of sample stimulus
Other things about the sample stimulus
Factors that affect memory in delayed-matching-to-sample
Nature of sample stimulus
- some stimuli are easier to remember than others
Duration of exposure to sample stimulus
- longer exposure = better recall
Retention interval between sample and choice task
- longer interval = poorer recall
What is learned in Delayed-Matching-To-Sample?
One rule: choose same as sample
Multiples rules: if x, then choose x; if y, then choose y; if z, then choose z, etc

Premack Experiment - Chimps

Training: delayed-matching-to-sample with set stimuli
Testing: Used new stimuli (chimp had not seen these presented before)
Predictions:
-If using “same as” rule, then should get answer correct during testing
-if using specific rules, then should not get answer correct during testing
Results: Answered correctly to new stimul

Memory for Places (Spatial Location)

Morris water maze - rats use spatial cues to find platform in a circular water dish

Radial Arm Maze

-Rats search radial arms in random pattern
-Rats search the radial arm maze in a different pattern every time (no set strategy)
-Rats do not typically revisit arms that they have already searched (make very few errors)
Question: How are they searching/solving the maze without making errors?
Phase I: allow rats to only search 4 arms of 8-arm radial maze
Four-hour delay (rotate maze 90 degrees) – changes spatial location of arms
Phase II: allow rat access to all arms
Predictions:
-If rat is using scent cues to solve the radial arm maze, then it will visit the arms that it did not visit yet, even though the spatial location of those arms has been moved.
-if rat is using spatial location (cues on the walls) to solve the maze, then it will go to the arms that are in the spatial location that it has not yet visited (causing it to revisit arms that were moved to a new spatial location).
Results:
Rats go to the spatial location that they had not yet visited.

Aquisition/Encoding

Process of taking information in through your senses and translating it into a form that your brain can “write down” and store for later use

-in humans, influenced by selective attention, levels of processing, elaboration, and mental imagery

Selective Attention

Focusing on a specific aspect of experience while ignoring others

-constantly working

-stimuli compete for our attention

-we can only fully attend to one thing at a time

-inattention leads to encoding failure

Levels of Processing

A continuum of memory processing and ranging from shallow to deep

shallow - physical features analyzed

intermediate - recognition and labeling

deep - meaningful characteristics, deep processing leads to better memory

Elaboration

The “web” of connections, associations, and relevant meanings given to a stimulus

Mental Imagery

creating a mental “story” or scene around stimuli that we would like to remember

Dual code hypothesis

memory is stored in two ways: verbal code and picture code

-mental images are remembered better because it contains both picture and verbal codes

Grill study Imagery in Non-Human Animals

bitter tastes → yuck face
Conditioned Taste Aversion:
Taste→ illness, Taste→yuck face
Study #1 by P. Holland
Phase I:
Tone → Flavor 1
Noise → Flavor 2
Phase II:
Flavor 2 → illness

Test:
Reaction to Tone? Reaction to Noise
Results:
Noise → Yuck face
Tone → no yuck face
Study #2 by P. Holland (Mediated Acquisition)
Phase I:
Tone → Flavor 1
Noise → Flavor 2
Phase II:
Tone → “mental image of flavor 1” → Nothing
Noise → “mental image of flavor 2” → illness
Test:
Reaction to F1? Reaction to F2?
Results:
Avoid Flavor 2, even though it was never paired directly with illness.

mental time travel

brain mechanism that allows us to mentally move forward or backward in time, prospective and retrospective coding rely on the same brain areas

retrospective coding

memories for past events

prospective coding

remembering plans for future action

Mental Time Travel in Animals

Rats – solving radial arm maze – use retrospective coding, then switch to prospective coding by the end.
Chickadees – will eat less sunflower seeds if they think that mealworms are coming later (can plan for future meal)

Directed Forgetting

The accuracy of recall can be modified by cues or instructions indicating that something should (or should not) be remembered.
R-cue - “remember this!” cue
F-cue - “forget this!” cue
Significance: memory is an active process that can be brought under stimulus control.

Directed Forgetting in Animals

Pigeons remembered items that were cued as “remember this” and did not remember items that were cued as “forget this!”

Ebbinghaus Forgetting Curve

remembered 100 nonsense words, after 19 minutes forgot 49, forgetting occurs fairly quickly

-after forgetting, relearn list faster after

Retroactive Interference

new information interferes with our ability to remember old information

-get a new phone #, memorize it, can’t remember old #

Proactive Interference

old information interferes with our ability to remember new information

-all times took daily pill in past interferes with remembering if you took it today

Amnesia

failure to retrieve large portions of information that they normally would remember

infantile amnesia

no early memories, like being born or being a baby

Retrograde amnesia

incident happens, loss of past memories

-usually caused by head trauma, brain infection, trauma, extreme hypothermia

-often temporary, much of lost memories return, except those closest to traumatic event

Anterograde Amnesia

incident happens, no longer form new long term memories

-caused by anything that harms hippocampus (head trauma, seizure, brain infection, drinking too much alcohol)

hippocampus

necessary for consolidation of new memories

amygdala

boosts memories for significant events with emotion

memory consolidation

long term memories developed over a period of time

-as time passes, memories become more consolidated or more resistant to forgetting

Primacy Effect

remember the items at the beginning of the list
These items are recalled because they are in long-term-memory

Recency Effect

remember the items at the end of the list
These items are recalled because they are still in short-term memory

Serial Position Effects in Animals

Rats (Study by Kesner)
List was order in which they solved a radial-arm maze
Asked to recall which arm (of a choice of two) was visited first.
1 v 2: did well (remembered which arm they visited first v. second.)
7 v 8: did well (remembered which arm they visited second to last v. last).
4 v 5: did not remember which arm they visited 4th v 5th
Significance: rats show same primacy and recency effects as humans