Tolman Flashcards
Tolman vs Watson
Tolman rejected Watson’s behaviorism as molecular resting upon factors of physics and physiology instead of psychology
He also believed that Watson ignored goal-seeking behaviors
Tolman is the most Molar theorist– he believed that molar behavior has four properties that should be obvious to all intelligent observers…
- Goal-directed
o Behavior involoves getting toward something
• This is judged by obeserving what the organism is doing - Makes Use of Environmental Supports
• aka, “means objects” that are used to get to goal
o e.g. use of paths and tools—this marks the behavior as purposive and cognitive - Selective preference for short or easy means
“Robinson’s principle of least effort (1926)” - Behavior is Teachable / Docile / Flexible
• If behavior is rigid, mechanical, stereotyped like spinal reflex, it should only be analyzed on the molecular level (classical conditioning)
Tolman’s Intervening Variables
The simple S-R approach provided only a limited account of human behavior, and could not account for individual differences or for the interesting or complex aspects of human behavior
He believed that additional elements that had to be interposed– Intervening Variables
Demand Appetite Differentiation Hypotheses Bias Motor Skill
However, Tolman never leaves variables completely interior—he always correlates to external, environmental variables
Demand
What is the organism working for? [similar to Drive]
Related to the maintenance schedule of deprivation and satiation
Skinnerians define reinforcers as what the animal will work to get , [instead of assuming it’s hungry]
Appetite
The appropriateness of the goal object
e.g. does it reduce hunger?
Differentiation
Types and modes of stimuli provided
[aka discriminative stimuli]
Cues to get to goal
Motor Skill
Kind of physical response necessary
The organism must know how to make this response, i.e. biologically capable of response
e. g. RO too slow to ever play shortstop effectively
e. g. Guinea pig fused backbone can’t press lever
Hypotheses
Cumulative number and nature of trials
Expectancy based on past reinforcement history
e.g. direct experience in maze
Bias
Pattern of succeeding and preceding maze units
The maze itself, completely external
e.g. If maze more successful lefts than rights, biased toward making lefts
Expectancy Principle
Tolman’s attempt to make contiguity less mechanical and more cognitive and purposive
When one stimulus (sign) is followed by a second stimulus (significant), the learner acquires an association between these stimuli
The organism learns to expect the significant when the sign is presented
Learning occurs throughout a series of paired experiences–contiguity
Sign will attach to any meaningful significant
• e.g. food, getting kicked by owner
The nature of the significant doesn’t matter, as long as it’s meaningful
To Tolman, punishment works exactly the same
Latent Learning
Learning which is not apparent in the learner’s behavior at the time of learning, but which manifests later when a suitable motivation and circumstances appear
**Reinforcement is not necessary for learning to occur
Reinforcement is required only for performance
e.g. Cognitive Map
Tolman and Contiguity
He was a contiguity psychologist who thought of associations being formed between stimuli [S-S]
Frequency of pairing is the only learning principle
No need for reward or reinforcement in this theory, it is simply a contiguity theory
E.g.:
Bell—Food—Approach
Approach based on how hungry you are and how it is appropriate to your appetite
Assuming it is a edible food, it is a significant
Bell is a sign that the significant is coming
Bell substitutes for the food
The sign tells you to expect the significant
The animal salivates to the bell because it expect the meat
S-O-R
O = Organism
The organism acts because it expects a reward, based on its history with reinforcement
However, this expectancy doesn’t matter when predicting behavior
**The true motivation is unconscious to the organism
Purposive behaviorism
Behavior is regulated in accordance with objectively determinable ends
The purposes are those that any observer would recognize in the behavior
Problem: you must know the past reinforcement history to understand
e.g., during WWII, Russians trained dogs to run into tanks wearing land mines
•If you believe Tolman, Dog’s purpose is to get blown up
The problem with this analysis, why does the dog have this goal—conditioned–learning history
Emphasis on expectancy, you must explain where expectancy comes from
It is internal, but it is related to learning history
Motivations are unconscious—one instance that behaviorists and psychoanalysts take a similar view—you don’t know your own learning history
Tolman’s criticism mostly came from his loose language
