Chapter 3

The Effects of Behavior:  Thorndike and Hull

There is no easily demonstrated, high level reasoning among cats… or monkeys.
There appears to be no real imitation, based on understanding and ideas, among animals.
Behavior is instinctual or learned through associations that have previously had satisfying consequences.
Animals and humans learn through trial and error, making a number of responses, one or more of which leads to a solution or “a satisfying state of affairs.”

Continguity vs. Reinforcement
contiguity = an association is formed between stimuli, or between stimuli and responses, because they are presented in close temporal proximity.
Pavlov, Watson, and Guthrie used contiguity to explain learning
reinforcement = learning occurs because of consequences of behavior, or that it leads to pleasant consequences, the elimination of something unpleasant, or both.
Thorndike believed that contiguity was only part of the story and that reinforcement was the rest.

Thorndike’s Pre-1930s theory: Emphasis on practice
Believed statements about consciousness are uncertain and difficult
Emphasized experimentation rather than introspection
Learning consists of the formation of bonds between stimuli and responses
Learning involves stamping in of stimulus-response (S-R) connections
Forgetting involves stamping out of connections.

Law of Exercise
Law of exercise: bonds between stimuli and responses are strengthened through being exercised “frequently,” “recently,” and “vigorously.”
Watson supported the laws of frequency and recency; Guthrie adhered to the law of recency, but NOT the law of frequency.
Thorndike repealed the law completely after 1930. Through experimentation, he discovered that responses may be highly variable and that repetition does not cause learning.

Law of Effect
Whether a connection is stamped in or not depends more on its consequences (its effects) than on repetition.
Responses just before a satisfying state of affairs are more likely to be repeated.
Responses just before an annoying state of affairs are more likely not to be repeated
Like Watson and Guthrie, Thorndike focuses on the S-R connection. Watson and Guthrie were contiguity theorists, however, and Thorndike was clearly a reinforcement theorist.
Hull repealed half of the law after 1930; responses leading to satisfying states tend to be stamped in, but responses leading to annoying states do not tend to be stamped out. In fact, they do relatively nothing.

Law of Readiness
Law of Readiness: Certain behaviors are more likely to be learned (stamped in) than others
related to learner’s maturation and to previous learning
a pleasant state of affairs results when a learner is ready to learn and is allowed to do so, but being forced to learn when not ready or prevented from learning when ready leads to an annoying state.
Watson and Guthrie both believed that biology plays a role only in the sense that it determines learning potential (or physical boundaries of behavior). With Thorndike, motivation and development, as evident in satisfying states or annoying states, also influence learning behavior.

Law of Multiple Responses
Multiple responses: in any given situation, the organism will respond in a variety of ways if its first response does not lead immediately to a more satisfying state of affairs
Pavlov, Watson, and Guthrie all assume that learning is inherent in an association between two stimuli or between stimuli and response. By using the concept of trial and error to learning, Thorndike is implying that efforts will cease with a satisfying state has been achieved (reinforcement). For early behaviorists, the resulting state had nothing to do with learning.

Law of Set or attitude: Learning is partly a function of attitude or predisposition to react in a given way.
Early behaviorists of Pavlov, Watson, and Guthrie address predispositions only in terms of biologically determined potentials and limits to learning. Thorndike is addressing the role of culture or societal influences on “set,” while early behaviorists focus on experience in terms of an individual. Also, Thorndike is depending on reinforcement to stamp in a response encouraged by culture, and early behaviorists ignore reinforcement.

Law of Prepotency of Elements
Prepotency of elements: It is possible for a learner to react only to the significant (prepotent) elements in a problem situation and be undistracted by irrelevant aspects of the situation.
Watson and Guthrie both focus on learning by doing, or by responding to whatever stimuli is presented; both theorists do not identify “right” or “wrong” responses but simply focus on the reflexive response to a given stimuli. Thorndike’s Prepotency of Elements implies that individuals can filter through numerous stimuli to focus only on the important aspects of a given situation; and again, identifying “important” aspects implies a connection to outcome, an idea inherent in reinforcement theory and not the contiguity theory of early behaviorists.

Response by Analogy
Response by analogy: A person placed in a novel situation may react with responses that might be employed for other situations that are in some ways similar or that share identical elements.
This idea has some characteristics of early behavioristic concepts but has distinction in its “hinting” of cognition.

Associative Shifting
Associative Shifting: It is possible to shift a response from one stimulus to another.
This law is the premise of classical conditioning, utilized by both Pavlov and Watson in their theories, though Watson focuses primarily on S-R association.

Learning by Ideas
This concept gave cognitive concerns like analysis, abstraction, and meaningfulness more importance in learning.
principle of belongingness: if two or more elements are seen as belonging together, they are more easily learned
spread of effect: when a response is followed by a satisfying state of affairs, other related responses also seem to be affected
Cognitive aspects, to which this concept hints, were avoided in early behavioristic theories.

Critique
S-R bonds are stamped in because of the satisfying nature of their consequences, influenced as well by the individual’s sense of what goes with what.

Importance of trial and error, predetermined attitude, responses learned in other similar situations, and most important aspects of a situation.

Theorizing based on informal observation, though he introduced controlled investigations of animals and people to verify predictions.

Used vague, internal states as basis for explaining behavior (satisfying and annoying)

Significance of reinforcement, law of effect, has been long-lasting.

Contributed to practical application of things such as teaching.

Was willing to examine other views, change his thinking, and admit that much was not explained by his theory.

Hull’s Formal Theory Building
Most ambitious of all the connectionist theorist

Most influential theorist of his time

He proposed that the ideal theory would be a logical structure of postulates and theorems

Rather than being general ideas about learning such as Watson’s and Guthrie’s, he wanted to be specific enough that his ideas could be tested and proven or disproven.

Hull: A hypthetico-deductive system
Hull believed science has 2 essential aspects, one being the facts, the other being the organization of facts into a coherent, logical system or theory.
Therefore, Hull developed a hypothetico-deductive system, a system of laws deduced logically that govern human behavior.
Entire system is based on 17 postulates from which 133 theorems and corollaries were derived
Hull was an S-O-R theorist, a neobehaviorist.

The Four-Stage Analysis
1. Independent variables
Deprivation of (food, water, sex, etc.)
Strength of stimulation
Magnitude of Reward
Number of reinforced training trials
Work involved in responding

2. & 3. Intervening Variables
D = Drive SER = excitatory potential
K = incentive motivation SoER = net reaction potential
SHR = habit strength SIR = Aggregate Inhibitory Potential
V = Stimulus Intensity SLR = Reaction Threshold
SOR = Behavioral Oscillation
4. Dependent Variables
Amplitude
Speed (response latency)
Resistance to Extinction

Fractional Antedating Goal Reactions
Fractional antedating goal response is a conditioned response made before the actual goal reaction.
This concept serves the same purpose as Guthrie’s movement produced stimuli (MPS), but Hull’s fractional antedating goal reactions are learned with reinforcement. The term attempts to explain what Thorndike terms “learning by ideas.”
Hull was attempting to use fractional antedating goal reactions to explain in precise, measurable terms behaviors that most people might explain using more vague, mentalistic terms. He explained that although fractional antedating goal responses are conditioned mechanisms, because they occur before a response, they constitute on the part of the organism as “foresight” or “foreknowledge.”

Habit-Family Hierarchies
Habit-family hierarchies
a number of different learned responses to the same stimulus
hierarchy – one alternative is usually preferred over another because it has been rewarded more in the past and has more reaction potential
A hierarchy of responses implies the importance of outcome. Since early behaviorists ignore reinforcement, and thus outcome, they have no reason to conceptualize numerous responses that are more or less preferred. Thorndike, however, addresses multiple responses by explaining that due to trial and error, responses will vary until one eliciting a satisfying state is given. Of course, Thorndike is a reinforcement theorist.

Critique
clear explanation that behavior is lawful and predictable

predict output given knowledge about input; mathematical formulas are more complex

does reflect facts well; also makes inferences termed logical constructs (such as reaction potential, aggregate inhibitory potential, behavioral oscillation, incentive motivation, etc.)

intervening variables are useful within Hull’s system but not so much in practical application

advanced psychological theory by introducing some cognitive variables

systematic experimentation and application of logic have profoundly influenced how psychological investigations are conducted.

Hull, along with Thorndike and Skinner, is credited for popularizing and systematizing notion of reinforcement.

Hull was personally ambitious but provided little research to support his corollaries.

Education Implications of Thorndike and Hull
Thorndike:
stressed importance of students’ attitudes, most important aspects of a situation, and teaching for transfer
defined and established educational psychology as a field
popularized use of tests and statistical methods in education and psychology
changed child psychology into an objective discipline
Hull - popularized notion that reinforcement is centrally involved in learning
Both - Recognition of the importance of the consequences of behavior has profoundly affected practices in schools.