Download Clark Leonard Hull (1884-1952) - Chapter 6 and more Slides Behavioural Science in PDF only on Docsity!
1
Clark Leonard Hull
Chapter 6
2
Clark Leonard Hull
- Born Mar. 20, 1884 Susquehanna, Pennsylvania.
- Belonged to poor household. Contracted polio at 24.
- Completed his PhD (1918) from University of Wisconsin. Studied Aptitudes and hypnosis. (1884-1952) vlp.mpiwg - berlin.mpg.de 3
Clark Leonard Hull
- Provoked by Pavlov to study conditioning.
- Taught at Wisconsin and then went to Yale.
- Wrote Principles of Behavior (1943), and A Behavior System (1952).
- Became the most cited psychologist in the 40s and 50s. Died in 1952. (1884-1952) vlp.mpiwg - berlin.mpg.de
4
Watson’s Behaviorism
- Watson’s behaviorism radically denounces mind and consciousness and focuses on overt measurable behavior.
- For Watson, the organism is empty and brain physiology merely connects stimulus with response. Stimulus Response Organism 5
Disagreement with Watson
- Hull believed that Watson’s rejection of unobservable events inside the organism (body and brain) was unscientific.
- Hull argued that physicists and chemists make inferences about things that they have not actually seen or observed, e.g., gravity, forces, atoms, molecules etc., but such phenomena exist. 6
Mediating Events
- Hull proposed that since we cannot observe the mediating events inside the organism, it does not mean that they do not exist.
- In order to study these mediating events we need to operationally define them. In other words operational definitions are those that can quantify these events. Stimulus Response Organism Mediating events
10
Neobehaviorism
Hull thus evolved a new brand of behaviorism (neobehaviorism) which uses intervening variables as hypothesized physiological processes, to explain behavior. 11
Hypothetico-Deductive System
Hull developed a dynamic system that framed hypotheses and refined them to support and modify learning theory.
- Reviewed research on learning.
- Summarized all those findings.
- Developed hypotheses from those summarized findings. Hull formulated 16 (1943) postulates (18 postulates, 1951) and many theorems to explain his learning theory. 12
Postulate 1: Stimulus Trace
External stimulation triggers a sensory impulse that continues for a few seconds after the stimulating event has terminated. This impulse is the stimulus trace. Organism Stimulus (S) Response (R) Stimulus trace ( s ) Response trace ( r ) S R (Traditional Behaviorism) S s r R (Hull’s Behaviorism)
13
Postulate 2: Impulses Interact
Each organism is bombarded by many stimuli and thus many sensory traces are generated. These traces interact with one another, and represent complexity of stimulation. Organism Stimuli (S) Response (R) Stimulus traces Response trace ( r ) s - S s - r R 14
Postulate 3: Unlearnt Behavior
Organisms are born with hierarchy of unlearnt responses. Unlearnt behaviors are triggered first when the need arises. If these behaviors fail organism learns new behaviors to reduce the need. Organism Stimuli (S) Response (R) Unlearnt Behavior 15
Postulate 3: Unlearnt Behavior
Need Hierarchy of Behaviors Behavior Cold Shivering Unlearnt Crouching Moving/running Eating/drinking Urinating/defecating Wearing clothes Learnt Turning heat on
19
Postulate 5: Stimulus Generalization
Similar stimuli will evoke the same response or habit. Prior experience will affect current learning. Hull called it, Generalized Habit Strength ( S^ ― HR ). This is similar to Pavlov’s stimulus generalization, and to Thorndike’s identical elements theory, where identical elements in stimuli across situations evoke the same or similar response. 20
Postulate 6: Drive Stimuli
A biological deficit produces a drive (D) which is associated with drive stimuli. Drive Internal Stimulus External Stimulus Hunger Pangs Smell of food Thirst Parched throat Sight of water 21
Postulate 7: Reaction Potential
The probability of a learned response occurring at a particular moment is called reaction potential ( SER ), and is the product of habit strength and drive. If habit strength or drive equals zero, reaction potential equals zero.
SER =^ SHR X^ D
22
Postulate 8: Reactive Inhibition
Responding causes fatigue, and fatigue eventually inhibits responding. This is called Reactive Inhibition ( IR ). Fatigue is a negative drive state. Makes the animal not to respond. 23
Rest from Fatigue
Extinction of a response not only results from lack of reinforcement but also from fatigue. When rest ameliorates fatigue, extinguished response spontaneously recovers. -2^0 24 68 10 1214 1618 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Trial CR (Saliva, ml) Extingushed^ CR Spontaneous Recovery Rest 24
Reminiscence Effect
Mass practice on a motor task (rotating disk) increases fatigue, and thus results in poorer performance compared to distributed practice. Rotating disk Stylus Target 0 100 200 300 400 500 600 1 ♫ 2 ♫ 3 ♫ 4 ♫ 5 ♫ 6 ♫ 7 ♫ 8 ♫ 9 ♫ 10 ♫ 11 ♫ 12 ♫ 13 ♫ 14 ♫ 15 Successive Trials Mean Time on Target (Seconds) (^) Distributed Massed Both (Kimble and Garmezy, 1968)
28
Postulate 11: Reaction Threshold
The learned response will only be emitted if (^) SER is greater than reaction threshold (^) SLR. —
Postulate 12: Response Occurrence
The probability that a learned response will occur as a combined function of if (^) SE — R, (^) SOR , and (^) SLR. 29
Postulate 13: Response Latency
The greater the value momentary effective reaction potential (^) SER the shorter the latency (time) (^) StR between S and R. —
Postulate 14: Resistance to Extinction
The value (^) S ER will determine resistance to extinction. The greater its value the greater the resistance to extinction. — 30
Postulate 15: Response Amplitude
Some learnt responses occur in degrees like, salivation or GSR. The amplitude (A) of a conditioned response varies directly with (^) SE^ — R.
Postulate 16: Incompatible
Responses
When two or more incompatible responses are elicited in a situation, one with the greatest (^) SER will occur. —
31
Changes in Hull’s Theories
From 1943 to 1952
32
Changes after 1951
In 1943 version of his theory, Hull proposed that reinforcement affected learning. Experiments by Crespi (1942, 1944) and Zeeman (1949) suggested that performance, not learning was altered by the magnitude of reinforcement. Running Speed (performance) in a straight runaway changes as a function of reinforcement, not learning in the animal. 33
Performance ( K )
Performance ( K ) drops dramatically when animals trained on a large reinforcement were switched to smaller reinforcement and vice versa. Thus momentary effective reaction potential required K as a product component to habit strength and drive. S ER =^ ( S HR X D X^ K )^ -^ ( IR +^ S IR )^ –^ SOR — Hull concluded that organisms learn rapidly for a small or a large incentive, but perform differently for these incentives.
37 Fractional Antedating Goal Response ( r G)
- Fractional antedating goal responses ( r G) are small partial responses (salivating, chewing, etc.,) made by the organism prior to making the actual goal responses (eating, chewing, salivating).
- r G are not only responses but are also cues or stimuli ( s G) to subsequent responses. This is similar to Guthrie’s movement-produced stimuli. 38 Fractional Antedating Goal Response ( r G) 39
Habit Family Hierarchy
The organism can generate a number of possible overt responses s G (habit family hierarchy) in any particular situation, and these serve as alternate ways of reaching a goal. The set of responses, that brings about reinforcement most rapidly and with the least amount of effort, will stick.
40
Habit Family Hierarchy
To test this idea Tolman and Gleitman (1949?) used a 3-path maze. Path 1 required least amount of effort bringing the reinforcement sooner, followed by path 2 and path 3. Their results confirmed habit family hierarchy. 41
Evaluation
Contributions Criticisms Developed a systematic behavior theory. Theory had little value beyond laboratory. Popularized behaviorist approach. Lack of internal inconsistency in the theory, e.g., the phenomenon of extinction can be explained by various postulates. Precise definitions of variables and empirical verifications. e.g., definition of reinforcement. Theory not revised in the face of contradictory data. 42
O. Hobart Mowrer
- Born in Unionville, Missouri in 1907.
- Did his PhD from John Hopkins University in
- Mowrer became a colleague of Hull at Yale in 1940. (1907-1982 )
46
Two-Process Theory
Classical conditioning: Sign Learning Light Electric Shock Fear/Pain CS US UR Light Run away Escape Pain CS CR US Instrumental conditioning: Solution Learning 47
Sign & Solution Learning
- In avoidance learning, light (sign/warning) tells animal to avoid the shock because it is painful. Through classical conditioning the animal associates light with pain. Mowrer referred to this as sign learning.
- Once the animal is warned about the shock and the resulting pain, it must perform a behavior (run away) to avoid the pain. Avoiding pain becomes negative reinforcement what Mowrer called solution learning. 48
Emotions
- In 1960, Mowrer went on to note that many emotions (other than fear) can be explained with his two-factor theory.
- One can experience the emotion of hope if a bell (reinforcement) sounds just prior to food or of disappointment if the bell (reinforcement) sounds just prior to the removal of food.
49
Reinforcers
- Mowrer classified unconditioned stimuli into two kinds of reinforcers. Decremental reinforcers, were those that reduced a drive, e.g., food which reduces hunger.
- And incremental reinforcers, increased a drive e.g., like shock which increases the need to avoid it. 50
Mowrer’s Theory
- Eventually Mowrer considered all learning sign learning, because internal responses (proprioceptive) served as stimuli that gave a “sign” of what to expect and therefore solution learning was not needed.
- Differing with Hull, Mowrer suggested that learning can occur both with drive induction (onset) and drive reduction (termination). 51
Kenneth Wartinbee Spence
- Born May 6, 1907 in Chicago, Illinois.
- Student of Hull. Did his Ph.D. with him from Yale (1933).
- Hull and Spence had tremendous influence on each other, therefore Hull learning theory is sometimes called Hull- Spence theory of learning. (1907-1982 ) www.nap.edu
55
Latent Learning
Latent learning is implicit learning that takes place without reinforcement and remains dormant until reinforcement reveals it. Tolman and Hoznik (1930) did pioneering experiments in this area, and Spence and Lippitt (1940) replicated and confirmed these experiments later on. 56
Latent Learning
Spence and Lippitt (1946) took sated and quenched rats and allowed them to explore a Y- maze with food and water in each arm. They were later divided in two groups and made them hungry and thirsty. Since the rats had latently learnt the maze hungry rats went to the arm with food (left) and thirsty rats to the arm with water (right). 57
Arguments on Latent Learning
- Hull suggested that removing the rats from the arm of the Y-maze was sufficient (though small), reinforcement to cause learning.
- Spence disagreed and suggested that this type of learning occurred independent of reinforcement.
- Animals learns a response by making it (contiguity: Guthrie, Aristotle) and making it many times (frequency: Aristotle).
58
Latent Learning
For Hull, in the absence of drive ( D ) or incentive ( K ) a learnt response could not be emitted. Spence argued otherwise. If one has eaten frequently at a location at 6:00. Whenever at that location, at 6:00, will eat even though not hungry ( D = 0) or not given an incentive ( K = 0, latent learning). Spence changed Hull equation as follows: S ER =^ SHR X (D + K )^ -^ ( IR +^ SIR )]^ –^ SOR — 59
Frustration-Competition Theory
- Hull suggested when K = 0, reactive inhibition ( IR ) and conditioned inhibition ( SIR ) become larger and the response extinguished.
- Spence argued when a regular reinforcer was removed and K equaled 0, the animal got frustrated for not receiving a reinforcer, and emitted responses incompatible with the learnt response which gets extinguished. 60
Types of Frustration
- Primary Frustration: Not finding the reinforcement (food) in the goal box.
- Fractional anticipatory frustration reaction ( r F) A frustration caused by anticipating (Tolman’s beliefs ) no reinforcement in the goal box. All such responses (rF) become stimuli (sF) for resisting the goal box.
