The most elementary form of Hull-Spence theory is SER = (SHR ) X (D) , where
SER, EXCITATORY POTENTIAL, refers to the probability, strength or speed of a response;
SHR--HABIT STRENGTH, the learning portion of the theory (the acquired habit of emitting a certain response in a certain situation). Habit strength (measured in HABS) increases with each reinforced response. Drive reduction must occur before a response is reinforced; and
D--GENERALIZED DRIVE, the motivational portion of the theory.
So, the likelyhood of an organism emmiting a certain response in a certain situation, SER, is equal to the number of times that response has been reinforced in that situation, SHR, times the generalized motivational state of the organism, D. D doen not direct behavior; D simple motivates behavior.
Note that Excitatory Potential, Habit Strength, and Drive are all intervening variables or hypothetical constructs.
D is a function of rE --the persistent emotional response. The magnitude of can be estimated by knowledge of the past history of the organism (i.e., grounded on the stimulus side). rE is a function of SU (The magnitude of any aversive stimulus received), and Sum SU (The sum of aversive stimuli received). rE can also be estimated by the emotional responsiveness of a human person--Ra. rE in Hull-Spence theory is estimated (i.e., grounded on the response side) by a person’s score on the Manifest Anxiety Scale (MAS). Thus, a person with a high MAS score can be said to have an above average amount of D, compared with a person with a low score on the MAS--in the same way as a person who had received many electric shocks can be said to have more D than a person that had received no shocks.
Classical conditioning of human eyeblink responses typically have used a tone as a CS and an air-puff to the cornea as the UCS. According to the theory the acquisition of conditioned eyeblinks should occur more rapidly when more aversive air-puffs (2.0 lbs/sq. In) are used as UCSs than when less aversive air-puffs are used (.6 lbs/sq in.) Replicated research supports the theory.
In a similar fashion, subjects with high scores on the MAS (HiA) should show more rapid acquisition of conditioned eyeblinks than LoA subjects, in 1964 Spence reported that 21 of 25 studies showed HiA>LoA on conditioned eyeblink tasks.
Development of the Manifest Anxiety Scale
Janet Taylor (Before she married K. Spence) gave 200 MMPI items to 5 clinical psychologists and asked them to pick items that reflected Cameron’s (the author of a text on psychopathology) description of chronic anxiety patients. The clinicians had 80% agreement on 65 items, which comprised the initial scale, which has later revisions to shorter forms. The 50 item scale has a split-half reliability of .92 and test retest reliabilities of .81 to .89 over a 3 week to 17 month period.
Subsequent research tested Hull-Spence anxiety theory using more complex learning tasks. One way of defining task complexity involved the notion of the HABIT FAMILY HIERARCHY in Hull-Spence theory. Specifically, more than 1 response can be conditioned to the same stimulus. Imagine a food-deprived rat trained in a complex maze (the Stimulus). 60% of the time the rat is reinforced for turning Right (R-Right); 30% of the time the food is in the goal box that is straight ahead (R-Forward); 10% of the time the food is in the left goal box (R-Left). Over many training trials the habit strength of R-Right would be stronger than R-F would be stronger than R-L. These three habits could be said to be arranged in a hierarchy according to thier strength. The same principal can apply to human learning. For example, in the Stroop test, reading is a stronger habit than color naming. If we call reading the Stroop test the correct response, the Stroop is a much easier task than if we call color naming the correct response. Thus easy tasks are those that are defined as having the dominant habit as the correct response. Difficult tasks have a subordinant habit as the correct response. With extended training on a difficult task the subordinant habit becomes stronger over time (with continued reinforcement) and can surpass the original dominant habit. When this happens the task becomes “easy” because the correct response is now dominant. Hull-Spence theory predicts that HI-A > LO-A on easy tasks. At the beginning of difficult tasks, the theory predicts that LO-A > HI-A. As learning on a difficult task progresses, however, there will be a cross-over effect such that when the task becomes easy, HI-A > LO-A.
Go now and read the notes with the STANDISH AND CHAMPION reading, which is a good test of this theory. Then come back here for the REVISED VERSION OF THE SPENCE & SPENCE, anxiety theory.
YOUR FIRST EXPERIMENT IS A TEST OF THE REVISED VERSION, NOT THE ORIGINAL VERSION PRESENTED THUS FAR. Don’t start to write your intros yet.
By 1966 there were some research reports that did not support the 1st version of Spence-Spence (S-S) theory. In stimulus generalization studies, Mednick could only find results consistent with the theory when using naive subjects. Veterans of psych research studies did not behave according to the theory. S-S also found that MAS effects were more likely to occur when using naive Ss, dim rooms, and conditions that made subjects uneasy. So the anxiety theory was revised using other aspects of Hull-Spence theory. The revised theory incorporates a RESPONSE INTERFERENCE HYPOTHESIS. Before we consider this hypothesis it is important to recognize that this revision weakens the theory. In the first version, Hi and Lo MAS folks were expected to behave differently in ALL situations. In the revised version, Hi and Lo MAS folks are only expected to behave differently in situations in which:
THERE IS SOME DEGREE OF THREAT
Hi-A subjects, in addition to having elevations of D, are also assumed to have lower thresholds in recognizing and responding to threatening situations than Lo-A folks. Remember that HiA Ss also have higher levels of the persistent emotional response, rE, which produce drive stimuli (the stimulus components of a drive state). These drive stimuli have the capacity to evoke responses (learned, unlearned [physiological], overt, covert) that are TASK-IRRELEVANT. For example, in a threatening situation the HiA person would have heightened autonomic reactions (high heart rate, start sweating, etc.); become aware that he/she is nervous, engage in covert vocalizations such as “I’m not doing well. What are they thinking of me. What am I doing here. I’v got to get out of here.” ; and have other reactions that are deemed “irrelevant” to the task at hand. The LoA person would have fewer of these task-irrelevant responses.
These task irrelevant responses may be compatable with the behavior required in the threatening setting; if so they would facilitate the desired behavior and HiA > LoA. This is often the case in simple (easy) tasks.
However, when the taks irrelevant responses compete (interfere) with desired behavior, then the HI A < Lo A. This is often the case in difficult tasks.
Thus in the revised version of the theory. MAS differences are dependent on two conditions:
1. THE PRESENCE OF THREAT (otherwise HiA = LoA), and
2. The degree to which task irrelevent responses facilitate or compete with desired behavior.
2a. If the task is easy and the task irrelevant responses facilitate desired behavior: HiA > LoA.
2b. If the task is difficult and the task irrelevant responses compete with desired behavior:
HiA < LoA.
NOW YOU ARE READY TO WRITE YOU INTRO TO THE PAPER:
Assume that the Stroop test is a difficult test and that Hi-A Ss under threat (loud noises) would have more interfeering responses than Lo-A Ss under threat.
FROM HERE YOU SHOULD BE ABLE TO FORMULATE THE HYPOTHESES FOR THE FIRST PAPER.