SMOKER PSYCHOLOGY PROGRAM REVIEW
SMOKER PSYCHOLOGY PROGRAM REVIEW
Philip Morris, October 10, 1977Bates #: 2060564018/4026
[Handwritten in top left corner: "Bill Dunn, 10/19/77"]
SMOKER PSYCHOLOGY PROGRAM REVIEW
I've taken this as an occasion to review our basic premises - to ask myself why are we doing the things we're doing. It would be much easier just to do an inventory of our projects. It isn't easy to examine the rationale for one's existence.
First, every undertaking has to have some mission. I would state our charter from Philip Morris in this fashion:
Study the psychology of the smoker in search of information that can increase corporate profits.
Our charter is unique in that we are given the smoker to study. Other R&D charters specify the cigarette in terms of component parts (tobacco material, flavor) some phase of its development (new products, tobacco services) or the professional discipline brought to bear upon it. We get the whole smoker to ourselves. But there isn't that much freedom, really, because the charter demands that the research have some promise for dollar payout.
Over the years we have more unconsciously than consciously been applying two criteria in selecting projects:
a) Some better than chance probability of gaining new knowledge
b) The prospective new knowledge could have corporate usefulness.
I invite you to apply these criteria to assessing the Behavioral Research Program.
With this charter and these two criteria, how might we proceed:
1) Product development
a) cost reduction - in 1965 we said total utilization
in 1966 we said forget aging
b) product improvement - nicotine/tar ratio
puff patterns
dilution hole placement . . .
2060564018
Plaintiff's Exhibit 5209
-2-
2) Smoker persuasionWe've devoted considerable effort in the past to elucidating the benefits of smoking and would likely wish to continue in that direction. But there's diminished likelihood of payout on this front because of increased industry-wide legal concern over the counter-thrust of antismoking groups. It is feared that any pronouncement about smoking from the industry will provide an opportunity for attack. There is also understandable legal concern about any industry-endorsed reference to the pharmacological effects of smoke, however beneficial that effect might be. The risk is great that such reference might be used to advantage by those seeking to persuade the public that cigarette smoke contains a dependency-producing drug classifiable with the amphetamines and opiates.
a) advertising - Virginia Slims, active charcoal theme
b) public relations - St. Martin, benefits, SEX-1
We can therefore no longer reasonably justify our psychophysiological study of smoking in terms of its potential payout for a PR promotion of the benefits derived from smoking. But we can and do justify the research effort on the grounds of potential contribution to product development.
There is a general realm of psychological inquiry that would not make our lawyers nervous were the findings to be made public. This is the socio-psychological aspects of smoking. Dr. Wakeham has gone out on the extreme end of the spectrum. He is talking with a cultural anthropologist at Harvard about a contractual study of the attitudes of smokers and nonsmokers towards one another. The publication of this kind of material would likely be sanctioned.
2060564019
-3-
We are not planning to devote much effort to studying smoking in its social context, although we have one project that is of this nature. We'd welcome any feedback from you regarding our priority decisions. Should we be expanding our effort into the sociopsychological sphere?
Meanwhile, back in the lab, we continue to ask what it is that the smoker derives from smoking.
You've no doubt heard me at one time or another state that we are seeking to understand the reinforcement mechanism in smoking. It occurs to me that some explanation is in order. What is a reinforcement mechanism?
I'll try to describe the conceptual model that we are using.
There are some distinctive features about the cigarette market. We sell a product that is bought often and used from 10 to 60 times a day, day in, day out. We provide a product that is the essential commodity in a habitual act. It is inevitable therefore that we focus upon habit-formation and habit maintenance, and that we make use of a conceptual model specifically designed to explain habit formation. The model is known within psychology as the Skinnerian operant conditioning model.
You are all familiar with Pavlov's dog. Present food, he salivates. Ring a bell each time you present food, and in due time you can ring the bell without the food and he will salivate. This is classical conditioning. The bell, a conditioned stimulus, will elicit the unconditioned response.
Operant conditioning is quite different. Consider the classical operant experiment. Put a hungry rat in a cage. Put a lever in the cage. Make it so that a push on the lever delivers a pellet of good. In its random movements the rat will push the lever. He'll get one pellet. The rat will soon associate
2060564020
-4-
lever-pushing with the food pellet. In time he will be rapidly pushing the lever for food. This is the paradigm of operant conditioning. What are the essential elements. The lever pressing is the operant behavior. The food pellet is the rewarding stimulus. The rewarding stimulus reinforces the operant behavior. The word operant refers to the operations on the environment which lead to reward.
As a training method, nothing is new. Animal trainers for generations have shaped the behavior of their animals by rewarding the right movements with food. Skinner has trained his pigeons to perform grotesque movements quite foreign to their normal repertoire by the simple technique. You have trained your dog to roll over in the same fashion. What is new is the conceptual model and its generalization to human behavior, particularly that behavior which has become habitual.
Consider the smoker. Smoking the cigarette is the lever press. The effect of that smoking act upon his person is the reward. The effect reinforces the smoking act. He comes to push the lever 10 to 60 times per day. Our task is to understand the reinforcing mechanism, or process, whereby the habit is established and maintained. Such is our model. Conceptually simple on the surface, but the complexities lurk in numbers beneath the surface.
First, consider the operant behavior. It is not a simple lever press. The smoking act has many dimensions. The motor aspects, drawing the smoke into the mouth, inhaling the smoke into the lungs, and consider the complex composition of the smoke. What are the essential elements of this operant behavior?
Now consider the many effects upon the person of the smoker. Taste and olfactory sensations are elicited, sensations are elicited when the smoke is drawn down into the lungs, psychosocial symbolism is imparted to the person,
2060564021
-5-
and a large number of chemical compounds pass rapidly into the bloodstream. Which of these many events are rewarding? Which are essential for reinforcing the act?
Our task would be much easier if the essential parts of the smoking act were cleanly identified, or if the essential rewards were identified. Then we'd only have to fret with the one or the other. But we're being double-ended.
This, then, is how we have translated our corporate charter into a scientific model.
As I told you earlier, we are forced into making some judgmental decisions. One important decision was to assume that Component X of the smoking act is likely to be found among the chemical compounds being introduced into the bloodstream. This assumption has in large measure provided the structure and direction of our research program. The assumption does not rule out other components of the act as important variables, but it does attribute a primary role to the active chemical compound. Without the chemical compound, the cigarette market would collapse, P.M. would collapse, and we'd all lose our jobs and our consulting fees.
The questions which give substance to our research effort can now be ground out:
1. Is nicotine Component X? For much of our research, we beg that question, and proceed with:
If so:
a) What is the lower delivery level limit beyond which the smoking act is not reinforced?
2060564022
-6-
b) Within what limits can we vary nicotine concentration relative to other smoke constituents?
1) What is the optimum nicotine/tar ratio?c) Given a fixed quantity of nicotine in the tobacco, what factors in the cigarette design determine its availability for delivery to the smoker? . .
d) At what point in the delivery sequence is nicotine to be meaningfully monitored? mainstream? respiratory tract retention? bloodstream?
e) Does the smoker seek spike effects, or bloodstream constancy?
f) Can some moiety of the nicotine molecule be Component X?
g) How important is the form of the delivered nicotine? (salt vs. free base? pH? particle size?)
2. Is Component X some other compound than nicotine? CO? some other alkaloid?
3. What is the nature of the reinforcement?
a) What is its locus? c.n.s.? specific brain size? synapse? peripheral? autonomic?2060564023
-7-
b) What is the psychological correlate? increased alertness? reduced aggressivity? reduced anxiety? subjective well-being? damped emotional responsivity?4. Is the industry placing itself at risk by lowering delivery levels?
c) Is the reinforcement mediated by one of the known physiological effects of smoke, or one yet to be isolated?
a) Is there a relationship between brand delivery level and quitting rate?
5. If we could preserve all the stimulus properties of smoke and block all pharmacological action, would the smoker cease to smoke?
6. Is cigarette smoke the vehicle of a dependency-producing drug? What are the fundamental differences between the habit of tobacco smoking and heroin injection?
7. Is there some personal or psychophysiological attribute that makes for stronger reinforcement of the smoking act?
8. Are there other factors not related to Component X that influence the frequency of the smoking act? e.g. the social utility of the act?
Many of these questions we are not qualified to try to answer. And there are other constraints:
1) noninvasive2060564024
2) nonobtrusive
3) ethics
-8-
Now let's take a look at the research program.
Because of a strong conviction about the central role of the pharmacological effects of inhaled smoke, much of our time and thought has been and will be devoted to the questions we've listed having to do with these effects. We are investigating both whole smoke as the independent variable and nicotine as the independent variable.
We reason in this fashion:
If the operant behavior (smoking) is being reinforced by the psychopharmacological effect of some component of the smoke, then we should be able to:
1. Observe quantitative changes in smoke-intake as a function of changes in the concentration of that agent in the smoke (the titration hypothesis)
2. Identify the reinforcing psychopharmacological effect at both the physiological and psychological levels.
3. Observe extinction of the smoking act in the absence of the agent in the smoke.
Our present program addresses itself to all three of these contingencies:
1. a) We have investigated the titration hypothesis for a long while.
At no time have we gotten clean supporting evidence. We are investigating two sub-hypotheses at present:
1. Additional regulation is occurring at the inhalation level.2060564025
2. Other factors than dosage regulations trigger the smoking act, so that the smoker is super-dosed. The act has taken on utility function of its own through secondary reinforcement.
-9-
b) We have requested denicotinized cigarettes in order to independently vary nicotine delivery.
2. a) We are establishing an EEG laboratory in search of the reinforcing event. Brain waves are neurophysiological phenomena, but they are legitimate subject matter for us in that brain events underlie behavioral events. Smoke related changes in brain waves can give us clues as to smoke related psychological changes.
b) Learned helplessness
c) Performance of rats under stress
d) Learned discrimination between saline and nicotine injection.
3. a) Self-administration of nicotine by rats.
b) Quitting rate as a function of cigarette TPM delivery level.
c) Monitoring of cigarette acceptability as a function of delivery level.
2060564026
CITATION:
DOCUMENT ID: 2060564018/4026
OTHER NUMBER: LIT0000046538
DOCUMENT DATE: 19771019
PRIMARY TYPE: REPT, REPORT, OTHER
PERSON MENTIONED: PAVLOV; SKINNER; WAKEHAM
PERSON NOTED: D,W
ORG MENTIONED: EEG LAB; HARVARD; TPM
CHARACTERISTICS: CONF, CONFIDENTIAL; ILLE, ILLEGIBLE; MARG, MARGINALIA
FILENAME: 2060563811/2060564105/MISSING;
SITE: R461
AREA: ELLIS,CATHY/OFFICE
TITLE: SMOKER PSYCHOLOGY PROGRAM REVIEW
MENTIONED BRAND: ST MARTIN; VIRGINIA SLIMS
PAGE COUNT: 9
DATE LOADED: 19990417
LITIGATION USAGE: IWOH/PRODUCED
***********************
Go To: Tobacco BBS HomePage / Resources Page / Health Page / Documents Page / Culture Page / Activism Page
***********************
END OF DOCUMENT