Covert Sensitization and Behavior

The process of covert conditioning attempts to bring about a change in maladaptive behaviors, whether overt or covert in nature, by manipulating private events such as thoughts, feelings or images. J. F. Cautela (1966) developed a procedure known as 'covert sensitization' for treating maladaptive approach behavior. The procedure is covert as neither the undesirable behavior or aversive stimulus is presented as the subject is 'sensitized' to a particular avoidance response in relation to the defective behavior. Relaxation is part of covert sensitization, in the same manner as it has been in the desensitization procedure (Wolpe, 1958, pp. 139-155). Desensitization works in the opposite direction by encouraging certain behaviors in the treatment of maladaptive avoidance responses, whereas covert sensitization treats maladaptive approach responses in behavior. The subject learns to pair aversive consequences with the imaginal representation of the maladaptive behavior. Understandably, it has therefore been suggested that anyone applying the techniques of covert conditioning be familiar with the psychology of learning (Cautela & Kearney, 1986).

Covert Sensitization has been used in the treatment of a wide range of behaviors such as alcoholism, overeating and deviant sexual behavior (Lichstein and Hung, 1980). An overt example of the same principles is a shock administered as an aversive stimuli in the presence of a fetish object (Marks, Rachman & Gelder, 1965). However, covert sensitization has also shown to be an effective component of treatment for sexual deviations (Callahan and Leitenberg, 1973; Cautela and Wisocki, 1971). Despite its popularity as a clinical technique in such settings, however, there is still little empirical data and theoretical guidelines surrounding covert sensitization (Clarke & Hayes, 1984). One assumption that has been made is that all covert aversive consequences are of equal relevance, known as the 'equipotentiality premise'. The consideration was seen to be; in light of equally-aversive covert stimuli; the intensity and degree of discomfort of these stimuli. Wilson and Davidson (1969) suggested that this may not be the case and that it was in fact the quality of stimuli that was important. This was in light of studies in animal taste-aversion which demonstrated that tastes associate with toxicosis malaise, and audio-visual cues with shock (e.g., Garcia and Koelling, 1966). It seems that the more relevant or contingent the consequence is to the actual behavior under scrutiny, the more likely that the employed aversive technique will succeed. We would therefore need to investigate whether or not such a possibility were true. That is, whether the quality rather than the quantity of aversive stimuli were the important factor. This would be achieved by attempting to gauge a difference in behavior after the covert use of a contingent aversive image in association with the behavior in question. In the present study our query is therefore whether the application of a contingent imagery has any effect on a specific behavior (in this case, consumption and reported taste of green cordial). We would expect to find that relevant aversive imagery has a greater effect than aversive imagery of a different quality, or of no particular relevance.

The presence of two different dependent variables also provides two levels of behavior from which we can develop conclusions. Not only can it be determined whether or not future consumption of the liquid has decreased thanks to the aversive technique, any perceived changes in 'approach' or 'attitude' towards consumption may be investigated. As covert sensitization, being a covert conditioning technique, seeks to manipulate private events such as thoughts and feelings forming the base of behavior, we would expect to notice a perceived change in palatability rating as a possible manifestation of this.

Method

Subjects

100 subjects are randomly assigned to three different imagery conditions, being second year psychology students consenting to be a part of this experiment after being requested to do so. 39 to the Phobic Scene instruction group, 31 to Attributed Nausea and 30 to Contingent Nausea.

Apparatus

2 cups of green cordial for each participant.
The Test Rating Scale presents two different scales for taste rating on a -10 to +10 scale, with a halfway point marked 0; -10 meaning 'really awful stuff', 0 being the equivalent of 'I can take it or leave it' and +10 being the highest possible rating of 'really great stuff'. An example of the Test Rating Scale is presented in Appendix A.

Procedure

Each participant is randomly assigned to one of the three imagery instruction groups, which will serve as the Independent Variable (IV) under manipulation by the experimenter. These groups are identified as the Phobic Scene group (PS), Attributed Nausea group (AN) or Contingent Nausea group (CN). The participant is then read the introductory portion of the prepared instructions (found in Appendix B) to inform them of the procedure in which they are about to be participating. This serves as a Plain Language Statement. The participant is also given the choice whether or not to proceed with the session at this point, and invited to explain (by writing on the rating scale) why they may have chosen not to at the conclusion of the experiment. This insures that the participant has given their consent.

A 10 minute relaxation period is then read out to the subject before the actual covert sensitization procedure takes place. This may serve as a means of reducing the possibility of confounding elements such as stress, apprehension, or other covert behaviors which may impinge upon the effective application (and identification) of aversive imagery.

The participant is then invited to drink from the first cup (but not all of it) and give the cordial a taste-rating on the scale provided.

Either the PS, AN or CN imagery is the read to the subject before a further minute is spent encouraging a state of relaxation. An example of the imagery scenes is to be found in Appendix C. The participant is then told to visualize consuming the liquid and then undergoing the scene described to them dependent on the instruction group into which they have been assigned. This procedure of relaxation and visualization is then repeated three times for a total of four covert sensitization pairings.

The subject is then invited to drink from cup 2 and provide another taste rating.

Those who chose not to participate are now asked to identify their reasons for this choice under the rating scales.

A measurement in the difference between the volume drunk from cup 1 and cup 2 is now taken.

Our data is then calculated as a comparison between palatability ratings given pre- and post-test (TASTE), and a comparison between the volume drunk pre- and post-test in millilitres (SIP), in order to determine if any changes in the behavior under scrutiny occurred as a result of the imagery condition.

Results

The means and standard deviations for the changes between the samples are shown in Table 1, providing a basic indication of the average change in the amount drunk (SIP) or palatability rating (TASTE). The analysis of variance revealed that there was a statistically significant change in the amount drunk (SIP) pre- and post-test (F(2, 97) = 15.00, p < .05), but no significant change in taste rating (TASTE) pre- and post-test (F(2, 97) = 0.95, p > .05). After determining that only a change in amount consumed proved to be a statistically significant change in behavior, it was then necessary to determine exactly where the significance exists between the three treatment means for SIP. Tukey's HSD (Honestly Significant Difference) Test was applied, the results of which can be found within Table 2. Although a significant mean difference exists between the two nausea conditions and the Phobic Scene condition, there is no statistical evidence to suggest a difference between the Attributed Nausea and Contingent Nausea instruction groups.

Table 1. Descriptive Statistics - Mean & Standard Deviation.

Table 2. Multiple Comparisons - Dependent Variable: post-pre (change in amount drunk) Tukey HSD *. The mean difference is significant at the .05 level.

Discussion

It would seem that a change in overt behavior has occurred and with significant difference between the three established imagery conditions. By eyeballing the means and standard deviations collected it would appear that both our dependent variables, TASTE and SIP, changed as a result of the imaginal aversive stimuli, but this is not the case. An analysis of variance demonstrated that only the change in amount drunk was statistically significant. This suggests that although a change in behavior did occur, the exact nature of this change is itself of a specific quality, just as it was predicted that different aversive stimuli were of different quality in relation to one another.

The aversive stimuli also provided results that added extra complexity to the expected data. It would seem that a change in behavior has occurred and with significant difference between the three established imagery conditions. This result serves us immediately in denying that the 'equipotentiality premise' has any credence in relation to aversive stimuli. However, no significant difference was found between the attributed and contingent nausea instruction groups, which would suggest that statistically they are of the same quality in relation to effectiveness as an aversive stimulus. The fact that the Phobic Scene was completely different from these two nauseating aversive stimuli suggests that it may have been less effective than the two similar nausea conditions, which are more relevant to the behavior in question than the phobic imagery. We can assert that our principle assertion that relevant, covert, aversive techniques have a greater effectiveness in the treatment of maladaptive behaviors.

It needs to be recognized, however, that although the aversive stimuli does need to be relevant to the behavior under scrutiny, it does not need to be rigorously specific. Consumption of a liquid certainly proved to be associated with toxicosis malaise more than with any other aversive stimuli, but with no statistically significant difference between attributed or contingent malaise (as demonstrated in Table 2). Further research would need to investigate the possibility that aversive conditions of a similar relevance to the behavior are both equally effective in obtaining the desired results.

It is also noteworthy that attempts to identify an actual change in covert behavior rather than overt expressions of it have yielded no result if we determine change in palatability as a manifestation of altered perception of the liquid. Further investigation into whether covert sensitization actually targets covert behaviors or only the outward, overt manifestations of these private events also needs to be pursued.

References

Clarke, J. C., & Hayes, K. (1984). Covert sensitization and the equipotentiality premise. Behavior Research and Therapy, 22, 451-454.
Cautela, J. F. (1967). Covert Sensitization. Psychological Reports, 20, 459-468.
Cautela, J. F., & Kearney, A. J. (1990). Behavior analysis, cognitive therapy, and covert conditioning. Journal of Behavior Therapy and Experimental Psychiatry, 21, 83-90.
Gravetter, F. J., & Wallnau, L. B. (1996). Statistics for the Behavioral Sciences (4th ed.). St Paul, MN: West Publishing. pp. 371-409.
Miller, W. R., & Dougher, M. J. (1989). Covert sensitization: Alternative treatment procedures for alcoholism. Behavioral Psychotherapy, 17, 203-220.
O'Shea, R. P. (1993). Writing for Psychology: An introductory guide for students. Sydney: Harcourt, Brace & Javanovich.

Appendix A

Taste Rating Scale

Taste Rating 1: Rate the taste of the liquid on the following scale:

-10!___!___!___!___!___!___!___!___!___!0___!___!___!___!___!___!___!___!___!___!+10

(Where -10 is the lowest possible rating = "really awful stuff", 0 = "I can take it or leave it" and +10 is the highest possible rating = "really great stuff")

Taste Rating 2: Rate the taste of the liquid on the following scale:

-10!___!___!___!___!___!___!___!___!___!0___!___!___!___!___!___!___!___!___!___!+10

(Where -10 is the lowest possible rating = "really awful stuff", 0 = "I can take it or leave it" and +10 is the highest possible rating = "really great stuff")

Appendix B

Instructions

"In front of you are two glasses and a rating scale. Please do not touch them until you are asked to do so."
"During this session we will be conducting an experiment on sovert sensitization. Covert sensitization is a method of changing behaviour or preferences about a particular substance by using imagery techniques. We will attempt to change your preferences about the taste of green cordial. This may result in some temporary discomfort. If you do not which to participate, you can just sit and relax during the session, and at the end of the session you should write on the questionnaire the reason why you did not participate."
"This session will consist of a few different things. First, there will be a relaxation period. Second, you will taste the cordial. Third, we will describe a scene to you. Fourth, you will imagine both the scene and drinking the cordial, a number of times. Fifth, you will then have to drink more cordial from the second cup."
"Let's start with the relaxation period. Sit up fairly straight in your chair, with your hands in your lap, your legs uncrossed and both feet on the floor. Close your eyes and focus on your breathing. Take in a deep, relaxed breath, hold it for a split second, then breath out. In (1 sec), out, (1 sec). In, out. One, two. One, two. Now that you have a nice easy breathing pattern, be sure that you don't let the tension you will soon be producing interfere with your breathing. I'd like you to continue as though your breathing didn't "know" anything about the tensing and relaxing. Good. All right, in a minute I'm going to ask you to make your right hand into a fist, fingers flat against your palms so your fingers don't dig into them. Then I'll ask you to clench your fist hard, and breath out as you unclench your fist. Let all that tension go as you breathe out. But remember, you are feeling alert and relaxed, not sleepy. Okay, we'll start. As I say "one", make your hand into a tight fist ... as I say "two", unclench your fist and breathe out as you do it." One (2 sec), two (2 sec)."
"Do this for a few minutes."
Then,
"Alright, now I'd like you to stop clenching and unclenching your fist, and just concentrate on your breathing. In, out. In, out. One, two, one, two. Remember, you are feeling alert and relaxed, not sleepy.."
Do this for a few more minutes. The whole procedure should take about 10 minutes.
"Now open your eyes. Taste the cordial in the cup labelled '1' in front of you, but don't drink all of the cordial in the cup.
"Now, give the cordial a taste-rating on the scale provided.
"Now, close your eyes again, I'm going to read a scene to you"
Read either the PI, AN or CN scene, slowly, with pauses of a few seconds between utterances.
"Okay, now I'd like you to just relax again ... close your eyes, and concentrate on deep and easy breathing. You are alert and relaxed, not sleepy." (One minute).
"Now for the next couple of minutes, I'd like you to again visualize yourself consuming the liquid, and then undergoing the consequence that I described to you in the scene. Remember, first imagine drinking, then imagine the scene, imagine ... (PI: large hairy spiders crawling on you, your heart is racing, your whole body is sweating; AN: the queasy, churning feelings in your stomach, the unpleasant taste in your mouth; CN: the thick phlegm at the back of your mouth, the unpleasant taste of the liquid, your wanting to vomit up the liquid" (two minutes).
Repeat the above two paragraphs (relaxation, imagine pairing) three times. That is, there is a total of 4 covert sensitization pairings.
"Okay, now I'd like you to taste the cordial again, this time from the glass labelled '2'."
"Now, rate the taste of the cordial on the second rating scale."
"Those people who did not participate, please note the reason for your non-participation under the rating scales."

Appendix C

IMAGERY SCENES

I'd like you vividly to imagine that you are tasting the fluid that you previously rated. "See" yourself tasting it, capture the exact taste, colour and consistency. Use all your senses, you are actually drinking it, tasting it, swallowing it, feel the cup in your hand, and note the temperature, texture, smell, and especially the taste of the fluid. Now I would like you to imagine the following scene:

Phobic Scene

You are on a bush walk, enjoying the scenery. You suddenly trip over a root, and tumble down the side of an embankment, crashing through the bushes. Before you can stop yourself, you slide into a dark hole at the bottom of the embankment. You find yourself wedged in tight, unable to move. You try to wiggle free, but you can't move. You soon become aware of sticky cobwebs covering your face and body, but you can't move to brush them away. As your eyes adapt you sense the movement above you in the dark. You hear small scuttling, rustling noises, only centimetres from your head. You start to sweat and strain to move, but there is no escape. You feel something hairy drop onto your legs and start to walk along them. This gives you a tickling, itchy feeling, but you can't move to stop it. With a start you become aware of masses of large hairy spiders crawling just above you on the ceiling. One by one they start to drop onto your legs, arms and chest. You try to keep absolutely still, but your heart is racing, and your whole body is sweating. You feel the spiders moving in your clothing, crawling between your clothing and your skin. A spider falls on your face, you shut your eyes and mouth tight. You're rigid with fear as the spiders climb all over you.

Attributed Nausea Scene

You are in an amusement park, and you purchase a ticket for the ride called "The Octopus". You climb the stairs to the platform and are strapped into one of the "arms". Looking down you can see the faces of the people below; they look small, and you start to feel slightly uneasy. A cold sweat breaks out on your forehead. With a sudden jerk the ride starts and you are pressed against the back of the seat. The cars move around in a circle, high above the ground, faster and faster and faster. All the colors blur and run together, you can't focus on anything and your stomach feels queasy. Then the motion changes - as well as spinning around, the arms also shoot up and plunge down. You're no longer enjoying this; you feel to sick. You retch and bitter tasting bile rises in your mouth and you can feel and taste it on your tongue. Your head feels hot and you feel dizzy. You feel as though you might vomit, and your stomach churns. You wish for the motion to end. At last it does. You stagger back down to the ground but the world still seems to be spinning. The nausea persists.

Contingent Nausea Scene

You are drinking the cordial. As you swallow, your stomach starts to churn and you really feel sick. You can notice a cold sweat break out across your back, on your forehead, and on the back of your hands as the nausea increases. You notice that the taste of the cordial feels unpleasant in your mouth and on your tongue and you cannot swallow as the unpleasant taste of the cordial blocks the back of your throat. Your head feels hot and light, and slightly headachey. Your eyes refuse to focus properly. Imagine that you stare at the unfinished liquid in the cup, and you notice the smell of the cordial seems to be increasing your sickness. Now phlegm is gathering at the back of your throat, and you feel as if you might vomit up this foul cordial, but you don't get past being on the verge and the sick queasiness remains in the pit of your stomach. Your stomach churns. The cordial seems to be coagulating in your stomach and throat, refusing to move, slowly choking you, but regardless of your heaving and gagging you cannot vomit, or rid yourself of the foul taste, the unpleasant odour, the nausea, the sweating ... .