The Use of Respondent Conditioning in Improving Health Outcomes

Emily Mandel, M.S., BCBA, LABA

bSci21 Contributing Writer

Approximately 133 million Americans (50% of the overall population) suffer from chronic illness, which results in 7 in 10 deaths each year (Center for Disease Control and Prevention). In addition, chronic illnesses account for over 75% of healthcare costs, largely in part due to invasive procedures that cost patients and insurance companies trillions of dollars per year.

Some research has been conducted to investigate the management of chronic illness with non-invasive procedures. These practices, if proven effective and implemented on a national scale, could potentially minimize the use painful procedures, as well as decrease overall healthcare spending. One such technique is the use of respondent conditioning.

Respondent conditioning “occurs when an organism responds to a new event based on a history of pairing with a biologically important stimulus” (Pierce & Cheney, 494). That is, there is a pairing of two antecedent stimuli, an unconditioned stimulus that naturally elicits responding, and a neutral (eventually conditioned) stimulus, which takes on an eliciting effect after repeated pairings with the unconditioned stimulus.

There is a small and slowly developing area of research that investigates the use of respondent conditioning in manipulating physiological reflexes that are damaging to an organism’s health, such as high blood pressure and sexual dysfunction. The following articles demonstrate how respondent conditioning can be used to improve health outcomes across a large range of respondent behaviors.

A study conducted by Letourneau and O’Donohue (1997) investigated classical conditioning of female sexual arousal. Experimenters placed participants into two groups, one intervention group and one control group. In the intervention condition, clips of erotic videos were paired with an amber light. In the control condition, participants were also exposed to the amber light and the erotic video clips, but these stimuli were not presented simultaneously. There were 2 dependent variables: subjective participant ratings, as well as a measure of change in vaginal pulse amplitude. Subjects in the intervention group did demonstrate increased arousal, but the changes from baseline were not drastic. This suggests that, while the treatment was effective, the independent variables could have been presented differently (such as increased magnitude or duration) in order to elicit a more response with greater magnitude.

In an experiment carried out by Plaud and Martini (1999), investigators used a penile plethysmography (a tool that measures expansion and shrinkage of an organ) and 2 control procedures to measure male sexual arousal. There were 9 participants in the study. The study lasted 3 weeks and consisted of 3 sessions. There were 15 stimulus periods and 15 detumescence (decrease in penile expansion) periods in each session. There were 3 conditions with 3 participants in each. In the first condition (short delay conditioning), experimenters presented the conditioned stimulus for 15 seconds, after which there was an immediate presentation of the unconditioned stimulus for 30 seconds. In the second condition, experimenters used a backward conditioning procedure (the unconditioned stimulus was presented immediately before the conditioned stimulus). The third condition was a “random control condition,” in which the conditioned stimulus and unconditioned stimulus were presented in a random order. The results of this study were that the short delay condition produced penile tumescence (swelling), but the other conditions did not. These results suggest that the use of short, rather than long, delay procedures are potentially more effective at eliciting respondent sexual behaviors, and future research should expand upon short delay respondent conditioning (possibly through the use of parametric analyses of delay length).

A study conducted by Stockhurst, Steingruber, and Scherbaum (2000) aimed to determine whether blood glucose could be decreased through the use of respondent conditioning. In Experiment 1, there were 2 groups: a control group that was given a placebo repeatedly paired with the conditioned stimulus, and a group that was given insulin repeatedly paired with the conditioned stimulus. The group that received the insulin and the conditioned stimulus experienced a decrease in blood glucose in the presence of the conditioned stimulus alone, while the participants in the control group did not experience a blood glucose decrease in the presence of the conditioned stimulus. Examiners then conducted a second experiment. Subjects either received the conditioned stimulus paired with insulin, the conditioned stimulus paired with glucose, or the conditioned stimulus paired with the placebo. The participants who were given the pairing of the insulin and the conditioned stimulus experienced a decrease in blood glucose similar to that in the first experiment. The participants in the other conditions did not experience a significant change in blood glucose. These results demonstrate that classical conditioning can be used to change physiological responding such as glucose production and change in glucose levels within the body.

A study conducted by Whitehead and Blackwell (1976) was conducted to determine whether classical conditioning could produce a decrease in systolic blood pressure. Participants in two of the groups did not have hypertension, while participants in the third group did. Tilting the head forward 15 degrees has been proven to produce small decreases in systolic blood pressure. Experimenters paired the presentation of an auditory conditioned stimulus with the head-tilting procedure. The results were that even participants in the hypertensive group experienced a decrease in systolic blood pressure when the conditioned stimulus was presented on its own, following repeated pairings with the unconditioned stimulus (the head-tilting procedure). The control group demonstrated that the reflex’s “sensitization” by tilting the head multiple times could not account for decreases in blood pressure similar to the ones the experimental subjects experienced when exposed to the conditioned stimulus. These results demonstrate that respondent conditioning can elicit decreases in systolic blood pressure, but future research should aim to manipulate the independent variables in a way that would elicit a greater decrease in blood pressure.

The above studies demonstrate that respondent conditioning can be used to change certain health outcomes. While these interventions were effective, they often produced only small physiological changes, and maintenance probes were rarely performed. Future research should aim to elicit a greater change in the dependent variable, as well as to program and test for maintenance and generalization. Behavior analysts should conduct further research within this area with the aim of offering the public less invasive alternatives to commonly accepted treatments, and improving overall public health.

How do you think respondent conditioning could contributing to health outcomes?  Let us know in the comments below and be sure to subscribe to bSci21 via email to receive the latest articles directly to your inbox!

Centers for Disease Control and Prevention (December 17, 2009). Chronic Diseases – The Power to Prevent, The Call to Control. Retrieved from http://www.cdc.gov/chronicdisease/resources/publications/aag/chronic.htm

Letourneau, E.J. & O’Donohue, W. (1997). Classical conditioning of female sexual arousal. Archives of Sexual Behavior, 26, 63-78.

Pierce, W.D. & Cheney, C.D. (2013). Behavior Analysis and Learning. New York, NY: Psychology Press.

Plaud, J.J. & Martini, J.R. (1999). The respondent conditioning of male sexual arousal. Behavior Modification, 23(2), 254-268.

Stockhurst, U, Steingruber, H.J., & Scherbaum, W.A. (2000). Classically conditioned responses following repeated insulin and glucose administration in humans. Behavioural Brain Research, 110(1-2), 143-159.

Whitehead, W.E. & Blackwell, B. (1976). Classical conditioning of decreases in human systolic blood pressure. Journal of Applied Behavior Analysis, 9(2), 153-157.


Emily Mandel, M.S., BCBA, LABA, is a behavior clinician in the Greater Denver Area who works with children with a range of developmental and social-emotional disabilities. She has over 4 years of experience delivering therapeutic services both in-home and in school settings. Though she is predominantly focused on the utilization of Applied Behavior Analysis in treating individuals with disabilities, Emily enjoys examining topics such as religion, medicine, politics, and social constructs, through a behavioral lens. You can contact her at [email protected].