Verified by 
What Is Biofeedback?
I distinguish between biofeedback training and therapy. Biofeedback training refers to the methods and procedures used to increase control over physiological responses. Biofeedback therapy refers to clinical applications of biofeedback training. I use the term biofeedback to refer to both biofeedback training and therapy.
Biofeedback uses specialized instruments to measure and display a variety of physiological responses including skin temperature, skin conductance, muscle tension, and brain waves. Conscious control over these physiological responses gradually emerges with training.
How Popular Is Biofeedback?
Biofeedback and biofeedback therapy continue to be popular. Searching the National Library of Medicine (Pub Med) for publications regarding “biofeedback” returned 11,030 citations on 02/08/20. Searching for the term “Biofeedback Therapy” returned 9,362 citations on the same day. Table 1 in Frank et al. (2010) lists medical conditions for which empirical evidence shows that biofeedback is efficacious, probably efficacious, possible efficacious, and not empirically supported. Taken together, this evidence supports the claim that biofeedback works for a variety of disorders. However, the question of how it works remains mostly unanswered. The aim of this post is to better answer this question.
How Does Biofeedback Work?
Operant conditioning is the current and dominant explanation for how biofeedback works. This explanation had already become widely accepted by 1977 when it was promoted by Black, Cott, and Pavloski. Operant conditioning of behavior was popularized by B. F. Skinner, who studied the behavior of rats and pigeons. He reported that the frequency of bar presses by rats and key pecks by pigeons increased when they were followed by access to food. Skinner argued that access to food reinforced, strengthened behavior. This reasoning was extended to biofeedback as follows.
Behavior in the biofeedback context refers to physiological responses such as skin temperature, skin conductance, muscle tension, and brain waves. Reinforcers in the biofeedback context refers to the auditory and visual displays of these physiological responses provided by the biofeedback equipment. For example, elements in a light bar can appear to move to the right as hand temperature increases indicating greater blood flow due to increased relaxation. Apparent movement to the left would indicate reduced blood flow due to decreased relaxation. Alternatively, a color display might be used. A redder display might indicate a warmer hand whereas a bluer display might indicate a colder hand. Sound can also be used. A lower frequency tone could be used to reflect a warmer hand and a higher frequency tone could be used to reflect a colder hand.
Reinforcement is thought to be mostly automatic when applied to the behavior of animals. Reinforcement is thought to be mostly cognitive when applied to the behavior of people. Reinforcement is understood to be largely conscious and purposeful when applied to biofeedback.
The operant conditioning explanations of both behavior and biofeedback are incomplete because they lack mechanism information. These explanations are successful in so far as they specify conditions that are sufficient for behavior change to occur. Hence, they are called conditioning explanations. These explanations are incomplete because they lack mechanism information that can explain how and why these conditions are sufficient for behavior and physiological responses to change.
Pennington (2014) asked and answered the question, “What does it mean to explain something from a natural science perspective? Basically, it means that we identify the cause of that thing in terms of relevant mechanisms." Connectionist neural network models provide some of the mechanism information that is missing from operant conditioning explanations.
The figure at left illustrates a simple three-layer connectionist neural network model. The top layer is labeled “S” because it represents sensory neurons that code for stimuli that the network perceives. The middle layer is labeled “O” because it represents one or more layers of neurons that constitute most of the organism’s brain. The bottom layer is labeled “R” because it codes for responses that the network makes. These responses are the bar presses or key pecks that animals make and the physiological responses that people make during biofeedback.
Normal processing occurs when activation cascades from the S to O to R layers. Our figure shows that biofeedback equipment effectively connects neurons in the R layer with neurons in the S layer. Biofeedback equipment temporarily augments the person’s nervous system by providing sensors that detect physiological responses that the person might not otherwise be aware of and displays them in a way that they can easily perceive.
Our figure still does not explain how biofeedback enables the person to better control their biological responses. That is because the synapses that connect pairs of simulated neurons are not included in this figure for simplicity's sake. You can mentally insert these synapses now by imagining that each arrow is broken in the middle and that the two parts of the arrow are separated by a small space. Now imagine that neurotransmitters are released where the break begins and that they disperse across the gap between arrow segments and attach in lock-and-key fashion to receptors that are located where the break ends. Now the arrows represent synapses.
Synapses are crucial to explaining behavior change and are integral to connectionist neural network models because they change during learning and memory formation. This process is called experience-dependent synaptic plasticity (Bear, Connors, & Paradiso, 2007). Connectionist neural network models simulate these changes.
We are now ready to explain how biofeedback works. Biofeedback equipment temporarily augments the person's neural network system by adding sensors to detect physiological events that might otherwise not be perceived and to display these events in ways that are readily perceived and encoded by sensory neurons. Biofeedback enables a processing loop that activates experience-dependent synaptic plasticity mechanisms that modify synaptic excitation/inhibition levels across the relevant neural networks thereby enabling the person to better control their skin temperature, skin conductance, muscle tension and/or brain waves.
One final note: The synaptic modifications that are learned during biofeedback have similar properties to other learned synaptic changes. Continued practice is needed to keep them. Otherwise, they tend to decay.
References
Bear, M. F., Connors, B. W., & Paradiso, M. A. (2007). Neuroscience: Exploring the brain (3rd ed.). Baltimore, MD: Lippincott, Williams & Wilkins.
Black, A. H., Cott, A., & Pavloski, C. R. (1977). The operant learning theory approach to biofeedback training. In G. Schwartz & J. Beatty (Eds.) Biofeedback: theory and research (pp. 89-127). NY: Academic Press.
Frank, D. L., Khorshid, L., Kiffer, J. F., Moravec, C. S., & McKee, M. G. (2010). Biofeedback in medicine: Who, why, and how? Mental Health Family Medicine, 7(2), 85-91.
Pennington, B. F. (2014). Explaining abnormal behavior: A cognitive neuroscience perspective. New York: Guilford.
