In the 1960s, a major question emerging from the fields of psychophysiology and learning theory was whether operant procedures could directly change autonomically mediated responses. Miller and DiCara (1967) researched this question using curarized animals and demonstrated that operant procedures could produce bidirectional changes in heart rate, blood pressure, and glandular activity. These findings challenged the view that smooth muscle learning was possible only through classical conditioning. However, satisfactory replication of this research has yet to occur. Thus, although numerous studies have shown that operant procedures can produce autonomic changes in noncurarized animals and humans (Bower & Hilgard, 1981), none have really employed adequate controls for skeletal muscle and central nervous system mediation.
Although specification of the mechanisms of visceral learning is of theoretical importance, it seems less so in clinical applications, because skeletal and visceral responses are inextricably linked as part of the centrally integrated response patterns (Miller, 1978). The colorectal system provides an example in which the autonomic nervous system and the somatic branch interact to maintain bowel function. This integration provides the opportunity for the use of operant procedures to alter disordered bowel function.
Reviews of the literature report that, overall, biofeedback appears to be highly effective in reducing fecal incontinence (Enck, 1993; Norton, Hosker, & Brazzelli, 2000; Heymen, Jones, Ringel, Scarlett, & Whitehead, 2001). However, comparisons among studies are difficult, owing to small sample sizes and inconsistencies with regard to methodology, selection criteria, and outcome assessment. Improvements in bowel control after biofeedback are reported to range between 64% and 70%, depending upon the methodology used. However, in light of the methodological shortcomings, these outcomes must be interpreted with caution.
The use of operant procedures to improve anorectal physiology was first reported in a single-case pediatric study (Kohlenberg, 1973). However, the seminal study of Engel, Nikoomanesh, and Schuster (1974) defined the method that was replicated by many subsequent reports. Engel et al. (1974) used a manometric three-balloon probe (described below), which simultaneously measures and allows for the reinforcement of three specific anorectal responses