A Shocking History: Electric-Shock Punishment as Behavioral Treatment

Andy Lattal
West Virginia University
Cambridge Center for Behavioral Studies, Member, Board of Directors

The use of electric shock delivered to the skin as a means of inducing behavior changes in humans undergoing behavioral treatments has not only a long history but also has been very much in recent behavior-analytic news. In 2021, the Association for Behavior Analysis International (ABAI) assembled a task force to consider what position the organization should take on the question of electric shock so used. After reviewing all aspects of the issue, including ABAI’s scientific and humanitarian values, the task force’s conclusion was as follows:

“In accordance with these values, we strongly oppose the use of contingent electric skin shock (CESS) except in the most extraordinary of circumstances.“

This was a conclusion with which some members of the task force and some members of the ABAI executive council did not agree. As a result, the question was put to the voting members of ABAI. The outcome was that the following was selected to be the official position of ABAI by the majority of the members who voted:

“In accordance with these values, we strongly oppose the use of contingent electric skin shock (CESS) under any condition.”

The present author added the italics to highlight the difference between the two statements.

Behind this controversy are historical questions about how CESS entered the behavior-analytic treatment armamentarium and the technology that led to its adoption.

In the laboratory, electric shock was used in Pavlovian conditioning in the early 20th century. Skinner (1938) used only a forceful bar slap as a punisher in the research described in Behavior of Organisms, but in 1941, Estes and Skinner published the first paper on what they called “conditioned anxiety,” a phenomenon later described as conditioned suppression. As an animal’s responding was maintained by some schedule of reinforcement, a different stimulus was introduced for a short time and followed by a brief, response-independent electric shock when that stimulus terminates. Over repeated introductions, responding during the distinct stimulus was suppressed relative to that occurring in its absence. This, of course, is not a punishment procedure because the shock is delivered independently of responding. Nonetheless, it was one of the earliest examples of using electric shock in behavior-analytic research. As punishment research and technology accumulated (Azrin & Holz, 1966, provided an outstanding scientific review of the topic), its translation into treatment seemed inevitable.

The interdependence of science application and technology is a frequent topic in the history of science (e.g., Lattal, 2008). Skinner’s operant conditioning and the use of CESS, for example, were enabled by the technology of electricity and electronics. Of course, it took others trained in behavior analysis to harness the technology for use in basic and applied behavior analysis.

Ivar Lovaas was among the first to use CESS as a behavior-analytically grounded treatment (e.g., Lovaas et al., 1965). Lovaas’s work was controversial from its inception, and the use of CESS as treatment became even more controversial with the appearance of the self-injurious behavior inhibiting system (SIBIS ). Although the latter reduced self-injury, it became a catalyst for critics of the use of CESS and, for many others, behavior-analytic treatment of any sort, whether involving reinforcement or punishment.

An early version of the SIBIS device was developed at the Johns Hopkins University Applied Physics Laboratory (JHUAPL) and described by Friauf (1973). The device was housed in a lightweight helmet, shown in Figure 1, worn by the person whose behavior needed to change. It could deliver an 80-millisecond electric shock automatically to electrodes, placed on the person’s back, following a sudden arm movement toward the head. An accelerometer incorporated into the device detected the movement and translated it into a shock. A later version, also developed at JHUAPL, was patented by Fischell, Newman, Riblet, and Powell, and described by Newman (1984; see also Jones, 1985). This device delivered a 100-200 millisecond shock to electrodes placed on the arm. It was Arnold L. Newman’s first-listed patent (Arnold L. Newman Inventions, Patents and Patent Applications – Justia Patents Search), seemingly launching a well-regarded series of patents of more than 25 such medical devices. Linscheid (1993) noted that the funding for both systems came in part from a foundation established by the parents of a young girl who suffered from self-injurious behavior.

Figure 1. The “aversive stimulator” described by Friauf (1973).

Linscheid et al. (1990) presented several case studies evaluating the effectiveness of a modified version of the device described by Newman (1984) in controlling self-injurious behavior, along with caveats and cautions about its use. The SIBIS device and other forms of CESS, however,  continued to be debated, both inside (e.g., Iwata, 1988) and outside of behavior analysis. One result of the debate was the ABAI position statement quoted above. Another was the attempted banning by the Federal Drug Administration in 2020 of electric shock delivery as part of any behavioral treatment program. As of this writing (April, 2025), the proposed ban, having gone through both legislative and legal reviews and challenges, was expected to receive a final ruling in the autumn of 2025 (https://www.disabilityscoop.com/2024/12/09/fda-under-pressure-to-finalize-ban-on-shock-devices-for-people-with-developmental-disabilities/31198/).    

Behavior-analytic treatments have from their inception focused on relieving human suffering. Reasonable people have disagreed as to what that focus means and how it should be achieved. Technologies acceptable to behavior analysts, caregivers, and the general public change over time as scientific and general cultural variables and values change. Well-intended technologies fade into the mists of time for good and bad reasons as others emerge. The hope of everyone is that scientifically and societally acceptable and effective behavior-analytic treatments emerge that can help those individuals afflicted by the kinds of behavior CESS was designed to ameliorate. Behavior-analytic science must seek a better understanding of how behavior is determined, wherever such a search takes it. At the same time, translating that research into procedures intended to help others requires sensitivity to cultural values of their communities so that our science can optimize its desired impact.

Figure 2. Diagram of the implementation of the device described by Newman (1984).

References