What Psychology as a Science Owes Neal Miller: The Example of His Biofeedback Research

Abstract

Neal Miller did more to make psychology a science than any other investigator. His importance does not lie with any specific discoveries that he made, but rather with his way of doing scientific research, which involved pursuing a line of logic systematically through sequences of experiments, and paying attention to several alternate hypotheses that could answer each of the experimental questions addressed. His approach was a model of what has been called ''Strong Inference'' and that is characteristically used in the hard sciences. His biofeedback research is used as a case history of his method of approach.Neal Miller was a great scientific thinker. He was arguably the most important scientific thinker in psychology of the 20th century. He did more to make psychology a science than any other investigator. The only other person who is in contention for this honor is B. F. Skinner. Important as Skinner was, he was not as important as Miller for reasons that I will outline below. Neal Miller is of greatest significance not for any specific discovery he made, but rather for his way of doing scientific research. This is less obvious than the contribution of specific discoveries; but because it changes our way of doing science, it usually makes a more long-lasting and general contribution than any specific discovery.Neal Miller worked in many areas that were quite different from one another. The full range of these areas are described in large part by a companion piece to this article, appearing in this issue of Biofeedback by Edgar E. Coons and Sarah Leibowitz. These include human factors in World War II aviation, learning and behavior, an experimental analysis of psychoanalytic theory in behavioral terms, hypothalamic mechanisms of different drives such as hunger and thirst, central nervous systems of reward and intracranial self-stimulation, then biofeedback, and finally applications of biofeedback to ameliorate human pathology. He not only worked in these areas, but he also did definitive work in every one of them. Every time out he and his students rang the bell. This is a remarkable record. After all, the areas he worked in were widely divergent. What was the common thread that permitted him to do this? Neal Miller had great clarity of thought. His analyses almost always involve lines of logic that are transparent, simple, and pellucidly clear. However, this by itself is not the reason for his wide-ranging success; other scientific psychologists also had this. Neither does the element underlying Miller's success have to do with a fortunate choice of subject matter; Miller worked in many divergent areas that numerous other investigators worked in. Moreover, the different areas do not really have a commonality. The connection is in Neal Miller's extremely effective method of approaching and solving scientific problems. He is the prime example in psychology of a scientist who invariably and systematically used a method of doing research that has been called Strong Inference. Method of Strong InferenceStrong Infere...

Turning negative into positives! Exploiting ‘negative’ results in Brain–Machine Interface (BMI) research

“…These include, among others, the choice of the experimental conditions to be tested, the types of subjects that are included in the study and the metrics used to assess performance. For sake of space, we briefly discuss some general recommendations (relevant material is found in previous publications on this topic [47,48,[51][52][53]).…”

“…The approach of inductive inference goes back to Francis Bacon [63] (cf. [53,62]). According to Platt, four steps are crucial for strong inference.…”

“…Taub describes three additional steps that were formulated by T.C. Chamberlin [53,64]; Table 1 summarizes those steps and lists an example from the BMI field.…”

“…It is worth thinking of as many alternative hypotheses as possible which reflects the whole logical structure of a problem. The deducted sequence of experiments allows for a successive approximation (p. 111, [53]) to the phenomenon and guides interpretation of results. Those steps are familiar to all of us, but the difference arises from their Table 1.…”

“…Those steps are familiar to all of us, but the difference arises from their Table 1. The seven/four steps constituting strong inference according to [53] and [62] Step Content Example for BMI systematic application. Applying this method more often and stringently in the field of BMI would increase the quality of studies and facilitate their subsequent publication independently of the results being positive or negative.…”