Operant Learning & Cognitive Psychology: the Behavioral Science Behind PerfectCoaches
My goal in this blogpost is to discuss the behavioral science behind PerfectCoaches. The ideas have evolved during a career that began as a research assistant in a neuroscience lab, wended its way through a study of Sigmund Freud and years of technology consulting, and brought me where I am today: Marketing a scientifically designed process for improving the performance of people and organizations.
Let’s start in the present. My patent for a User Attribute Analysis System, Patent Number 10,249,212, was published by the U.S. Patent and Trademark Office on April 2, 2019. The patent addresses the underlying process for the PerfectCoaches software, including the module called AMI for Adaptive Motivational Interaction. Using machine learning and artificial intelligence, AMI provides feedback in the form of Socratic-style questions asking PerfectCoaches users about their experience in learning new behaviors. A first implementation, in which AMI serves as a productivity tool for people performing the role of virtual motivation coaching, is in production today.
Behavioral focus and feedback. Now, back to the beginning. From a scientific point of view, PerfectCoaches implements self-managed behavioral modification via a cyclical process involving:
· behavioral focus, and
The method’s use of behavioral focus and feedback derive from theories of operant learning first advocated by John B. Watson and later by B.F. Skinner and other so-called behaviorists who emphasized reinforcement as the foundation of habits. The virtual motivation coach can’t be present to actually reinforce habit formation directly; rather, the feedback from the motivation coach reinforces the user’s mindfulness of the behavior being mastered.
Operant learning is a powerful tool. It enables us to "operate" on our environment to generate consequences. I first learned about it as a research assistant in the neuroscience laboratory of what was then called Friends of Scientific Research. There, as a 17 year old college freshman, I began training Rhesus monkeys in a series of learning experiments. Raisins served as reinforcers (rewards), and the box-like apparatus and methods used were classically Skinnerian. For a high level introduction to B.F. Skinner’s view of learning, which influenced me greatly, visit https://www.simplypsychology.org/operant-conditioning.html
Self Awareness. While studying group dynamics in graduate school, I concluded that a practical deficiency of orthodox, stimulus-response behaviorism as an approach to human habit formation is that it fails to account for motivation. Whereas a Rhesus monkey in a testing apparatus may be “motivated” by raisins, the picture for humans interacting with each other is more complex. There has to be a role for cognitive psychology, i.e., “the scientific study of the minds as an information processor” (see https://www.simplypsychology.org/cognitive.html). For me self-awareness, a cognitive activity harnessing the power of personal motivation, filled the gap. In a "self-awareness snapshot" users ask themselves seven questions including: Who am I? What do I do? Where do I want to go? and How do I change?
My book Life’s Seven PerfectCoaches describes a thought experiment that poses these fundamental questions about life and also invites users to imagine coaches asking the questions. Not only that, users are asked to draw a mental picture of “perfect coaches,” perfectly patient, disciplined, and accepting of every person using the method. The Coaches are the questions and the questions are the Coaches.
Self Awareness and the Human Connection. The sociologist Charles H. Cooley compared self-awareness to a "looking glass," where we see ourselves as we believe others see us. Our sense of who we are and, importantly, who we should be, arises from connection to others. Sigmund Freud explored this in depth through the concept of identification, i.e., the process where a child wants to become like its mother or father. Not only were Freud's writings sometimes controversial, it was difficult to discern how the variables he discussed could be measured in a scientifically useful way. For my PhD dissertation I addressed this problem, using multi-variate regression analysis to measure and validate key elements of Freud’s development theory.
My dissertation concluded that “To the extent that one feels comfortable with these operational definitions, one may be willing to conclude at least tentatively from this study that parental status, nurturance, power, and similarity-to-self rank in that order as determinants of identificatory motive.” In other words, the capacity of a person to identify with, i.e., to take as a model, another person is a function primarily of that other person’s capacity to nurture and exert power.
On to the possibilities of system animism. Drawing on the knowledge of computing I acquired in graduate school—the University of Maryland actually let me use the Fortran programming language to satisfy my foreign language requirement—my career veered away from academic social psychology. I eventually entered the world of technology and management consulting where I continued to refine my understanding of how individual behavior is reinforced in natural settings. I wrote my college textbook book Office Automation: Tools and Methods for System Building (John Wiley, 1985) in order to place human information processing and computer-based information processing in a single paradigm depicting modern organizations as systems in which humans and computers exchanged information in order to achieve business goals.
The book’s concluding discussion of the future of information technology is quoted at length in the attached excerpt. Of particular importance is the idea that computer systems “are becoming more animated, in the original Latin sense of animalis as “living, animate, from anima for breath, air, soul.” I illustrated this point by citing fictional accounts of humans interacting with machines as if the machine itself was human.
This brings us to the line of intellectual inquiry that will complete my journey. Given my interest in both human social interaction and machine computing, I have always been fascinated by the so-called Turing test. It’s a 1950s era concept in which the British scientist Alan Turing speculated about “a test of a machine's ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human.” The “test” is a gateway into the fascinating array of philosophical and scientific issues raised by what is now known as artificial intelligence (see https://en.wikipedia.org/wiki/Turing_test).
Could AMI, our Adaptive Motivational Interaction module, ever pass the Turing Test? The current version certainly can’t. However, AMI only asks questions, sustaining a Socratic-style dialog recommended as a modern approach to counselling; this makes it easier to model AMI's linguistic style. More importantly, AMI learns. It “remembers” every interaction between a user and the human virtual motivation coaches. Not only that, AMI’s heuristic algorithm, written by me in pseudocode, will continuously improve its ability to exhibit the “status, nurturance, power, and similarity-to-self” my dissertation claimed were key to Sigmund Freud’s theory of identification.
Perhaps AMI could even be like the fictional electronic grandmother described in my book, “assisting the people around her and helping them to realize their full human potential.” But to do so with maximum effect it would have to pass the Turing Test. So my intellectual journey ends with a question: Will AMI ever pass that test, even when interacting with people via wrist watches and Augmented Reality glasses as envisioned in U.S. Patent Number 10,249,212? Perhaps, perhaps not. Check out our AMI concept video, meet AMI@Perfectcoaches, and give it some thought. I think about it every day.
—Dr. V. Douglas Hines, creator of PerfectCoaches