The overarching point of this article will be to convey that behavioural psychology may be out-of-fashion, but still has many things to contribute to modern psychology.
All science is ultimately born of philosophy (see Pepper, 1942) and therefore there is no reason why this science should play second fiddle to any other. However, in the rat race to make strides in the science of behaviour, principles of science are often discarded in favour of convenience. The dominant school of thought in psychology at present is cognitivism. This school adopts a predominantly top-down approach to psychology. This involves simplifying phenomena into their perceived component parts, in order to study them. This phenomenon may constitute a set back for the way research is conducted and human behavior ultimately conceptualized. Allow me to illustrate why this is a problem.
oHere is human behaviour, broadly:
Here are emotion and attention, respectively:
The difficulty arises with the categorization of behaviour. This has been recently brought to light with the rejection of the new DSM by the British Psychological Society. One can see here that it would be easy for a psychologist to include or exclude an extra aspect of behaviour when trying to group aspects of cognition as particular phenomena. Those phenomena are themselves cultural interpretations and do not ‘exist’ beyond their usefulness as models for the study of behaviour. However, these illustrations represent a bottom-up view of how distortions of understanding may arise within a top-down tradition. Allow me, once more, to demonstrate the latter example as a scientist with a top-down approach to psychology might see it:
This shows even more vulnerability to error. The intangibility of mental processes means that they are always difficult to define, model, and test. It also means that notions are more difficult to refute because they are not directly measurable. It follows that mentalist approaches to psychology take an ontological leap whereby mental structures are taken to be real things, rather than just being taken as ‘useful’. This, of course, is a fallacy and an error in scientific method.
The alternative is the bottom-up approach. I would argue that in psychology, behaviourism would appear to represent this well. A disregard for the value of the bottom-up building block approach emerged as a radical reaction to disagreements regarding early theories from behaviour analysis. Powerful studies by Watson and Thorndike in the early 20th century made people take notice of the behavioural approach. However, later behaviourists did not significantly develop the field. Pavlov trained animals to, quite reliably, salivate. However, this training did not apply to humans as a functional breakthrough, but provided support for psychosomatics. Even the classic behavioural studies which conditioned fear in Little Hans and Albert, as well as obedience (Milgram) and helplessness (Seligman) were useless beyond learning about those specific types of behaviour. Skinner’s (1957) Verbal Behaviour presented a powerful programme of the basic study of behaviour, as applicable to all contexts. However, its acceptance would not endure. This lead to a widespread conclusion that laboratory experiments could not shed light on understanding, explaining, predicting and controlling behaviour in a way that could be considered applicable in the real world. Hayes, Barnes-Holmes and Roche (2001, pp. xi-xii) put it like this:
“(Verbal Behaviour) shouted ‘victory’!
But something was wrong. The (behavioural) victory did not look like other victories in behaviour analysis. The research that followed was barely a trickle. The applications were relatively few. The new methods, questions, and preparations never appeared. Criticisms were answered with rhetoric, not data. The field of psychology looked, paused, shook its head, and moved on to the direct analysis of human language and cognition. In basic psychology at least, behaviour analysis was elbowed to the sideline.”
Skinner’s approach did not provide data on how novel behaviours occur. His experimental work was based on a strictly action-reaction model. Novel language was not well addressed. In addition, many simply did not like how Skinner described a world of operants as being responsible for language and cognition because it refuted the valued notion of free will (it is worth noting that the concept of free will was also fundamental to some religious beliefs which were societally more influential at the time). Others believed that his trying to isolate behaviours in lesser evolved animals was irrelevant to humans. The issue I would like to address is that behavioural psychology did not, in fact, live and die with Burrhus Frederic Skinner.
Rather than trying to wrestle with notions of ontological truth posed by the hard sciences or adopting a top-down, mechanistic approach as is popularly used nowadays, an alternative is possible. Functional-contextual psychology’s (see Hayes, 1993) truth criterion of a given theory is that it is useful for purposes of what the scientist wants to measure. It does not make an ontological leap in the same fashion as mentalist or mechanistic theories. Theories in this behavioural approach also appeal to the individual’s ongoing context, i.e., a network of learning and reinforcement that also influences future behaviours – embracing biases and other dispositions. In other words, as a philosophy of science, it can offer something different to what the more popular mentalist theories of psychology can offer. The main psychological theory to emerge from this world view is the one with which I consciously chose to pursue research.
Relational Frame Theory (Hayes, 1991)
This theory is a modern, functional-contextual account of human language and cognition (in that cognition entails the same processes as language, but not expressed communicatively). It proposes that the key skill in language is derived relational responding. For example, if X is more than Y, then I can derive that Y is less than X. RFT describes how deriving relations between events (e.g., words, objects) is learned through early social interactions between children and their caregivers, and that this ability is crucial to the development of language and manifestations of behaviour interpreted as being cognitive in nature. Evidence in favour of the importance of derived relational ability has been provided by researchers who have shown that training children in derived relational responding can substantially boost IQ through targeting a generalised account of behaviour (Cassidy, Roche and Hayes, 2011) – a feat which cognitivism has not managed to do. This theory accounts for the generative instances of language and cognition through combinatorial derived relational responding and analogical processes. It has the potential to provide hard data with minimal inference on almost all aspects of language and cognition, including analogy, metaphor and stories, thinking and problem solving, understanding and verbal regulation, rules, psychological development, education, social processes, psychopathology (i.e., in Acceptance Commitment Therapy) and religion, spirituality and transcendence (see Hayes et al., 2001). The concept of behaviour in context based on a primitive and consistent model such as RFT, means that the process of language and cognition can be targeted through manipulating the individual’s environment. In short, this theory comes from what is an empirical and philosophically sound scientific disposition, but also fully addresses the limitations of why behaviourism fell out of favour with psychological scientists in the first place.
RFT may not be the only such behavioural account of language and cognition which builds on previously identified limitations of both behaviourism and cognitivism, but it just the one with which I am most familiar. However, with the cognitive tradition in full flow, it would be easy to jump into research in psychology without giving due consideration to the philosophy of science from which psychology comes, and to assume that because behavioural science is not popular, that it is not adequate. Upon close inspection, it might even be argued that it is the way forward.
Many scientists (but especially aspiring scientists) would argue that science is systematically finding out about how the world is. However, theories and models are simply useful within a context. Data in hard sciences provide evidence of this, but rarely constitute proof of their ontological status (Laudan, 1981). The contextual behavioural approach to psychology (and indeed, medicine) provides a different account that is arguably at least as valid as any popular hard-scientific approach to science. To be a psychologist, one may study the ultimately hypothetical psyche, but to be a behavioural scientist, one must recognise that when it comes down to it, one can only study behaviour. This an old argument, but the point still stands.
Cognitivism is, of course, far from useless. Indeed, it can often be quite practical. On the other hand, it is limited as a science, in principle. To address some of those limitations, the new wave of behaviourism is useful. And it exists.
Cassidy, S., Roche, B., & Hayes, S. C. (2011). A relational frame training intervention to raise intelligence quotients: A pilot study. The Psychological Record, 61(1), 173-198.
Hayes, S. C. (1991). A relational control theory of stimulus equivalence. In Dialogues on Verbal Behaviour, L. J. Hayes & P. N. Chase (eds.), Context Press: Reno, NV, pp. 19-40.
Hayes, S. C. (1993). Analytic goals and the varieties of scientific contextualism. In S. C. Hayes, L. J. Hayes, H. W. Reese, & T. R. Sarbin (Eds.), Varieties of scientific contextualism (pp. 11-27). Reno, NV: Context Press.
Hayes, S. C., Barnes-Holmes, D. & Roche, B. (Eds.). (2001). Relational Frame Theory: A Post-Skinnerian account of human language and cognition. New York: Plenum Press.
Laudan, L. (1981). A confutation of convergent realism. Philosophy of Science, 48(1), 19-49.
Pepper, S. C. (1942). World hypotheses: A study in evidence. Berkeley, CA: University of California Press.
Skinner, B. F. (1957). Verbal behavior. New York: Appleton-Century-Crofts.