Up-dating RFT: Cooperation Came First, the ROE as a Unit of Analysis, and Engineering Prosocial Behavior

In this month’s blog I am thrilled to have RFT father Dermot Barnes-Holmes talking RFT and prosocial behavior in the time of a pandemic. Dermot was my PhD supervisor many years ago and I am very thankful for all his wisdom and guidance over the years. Joining Dermot is a researcher in the area of behavior analysis and developmental psychology Maithri Silvaraman.

Thank you Dermot and Maithri for joining the ABAI Symbolic Language and Thought Blog series. -Prof. Louise McHugh, University College Dublin, ABAI Symbolic Language and Thought Blog series host 2020.

Bio: Prof. Dermot Barnes-Holmes received his D.Phil. in behavioral analysis and behavioral biology from the University of Ulster, Coleraine, N. Ireland. He currently serves as Senior Full Professor and Odysseus Laureate at Ghent University, Belgium, having previously served on the faculties of the National University of Ireland, Maynooth, and University College Cork. Dr. Barnes-Holmes is an extraordinarily prolific researcher who has made extensive contributions to the behavior analytic literature, especially in the areas of language and cognition. The competitive and prestigious Odysseus Laureate awarded to Dr. Barnes-Holmes in 2015 is just the most recent recognition of the esteem in which his work is held among behavioral scientists internationally.

Bio: Maithri Sivaraman is a BCBA with a Masters in Psychology from the University of Madras and holds a Graduate Certificate in ABA from the University of North Texas. She is currently a doctoral student in Psychology at Ghent University, Belgium. Prior to this position, Maithri provided behavior analytic services to children with autism and other developmental disabilities in Chennai, India. She is the recipient of a dissemination grant from the Behavior Analysis Certification Board (BACB) to train caregivers in function-based assessments and intervention for problem behavior in India. She has presented papers at international conferences, published articles in peer-reviewed journals and has authored a column for the ‘Autism Network’, India’s quarterly autism journal. She is the International Dissemination Coordinator of the Association for Science in Autism Treatment (ASAT) and a member of the Distinguished Scholars Group of the Cambridge Center for Behavioral Studies.

In this blog we will reflect upon how an up-dated version of relational frame theory (RFT) views the evolution of human language and cognition. Specifically, it has been argued that the human propensity for cooperation drove the evolution of the basic, functional-analytic units of arbitrarily applicable relational responding (AARR); these units may be conceptualized as creating dynamical, non-linear patterns of orienting, evoking, and relating (or ROEing), which are involved in the vast majority of human psychological events. The blog will also suggest that connecting (an up-dated) RFT to a multi-level evolution science may be useful in seeking to encourage and engender prosocial behaviors at a period in human history when doing so seems to be essential to our on-going survival as a species.

Human Evolution, Climate Change, and Plagues of Global Warming

For approximately 2 million years, humans lived as hunter-gatherers; only 10,000 years ago, during the Neolithic revolution, did people first begin to discover how to cultivate crops and to domesticate animals. In doing so, one might argue that a subtle shift occurred from a human sense or perception of being owned by the land (and waters) through which we roamed, to one in which we owned and controlled the land and animals upon which our survival depended. The change in human behavior, from hunter-gatherer to “farmer,” was a fundamental one in the history of our species, but it had a limited impact on the ecosystem of the planet we call earth and home. A far more dramatic impact emerged, of course, during the industrial revolution, which commenced approximately 240 years ago.

A likely assumption is that that this revolution marked the beginning of a sudden and sharp rise in the release of carbon into the atmosphere of our planet and with it most if not all of the repercussions for climate change. In fact, there may even be a sense that the worst of the damage was done earlier in the industrial revolution, when the poet William Blake wrote of England’s “dark satanic mills” as a metaphor for the industrial scarring of the nation’s “green and pleasant” land. Alas, both assumptions are ill-founded. First, the “mills” to which Blake referred were the orthodox churches of the establishment. Second, and perhaps most worryingly, over half of the carbon pumped into the atmosphere by the burning of fossil fuels has been released in just the past three decades. In effect, we have inflicted as much damage to mother earth, and its ability to sustain human life and civilization, since Al Gore first began working on climate change (emerging 14 years ago as his “inconvenient truth”), than in the centuries and indeed millennia that preceded his unwelcome message. We have, one could argue, walked into climate change with our eyes wide shut. Having being warned of a perilous cliff edge ahead of us, collectively we put the “peddle to the metal” and have driven with increasing speed towards our own doom. Clearly, information alone is not enough to bring about the behavioral change that is required to reduce, much less reverse, recent levels of carbon emissions.

If information about climate change has had little impact on our behavior, then what is the alternative? A behavioral science (e.g., Hayes, Barnes-Holmes, & Wilson, 2012) solution involves embracing a paradox, and the paradox is this. The human species, like all forms of life on earth, are the products of evolution, and as such the current threat of climate change to our survival as a species could be seen as a natural product of evolution itself. If that is the case, then perhaps we, as a species, should turn to evolution to find a solution to the problem of climate change. In doing so, however, it will be important to conceive of evolution, as not simply impacting upon genetic change. Instead, a modern evolutionary science will need to be embraced, which involves focusing on the dimensions of variation, selection, and retention across multiple levels, including the symbolic and cultural levels that characterize the human species in particular (Wilson & Wilson, 2007). Perhaps it is only in harnessing the processes of human evolution, when conceived of in multi-level and multi-dimensional terms, will our collective foot be drawn back from the peddle that is speeding us towards the point of no return. Or in the words of Atkins, Wilson, and Hayes (2019);

“Multilevel selection is like a tug-of-war, with within-group selection pulling toward self-serving traits and between-group selection pulling toward group-serving traits. This is why we see a mix of prosocial and disruptive self-serving behaviors in human life and why – with sufficient know-how – we can tilt the playing field in the direction of prosocial behaviors.” (p. 13).

At this point, it seems important to emphasize that the recent corona virus pandemic has only served to underscore the urgent need to address the broader issue of climate change and sustainability. Indeed, it would be a profound error to view the current pandemic as unrelated to the broader issue of climate change. For example, in a recent volume on this issue, The Uninhabitable Earth, Wallace-Wells (2019) entitled a chapter, “Plagues of Warming.” The link between climate change and such plagues was cogently summarized in the following passage:

“. . .global warming will scramble [our] ecosystems, meaning it will help disease trespass those limits as surely as Cortés did. The footprint of every mosquito-borne illness is presently circumscribed, but those borders are disappearing rapidly, as the tropics expand—the current rate is thirty miles per decade. In Brazil, for generations, yellow fever sat in the Amazon basin, where the Haemagogus and Sabethes mosquitoes thrived, making the disease a concern for those who lived, worked, or traveled deep into the jungle, but only for them; in 2016, it left the Amazon, as more and more mosquitoes fanned out of the rain forest; and by 2017 it had reached the country’s megalopolises, São Paulo and Rio de Janeiro—more than thirty million people, many of them living in shantytowns, facing the arrival of a disease that kills between 3 and 8 percent of those infected.” (p. 111).

The “Plagues of Warming” chapter considers numerous other ways in which climate change will lead to the spreading and release of a wide range of potentially lethal diseases throughout the world; and perhaps somewhat prophetically, in the context of the current Covid-19 pandemic, the next chapter in the book is entitled “Economic Collapse.”

Evolutionary Science, Human Cooperation, and an up-dated RFT

Wilson (2007) summarized human evolution as the “three C’s”: cognition, culture, and cooperation. While all three of these were considered in early renditions of RFT, it appears that cooperation was somewhat underplayed, if not largely ignored. According to the first book-length treatment of RFT, Hayes, Barnes-Holmes, and Roche (2001) suggested that mutual entailment (the bi-directional relational responding that may occur between two stimuli) in a listener could enhance or support avoidance of predators even if entailment was not yet present as part of a vocal or speaking repertoire. In addition, it was argued that this small difference could generate a group of listeners who could then reinforce mutually entailed responses in a speaking/vocal repertoire. Upon reflection, this account relies heavily on the evolution of mutual entailment as an adaptation of cognition in listening responses, and then spreads to speaking or vocal responses, thereby leading to increased social cooperation throughout the wider group or culture. In contrast to this account, Hayes and Sanford (2014) argued that it is more evolutionarily viable to assume that human cooperation was the primary driver in the evolution of mutual entailing, rather than the other way around. Indeed, as Hayes and Sanford point out, there is a wealth of empirical data supporting the argument that human cooperation was established by multilevel selection of cooperation itself, because it provided advantages for human group competition, which occurred alongside the cultural suppression of individual selfishness.

From an up-dated RFT perspective, a critical feature of human cooperation involves pointing and grunting, for example, which provided humans with highly important behavioral skills, which have been labelled as “joint attention,” “social referencing,” and perspective-taking.” It has been suggested that these skills increase the likelihood that more advanced forms of cooperation, involving the emission of specific vocal sounds, will be selected or reinforced, as is the case with young children. For example, if a young child orients to a care-giver and then towards a toy while emitting a nonspecific vocal sound (e.g., “eh”), and tries to reach for the toy, the care-giver may reinforce this cooperative act by giving the toy to the child. In the words of Hayes and Sanford (2014): “The entire exchange will build cooperation, perspective taking, and joint attention as patterns that are maintained within the group because it is a functionally useful communication exchange. If we unpack this highly likely sequence it means that in the context of high levels of cooperation, and adequate skills in joint attention, social referencing, and perspective taking, any characteristic vocalization in the presence of a desired object would likely lead to reinforced instances of symmetry or mutual entailment” (p. 122).

According to an up-dated version of RFT, which focuses on cooperation as a key driver of derived relational responding itself, the critical behavioral history does not begin with speaking or even simply listening. Rather it begins with orienting, or more precisely bi-directional orienting as a child orients back and forth between a care-giver and an object or stimulus that the care-giver is oriented towards. In an up-dated version of RFT this behavioral pattern may be referred to as mutually entailed orienting, which we suggest is one of the core elements of the generic unit of arbitrarily applicable relational responding (AARR). We make this claim because the behavior is deemed to be part and parcel of the evolution of a level and type of cooperation that is unique to the human species. In this sense, AARR (or the symbolic inheritance stream) does not begin with listening and speaking, it starts with one of the most basic of human cooperative acts (i.e., mutually entailed orienting).

The critical importance of mutually entailed orienting cannot be underestimated because it allows care-givers to establish appetitive and aversive evoking functions for stimuli in the child’s environment. Once a care-giver and an infant are engaging in mutually entailed orienting, the care-giver can now orient the child towards a particular stimulus and encourage the child to approach “safe” and avoid “dangerous” stimuli. Mutually entailed orienting may thus also involve establishing specific orienting and evoking functions for particular stimuli. For example, if a caregiver shouts loudly when the child approaches a dangerous stimulus (e.g., an insect with a powerful venom) that stimulus will likely acquire strong orienting and (aversive) evoking properties for the child. Furthermore, when an infant engages in mutually entailed orienting, even items that are simply oriented towards by the caregiver, without issuing any sort of warning signal, may acquire relatively positive evoking (approach) functions for the infant. Mutually entailed orienting is thus more accurately labelled mutually entailed orienting and evoking. As a listening repertoire develops, mutually entailing orienting and evoking functions for particular stimuli become related, in an arbitrarily applicable manner, to specific sounds (i.e., words). Gradually, therefore, a new response unit involving relating, orienting, and evoking is established for the child. In an up-dated version of RFT this response unit is labelled the ROE (pronounced “row”). The reader should note that the term mutually entailed (orienting and evoking) is employed to denote this type of human infant learning because it typically occurs in parallel with establishing a basic listener repertoire (e.g., a care-giver rarely engages a child in mutually entailed orienting and evoking without also emitting language-appropriate sounds, such as “Look, it’s teddy”, when orienting the child towards a toy teddy-bear). Employing the term mutual entailment (for orienting and evoking) also serves to highlight that this type of learning is viewed, by an up-dated version of RFT, as emerging from an evolutionary history of cooperation that appears to be unique to the human species.

Mutually entailed orienting/evoking are not simply new terms for joint attention and social referencing. The latter are seen, at best, as precursors for AARR, and thus have no “technical weight” within RFT itself. By introducing these new concepts (mutually entailed orienting/evoking), an up-dated version of RFT seeks to establish explanatory depth. Specifically, it links the behavioral account of human language and cognition more directly to the evolution science argument that human cooperation drove, at least initially, the evolution of human language and cognition itself.

In addition, the term mutually entailed orienting/evoking aims to establish a functional-analytic-abstractive quality to the conceptual analysis of the behavioral topographies usually associated with the terms, joint attention and social referencing (and perspective-taking more generally). To appreciate the point being made here, imagine you trained a dog to fetch an object and bring it to you by pointing at it, or even simply gazing at it, and shouting “fetch”. One might argue that this interaction was clearly cooperative and involved at least some element of joint attention (and perhaps even social referencing), because the dog and you both needed to attend to the same object for the dog to fetch it for you. According to an up-dated version of RFT, however, this interaction would not be defined as mutually entailed orienting, for the dog, unless it was functioning as part of a phylogenetic and ontogenetic history of AARR for that animal. Or to put it another way, if the cooperative interaction is part of an evolutionary history that leads to the establishment of AARR (and ROEing) for the dog, then the dog could be considered as engaging in mutually entailed orienting; if there is little or no evidence of AARR in the dog’s behavioral repertoire in the past (as a species) or in the future as an individual organism, then the term mutually entailed orienting/evoking should not be applied to the dog’s behavior in this example of joint attention (or social referencing).

In contrast to dogs, and other non-human species, mutually entailed orienting, evoking, and relating are typically established through ongoing interactions between the human child and its caregivers. According to an up-dated version of RFT, extended cooperation further facilitates the adaptation of the species, by allowing for more complex adaptations of the functional units, such as combinatorial entailment. This increasing complexity in derived relational responding involves the use of symbols and the ability to problem-solve in the natural and social environment. According to this up-dated version of RFT, therefore, cooperation facilitates more useful forms of cognition, rather than cognition producing more useful forms of cooperation, although it is important to appreciate that the relationship is likely non-linear and dynamical (i.e., cooperation generates increasingly advanced cognition, which in turn feeds back into generating increasingly complex forms of cooperation).

Once the generic response unit of AARR (i.e., the ROE) is established, it allows for the evolution of increasingly complex relational responding inside the ROE, such as relational networking, the relating of relations (e.g., analogy and metaphor), and the relating of entire relational networks to other relational networks (e.g., extracting common themes from different narratives). An up-dated version of RFT has proposed a new multi-level framework for conceptualizing this increasing complexity in relational responding in terms of five levels of relational development; (i) mutually entailing, (ii) combinatorial entailing, (iii) relational networking, (iv) relating relations, and (v) relating relational networks. Each level of the framework intersects which four dimensions; (i) coherence, (ii) complexity, (iii), derivation, and (v) flexibility, thus yielding 20 units of analysis.

As an illustrative example, a mutually entailed relation (e.g., “hornets are dangerous”) may be conceptualized as varying in coherence, complexity, derivation, and flexibility. In very general terms, the relation between hornets and danger may be relatively high in coherence if the statement coheres with similar assertions (e.g., “a small number of hornet stings can kill”); relatively low in complexity if understanding the statement involves a limited number of other relational responses (e.g., the words “hornet” and “dangerous” are directly related to the referents hornet and danger); relatively low in derivation (e.g., if similar statements have been heard many times in the past); and low in flexibility (e.g., if it is difficult to modify or “challenge” the perceived truth of the statement). Critically, this relational activity is seen to interact in a non-linear and dynamic manner with the orienting and evoking functions of stimulating events for humans as they navigate their environments. For example, the statement (“hornets are dangerous”) may increase orienting and (aversive) evoking functions for hornets if the statement is uttered just before entering an area where they are commonly found. This up-dated RFT framework for conceptualizing the dynamical interplay among relating, orienting, and evoking (i.e., ROEing) has been defined as hyper-dimensional and multi-level (i.e., the HDML framework; see Barnes-Holmes, 2018; Barnes-Holmes, Barnes-Holmes, & McEnteggart, 2020, for more detailed treatments).

From an up-dated RFT perspective, the set of relational abilities, and associated orienting and evoking functions contained within the ROE, evolved into complex forms of communication and problem-solving in only a few thousand years. Indeed, as argued above, the ability to engage in ROEing appears to be a defining characteristic of the human species, and allows us to predict and influence our environment in increasingly sophisticated and powerful ways. From this perspective, once ROEing evolves, the natural environment becomes thick and rich with stimuli that are symbolic, rather than direct-acting, as they appear to be for non-human species. For example, symbolic stimuli can be used to form new meanings and to construct new realities detached from direct experience (e.g., fiction, poetry, metaphor). As such, the transmission of behaviors, from one individual to another and from one generation to the next, is increased dramatically. This ultimately leads to greater variation in behavior and the potential for the acquisition of new behaviors that serve to increase survival at multiple levels — individuals, groups, and species.

The Downside of Cooperation and What Can We Do About it?

The complexity, richness, and sheer cognitive power that the evolution of ROEing has involved allowed a relatively weak and slow moving primate to cooperate in the building of cities, cultures, and entire civilizations with all of the art, science, and technology that many of us enjoy today. But of course, the same behavioral process of ROEing has been involved in humans acquiring the ability to cooperate in waging wars on a global scale, creating weapons of mass destruction, and impacting upon the planet’s ecosystems to the extent that we have now extinguished many other species and even threaten, quite seriously, our own survival. Metaphorically, it seems as if planet earth is beginning to wage war on us, and unless there is a dramatic and sustained change in our behavior, at all levels, we are destined to lose that war, possibly within the next two to three hundred years.

How might we respond to this very serious challenge to our survival? One behavioral science answer to this question involves drawing on the very unique human characteristics that generated the current crisis in the first place; the highly evolved human capacity for cooperation and the ability to engage in ROEing as a result of that driving force of cooperation itself. A critically important part of this behavioral science attempt to promote the types of prosocial behaviors that will be required to ensure sustainability at a global scale involves drawing on the collaborative work of Elinor Ostrom (winner of a Nobel Prize for economics) and David Sloan Wilson (who has been at the forefront of developing a multi-level, multi-dimensional view of evolution processes). Specifically, with their colleagues they generalized a set of core design principles (CDPs) that was originally identified by Ostrom for human groups who were managing common-pool resources (CPRs), such as forests and fisheries. Ostrom had proposed a broad set of institutional arrangements that succeeded in managing these CPRs against those arrangements that did not. Critically, Ostrom, Sloan-Wilson, and colleagues showed that these principles may be generalized in terms of foundational evolutionary principles and perhaps most importantly, from a behavioral science perspective, employed as a guide for improving the prosocial behaviors of a wide range of groups at multiple levels (Wilson, Ostrom, & Cox, 2013).

Very briefly, it appears that human groups engage in successful cooperation in the context of specific clusters of complex relational networks (i.e., CDPs). When each of these CDPs influence the behavior of most if not all of the members of a group, individuals within the group tend to act in the interests of the group rather than the selfish interests of an individual. Bringing these CDPs to bear on human groups, at all levels — families, villages, towns, cities, states, nations, continents, and ultimately at a global scale — offers the human species a way in which to harness its capacity for highly complex forms of cooperation to ensure the ongoing and sustainable survival of the species. In a sense, we need to create a new symbolic and cultural inheritance stream – one that selects for cooperation at all levels of human organization on the planet.

Of course, this is far more easily said than done. However, it appears that the outline of a scientifically-based roadmap on how to begin this work is emerging. It is a roadmap derived from a modern, multi-level, multi-dimensional view of evolution science and a multi-level, multi-dimensional view of the evolution of human language and cognition within that broader evolution science framework. It is a framework that seeks to explain how the evolution of human cooperation drove the evolution of increasingly complex patterns of human language and cognition, which were both successful in helping the human species survive and thrive, and also bring it to brink of its own self-destruction. But most importantly of all, it is a framework that seeks to provide scientifically-based principles, including behavioral principles, for managing the human capacity for cooperation in a manner that steers our species away from the rapidly approaching cliff edge towards which we appear to be hurtling.


Atkins, P. W. B., Wilson, D. S., & Hayes, S. C. (2019). Prosocial: Using evolutionary science to build productive, equitable, and collaborative groups. Oakland, CA: New Harbinger.

Barnes-Holmes, D. (2018). The double edged sword of human language and cognition: Shall we be Olympians or fallen angels? [Blog post]. Retrieved from https://science.abainternational.org/the-doubleedged-sword-of-human-language-and-cognition-shall-we-beolympians-or-fallen-angels/rrehfeldtabainternational-org/

Barnes-Holmes, D., Barnes-Holmes, Y., & McEnteggart, C. (2020). Updating RFT (More field than frame) and its implications for process-based therapy. The Psychological Record. https://doi.org/10.1007/s40732-019-00372-3

Hayes, S. C., Barnes-Holmes, D., & Roche, B. (2001). Relational frame theory: A post-

Skinnerian account of human language and cognition. New York, NY: Plenum.

Hayes, S. C., Barnes-Holmes, D., & Wilson, K. G. (2012). Contextual behavioral science: Creating a science more adequate to the challenge of the human condition. Journal of Contextual Behavioral Science, 1(1-2), 1-16. doi:10.1016/j.jcbs.2012.09.004

Hayes, S. C., & Sanford, B. T. (2014). Cooperation came first: Evolution and human cognition. Journal of the Experimental Analysis of Behavior, 101(1), 112–129.

Wallace-Wells, D. (2019). The uninhabitable earth: A story of the future. London, UK: Penguin.

Wilson, D. S. (2007). Evolution for everyone. How Darwin’s theory can change the way we  think about our lives. New York: Delacorte.

Wilson, D, S., Ostrom, E., & Cox, M. E. (2013). Generalizing the core design principles for the efficacy of groups. Journal of Economic Behavior & Organization, 90, 21-32.