Could We Have Identified the Basic Building Blocks of Intelligence?
Much research has now shown that intelligence levels are malleable and this is covered in this blog (IQ bootcamp) and elsewhere. It has also become the trumpet call of just about every brain health and brain training company on the internet, so it is hardly a novel idea at this stage, even if it remains controversial.
What is less well known is that behavioral researchers have hit upon what they believe to be the basic building blocks of intelligence. These are a set of skills, known as relational framing skills (or relational skills for short), that appear to function as behavioral precursors for a wide range of intellectual skills. While brain training software aims to stimulate neural connections in key locations associated with intellectual activity, a relational skills training approach simply deconstructs the intellectual activity into its component skills – and trains those instead.
Behavior analysts have worked out a method for training relational skills – called Multiple Exemplar Training (MET), that they have become quite excited about in recent years. MET has been used to establish and improve complex repertoires of mathematical skills and basic reading skills in children. Perhaps most excitingly, it is increasingly being considered a viable method for training to a high level of fluency the most important of the relational skills, and in so doing appears to provide a promising way of elevating the entire level of intellectual skill displayed by a child (i.e., most or all of those skills measured on standard IQ tests).
In case this sounds too good to be true, it might be interesting to note that precisely because behavior analyst have been fairly disinterested in IQ as a hypothetical construct, it has gone relatively unheralded that for decades now there have been published reports of large and permanent IQ gains in children who have undergone Applied Behavior Analysis (ABA) interventions for intellectual difficulties. These increases in general intellectual functioning dwarf anything hinted at (but crucially not promised) by the recent glut of brain training companies that seem to have materialized out of thin air in recent years. For example, the late pioneer of treatments for autism O. Ivar Lovaas (1987) reported IQ gains up to 30 points (roughly two standard deviations) following a three-year ABA programme for autistic children, and several other studies in that tradition have had similar effects.
What is a Relational Skill?
Current behavioral research in the field of Relational Frame Theory (Hayes, Barnes-Holmes & Roche, 2001) has shown that understanding relations (or “concepts” if it helps the reader), such as “more than,” “less than,” “opposite,” “same,” “before,” “after,” “here-there,” is crucial for our intellectual development. As an example of an abstract relational skill that children must acquire, consider how money works. The value of a coin is unrelated to its physical size. The size of a coin can be determined simply by looking at it or feeling it. It is extant and non-arbitrary (leaving aside for now, debates about the difference between sensation and perception). The important point is that the value of a coin is very much not a physical property of the coin and it is not discernible from interacting with it directly. Coin value is an abstract relational property and using it requires a basic grasp of some algebraic concepts, which is precisely why many children cannot use money. Children’s relational skills are not well enough elaborated to allow them to deal with abstract and arbitrary relations between symbols, such as coins.
Parents and caregivers inadvertently teach children relational concepts such as the “sameness” in normal language interaction. This is not simply to say that parents teach children how to use words. The critical issue here is that parents teach children a somewhat elaborated concept when they teach children to use the word “same” well. To be more specific, a parent will not just teach a child one word for a television set, they may in fact use two. On one occasion they may refer to it as the “TV” and on another as “the box.” The child will have to be explicitly told in the early years that given this information, “TV” and “box” refer to the same thing. Any confusion shown by the child is met with assurance from the parent that whenever multiple words are used for the same thing those words have the same meaning as each other. This is just one way in which a child learns to understand what “same” means and how “same” relations can be derived across multiple words and objects in logical ways. This is a skill required for vocabulary expansion, amongst other things. If it were not for this skill, each and every word in the child’s vocabulary would have to be taught individually and related to each other word individually (i.e., billions of individual learning tasks).
Isn’t this just syllogistic reasoning?
Relational skills are not to be confused with simple syllogistic reasoning. In fact, not only are these not equivalent, but from our point of view, it is the relational skill that allows syllogistic reasoning to take place, and not vice versa (reasoning itself has to be taught through multiple exemplars). In any case, even those relational skills that are very similar to syllogistic reasoning can be taught to levels of complexity and fluency not reached by ordinary educational or cultural experiences.
The correct term to refer to syllogistic reasoning, as we teach it, is “stimulus equivalence,” which is a mathematical concept that extends the range of conditions that need to be met over and above mere derivation between first and third terms. Specifically, in a syllogism; If A=B and B=C, it can derived that A=C. But in stimulus equivalence it is also necessary to show that the individual can derive that B=A, C=B (symmetry), A=C (transitivity). We usually go a step further and teach children to derive that C=A (combined symmetry and transitivity).
In addition, Relational Frame Theory studies new behavioral units never studied by the ancient Greeks when they discussed syllogisms. For example, the relation of “oppositeness” is crucial in mathematics, for understanding concepts like negative numbers. The understanding does not come first - the behavioral unit does. Once the individual learns to respond to C as the same A given A Opposite to B and B opposite to C as conditions, they are able to derive same relations from other topographically varying forms of combined opposite relations. Dozens of published studies show how the word opposite itself acquires its contextually controlling functions over the other words in the argument in the first place, so this is not a tautology. We understand from the ground up how relational terms acquire their first narrow and tentative functions (i.e., meanings) and also how this elaborates across time and examples of its use.
It gets even more complex. For instance, the opposite of an opposite of an opposite is in fact an opposite relation, not a same relation, as many adults initially think. And there are more relations that we have identified as important for other intellectual skills, such as temporal ones: If A is before B then B is in fact after A (the relation is not symmetrical). Hierarchical ones: If A is a type of B the B is not a type of A. Deictic ones: If I were you and you were me, then you would be here and I would be there (this is crucial for spatial reasoning and perspective taking, especially in autistic populations).
Relational Skills Training is not a “trick” or a shortcut
There is no “trick” to relational skills training. It was not developed to increase IQ. It is simply a straightforward teaching method for targeting precisely those foundational skills upon which intellectual activity depends. When we use this method of intellectual skills training we are not trying to tangentially enhance intellectual skills vis-à-vis some other process (e.g., neurogenesis; although behavior analysts have now studied the neural correlates of relational activity.). We are targeting the relational skills repertoire where it is most deficient.
Several studies published in the Journal of Applied Behavior Analysis and the Journal of the Experimental Analysis of Behavior confirm that children do need to be taught relational skills and that these will likely not emerge without the correct types of social interactions. On the whole we do not do a great job of establishing these skills in children, which is why so many adults, if you ask them (as your next nerdy party piece), think that the opposite of an opposite is an opposite! How can a parent who thinks this, teach relational concepts well. And why does the child not correct the parent using intuition alone if the answer to this poser is self-evident and hard wired into human cognition? The answer is because it is not self-evident at all – it is conventional. The rules of logical reasoning were invented, not discovered – and they need to be taught and taught well.
Plenty of studies have helped us triangulate in on the conclusion that relational skills underlie intelligence and the two have been shown to correlate well (see for example research by Dr. Denis O’Hora of the National University of Ireland, Galway).
Deriving relations is language itself
Emergent stimulus equivalence is now considered to be so completely synonymous with language acquisition, that Relational Frame Theorists refer to the stimulus equivalence effect as a “verbal process.” Deriving relations is language itself, and has never been satisfactorily shown in animals although many have tried with rats, sea lions, chimps and parrots.
Relational skills training has been shown in published research to impact intellectual ability scores (measured using the WISC) and in independent research into relational skills have shown that our ability to understand abstract relations corresponds to scores on standard IQ tests (e.g., the WAIS and Kaufman’s brief intelligence test). One published research paper (Cassidy, Roche & Hayes, 2011) described how a range of different children (four normally developing and eight educationally challenged) were provided with a fully automated relational skills training method on a computer in once to twice weekly sessions lasting approximately 90 minutes across several months. IQ tests (WISC III) were administered before the relational training and several weeks following the completion of training. At the outset of the study, the four normal children had an average IQ of 105 (ranging from 96-119). This is typical of normally developing children. Nevertheless, this average IQ was raised to over 130, which is called high functioning or exceptional. Children in this intellectual range are often referred to as gifted. The lowest IQ among the normally developing students following the intervention was 128 and the highest was 137. This means that these children’s intellectual ability was moved from average range to within the top 2 percent of the population. Four further typically developing children who had average range IQs were also tracked across the period of the study but they did not receive the training. Their IQs showed no change over the intervention period, as expected.
Eight further educationally challenged children started the program with an average IQ of 82 (using the WISC IV), well below the average score of around 100. Following the intervention, these IQs were moved on average to 96, well within the average IQ range. While all IQs improved, three remained below average. A further three children had their IQ moved into the average range, while two had their IQ raised to high average ranges. These raised IQs have been maintained four years since the trial (see Roche, Cassidy and Stewart, 2013).
So could it be that Relational Frame Theory has identified some basic behavioral units that constitute the building blocks of intelligence? I fear this will be answered with an automatic “no” by many readers. I have presented these findings at enough conferences to know how psychologists outside behavioral psychology think! Usually the resistance has something to do with a loyalty to Intelligence as construct and its completely successful reification by mainstream psychology. If, however, one can entertain the notion that intelligence is merely a measure of the fluency of a skill set, whether that is limited by biology or not, one can begin to understand the simple logic of asking “What is that skill set, and can we teach it?” Once you make this move you are well on your way to thinking like a behavior analyst, and the sacredness of intelligence as a stable trait begins to crumble. In essence, the question becomes, not “Can we really teach a person to behave more intelligently?” but “How do we best we teach them?”