When a person “doesn’t know right from left”, they are metaphorically confused, or unable to navigate the world. In the non-metaphorical meaning of the phrase, we’re talking about a person whose concepts of right and left are somehow undeveloped – and telling left from right is an ability that we definitely need in order to get around.

Direction words like “right” and “left” are two examples of the simple relations that are exploited by cognitive systems. Humans, rats, ants, and many other animals can all understand and exploit these kinds of relations.

One of the most defining aspects of human cognition is how much we rely on relations to understand different aspects of the world. Cereal goes in the bowl, Halloween is a holiday that takes place in October, plates can be stacked on top of each other, and so on.

Relations are supported by a cognitive process known as binding, which helps us pair one thing with another in either short-term or long-term memory. While we do not have a single mechanistic theory about how two items are bound together, we know that binding is important in most cognitive theories, from theoretical linguistics to studies of the hippocampus and amnesia. For example, there is much evidence that items on a short list are bound to contextual or positional representations in short-term memory.

Relations are so common in our day-to-day life that it might be easy to miss them, but deficits in relational binding can mean we no longer remember where we put our keys; we might forget who we heard a story from; or we might get lost on the way to the grocery store because we turn right rather than left at an intersection.

What happens when our ability to bind objects to each other goes wrong? Is there anything we can do to fix it? Especially as we get older, we suffer from worse relational memory, to the point that science and industry are putting out more and more possible remedies – including exercise, medicine, everyday puzzle solving, and “brain games.” At the core of the last two ideas is the belief that practice with reasoning can help diminish the effects of age on cognitive abilities.

In a recent study out in the Psychonomic Society’s journal Memory and Cognition, researchers D’Angelo, Noly-Gandon, Kacollja, Barense, and Ryan sought to identify whether practicing binding using a technique called unitization could reduce age-related relational memory problems.

Unitization is a task or process where several items, like the words “cat”, “armchair”, and “pizza” are put together and treated as a single unit. When people create these units, they invoke a structure that links the meanings of these words, or a mental image (e.g. a cat on an armchair eating pizza). Once these units are created, only one item is left to be remembered, and there is no longer a relation that must be held in memory. In prior work, D’Angelo and colleagues identified at least three components to unitization – fusion, motion, and action-consequence sequences.

These three strategies are all slightly different ways of thinking about pairs of items. In fusion, two items are combined into a single “image”, like a needle and a pincushion; for motion, one object might move toward the other. In the action-consequence case, we can imagine one object doing something to another, such as a needle piercing a pincushion. Conceptually, action-consequence unitization captures the idea that a relation can be transitive or directional, while fusion and motion represent symmetric relationships.

To study whether unitization or any one part of it can be used to improve relational memory, D’Angelo and colleagues have used the transverse patterning (TP) paradigm, which is a lot like the game rock-paper-scissors, except with abstract shapes A, B, and C. In this task, A wins over B, B wins over C, but C wins over A. D’Angelo and colleagues taught a group of 80 healthy older adults to do the transverse patterning task with different instructions and displays to teach new strategies to see whether one strategy helps with relational memory more than others.

In this study, the authors compared older adults’ accuracy at learning the relations through (1) trial-and-error, (2) learning them with use of a fusion metaphor (the darker object in this fused object wins), (3) a motion metaphor (the winner moves), (4) an action-consequence metaphor (the dark object “squishes” the light object), and (5) a full unitization strategy (the winner moves toward the other, “squishes” it, and the two objects fuse). The four non-standard strategies are reproduced above. For every trial, participants were asked to identify which object won, and were reminded of the relationship between the two items in all the experimental conditions. After making a decision during the training phase, participants received feedback about which item won. There were no initial differences in accuracy during the training phase across all five conditions.

After training, there were two test phases; one took place immediately, and another at a delay. Unitization strategies are known to help memory for relations, so the authors expected unitization training to help during the test phase. At the short term test phase, giving participants a strategy during training helped them identify which shape would win. At delay, however, the motion and fusion strategies did not help, while the unitization and action/consequence strategies did. Surprisingly, action/consequence training was as helpful as the unitization training, suggesting that identifying a transitive action helps us remember relations. The authors propose that understanding action/consequence sequences might be the entire source of the benefits of unitization.

While this strategy might not drive the results in all studies where unitization has proved to be helpful, and even though many different strategies could be used depending on the exact task, these results suggest that simply using prior knowledge about the ordering of actions like “A squishes B” can help people remember the relationship between two items. This kind of strategy could be useful to get around the problems of weakening relational memory skills as we age – without having to buy volumes of puzzles or subscribe to brain game programs.

Article Highlighted in this Post

D’Angelo, M. C., Noly-Gandon, A., Kacollja, A., Barense, M. D., & Ryan, J. D. (2017). Breaking down unitization: Is the whole greater than the sum of its parts? Memory & Cognition, 1-13. DOI: 10.3758/s13421-017-0736-x.