By guest blogger Bradley Busch
As father to an 18-month-old toddler, I would love to know exactly what my son is thinking. Along with many parents, one of the ways I try to find out is to ask him questions of the variety “Do you want X or Y?” But does his answer to this type of question actually reveal his preference or is it a more a reflection of a quirky cognitive bias that is more powerful in children than adults?
Two competing effects influence how adults respond to binary choices. The first is called “The Primary Effect” which describes the way that the first option we hear tends to stick in our minds. For example, one study found that adults are more likely to choose “heads” when asked if a coin toss is going to be “heads or tails”. The second effect, which sometimes contrasts with the Primary Effect, is called “The Recency Effect”. This captures the way that the last thing we hear or experience can also have more weight in our memory (this is why popstars end their concerts on their best songs, so that everyone leaves thinking the whole gig was great). In adults, neither the Primary Effect or the Recency Effect is always the more pronounced, with evidence suggesting that one’s personality type, familiarity of the information, and how controversial the topic is, all play a mediating role.
Keen to test if the same is true in young children, or if one of the effects is more dominant, a team led by Emily Sumner conducted two experiments and published their findings in a recent paper in PLOS One fantastically titled “Cake or Broccoli: Recency Biases Children’s Verbal Responses”.
In the first experiment, they asked twenty-four toddlers a series of questions with two options for them to consider (i.e. “Should Rori bring a lunchbox or backpack to school?”). Later, after enough other questions had been asked to mask the repetition, the order of these options were flipped (i.e. “Should Quinn bring a backpack or a lunchbox to school?”). When answering verbally, which they usually did, the toddlers chose the second option 85.2 per cent of the time.
Interestingly, when the toddlers pointed to their preference instead of saying them, they chose the second option only 51.6 per cent of the time. It would appear that there is something about saying the answer that accentuates the Recency Effect. Further analysis of the data revealed that the Recency Effect was more powerful among younger toddlers (18-24 months old) compared with the older ones (24-48 months), suggesting its impact diminishes the older the children get, as they become more rational and logical.
In their second experiment, the researchers explored why the Recency Effect is so pronounced in young children. To do this, they had twenty-four 3-to-4 year olds choose names for different toys. Some of the options were one-syllable names (i.e. “Should we name this toy ‘Stog’ or ‘Meeb’?”) whereas others were longer (“Should we name this toy ‘Shallop’ or ‘Bingle?”). This went all the way up to four-syllable names. The researchers found that the longer the word was, the more likely the children were to choose the second of two options. The researchers speculated that the longer names placed more of a burden on children’s limited working memory capacity, resulting in the latter of the two options being fresher in their mind and therefore more prominent. This partially explains why the younger children in the first experiment acted on the Recency Effect more, as their working memory capacity is even more limited at that age.
This study has implications for both researchers in education and parents. For researchers, a popular way to assess young children’s comprehension and development is to have them do tasks which involve making these sort of fixed-choice decisions. However, as the authors of this study note, their findings “suggest that some of children’s earliest utterances may be the result of a recency bias, and do not necessarily indicate that the child comprehends the words they speak”.
For parents keen to gain an insight into their child’s thought processes and understand them better, it would probably pay dividends to vary the order of choices they give them. By ensuring that the same option is not always second, it will probably help them deduce what their child really prefers. Another effective strategy would also be having their child point to which of the two options they want, as this helped combat the recency bias in this study. Cunning parents could also consider exploiting the Recency Effect to encourage their child to express an apparent preference for a healthier option over the other. Perhaps if I do this, I can get my own toddler to eat more broccoli and less cake.
Post written by Bradley Busch (@Inner_Drive) for the BPS Research Digest. Bradley is a registered psychologist and director of InnerDrive. He works with schools, businesses and athletes and is the author of The Science of Learning and Release Your InnerDrive