Why prior knowledge doesn't always help

There's a tendency to think of prior knowledge as having a wholly positive impact on current and future learning, but learning is complex, and our brains are messy things.

Over fifty years ago, the American educational psychologist David Ausubel stated that ‘the most important single factor influencing learning is what the learner already knows’ (Ausubel, 1968). To paraphrase Ausubel, once we know that, we can then teach (and learn) accordingly.

According to Hambrick and Engle, this view espoused by Ausubel is ‘one of the most influential ideas to emerge in cognitive psychology in the past twenty-five years’ (Hambrick and Engle, 2002, p.340).

In the research literature, this notion is sometimes referred to as the Knowledge is Power hypothesis, or KiP (for example, Simonsmeier, 2022), but there remains a naïve assumption that relevant prior knowledge always facilitates learning. But Garvin Brod points out, Ausubel didn’t claim that prior knowledge was always helpful, just that it is a strong determinant of what a student will learn in a lesson (Brod, 2021 p24).

How prior knowledge helps (and hinders) learning.

Prior knowledge can help us learn more efficiently. However, there are also ways in which it's detrimental to learning.

Prior learning certainly has a positive impact on current learning, at least much of the time. At other times, it might hinder it. We can think of these as positive and negative mediators of prior learning.

Positive mediators of prior knowledge

Attention: Research finds that people with higher levels of prior learning in a particular area pay more attention to the important features of new to-be-learned information (for example, Tanaka et al., 2008; Yu et al., 2012). Our attentional system selects features that are familiar (as evidenced during during dichotic listening tests). This may also explain why experienced chess players can rapidly identify key pieces and positions on a chessboard, while novices often cannot identify the strategic significance.

In a classroom environment, this would lead to students with higher prior knowledge of a topic being more likely to focus their attention on new related information. This seems intuitive, seeing as the material is likely to make more sense within the context of what students already know.

Encoding and interpretation: Prior knowledge helps us understand and remember new information. For example, readers with background knowledge can better grasp and recall the context (Brod. 2013; van Kesteren 2014). Learners are employing knowledge from prior learning to make sense of new learning. The creation of these connections results in more resilient retention and better comprehension.

It’s easy to see how good foundational knowledge in mathematics, for example, would help students learn more efficiently later. For example, Sonja Hermann and her colleagues found that primary school children in Germany with lower initial scores in mathematics displayed greater learning gains than middle range and higher achievers. Importantly, however, the attainment gap never fully closed. Hermann suggests that the German systems, with its emphasis on foundational maths education, probably helped to mitigate the achievement gap.

Binding of new information into chunks: Binding is the process by which the brain’s cognitive architecture combines elements of an object into a single entity. By binding new information into chunks we allow for more efficient memorisation, processing, and retrieval. Chunking to overcome capacity limitations has been the mainstay of memory research since Miller’s digit-span experiments. You’ll recall that Miller discovered the capacity of short-term memory to be between 5 and 9 pieces, or chunks, of information (our immediate digit span, although it applies not only to digits). Miller wasn’t particularly specific about what he meant by this term chunk and, even today, it has an ambiguous quality to it.

Strategy development: If we have prior knowledge about how we can best learn, we are better equipped to employ strategies and solutions that make learning more efficient and effective.

Similarly, prior knowledge about the effectiveness and efficiency of problem-solving strategies can facilitate exploration, goal-directed behaviour and the construction of more advanced new strategies.

Strategy development broadly falls under the umbrella of metacognition and concerns the ability to think about our learning, plan, set goals and monitor our successes and setbacks.

Research finds that those with prior knowledge of problem-solving approaches can more efficiently explore solutions and create new strategies. This applies to all domains, but is particularly so in domains like mathematics, where understanding basic concepts helps in tackling more complex problems (see, for example, Schneider et al., 2011).

Source evaluation: Perhaps more than ever, the ability to validate the credibility of information sources is a vital skill that is often overlooked. Prior learning can help us evaluate the plausibility of new information, operating akin to a personal fact-checker. For example, if we have deep prior knowledge of climate change, we are more likely to be sceptical of claims denying its existence (e.g. Lombardi et al. 2016).

Negative mediators of prior knowledge

While there are many positive aspects of prior-knowledge, it’s not all good news.

Misconceptions and incomplete knowledge: Learning doesn’t always go the way we planned. Sometimes we remember some of what we learned about a topic while at other times we get on the wrong end the stick. This can then result in us reaching incorrect conclusions. For example, believing the Earth is flat (a misconception) can interfere with our understanding about related concepts like planetary motion and gravity (Vosniadou and Brewer, 1992).

Perceptual biases and inflexible learning: Past experiences with a particular problem can sometimes make it harder to find simpler solution, even if they are more efficient, a perceptual bias called the Einstellung effect. (See Lewandowsky and Kirsner, 2013; Bilalić et al., 2010; Sheridan and Reingold, 2013).

Automatisation and inflexibility: Excessive practice to automate a skill can lead to inflexibility. For example, musicians who have practiced a piece for years in a specific way may find it difficult to adapt to a different tempo or style, leading to inflexible behaviour (Johnson, 2003; Müller, 1999).

Increased possibility of interference: By having an extensive knowledge in a particular field, related concepts may interfere with each other, leading to confusion. Interference is also an important factor in forgetting, so we might remember a concept as relating to a particular topic when it actually relates to another. The likelihood is that some of this information has disappeared from memory entirely.

A 2007 study by Alan Castel highlights how prior knowledge can interfere with current learning (Castel et al., 2007). Castel and his team recruited forty undergraduates between the ages of 18 and 23. They gave participants two lists of words to learn.

The critical list comprised 11 animal names that were also the names of American football teams, for example, dolphins, falcons, and colts. The second list comprised body parts. After learning the lists, volunteers were given a filler task so that they couldn’t mentally rehearse the list but could create a delay between the learning session and the recall test. After the recall test, all volunteers were asked to complete a series of multiple-choice questions aimed at assessing their knowledge of football. They were then allocated to either a high-knowledge groups and a low-knowledge group.

The researchers found that participants with higher prior knowledge of football recalled more words from the list. However, they were also more likely to recall words that didn’t appear on the original list. Most notably, eagles, panthers, and cardinals - all NFL football teams.

Prior learning is important, but it still has its drawbacks.

While prior learning is important for future learning, it isn’t as simple as that. Sometimes, what we already know can impede what we learn in the present, as well as to prevent us from exploring new solutions that may well be more effective than those we’ve used in the past. Furthermore, human memory is fallible and often error-prone, despite what we like to believe.

---

I’m on Blue Sky. Come and say Hi!