Should learners take notes with laptop or pen and paper?

Students routinely take notes during university lectures. This helps encode information into memory and provides a source for later test revision. Increasingly, notes are produced on a laptop computer rather than with pen and paper. A newsworthy 2014 study found that handwritten notes produced superior learning, but other studies have found no benefit or even suggested laptop superiority.

Here, researchers directly replicated the procedures of the 2014 study, and then merged data from both experiments. They found that test performance was indistinguishable for the two methods, but that there were differences in note quality: laptop notes were longer and had more direct overlap with spoken lecture content.

The authors conclude that the evidence for superiority of laptop or pen and paper notes is too weak for prescriptive advice. Instead, they recommend more research into which method allows for notes that reflect later test material.

Morehead et al. (2019) How much mightier is the pen than the keyboard for note-taking? A replication and extension of Mueller and Oppenheimer (2014). Educational Psychology Review DOI: https://doi.org/10.1007/s10648-019-09468-2

Learning while sleeping

Sleep is important for consolidating memories formed earlier in the day, but can we also learn new information while in a deep slumber? To find out, researchers delivered pairs of words as participants slept. When they awoke, participants were given a memory task in which one word had to be linked to its pairing partner.

Although the effects were not large, the study showed this type of learning is possible during sleep. Importantly, learning only occurred if words were delivered at the peak of participants’ ongoing slow-wave sleep rhythm, as measured by EEG. Successful retrieval of word pairs (when awake) involved activation of the hippocampus and cortical language areas. The authors conclude that learning vocabulary during sleep is possible if the timing of word delivery matches the slow-wave sleep rhythm.

Züst et al. (2019) Implicit vocabulary learning during sleep is bound to slow-wave peaks. Current Biology 29(4): P541-553.E7 DOI: https://doi.org/10.1016/j.cub.2018.12.038

Achieving deep learning without explicit foundational knowledge

A goal of education is to instill deep rather than superficial knowledge in learners. How is such “higher order” learning – critical thinking, analysis, inference – best achieved? One influential theory is that the basic knowledge – the facts – must first be in place. Alternatively, pre-test training on higher order concepts may be the best preparation for a later higher order test.

This study counters the prevailing wisdom, finding that building a base of factual knowledge (via retrieval practice in the form of quizzes) did not improve students’ higher order learning (as later measured in a test of the same format). Instead, to produce better higher order learning, students should be encouraged to practice higher order questions, in which they must for example link concepts and draw inferences.

Agarwal (2019) Retrieval practice & Bloom’s taxonomy: Do students need fact knowledge before higher order learning? Journal of Educational Psychology 111(2): 189-209. DOI: http://dx.doi.org/10.1037/edu0000282

Dendritic signals needed for reward learning

By learning the link between our behavior and an ensuing reward, we position ourselves to receive future rewards. This is reward-based (or reinforcement) learning, a powerful and natural form of learning.

In this study, researchers sought to understand how individual neurons link rewards to ongoing behaviour. Mice were trained to seek a reward in response to a sensory cue, in this case the brushing of a whisker. As learning proceeded, the signal in the cell’s furthermost dendritic branches – where information from a host of other brain regions arrives – grew stronger. This suggests that these dendritic signals, and this inter-regional communication, are important for learning the link between reward (in this case a water droplet) and behavioral event (pole brushing whisker).

Lacefield et al. (2019) Reinforcement learning recruits somata and apical dendrites across layers of primary sensory cortex. Cell Reports 26(8): P2000-2008.E2 DOI: https://doi.org/10.1016/j.celrep.2019.01.093