How Do We Learn Things?

Repetition and similarity

All human learning requires repetition. The brain is a volatile knowledge storage device and very few humans can retain knowledge of any sort when only exposed to it once and that briefly. Psychology experiments have shown that certain processes need to be executed to move knowledge into “long term memory.” We have all experienced attempting to remember a phone number and failing, because we could not effectively memorize it. Some of the “tricks” of developing good memory retention involve repetition—saying something over and over—and associating whatever needs to be remember with some already learned information.

Clearly, repetition builds thought patterns, as any music teacher (or music student) can attest. Without copious repetition, it is very difficult for most people to learn to effectively play an instrument, learn a language, or play a sport. While the concepts of “short term” and “long term” memory are useful shorthand descriptions of how humans store knowledge in more permanent modalities, it is unlikely that these actually represent physical storage in locations in the brain. As already noted, the physical elements of the brain do not “store” knowledge so much as move it around in electrochemical-enabled patterns. Information in “short term memory” has not been associated with a persistent (self-reinforcing) pattern whereas that in “long term memory” has. Since repetition is so clearly necessary to make this transfer from short-to-long term, it is equally clearly a primary mechanism for generating such patterns. When “new” information is closely aligned with patterns that have already been developed, the effort and degree of repetition is less. Spanish speakers, for instance, have less difficulty learning Portuguese than in learning Mandarin Chinese.

Peekaboo

When sufficient repetition has built thought patterns that can be recalled later, the knowledge can be said to have been “learned.” When babies view the world, they initially have little ability to “recognize” what they see. There is evidence that they respond to images (and even graphical depictions) of faces. Some of the first repetition they experience, and learn to respond to, is the face of their primary caregiver. The evolutionary advantage of this behavior is pretty obvious. While they do have sufficient pattern structures built to “recognize” a face, they do not have a concept of continuity until their brains have developed somewhat further and they have been exposed to sufficient repetitions of the image appearing, disappearing, and reappearing. The game of “peekaboo” both surprises and delights babies because of this. The obvious delight a baby shows when a caregiver plays this game rewards both parties and encourages them to repeat the game. What the baby experiences is an image, followed by a different image, followed by the first image again. Since they have already been conditioned to respond positively to the caregiver’s face, seeing it again refreshes the pattern. When the face is hidden, the baby has to “reassess” what it is looking at. When the face reappears, the cycle starts again. Since babies have not yet acquired the concept of object permanence, when the face is gone it effectively no longer exists. Object permanence is a necessary skill to building thought patterns. It is interesting that parents reflexively start playing “peekaboo” by covering their own face, rather than covering something else and teaching their children object permanence using other objects. However, once a child has sufficiently mastered “peekaboo” with faces, the parent usually transitions to playing the same game using favorite toys, food, and other types of objects the baby will respond to.

In each of these stages, the brain is responding to two things: sequential repetition (of the same thing) and similarity (of the thing with other things). The repetition is building and then reinforcing the primary pattern that allows brains to “recognize” what is being presented. The similarity is building associative patterns and meta patterns that allow recognition across a wider range of objects.

Peekaboo Variations

As a thought experiment, and before we leave “peekaboo” behind, consider what would happen if the following occurred:

(a) The “peekaboo” game presented objects that the baby had no association or experience with: hand tools, raw vegetables, the US Declaration of Independence, a map, a telescope,…

(b) The “peekaboo” game presented entirely different objects each time, so each object was shown only once.

(c) The game showed the baby something it recognizes, but then hides it for a long time (all “peek”, no “aboo”).

(d) The game starts with the reveal (a game of “aboo-peek”).

Clearly, with these changes the effectiveness of the game in developing object permanence would be significantly compromised.

Alternatively, consider what happens in practice: once a baby has “mastered” peekaboo with the caregiver’s face, they usually move on to a favorite toy, and then a different toy, and then something else. This somewhat-the-same-somewhat-different repetition both reinforces the initial object permanence and helps build thought patterns which are now shared. Not only can a caregiver’s face disappear and reappear, but that can happen with other things. Gradually, the child builds thought patterns which reflect similarity and difference and this sets the stage for more extensive learning.