CUNY journalism teacher Emily Laber-Warren reports at Scientific American on an interesting psychology study out of Belgium that divided the 105 undergraduate participants into three groups where each member was to solve to solve a group of 70 word puzzles under three different conditions,. The first group just had to solve the puzzle in 25 seconds or less, the second group had to also remember two numbers flashed on a screen, and the third group had to remember four numbers flashed on the screen.

Each participant was asked to record whether the puzzle was solved by an Aha! insight or systematically, step by step. Laber-Warren explains:

The purpose of making people remember random numbers was to burden their mind with an unrelated task, which was expected to interfere with conscious problem-solving. “These cognitive resources, this pool that we can tap into to do anything consciously, is limited,” Stuyck says. The question was whether insightful thinking would be similarly affected.

Emily Laber-Warren, “Aha! Moments Pop Up from below the Level of Conscious Awareness” at Scientific American (January 26, 2022) The paper requires a subscription.

The word game was in Dutch but Laber-Warren gives, as an English equivalent: What word would pair naturally with a;ll three of these words: “artist,” “hatch” and “route.” The correct answer is “escape.”

Remarkably, the participants who relied on the Aha! insights were less likely to be distracted by the flashing numbers:

These participants accurately completed between 17 and 19 puzzles, on average, in all three groups. “Whether they don’t have the memory task or they have a low-demand memory task or a high-demand memory task, the number of puzzles they solve with insight remains constant,” Stuyck says. “That’s the most interesting result.”

Emily Laber-Warren, “Aha! Moments Pop Up from below the Level of Conscious Awareness” at Scientific American (January 26, 2022)

Indeed. Study leader Hans Stuyck noted that there is considerable debate in the psychology literature about whether reasoning can occur when we are unaware of it. This study would seem to suggest that it can.

There is certainly anecdotal evidence for that. Remarkable mathematician Srinivasa Ramanujan (1887–1920), for example, struggled with basic procedures but then a piercing insight would flash into his mind and, trying to explain, he simply said that God had told him.

The study’s findings can be seen as a subset of a general pattern by which creative insights are non-computational. Those who wonder whether computers will end up thinking like people — or even doing our thinking for us — might wish to ponder that. By their very nature, computers only compute but much human thinking, including flashes of remarkable insight, is non-computational.

You may also wish to read:

Why human creativity is not computable. There is a paradox involved with computers and human creativity, akin to Gödel’s Incompleteness Theorems or the Smallest Uninteresting Number.

A type of reasoning AI can’t replace. Abductive reasoning requires creativity, in addition to computation.

and

What? No AI overlords? What is Larson arguing and why does it matter? Information theorist William Dembski explains, computers can’t do some things by their very nature.