Skip to content
The methodOpen data
Back to the collection

Relating Natural Language Aptitude to Individual Differences in Learning Programming Languages

Chantel S. Prat, Tara M. Madhyastha, Malayka J. Mottarella, Chu-Hsuan Kuo

2020 · Scientific Reports

Programming languagesStatistics & probabilityPsychology
Read the paper
HN score
0
Mentions
6
Readers
6
Months
6
What’s in a score?

From the abstract

AbstractThis experiment employed an individual differences approach to test the hypothesis that learning modern programming languages resembles second “natural” language learning in adulthood. Behavioral and neural (resting-state EEG) indices of language aptitude were used along with numeracy and fluid cognitive measures (e.g., fluid reasoning, working memory, inhibitory control) as predictors. Rate of learning, programming accuracy, and post-test declarative knowledge were used as outcome measures in 36 individuals who participated in ten 45-minute Python training sessions. The resulting models explained 50–72% of the variance in learning outcomes, with language aptitude measures explaining significant variance in each outcome even when the other factors competed for variance. Across outcome variables, fluid reasoning and working-memory capacity explained 34% of the variance, followed by language aptitude (17%), resting-state EEG power in beta and low-gamma bands (10%), and numeracy (2%). These results provide a novel framework for understanding programming aptitude, suggesting that the importance of numeracy may be overestimated in modern programming education environments.

Metadata: Crossref. Abstracts retain their original authors’ rights.

In the conversation

Original HN comments, with labels from simple text rules. A label describes the whole comment and may miss context or irony. Read the discussion to make up your own mind. About the labels

6 selected excerpts · 6 total mentions

fellowniusmonk
Neutral mention · rule
I really don't understand people who are down on LLM. In terms of code output. I have gone from the productivity of being a Sr. Engineer to a team with .8 of a Sr. Engineer, 5 Jr. Engineers and One dude solely dedicated to reading/creating documentation. Unlike a lot of my fellow engineers who are also from traditional CS backgrounds and haven't worked in revenue restricted startup environments, I also have been VERY into interpreted languages like ruby in the past. Now compiled languages are even better, I think from a velocity perspective compiled languages are now incredibly on par for prototyping velocity and have had their last weakness removed. It's both exciting…
Read the full comment
and scary, I can't believe how people are still sleep walking in this environment and don't realize we are in a different world. Once again the human inability to "gut reason" about exponentials is going to screw us all over. One terribly overlooked thing I've noticed that I think explains the differing takes. Foundation of my position here: https://www.nature.com/articles/s41598-020-60661-8 Within the population that writes code there are a small number of successful people who approach the topic in a ~purely mathematical approach, and a small number of successful people that approach writing code in a ~purely linguistic approach. Most people fall somewhere in the middle. Those who are on the MOST extreme end of the mathematic side and are linguistically bereft HATE LLM's and effectively cannot use them. My guess is that HN population will tend to show stronger reactions against LLM's because it was heavily seeded with functional programmers which I think has a concentration of the successful extremely math focused. I worked for several years in a purely functional shop and that was my observation: Elixir, Haskell, Ramda. Just my speculation.
Original comment The Unreliability of LLMs and What Lies Ahead
HarHarVeryFunny
Neutral mention · rule
Sheesh - why so touchy?! Your response opened with addressing the headline, which anyone who had "RTFM" (RTFA) would have realized was unrelated to the body of the article. You then veered off into a tangent about the nature of math which again was not addressing the content of the article. The underlying nature article, linked from the posted story, makes it even more clear what is being discussed, with the abstract stating: > This experiment employed an individual differences approach to test the hypothesis that learning modern programming languages resembles second “natural” language learning in adulthood. https://www.nature.com/articles/s41598-020-60661-8
Original comment The language brain matters more for programming than the math brain? (2020)
isaacfrond
Neutral mention · rule
> Verbal skills are apparently more predictive of programming career success than math scores. This is an urban legend inspired by this article [1]. One problem with this research though is that it studies language learning. Not how good a programmer one becomes, and certainly not career success. [1]: Relating Natural Language Aptitude to Individual Differences in Learning Programming Languages. https://www.nature.com/articles/s41598-020-60661-8
Original comment ProofWiki: Online compendium of mathematical proofs
dnilasor
Neutral mention · rule
YEAH the quality of NLP is not nuanced/fine-grained enough to be able to differentiate offensive text...its comprehension is nowhere near as nuanced as a human brain. One thing that gives me a little hope here is understanding the current state of the way we hire for, support, and value software engineering skills. Knowing how to communicate (ex: https://www.nature.com/articles/s41598-020-60661-8) is really important to being good at building software. Also, using strong communication and making SMEs (ie in this case, linguists, anthropologists, and other language specialists) an integral part of the development and learning process is really important. I'm not convinced that Google (and other shops trying to solve this problem) have really been leaning into this - getting the right people at the table and setting a standard for great communication during dev. Try that and get back to us?
Original comment DeepMind tells Google it has no idea how to make AI less toxic
dlkf
Neutral mention · rule
Here's the original paper [0]. The thesis is that language ability is better predictor of coding ability than numerical ability. The problem here—as in so many psychology studies—is that you need to substitute "____ ability" with "performance on a test that is designed by psychologists to measure ____ ability, but probably measures test-taking ability instead." More specifically, they found that their language test (MLAT) is a better predictor of their coding test (Codecademy) than their numeracy test (Abbreviated Numeracy Scale). But there needs to be some argument as to why this isn't just a relic of the test formats, and I am not seeing it. After looking at some sample…
Read the full comment
questions from each [1], [2], the MLAT questions simply look more like Codecademy questions than ANS questions. We've learned something about the relationship between these tests but I'm not convinced we've learned anything about language, math, coding, or their relation to cognition. You might disagree, and I'd be interested to hear opposing points of view. I encourage you to take a look at the tests yourself. I should mention that I am not interested in gatekeeping, or saying that people who find math difficult shouldn't do CS. I was a bad math student in high school, and I would probably get a bad score on all these tests (not that I would take them without a gun to my head). I'm now happily employed as a data scientist. On the contrary, I think that if we are going to say "people like this are good at coding," regardless of what this is, our evidence should be more solid than this study. [0] https://www.nature.com/articles/s41598-020-60661-8 [1] http://lltf.net/mlat-sample-items/ [2] https://onlinelibrary.wiley.com/doi/10.1002/bdm.1751
Original comment Not a 'math person'? You may be better at learning to code than you think

Keep following the thread.

More in Programming languages

More from the same topics.

2018arXiv

Talent vs Luck: the role of randomness in success and failure

A. Pluchino. A. E. Biondo, A. Rapisarda

The largely dominant meritocratic paradigm of highly competitive Western cultures is rooted on the belief that success is due mainly, if not exclusively, to personal qualities such as talent, intelligence, skills, efforts or risk taking.

Psychology
16 mentions14 months
2.5HN score