Sharing a fun paper to geek out on with my fellow language nerds, “How children learn to communicate discriminatively” by Michael Ramscar. In this paper, the author makes an argument that the contrasting forces of “discriminability” and “regularity” both serve to make language something we pick up pretty much naturally, even if we don’t know all the words in the language.

“…the existence of regular and irregular forms represents a trade-off that balances the opposing communicative pressures of discriminability and learnability in the evolution of communicative codes. From this perspective, the existence of frequent, well-discriminated irregular forms serves to make important communicative contrasts more discriminable and thus also more learnable. By contrast, because regularity entails less discriminability, learners’ representations of lexico-morphological neighbourhoods will tend to be more generic, which causes the forms of large numbers of less frequent items to be learned implicitly, compensating for the incompleteness of individual experience.”

The language of this paper is, as you can see, a bit opaque, so much of this went just a bit over my head, but I found the arguments fascinating given the debates that happen about how to teach the “irregular” spelling of so many words in the English language. Here, the author seems to suggest (I may be over-extrapolating as I often tend to do, but this is what got me geeking out on it) that in fact there is some level of constructive tension between language forms that show up again and again, and the language forms that are more infrequent, but thus inherently gain more of our attention. This relates to the theory of “statistical learning” with which we not only learn language, but also when we map a language to its written form.

The author later provides what I thought was a very concrete thought experiment that demonstrates this principle when he moved from morphology to names:

Imagine that 33% of males are called John, and only 1% Cornelius. In this scenario, learning someone is named Cornelius is more informative than learning their name is John (Corneliuses are better discriminated by their names than Johns). On the other hand, Johns will be easier to remember (guessing ‘John’ will be correct 1/3 of the time). Further, although the memory advantage of John relies on its frequency, the memorability of Cornelius also benefits from this: Cornelius is easier to remember if the system contains fewer names (also, as discussed earlier, if John is easier to say than Cornelius, this will reduce the average effort of name articulation).

What is also interesting about the author’s argument in this paper connecting information theory to language learning is that these assertions are empirically testable:

“Whether these mathematical points about sampling and learning actually apply to human learners are empirical questions. This account makes clear predictions in regard to them: if learners are exposed to sets of geometrically distributed forms, they should acquire models of their probabilities that better approximate one another than when learning from other distributions. Conversely, if learning from geometric distributions does not produce convergence, it would suggest the probabilistic account of communication described here (indeed, any probabilistic account of communication) is false.”

There’s a lot more in the paper to nerd out on–I found the section on verbs especially interesting, for example, given that it connects to some other tidbits on the power and challenge of verbs I’ve come across before:

• More on a child’s linguistic development and verbs
• Also a tidbit on importance of verbs here

I’ll leave the rest to you!

#verbs #regularity #irregularity #learning #language #statisticallearning #probability #discriminability #informationtheory #form

As I was preparing for a session I was facilitating, I went down a rabbit hole on language use and cognition. I know saying “went down the rabbit hole” typically bears a negative connotation, but I gotta say, I love me some getting lost in meandering exploratory nerdy byalleyways. While rabbit holes may oft lead nowhere but to wasteful skimming of social news feeds, I believe that they can also lead to fortuitous and deeper connections that wouldn’t have occurred otherwise.

Case in point: whilst engaged in aforementioned spelunking, I discovered an absolutely wonderful paper synthesizing various theories of language, and I say wonderful because it manages to thread together varying theoretical perspectives from a stance of learning and curiosity, rather than pitting them against one another, as is so often the case. The paper is Essentials of a Theory of Language Cognition by Nick Ellis, and while it may be heady and academic, there’s something playful, even poetic, in the author’s use of language (so meta!).

By example, here’s a couple of gems:

“Language and usage are like the shoreline and the sea. Usage affects learning and it affects languages too. So, our understanding of language learning requires the detailed investigation of usage, its content, its participants, and its contexts—the micro level of human social action, interaction, and conversation; the meso level of sociocultural and educational institutions and communities; and the macro level of ideological structures.”

“Language is the quintessence of distributed cognition. Language is ever situated, either in the moment and the concrete context or by various means of mental extension to reflect prior or imaginary moments.”

Dear reader, you may or may not be aware that I have another (not updated any more) blog entitled, Schools & Ecosystems, wherein I geeked out about complex adaptive systems and how ecological concepts relate to the physical and social environment of schools. So you can imagine my nerdy delight when I discovered a connection in this paper between complex adaptive systems thinking and LANGUAGE! Oh my. It was like two previously schizophrenically disparate selves suddenly merged into one.

Here’s a couple of quotes regarding language as a complex adaptive system:

Language as a CAS [complex adaptive system] involves the following key features: The system consists of multiple agents (the speakers in the speech community) interacting with one another. The system is adaptive; that is, speakers’ behavior is based on their past interactions, and current and past interactions together feed forward into future behavior.

De Bot, Lowie, and Verspoor (DBL&V) present a persuasive case for language as a complex dynamic system where cognitive, social, and environmental factors continuously interact, where creative communicative behaviors emerge from socially co-regulated interactions, where there is little by way of linguistic universals as a starting point in the mind of ab initio language learners or discernable end state, where flux and individual variation abound, where cause-effect relationships are nonlinear, multivariate and interactive, and where language is not a collection of rules and target forms to be acquired, but rather a by-product of communicative processes.

In a previous post, we looked at some interesting findings from neuroscience that suggested language in the brain is mostly associated with parts for communication, rather than thinking. So this idea of language as a complex adaptive system that emerges based on social use within a particular community makes quite a bit of sense.

One of the other things that jumped out at me as a theme emerging from these various theories of language was the idea of language as an ecology: something dynamic and situated within a particular time, place, and community of relationships. It’s a beautiful — and more accurate — way to think of language that allows us to acknowledge the unique language ecologies we can each have as individuals and as members of communities — most especially for multilinguals who bring a rich repertoire of linguistic experiences and cultural knowledge.

This paper is also a wonderful companion to Annie Murphy Paul’s book, The Extended Mind, which isn’t focused on language per se, but connects to many of the theories in this paper and also has some Easter eggs for language focused nerds, such as an exploration of the use of gestures as a precursor and accelerator of language.

Somehow I had not stumbled across “usage-based” linguistic research or theory previously, so I’m excited to dig more into this realm. Seems like it has a lot to offer, especially as the reading research crowd begins to unpack more the language connection to reading (importance of phonology, morphology, incidental learning, statistical learning, etc).

#linguistics #language #ecology #complexadaptivesystem #interaction

Discuss...

I recently came across this fascinating study in which the researchers used social network analysis and found:

• Children’s language skills were significantly associated with friendship centrality and reciprocity
• In kindergarten, kids who enter school with lower language skills have fewer peers who nominate them as their friend
• Children at risk for specific language impairment (SLI)/developmental language disorder (DLD) were less central to their classroom networks
• The odds of a reciprocal friendship tie was more than 50% lower than peers not classified at risk
• And of children with or at risk for SLI/DLD, girls were significantly more central than boys, suggesting gender may play a role in friendship development in early elementary school, especially for children with lower communication skills

It really got me thinking, and also reminded me of a recent article on NPR about the experiences of students learning English during remote learning, in which a quote from a researcher caught my attention:

“Having one friend who speaks English well is a very, very good predictor of your grades,” says Marcelo Suárez-Orozco, who has spent years researching immigrant youth. Now the chancellor of the University of Massachusetts Boston, Suarez-Orozco previously co-authored a study with his wife about the process of learning English.

“Very few youth in our study could say they had one friend who was an English dominant speaker.”

–“Millions Of Kids Learn English At School. Teaching Them Remotely Hasn’t Been Easy” by Kavitha Cardoza

Now, as to WHAT study exactly it was that provided this data point is completely unclear, as apparently some articles don’t feel the need to provide citations, and I couldn’t figure out which of Suárez-Orozco’s many writings might have provided it. But if accurate, this seems like a highly critical point to consider alongside of the findings of that paper, both in terms of the needs of students learning a new language, as well as for students who may struggle with language due to a disability.

As a former special education teacher I well know how important relationships are for students, and furthermore, how central relationships are to the culture of a school, and this all brought me back to that.

For students who are developing language skills, having dynamic discussions and peer interactions is so powerful, and it makes complete sense that social relationships are interconnected there. Some of this includes not only knowing the language of academic discourse and the language of written texts, but furthermore the language that names emotions and identity. There is a “hidden curriculum” of school that relates to social norms, and all students benefit from explicit naming–and active co-construction of–those norms.

What can schools do to foster positive peer interaction and friendship for those students who need it the most?

#language #friendship #socialnetworks #peers #interaction #disability #socialemotional #relationships #culture

There was a fascinating summary thread I came across recently that I want to dig into, as there’s some really interesting and rich areas of tension to unpack. Here’s the thread

What especially caught my eye and made me ponder for days afterwards was this:

The language network does not overlap or build on nonlinguistic cognitive abilities . . . fMRI evidence from 32 experiments, with 64 conditions, and 761 participants across 1,007 scanning sessions suggests language is separate from thought – when processing non-linguistic stimuli other areas are activated compared to when processing linguistic stimuli . . . The language system does not share resources with other cognitive abilities.

Language is separate from thought. I really struggled to understand this . . . isn’t language how we think, whether conscious or not?

Fedorenko argues that there are properties of language that suggest is is not suitable for complex thought, but is well-suited for communication . . .For example, language processing is fundamentally predictive, something that wouldn’t be useful if language was primarily used for thought and not communication. Although the language network and other cognitive abilities seem to be distinct systems, they need to integrate in some way. Shedding light on this integration is a key direction for future research

Where language does intersect with other cognitive systems, however, according to this presentation, is “some exciting new research emerging that language is intimately linked with the system that supports social cognition, such as Theory of Mind.”

Another tantalizing tidbit in this thread relates to syntax and word meaning:

language does not rely on abstract syntax. Syntactic processing is distributed across the language network and “every syntax-responsive cell population or brain area is robustly sensitive to word meaning” . . . . In every region, even at the most fine-grained level of analysis shows that there are no selective responses to abstract syntactic structure – everything that responds to structure building also responds to word meaning.

Well now, I want to unpack that one a bit more! It seems to suggest that word meaning i.e. semantics i.e. vocabulary/morphology is higher leverage than syntactical structure.

All of this really got me thinking, about thought and cognition, about language . . . and especially about how adding in literacy — a writing system — complicates all of this . . . I mean, writing is a form of thought, right? I sometimes don’t think things, or know what I think about things, until I force myself to write it. Does reading and writing connect cognition and language in a way that language itself does not?

In pondering about this thread further, I threw out the following on Twitter:

Is working memory a component of the executive function construct? Or an inter-related but separate domain?

I got some great food for thought in response to this query — Corey Peltier, Courtney Ostaff, and Andrew Watson confirmed that working memory is typically understood as a component of executive function — the cognitive system of thought that would appear to be distinct from language.

Lisa Archibald then went in deep on the relation between working memory and language, and it’s worth digging into her specific points, as they bear challenges to some of the points made above in the earlier thread.

Key points she makes that I found very helpful:

• What is activated and therefore measured depends on the nature of the task
• Whether the brains scanned are children or adults matters, as adult brains are more specialized
• Just as with emerging reading/writing skills, language development requires more cognitive attention until we are fluent
• And similar to struggling readers and writers, students struggling with language (i.e. DLD / SLI) have to apply more cognitive energy to using language accurately, which makes meaning/content/thinking harder to get to

She also referred me to another thread from DLD and Me that gives a neat way of framing this as unity but diversity — i.e. there is a single pool of resources of executive function (unity) but there is a diversity of different types of tasks we’re trying to apply that pool of resources to

Whew! This is heady stuff. Share your thoughts and Discuss...!

#language #cognition #DLD #workingmemory #executivefunction #literacy #thought #brain

In my last post, we looked at a wonderful paper, “Universals in Learning to Read Across Languages and Writing Systems“, that outlines operating principles of reading and writing across languages, as well as some key variations. Continuing on this theme, I wanted to highlight another recent paper, “The universal language network: A cross-linguistic investigation spanning 45 languages and 11 language families.”

The project is cool — the researchers have started a cross-linguistic database of brain scans, and their initial findings demonstrate a strong universal neural basis for language across multiple languages. Here’s the key finding that stood out to me:

In summary, we have here established that several key properties of the neural architecture of language—including its topography, lateralization to the left hemisphere, strong within network functional integration, and selectivity for linguistic processing—hold across speakers of diverse languages spanning 11 language families; and the variability we observed across languages is lower than the inter-individual variability. The language brain network therefore appears well-suited to support the broadly common features of languages, shaped by biological and cultural evolution. (Ayyash et al., 2021)

I found out about this paper from this Twitter thread from one of the researchers, Ev Fedorenko, and her thread also provides a neat summary of the project.

As this database of brain scans across languages is built out, it will be interesting to see what specific variations between languages and neural architecture may arise. For example, another recent paper, “Difference Between Children and Adults in the Print-speech Coactivated Network,” examined the brain scans of native Chinese speakers and found some variations from past studies in the brains of developing readers, most likely due to the difference in writing systems in terms of the lack of grapheme-phoneme correspondence for Chinese characters, as well as how a single pronunciation can have many different meanings represented by different visual characters.

Taken together, our findings indicate that print-speech convergence is generally language-universal in adults, but it shows some language-specific features in developing readers. (He et al., 2021)

Overall, it’s fascinating to see how current research converges on the significant universality across languages in terms of how literacy develops, and exciting to see that specific differences between languages and writing systems are beginning to be studied with greater specificity.

As Perfetti and Verhoeven tidily pointed out in their paper:

The story of learning to read thus is one of universals and particulars: (i) Universals, because writing maps onto language, no matter the details of the system, creating a common challenge in learning that mapping, and because experience leads to familiarity-based identification across languages. (ii) Particulars, because it does matter for learning how different levels of language – morphemes, syllables, phonemes – are engaged; this in turn depends on the structure of the language and how its written form accommodates this structure. (Verhoeven & Perfetti, 2021)

#speech #language #literacy #universal #reading #multilingualism #orthography #brain #neuroscience #research

Discuss...