iiinova — where your digital self keeps evolving

Of course, this turns up to be useful when learning that a 🐕is called chien in 🇫🇷 and perro in 🇪🇸! PS. This was also a unique opportunity of plugging in the title one of the most memorable quotes from pirates of the Caribbean 🏴‍☠️🦜, because why not! #devpsy #devpsych #bilingualism #language

010 0

...but the rest is compatible with simple statistical learning: bilinguals select less the novel object => form a weaker novel_object-novel_word association => poorer retention of that new word-object mapping, but are better at forming a second association to the already name-known objects

Proportion of target selection for monolinguals and bilinguals between tasks. Bilinguals select less the novel object, hence form a weaker novel_object-novel_word association, hence poorer retention of that new word-object mapping, but are conversely better at forming a second association to the already name-known objects

It turns out that a good explanation is that learning mechanisms are essentially the same, but that bilinguals are less prone to select the novel objects when hearing novel words than monolingual peers (we already knew that from Byers-Heinlein’s work)...

More like guidelines than hard rules: Bilinguals accept lexical overlap despite using mutual exclusivity

Do bilingual children differ from monolinguals when learning new words? Monolinguals tend to start learning one word for one object, but surely this doesn’t hold for bilingual children, right? A brief dive into our latest paper with @mysia1.bsky.social : www.sciencedirect.com/science/arti...

sciencedirect.com

Mutual exclusivity (ME), the tendency that young children have to select a novel object upon hearing a novel label, gives children a basis to select a…

292 0

Julien Mayor @julienmayor.bsky.social · 27/05/2026

I like having the luxury of having experts providing input on the method and analytical strategy before freezing the modelling plans in RRs, but this is definitely more useful when this takes place earlier, in the planning stages...

000 0

Julien Mayor @julienmayor.bsky.social · 27/05/2026

hi Sam, RRs with existing data can work - but typically you are being asked to apply a stricter level of control on your analyses - e.g., see rr.peercommunityin.org/PCIRegistere... section 2.6 (Table 1) for PCI-RR

rr.peercommunityin.org

PCI Registered Reports

Peer Community in Registered Reports

100 0

Julien Mayor @julienmayor.bsky.social · 22/05/2026

Kudos to Luis Muñoz for the 1st article of his PhD being in the top 10% articles in Developmental Science – showing negative associations between pacifier use and young children’s vocabulary size! onlinelibrary.wiley.com/doi/full/10.... (stay tuned for follow-up papers very soon!) 2/2.

onlinelibrary.wiley.com

Sustained pacifier use is associated with smaller vocabulary sizes at 1 and 2 years of age: A cross‐sectional study

Pacifier use during childhood has been hypothesized to interfere with language processing, but, to date, there is limited evidence revealing detrimental effects of prolonged pacifier use on infant v...

040 0

Julien Mayor @julienmayor.bsky.social · 22/05/2026

Happy to see the hard work of our students getting recognition! 🎉 Kudos to Akvile Sinkeviciute for her MSc thesis (!) ending up in the top10% articles in Language Learning, showing that bilinguals perceive colours differently depending on the language they engage in! doi.org/10.1111/lang... 1/2

onlinelibrary.wiley.com

<em>Language Learning</em> | Language Learning Research Club Journal | Wiley Online Library

A one-page Accessible Summary of this article in non-technical language is freely available in the Supporting Information online and at https://oasis-database.org.

140 0

Julien Mayor @julienmayor.bsky.social · 07/05/2026

looking forward to reading your novels!

Full paper: onlinelibrary.wiley.com/doi/10.1111/... Thanks to the amazing Dean & Hana D'Souza, Angel Tovar, who have pushed a project that started nearly 15 years ago!

030 0

This perspective challenges (old) modular views of language in WS and highlights the critical role of non-linguistic domains, specifically visual processing, in shaping linguistic outcomes.

100 0

In sum, the comprehension-production gap is not necessarily a universal or invariant feature of language development. Its magnitude can be shaped by the interaction of domain-general processing constraints and sensory input characteristics.

100 0

Our model suggests WS visual maps are “exemplar-based.” Instead of a shared category prototype, the system treats items as isolated instances (see the scattered visual organisation in our WS model, below). This explains why production stays high while comprehension remains shallow.

Object category organisation in the auditory and visual maps. While auditory maps in both our model of typically-developing children and of williams syndrome have compact categories, the visual maps organisation present in typically-developing children is disrupted in the williams syndrome model.

100 0

To find “why,” we used a neural network model based on Self-Organising Maps (SOMs). We discovered that WS-like patterns emerge from selective difficulties in visual processing, rather than enhanced verbal skills.

Self-organising map model comprising an auditory map and a visual map, connected by Hebbian connections. Presentation of auditory and visual exemplars typically activates the best-matching unit, along with a neighbourhood.

100 0

We compared vocabularies of 67 children with WS to typically developing (TD), Down syndrome (DS), and Fragile X groups. While others showed typical asymmetry, children with WS had disproportionately higher production relative to their comprehension level.

Vocabulary size in production as a function of comprehension. Trajectories for Williams Syndrome sit above that of typically-developing children, as well as of children with Down or Fragile X syndromes

100 0

Julien Mayor @julienmayor.bsky.social · 28/01/2026

Young children usually understand more words than they can say: a “comprehension–production gap”. In a new study, we found that children with Williams syndrome (WS) have a unique profile: their gap is significantly reduced compared to other groups. 🧵 #LanguageDevelopment #WilliamsSyndrome #devpsyc

153 0

thanks for the clarification!

100 0

Julien Mayor @julienmayor.bsky.social · 28/01/2026

If I am not mistaken, the calibration is from EB, but our gaze-contingent attention getters are triggered via a python script. @audunrosslund.bsky.social do I get this right?

100 0

Julien Mayor @julienmayor.bsky.social · 27/01/2026

Interesting! We never had any problem with calibration with Eyelink (babies from 5 months of age) and we use a gaze-contingent attention getter system when babies look away from the screen too long which works like a charm!

Julien Mayor @julienmayor.bsky.social · 27/01/2026

One of the big factor was that, at the time of purchase, Tobii's native software did not include the possibility of programming gaze-contingent designs (maybe they do now?).

150 0

Julien Mayor @julienmayor.bsky.social · 27/01/2026

We are still very happy with our 6-year-old EyeLink 1000 Plus!

140 0

Nota Bene: This has been another fun project – with the inevitable @audunrosslund.bsky.social and @nataliakartushina.bsky.social! As usual – the data is openly available (+ pre-registration + author version): osf.io/s8zrm/. #LanguageAcquisition #IDS #DevPsy

osf.io

OSF

020 0

Final Takeaway 🎯 Our findings challenge the idea that acoustic modulation in IDS is a direct lexical teaching strategy. IDS likely serves broader goals: sustaining attention, conveying affect, and social bonding.

231 0

Our Bayesian analyses provided decisive evidence against the didactic effect, suggesting that parents do not adapt their speech based on their child individual word’s attributed knowledge. Twist: vowel dispersion increased for known words with age! This may suggest reinforcement, not instruction 🤔

120 0

Mike Frank @mcxfrank.bsky.social · 20/06/2025

New paper challenging the didactic role of Infant-Directed Speech (IDS) We hypothesised that parents use IDS as a didactic tool by acoustically emphasising unknown words. Analysing 38k speech segments (182 parents of 8-18m olds) => NO evidence for a didactic strategy! doi.org/10.1037/dev0002077

163 0

Reposted by Julien Mayor

Is anyone talking about the big design flaws in this study, e.g. 1) small N, btwn subject design, 2) v. heterogeneous sample, 3) writing prompt confounded with condition (really!), 4) vast number of measures with no preregistration? I can't believe this is getting so much press...

79620 0

Julien Mayor @julienmayor.bsky.social · 22/05/2025

Oh boy! Second language learning is not my field, but establishing a time to learn will depend on (among many others), (a) your age, (b) proximity to your first language, (c) what you mean by "learn" or "know" a language, (d) intrinsic motivation to learn, (e) cognitive ability, (f), (g), etc, etc!

130 0

Reposted by Julien Mayor

simine vazire @simine.com · 08/05/2025

My editorial on how journals can earn trust. We often use journal names as proxies for quality. This is bad bc it’s not valid. But it could be. Editors could make journal name a valid signal. And we could place value on journals that show us how they do that. journals.sagepub.com/doi/full/10....

212343 0

Julien Mayor @julienmayor.bsky.social · 07/05/2025

It looks like there are lines with twice the slope - two-tailed versus one-tailed? I wonder if this constrains your regression line in a weird way...

Julien Mayor @julienmayor.bsky.social · 06/05/2025

Forget just birth order for early language! (it's kinda U-shaped anyway...) We introduce a *better* measure (capturing more variance): a household "child-to-caregiver" ratio, where old-enough siblings are considered caregivers, rather than being another child competing for parental resources. 1/

22611 0

Reposted by Julien Mayor

Timo B. Roettger @timoroettger.bsky.social · 20/03/2025

🚨🚨🚨 I'm hiring a Postdoc for a 3-year position to come work with me at the University of Oslo using iterative learning experiments to understand the evolution of sound symbolism. Please share widely. 🔴 Deadline is May 31st '25 ⏲️ Desired starting date is Fall '25 shorturl.at/vSTZx

shorturl.at

Postdoctoral Research Fellow connected to "ICONIC" (277523) | University of Oslo

Job title: Postdoctoral Research Fellow connected to "ICONIC" (277523), Employer: University of Oslo, Deadline: Saturday, May 31, 2025

17070 0

Julien Mayor @julienmayor.bsky.social · 24/02/2025

One more week to apply for the Associate Professor position in Developmental Psychology! Great opportunity to join us in Oslo - academic freedom and easy access to nature... #DevSci #devpsy #openscience