🧠 neuroscience

The Interplay between Language Development, Short-Term Memory, and Auditory Associative Word Learning in Younger and Older Children

This study reveals that auditory associative word learning remains a challenging task for children beyond infancy, showing age-related declines in early learning effects and a specific reliance on visual-nonverbal short-term memory and musical training for successful acquisition in older children.

Original authors: Cosper, S. H., Bachmann, L., Sehmer, E., Steidel, A., Li, S.-C.

Published 2026-02-17

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Cosper, S. H., Bachmann, L., Sehmer, E., Steidel, A., Li, S.-C.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The Big Question: Why is it hard to learn "sound words"?

Imagine you are a baby. You hear a dog bark, and someone says "Dog." You instantly connect the sound of the bark with the word. It's easy! Babies are like super-sponges for sounds; they can learn to link a weird noise (like a door creaking) with a made-up word (like "Zorp") very quickly.

But as we grow up into adults, this ability seems to vanish. If you try to teach an adult that a specific door creak means "Zorp," they struggle. They are much better at learning that a picture of a door means "Zorp."

The Mystery: At what point do we lose that "baby superpower" for sounds? Do 5-year-olds still have it? Do 9-year-olds lose it? And what parts of our brain (like memory or language skills) help or hurt us?

The Experiment: A Game of "Sound Match"

The researchers set up a game to find out. They tested two groups of kids:

The "Little Explorers" (5–6 years old)
The "Big Explorers" (9–10 years old)

The Game:
The kids sat in a quiet room wearing a special cap with sensors (EEG) to measure their brainwaves.

The Training: They heard a sound (like a car horn) followed by a fake word (like "Bloop"). Sometimes the car horn always meant "Bloop" (Consistent). Sometimes it was random (Inconsistent).
The Test: Later, they heard the car horn and had to guess if the word "Bloop" was the right match.
The Bonus: They also played memory games (remembering sounds, words, and pictures) and took a language test.

What They Found: The "Sound Wall"

Here is the surprising part: Both groups struggled.

The Little Explorers (5–6): They couldn't really learn the sound-word pairs. Their brains showed a tiny flicker of understanding early on, but it faded away. They couldn't guess the right answers better than random chance.
The Big Explorers (9–10): They did slightly better at guessing, but their brains didn't show the "aha!" electrical signal (called the N400) that usually happens when someone learns a new word.

The Analogy: Think of learning these sound-words like trying to catch a slippery fish in a dark room.

Babies are like expert fishermen with a net; they catch the fish easily.
Adults are like people trying to catch the fish with their bare hands in the dark; it's nearly impossible.
The Kids in this study (5 to 10): They are in the middle. They aren't using a net anymore, but they haven't quite learned how to use their hands effectively yet. They are stuck in the "slippery fish" zone.

The Twist: The "Musical Superpower"

There was one group that showed a spark of learning: The 9-10-year-olds who played musical instruments.

The Musicians: Their brains showed a different kind of electrical signal when they heard the sounds. It wasn't the standard "I learned this!" signal, but it showed their brains were processing the sounds differently, perhaps because their ears are trained like high-definition microphones.
The Non-Musicians: They struggled just like everyone else.

The Metaphor: Musical training is like giving your brain a tuning fork. It helps you hear the subtle differences between sounds that other people miss. Even though they didn't "win" the game perfectly, their brains were working in a more sophisticated way.

The Secret Weapon: Visual Memory

The researchers also looked at what made the kids better at the game. They found a surprising link:

Good at remembering pictures = Good at learning sound-words.
Good at remembering sounds = Not necessarily better at learning sound-words.

The Analogy: Imagine your brain is a library.

When you try to learn a new sound-word, your brain tries to file it away.
It turns out, the kids who were good at filing away pictures (visual memory) were better at filing away sounds.
It's as if the "Visual Filing System" is so strong and organized that it helps the "Sound Filing System" work better, even though they are different types of files.

Why Does This Happen? (The "Why" Behind the "What")

The paper suggests a few reasons why learning sound-words gets hard as we grow:

The "Visual Takeover": As we get older, our brains start to prefer pictures over sounds. We rely on what we see more than what we hear. It's like the brain decides, "Why listen to a noise when I can just look at a photo?"
The "Brain Construction Site": Babies have a massive number of connections between brain cells (synapses). It's like a construction site with thousands of workers building bridges everywhere. As we get older, the construction slows down, and the workers get more specialized. We lose the ability to build random bridges between sounds and words as easily.
The "Noise" Problem: Our brains are busy. When we try to learn a sound-word, our brain might be trying to imagine the sound (auditory imagery) at the same time, which creates "static" or interference, making it hard to learn the new link.

The Bottom Line

This study tells us that learning to link sounds to words is actually quite hard for children, even more so than we thought. It's not just a baby thing; it's a skill that fades away before we even reach adulthood.

Musical training helps tune the brain to hear better.
Visual memory is the secret helper that supports sound learning.
Language skills generally get better with age, but they don't automatically fix the difficulty of learning sound-words.

In short: If you want to teach a child a new word based on a sound, don't just play the sound. Show them a picture, maybe play some music, and be patient—their brains are working hard to figure out the connection!

1. Problem Statement and Research Gap

The study addresses a critical developmental transition in language acquisition: the shift from successful auditory associative word learning in infancy to the significant difficulties adults face in the same task.

Background: Infants (10–12 months) can successfully map novel pseudowords to environmental sounds (auditory-associative learning). Conversely, young adults struggle with this sequential task, often requiring visual cues or simultaneous presentation to succeed.
The Gap: It is unclear when and why this shift occurs. Previous research suggests factors like sensory dominance (shifting from auditory to visual), short-term memory (STM) capacity, and language development play roles, but these have not been systematically tested in children to pinpoint the transition point.
Objective: To investigate auditory associative word learning in two distinct age groups (5–6 years and 9–10 years) to determine if they resemble infant-like success or adult-like difficulty. The study also explores the influence of STM (auditory-verbal, auditory-nonverbal, visual-nonverbal), language development, sex, and musical training on learning outcomes.

2. Methodology

Participants

Sample: 93 children total, divided into two groups:
- Younger Group (5–6 years): $N=46$ (22 males, 24 females).
- Older Group (9–10 years): $N=47$ (24 males, 23 females).
Inclusion Criteria: Native German speakers, right-handed, normal hearing/vision, no neurological conditions.
Musical Training: Defined as ongoing training for at least 1.5 years (4 children in the younger group; 19 in the older group).

Experimental Tasks

Participants completed five tasks over two visits:

Auditory Associative Word Learning (AAWL) Task (EEG):
- Stimuli: 16 environmental sounds paired with 16 disyllabic pseudowords.
- Design: Sequential presentation (Sound $\rightarrow$ 600ms pause $\rightarrow$ Pseudoword).
- Phases:
  - Training: 128 trials. 8 consistent pairings (repeated 8 times) and 64 inconsistent pairings (unique combinations).
  - Testing: 32 trials (16 matching, 16 violated pairings). Participants performed a forced-choice button press.
- Measurement: Event-Related Potentials (ERPs) focusing on the N400 component (400–800 ms) and a later window (800–1200 ms).
Short-Term Episodic Recognition Memory (STERM) Tasks:
- Three behavioral memory tasks (Presentation $\rightarrow$ $\to$ Recognition):
  - Auditory-Nonverbal (environmental sounds).
  - Auditory-Verbal (German words).
  - Visual-Nonverbal (pictures).
Language Development Assessment:
- SET 5-10: A standardized diagnostic battery assessing vocabulary, semantics, processing speed, comprehension, production, morphology, and auditory abilities.

Data Analysis

EEG: Preprocessing included filtering (0.3–30 Hz), ICA for artifact removal, and epoching. Statistical analysis used cluster-based permutation tests (FieldTrip) and Linear Mixed-Effects Models (lmeEEG) to compare conditions (consistent vs. inconsistent; matching vs. violated) across age, sex, and musical training.
Behavioral: Signal detection theory was applied. Due to response biases in children, corrected hit-rates were calculated rather than standard d-prime scores.
Correlations: Pearson correlations and LMMs were used to link STM, language scores, and demographic variables to AAWL performance.

3. Key Results

Behavioral Findings

Word Learning (AAWL):
- 5–6-year-olds: Performed at chance level (50.2%) in the testing phase. No evidence of successful associative learning.
- 9–10-year-olds: Performed significantly above chance (57.5%).
- Comparison: Older children significantly outperformed younger children, contradicting the hypothesis that younger children would show "infant-like" superior learning.
Short-Term Memory (STERM):
- Both age groups performed above chance in all three memory modalities.
- Visual-Nonverbal memory was the strongest modality for both groups.
- Age Effect: The 9–10-year-olds significantly outperformed the 5–6-year-olds in all three memory tasks.
Correlations:
- Visual-Nonverbal STM was the only memory measure positively correlated with AAWL performance.
- Language Development (SET score) was a significant predictor of overall behavioral performance but did not correlate with specific task scores.
- Musical Training: Non-musically trained 9–10-year-olds showed a trend-level N400 effect, while musically trained children showed different ERP patterns (LPC-like) but no behavioral advantage.

Electrophysiological (EEG) Findings

Training Phase:
- 5–6-year-olds: Showed a late negativity (822–1006 ms) in the first half of training, suggesting early sensitivity to consistent pairings. However, this effect disappeared in the second half.
- 9–10-year-olds: No significant group-level effects. However, musically trained children showed a significant positive cluster (LPC-like) in the second half of training, distinct from the non-musical group.
Testing Phase:
- No N400 Effects: Neither age group exhibited a significant N400-like response to violated pairings in the testing phase.
- Trend: Only non-musically trained 9–10-year-olds showed a trend-level negative response (400–516 ms), but it was not statistically significant.

4. Key Contributions

Developmental Trajectory: The study challenges the assumption that children retain "infant-like" ease in auditory associative learning. It suggests that the difficulty in sequential auditory word learning emerges early (by age 5–6) and persists into adulthood, rather than shifting gradually between ages 5 and 10.
Modality Specificity: It highlights that visual-nonverbal memory is a stronger predictor of auditory word learning success than auditory memory itself, suggesting a cross-modal reliance or that visual imagery aids in maintaining auditory associations.
Musical Training Nuance: While musical training altered neural processing (inducing LPC-like effects in the training phase), it did not translate to improved behavioral word learning in children, contrasting with findings in adults.
Methodological Rigor: The use of corrected hit-rates to account for response bias in young children and the combination of EEG with multiple STM modalities provides a robust framework for studying word learning in developing populations.

5. Significance and Implications

Theoretical Shift: The results suggest that the "window" for successful sequential auditory associative learning (as seen in infants) may close before age 5, possibly due to changes in synaptic density or the maturation of multisensory integration mechanisms that favor visual dominance.
Educational Context: The difficulty children face in this task implies that teaching methods relying solely on sequential auditory labeling (sound $\rightarrow$ word) may be less effective for children than methods utilizing visual support or simultaneous presentation.
Future Directions: The authors propose that future research should investigate younger children (2–4 years) to capture the transition point, explore the role of synaptic density in learning capacity, and test different stimulus synchronies (simultaneous vs. sequential) to see if they can rescue learning in older children.
Limitations: The study acknowledges that the fixed task order (AAWL first) may have induced fatigue or bias in younger children, and the definition of musical training was broad for the younger cohort.

In conclusion, the paper demonstrates that sequential auditory associative word learning is a cognitively demanding task for children aged 5–10, similar to adults, and is significantly supported by visual memory capabilities rather than auditory memory or language development alone.