← Back
Academic Essay • Dec 2025

Emergent
Neurotechnologies,
AI, & Language

How EEG decoding and Large Language Models are challenging our understanding of how we acquire second languages.

Joonas Luukkonen

University of Eastern Finland (Joensuu)

01

Context

"Bridging psychological constructs like aptitude with measurable neural markers."

Key Question

If language can be decoded directly from neural signals, what is the role of the learner?

The New Frontier

The nature of language and the process of acquiring it are being reconsidered. Recent demonstrations, such as EEG-based inner-speech decoding and AI-mediated literacy, illustrate how language can be externally represented in ways that challenge traditional understandings.

These developments raise fundamental questions for Second Language Acquisition (SLA): if language can be decoded directly from neural signals or generated by AI without understanding, how do cognitive processes and individual differences interact with the biological substrates of language?

02

AI vs. Human Cognition

The Symbol Grounding Problem

LLMs demonstrate remarkable proficiency in generating coherent text but operate solely on statistical associations. They illustrate Floridi's concept of "agency without intelligence".

In contrast, human language is embodied. Our understanding is grounded in sensorimotor experiences. This distinction is crucial for SLA theory, which emphasizes mapping linguistic forms to real-world meanings.

Statistical Form
Embodied Meaning

Grounding Viz

Dataset
Stats
"01011"
Token
"Apple"
"Apple"
Ungrounded

Agency without Intelligence: LLMs manipulate symbols based on statistical probability without understanding the object the symbol refers to.

03

Inside the Brain

EEG-based inner-speech decoding reveals that silent rehearsal is biologically instantiated. The dual-stream prediction model suggests that efference copies generated during inner speech are transmitted to auditory cortex regions.

Social Learning Network

Social learning recruits regions associated with multimodal integration: STS, pMTG, IPL, and TPJ. These activations support deep-elaborative encoding.

Successful Learners

High aptitude learners exhibit more coherent and integrated multi-path networks connecting IFG, MFG, and STG even before training begins.

Dual-Stream Prediction Model

Inner Speech
Motor Plan
Efference Copy
Auditory Pred.

Network Topology

Novice
Expert

Implications for Pedagogy

Social Learning

Instruction must prioritize semantic and pragmatic grounding. Role-play and storytelling promote deep encoding better than rote translation.

Embodied Practice

Silent rehearsal, visualization, and gesture-supported learning strengthen the form-meaning link through sensorimotor pathways.

Personalized AI

Neural diagnostics allow for tailored instruction. AI tools can support learners' "Ideal L2 Self" if used to enhance, not replace, cognitive effort.

References

  • [1] Abdulghani, M. M., Walters, W. L., & Abed, K. H. (2023). Imagined Speech Classification Using EEG and Deep Learning. Bioengineering, 10(6), 649.
  • [2] Barsalou, L. W. (2008). Grounded cognition. Annual Review of Psychology, 59, 617-645.
  • [3] Bender, E. M., & Koller, A. (2020). Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 5185-5198.
  • [4] Dörnyei, Z. (2005). The Psychology of the Language Learner: Individual Differences in Second Language Acquisition. Lawrence Erlbaum Associates.
  • [5] Ellis, N. C. (2008). The dynamics of second language emergence: Cycles of language use, language change, and language acquisition. Modern Language Journal, 92(2), 232-249.
  • [6] Floridi, L. (2023). AI as Agency Without Intelligence: on ChatGPT, Large Language Models, and Other Generative Models. Philosophy & Technology, 36, 15.
  • [7] Glenberg, A. M., et al. (2008). The grounding of language in action and perception. Cambridge Handbook of Situated Cognition, 12-28.
  • [8] Jeong, H., Li, P., Suzuki, W., Sugiura, M., & Kawashima, R. (2021). Neural mechanisms of language learning from social contexts. Brain and Language, 212, 104874.
  • [9] Yang, J., Gates, K. M., Molenaar, P., & Li, P. (2015). Neural changes underlying successful second language word learning: An fMRI study. Journal of Neurolinguistics, 33, 29-49.