Category: Lab News

We are pleased to share a new publication by Adam, Bingzi, and Youngah in Linguistics Vanguard: “Conservativity in the Scope of Phonological Generalization Learning.”

The study investigates how people learn phonological patterns, or regularities in the sound systems of languages. In many languages, sound changes known as phonological alternations occur when words combine with affixes, often reflecting broader restrictions on which sound sequences are allowed in the language. Because of this relationship, many theories of phonology assume that knowledge of alternations and knowledge of phonotactics (sound-pattern restrictions within words) are closely linked.

These findings suggest that learners are conservative when extending newly learned phonological generalisations to unfamiliar morphological contexts. Together with previous research, the results indicate that the relationship between phonological alternations and phonotactic knowledge may not be as tightly connected in learning as some theoretical models have proposed.

Chong, A. J., Yu, B., & Do, Y. (2026). Conservativity in the scope of phonological generalization learning. Linguistics Vanguard. open_in_new DOI

We are pleased to share a new publication by Xiaoyu, Yuyan, and Youngah in Language and Speech.

The paper, titled “Exploring Substantive Bias Through Incidental Phonological Learning: Accuracy and Conscious Status of the Resultant Knowledge,” investigates how substantive bias emerges in phonological learning by comparing the acquisition of phonetically natural patterns (vowel harmony) versus unnatural ones (vowel disharmony).

To address why many previous studies have struggled to find evidence for this bias, the team introduced a disguise task to minimize intentional learning. Moving beyond traditional accuracy rates, the study also comprehensively measured the conscious status of the participants’ resultant knowledge. The results showed that while overall learning rates were low across the board, participants exposed to the “unnatural” vowel disharmonic items actually had a higher likelihood of possessing conscious judgment knowledge compared to the harmonic group. The authors attribute this to the possibility that disharmonic items are perceptually more distinctive in a strictly incidental learning environment.

Overall, the findings demonstrate that substantive bias may not always manifest in basic accuracy rates, but rather through differences in knowledge consciousness, offering a fresh perspective on how we evaluate and understand the acquisition of phonological patterns.

Yu, X., Xue, Y., & Do, Y. (2026). Exploring substantive bias through incidental phonological learning: Accuracy and conscious status of the resultant knowledge. Language and Speech. open_in_new DOI

We are pleased to share a new publication by Shuhao and Youngah examining how predictability shapes sound discrimination in speech perception. Their paper, “Roles of predictability and acoustic distance in sound discrimination via contrastive learning,” is published in the proceedings of SCiL 2026 of the Association for Computational Linguistics.

The study investigates how predictability affects sound discrimination using a supervised contrastive learning framework, a machine learning approach that learns to distinguish between similar inputs by comparing them. The authors vary levels of predictability to examine whether its impact on discrimination is gradual or categorical, and they also explore how this effect interacts with acoustic distance (the degree of difference between sounds) and the presence of additional contrasts in a language.

The results show that only fully predictable sound patterns significantly reduce discrimination performance, suggesting a categorical effect rather than a gradual one. However, this reduction in sensitivity diminishes as the acoustic distance between sounds increases. In addition, the presence of other sound contrasts that share the same acoustic dimension improves discriminability, highlighting the importance of broader linguistic context in shaping speech perception.

Zhang, S., & Do, Y. (2026). Roles of predictability and acoustic distance in sound discrimination via contrastive learning. In Proceedings of the Society for Computation in Linguistics 2026 (pp.477–487). Association for Computational Linguistics. open_in_newDOI

We report a new publication by Frank and Youngah titled “The development of spectral and temporal encodings in speech sounds,” published in the Proceedings of the Society for Computation in Linguistics 2026 of the Association for Computational Linguistics.

In this study, the authors investigate how humans distinguish speech sounds by examining two key types of information: spectral properties (the frequency-based characteristics that define phonemes) and positional information (where sounds occur within a sequence). While prior neuroscience and behavioural research has shown that humans can process both, the developmental trajectory of these encodings remains unclear.

The study evaluates how representations learned by the model evolve over time using ABX discrimination tests, a method commonly used to assess perceptual similarity. The results show that the model develops a strong ability to distinguish spectral features, aligning with findings from neuroscience on auditory processing. In addition, the model demonstrates independent encoding of positional information, evidenced by its accurate temporal discrimination of speech sounds.

Tan, F. L. H., & Do, Y. (2026). The development of spectral and temporal encodings in speech sounds. In Proceedings of the Society for Computation in Linguistics 2026 (pp.113–126). Association for Computational Linguistics. open_in_newDOI

We report a new publication by Bonnie, Youngah, Arthur, and John in Topics in Cognitive Science. The paper titled “An expanded model for perceptual norming: Insights from Japanese ideophones.” It investigates how sensory experience is encoded in Japanese ideophones, a class of vivid words often described as sound‑symbolic.

Using perceptual strength ratings across 13 sensory dimensions, the study moves beyond the traditional five‑ or six‑sense model. The results show that so‑called visual dominance is driven mainly by movement, while other visual properties such as shape and light or colour behave differently and connect to other senses in distinct ways. For example, movement patterns with sound and internal bodily sensations, shape with touch, and colour with taste and smell. The study also finds meaningful structure within interoception, with pain separating from emotions and bodily feelings.

Overall, the findings demonstrate that finer‑grained sensory models reveal cross‑modal relationships that are hidden in coarser approaches, highlighting the value of expanded perceptual norming for the study of iconicity and meaning.

McLaren, B., Do, Y., Thompson, A.L., & Husman, J. (2026). An expanded model for perceptual norming: Insights from Japanese ideophones. Topics in Cognitive Science. open_in_new DOI