ELSSE Project

ELSSE Project
Eliciting language-specific signatures for expertise in the human brain

Principal Investigator: Dr Hilary Wynne
External Partner: Mr. Henry Emery, Latitude Aviation Services

This project has received funding through the University of Oxford/OUP’s John Fell Fund (2025-2026) and a British Academy/Leverhulme Small Research Grant (2025-2027)

The Question  

What does it mean to be an expert in terms of language? Using a specialised language (many which feature small vocabularies and

Pilots in the cockpit

restricted syntax) is often an integral part of being an expert in a field (e.g. medicine, aviation). Expertise comes, in part, from prior knowledge gained through training and experience. A growing body of research has focused on differences between experts and non-experts; however, evidence for the contribution of language to expertise is scarce. This collaborative project uses psycholinguistic and neurolinguistic experimentation to investigate how experts employ sentential contextual cues to process violations in specialised language.

News and Events

ICAEA 2025 conference: We will be giving a workshop entitled ‘Collaboration in emerging research technologies can provide access to wider expertise: Exploring the effects of non-standard phraseology with neurolinguistic tasks’ at the International Civil Aviation English Association’s next conference ‘Back to the future for Aviation English training: Current best practices and emerging technologies’ in Mexico City (6-8 October 2025).
Take part in a simulated experiment (via a mobile phone, tablet or computer), experience a hands-on introduction to some of the technologies we use, and take part in a discussion about how collaborations between subject matter experts, trainers, and academics could lead to faster advancements in the industry and field.

The Goals

•    To elicit crucial evidence that “the expert brain” is cued to use special (expert) information when processing specialised language, and that violations in the standard (expected) language have deliberate and compelling effects in these listeners.
•    By determining whether extremely constrained contexts result in greater benefits or greater costs for word processing, we will be able to inform the literature more clearly regarding the claims made for the role of context in the processing of normal language.
•    The topic of how to effectively train pilots and controllers is one that is at the forefront of discussions in the industry: incorrect phraseology is not only a problem for efficiency of operations but also an issue of safety – particularly when non-native speakers are involved.
•    If we can establish precisely what these violations do to the comprehender, we can demonstrate why it is so important to use standardised language in the field.

The Research

The following streams are currently funded and under investigation:

Stream 1: Eliciting Evidence of Violations to Expert Language in the Field of Aviation
Our initial stream examines the effects of hearing a violation to standard aviation phraseology through (1) human performance on behavioural psycholinguistic tasks and (2) measurements of brainwave activity during such tasks. 1) includes word recall tasks, judgement tasks, and signal detection paradigms. For 2), we use an electro-encephalograph (EEG) to record the electrical signals generated by the brain while participants listen to simulated ATC instructions. The goal of this research is to establish a working paradigm for future experiments, and to establish that violations to standard phraseology result in a specific brain response.

Preliminary behavioural response data (N=120) has established differences in the processing of specialised language by experts and non-experts: In this task, which combined sentence comprehension with an additional word recall task:
•    Non-experts were only able to perform the recall task correctly when the language they heard (e.g. Climb and *STAY AT four thousand feet) was non-standard (i.e. more natural).
•    Experts (pilots) were able to process standard phraseology (e.g. Climb and MAINTAIN four thousand feet) better than violations; when a violation was heard, this in turn made the word recall task much harder to perform.

Stream 2: Graded Violations to Expert Language in the Field of Aviation
Using the same tasks as above, this stream examines the effects of different types of violation to standard aviation phraseology. Research (Hagoort 2003; Neville et al. 1991) has shown that within-category violations elicit a smaller response than the between-category violations. This is in line with the prediction-based account of context effects. In this stream we will compare responses to graded violations (e.g. Remain this FREQUENCY/ *CHANNEL/ *PEPPER).