In high-stakes environments like operating rooms (ORs), healthcare professionals face intense cognitive demands. Also, environmental stressors that can contribute to significant work strain.
The complex soundscape of a typical OR is filled with alarms, equipment noises, and other ambient sounds. It presents a unique challenge, as these auditory elements can distract and increase cognitive load.
This study focuses on work strain in a simulated surgical environment using mobile EEG technology. The study focused on evaluating how varying cognitive demands affect auditory perception. Specifically by analyzing neural responses to soundscapes that an OR environment.
Performing surgery simulation and acquisition of mobile EEG
22 medical students participated in the study. During the experiment, they had a task to perform a laparoscopic surgery on a simulator.
Before the simulation, they were presented with a series of letters to memorize until the end of surgery. It was either 2 (low demand memory task) or 8 letters (high demand).
The reason authors chose this task as a distraction is because of previous research. It shows visual working memory abilities correlate with surgical performance. Also, the task didn’t interrupt the surgical procedure.
Through a speaker, researchers added some OR background noises such as ventilation, beeps, instruments etc. In addition to the realistic soundscape they presented a sequence of spoken letters. This was to increase work strain and add noise participants had to overcome and ignore.
In the end, participants also had to subjectively rate effort, frustration and distraction levels.
(a) Participant’s position between the simulation screen and letter-presenting screen. (b) One session explained. After presenting letters and surgery, participants are asked to insert the letter sequence they memorized. Also, subjective ratings of the block.During the entire procedure, experimenters are acquiring EEG using portable EEG device.
Analysis to understand EEG response to spoken words and OR soundscape
To determine responses to the spoken words, scientists extracted Event Related Potentials (ERPs). Derived ERPs are in response to the background noise letters. We’ll take a look at the results in the next chapter.
The study used TRF analysis to understand how continuous auditory stimuli (the OR soundscape) relate to ongoing EEG activity. TRF stands for Temporal Response Function.
To do this, we need to extract features of the audio signal. The chosen features are:
- Envelope of the raw OR playback;
- Envelope of the noise-reduced OR sounds;
- Onsets of the OR playback, when sound intensity sharply increased.
To predict EEG responses to each of these features, authors used a forward modeling approach. The model is computed individually for each participant. Check out the original paper to see how they trained and validated the model.
Once ERP and TRF are extracted, authors applied GED – generalized eigenvalue decomposition. It enhances SNR ratio and isolates components that reflect neural processes in response to audio. For each ERP and TRF time-window they computed a spatial filter across subjects and then apply it to the data. The GED component represents amplitude changes across the experimental conditions.
Finally, for each response they evaluated a condition difference using regression models. To be specific, they explored whether adding time as a predictor improved the model fit.
Two regression models (LMMs) are used: Model 1 included ’condition’ as a predictor, and model 2 ’condition’ and ’time’ as predictors.
You can see the selected models for each response in the table below.
Table shows the selected model for each subjective, behavioral, and neural response. ’Model’ lists the chosen model for each response. The stars indicate predictors below the Bonferroni-corrected p-value. A trend is marked with a dot.Perceived demand vs. Objective measures
Study found that participants felt more distracted during high demand (8 letters) task. But the objective measures of performance don’t show the difference between low vs. high. This highlights the need to assess both types of measures in high-stakes environments (such as ORs).
(a-c) Subjective ratings for low and high demand tasks. (d) Performance of the letters memory task (e-g) Surgery task performance. In solid lines, we can see the predicted values using regression models.Over time, participants improved in their surgical performance but didn’t report a reduction in task demand. That’s why in the table, we can see Model 1 was used for subjective measures, while Model 2 was used to predict surgical performance.
This indicates that, while people can perform better with practice, they might still perceive the task as demanding. It suggests that performance improvements alone might not fully ease work strain.
Neural adaptation over time and work strain management
EEG analysis revealed ERP adaptations in response to the repetitive spoken letters. Specifically, there was a decrease in N1 amplitude and an increase in P2 amplitude. This is likely due to neural adaptation to the ongoing stimuli.
Exploring time as a predictor significantly improved the models for the N1 and P2 amplitudes.
(a) Averaged first GED components for N1 ERP component. (b) Averaged first GED components for P2 component. In the main plot, we can see the waveform and the greyed area of N1 and P2 locations. In the left corner, topographies show which channels contributed most to the GED component. In the right corner, we also see the change in ERP amplitudes across time and conditions. Also, in solid lines we can see the predictions of used model 2.This neural adaptation may help individuals maintain focus despite ongoing distractions. This could suggest a potential mechanism for managing work strain.
TRF responses to OR soundscape didn’t significantly differ between conditions
Possible reason is that TRFs derived from the entire soundscape might not be sensitive enough to detect task demand effects.
TRF prediction accuracy was higher when using noise-reduced envelopes and sound onsets rather than raw audio. This could mean focusing on distinct auditory events (like beeps or tool sounds) may yield more sensitive measures.
EEG and soundscape interactions in applied settings
This research supports the use of mobile EEG to study cognitive processes in complex environments. The findings say that high performance can be maintained in demanding environments. But, there may be underlying costs to individuals’ well-being. This insight encourages further exploration of EEG to optimize auditory environments. Especially in settings where managing performance and work strain are critical, such as in ORs.
You can read the original study here: https://www.nature.com/articles/s41598-024-74946-9#Sec2
Curious if this setup is right for your study?
Our team works with researchers across neuroscience, sports science and clinical applications - we'll help you figure out if mobile EEG fits your study design.
