EEG Error Prediction as a Solution for Combining the Advantages of Retrieval Practice and Errorless Learning

Area of Science:

• Neuroscience
• Speech and Language Therapy
• Cognitive Rehabilitation

Background:

• Naming errors are frequent in aphasia, necessitating effective speech and language rehabilitation.
• Errorless learning shows promise for naming therapy, but retrieval practice may be crucial for long-term gains.
• Brain state, including arousal and focus, influences task performance and can be monitored via electroencephalography (EEG).

Purpose of the Study:

• To investigate the prediction of naming errors in aphasia using EEG spectral features.
• To develop a system for real-time monitoring of patient brain state during therapy.
• To correlate EEG responses with accuracy and error correction during object-naming tasks.

Main Methods:

• Eight participants with aphasia completed 900 object-naming trials across three sessions.
• Electroencephalography (EEG) was recorded, and response accuracy was scored for each trial.
• EEG spectral features were analyzed to identify correlations with naming accuracy and error correction.

Main Results:

• Significant correlations between EEG features and response accuracy/error correction were found for seven of eight participants.
• A predictive model generalized to unseen data, forecasting accuracy for the third session in seven participants.
• The model also predicted error correction categories in five participants, demonstrating robust generalization.

Conclusions:

• Individual EEG responses can predict error production during naming tasks in aphasia.
• Real-time error prediction could enable timely interventions, optimizing therapy strategies.
• This approach may integrate the benefits of errorless learning and retrieval practice for enhanced aphasia rehabilitation.