Dyslexia Diagnosis by EEG Temporal and Spectral Descriptors: An Anomaly Detection Approach

Diagnosis of learning difficulties is a challenging goal. There are huge number of factors involved in the evaluation procedure that present high variance among the population with the same difficulty. Diagnosis is usually performed by scoring subjects according to results obtained in different neuropsychological (performance-based) tests specifically designed to this end. One of the most frequent disorders is developmental dyslexia (DD), a specific difficulty in the acquisition of reading skills not related to mental age or inadequate schooling. Its prevalence is estimated between 5% and 12% of the population. Traditional tests for DD diagnosis aim to measure different behavioral variables involved in the reading process. In this paper, we propose a diagnostic method not based on behavioral variables but on involuntary neurophysiological responses to different auditory stimuli. The experiments performed use electroencephalography (EEG) signals to analyze the temporal behavior and the spectral content of the signal acquired from each electrode to extract relevant (temporal and spectral) features. Moreover, the relationship of the features extracted among electrodes allows to infer a connectivity-like model showing brain areas that process auditory stimuli in a synchronized way. Then an anomaly detection system based on the reconstruction residuals of an autoencoder using these features has been proposed. Hence, classification is performed by the proposed system based on the differences in the resulting connectivity models that have demonstrated to be a useful tool for differential diagnosis of DD as well as a method to step towards gaining a better knowledge of the brain processes involved in DD. The results corroborate that nonspeech stimulus modulated at specific frequencies related to the sampling processes developed in the brain to capture rhymes, syllables and phonemes produces effects in specific frequency bands that differentiate between controls and DD subjects. The proposed method showed relatively high sensitivity above 0.6, and up to 0.9 in some of the experiments.

Neurophysiological Indices of Speech and Nonspeech Stimulus Processing

Auditory event-related potentials (mismatch negativity and P300) and behavioral discrimination were measured to synthetically generated consonant-vowel (CV) speech and nonspeech contrasts in 10 young adults with normal auditory systems. Previous research has demonstrated that behavioral and P300 responses reflect a phonetic, categorical level of processing. The aims of the current investigation were (a) to examine whether the mismatch negativity (MMN) response is also influenced by the phonetic characteristics of a stimulus or if it reflects purely an acoustic level of processing and (b) to expand our understanding of the neurophysiology underlying categorical perception, a phenomenon crucial in the processing of speech. The CVs were 2 within-category stimuli and the nonspeech stimuli were 2 glides whose frequency ramps matched the formant transitions of the CV stimuli. Listeners exhibited better behavioral discrimination to the nonspeech versus speech stimuli in same/different and oddball behavioral paradigms. MMN responses were elicited by the nonspeech stimuli, but absent to CV speech stimuli. Larger amplitude and earlier P300s were elicited by the nonspeech stimuli, while smaller and longer latency P300s were elicited by the speech stimulus contrast. Results suggest that the 2 types of stimuli were processed differently when measured behaviorally, with MMN, or P300. The better discrimination and clearer neurophysiological representation of the frequency glide, nonspeech stimuli versus the CV speech stimuli of analogous acoustic content support (a) categorical perception representation at the level of the MMN generators and (b) parallel processing of acoustic (sensory) and phonetic (categorical) information at the level of the MMN generators.