Prediction of attention in autism from single-trial EEG using artificial neural networks

Two-layer and three-layer feed-forward artificial neural networks were trained to predict behvioural performance from singletrial EEG in autistic and normal subjects in a task involving response to rare stimuli and shifting of attention between vision and audition. Eyeblink artefacts were removed from the data using a frequency-domain filter. Performances of the networks on separate test sets varied across subjects but were usually at least 80%. The networks usually converged faster and attained a somewhat greater level of performance when input was presented in the frequency domain instead of in the time domain. Analysis of the network's failures in classifying the autistic auditory data turned up a variable response in which N270 and P700 were only occasionally present.

[1]  A J Gabor,et al.  Automated interictal EEG spike detection using artificial neural networks. , 1992, Electroencephalography and clinical neurophysiology.

[2]  T. Gasser,et al.  The transfer of EOG activity into the EEG for eyes open and closed. , 1985, Electroencephalography and clinical neurophysiology.

[3]  R Galambos,et al.  Autism: processing of novel auditory information assessed by event-related brain potentials. , 1984, Electroencephalography and clinical neurophysiology.

[4]  L. Kanner.  Autistic disturbances of affective contact. , 1968, Acta paedopsychiatrica.

[5]  B.H. Jansen.  Artificial neural nets for K-complex detection , 1990, IEEE Engineering in Medicine and Biology Magazine.

[6]  P. Lang,et al.  The effects of eye fixation and stimulus and response location on the contingent negative variation (CNV). , 1973, Biological psychology.

[7]  E. Donchin,et al.  Talking off the top of your head: toward a mental prosthesis utilizing event-related brain potentials. , 1988, Electroencephalography and clinical neurophysiology.

[8]  Jordan Grafman,et al.  Atypical cognitive deficits in developmental disorders : implications for brain function , 1994 .

[9]  Eric Courchesne,et al.  Parietal Damage and Narrow Spotlight Spatial Attention , 1994, Journal of Cognitive Neuroscience.

[10]  J Townsend,et al.  Impairment in shifting attention in autistic and cerebellar patients. , 1994, Behavioral neuroscience.

[11]  J. C. Woestenburg,et al.  The removal of the eye-movement artifact from the EEG by regression analysis in the frequency domain , 1983, Biological Psychology.

[12]  Terrence J. Sejnowski,et al.  Neural Network Analysis of Event Related Potentials and Electroencephalogram Predicts Vigilance , 1991, NIPS.

[13]  R. M. Siegel,et al.  Pattern recognition of the electroencephalogram by artificial neural networks. , 1993, Electroencephalography and clinical neurophysiology.

[14]  Eric Courchesne,et al.  Recent advances in autism , 1990 .

[15]  S. Makeig.  Auditory event-related dynamics of the EEG spectrum and effects of exposure to tones. , 1993, Electroencephalography and clinical neurophysiology.

[16]  F Y Wu,et al.  A neural network design for event-related potential diagnosis. , 1993, Computers in biology and medicine.

[17]  I. Bankman,et al.  Feature-based detection of the K-complex wave in the human electroencephalogram using neural networks , 1992, IEEE Transactions on Biomedical Engineering.