The library is applicable to both invasive and non-invasive recordings, including electroencephalography (EEG), electrocorticography (ECoG) and magnetoecnephalography (MEG).
stimulus: wavefiles containing continuous speech in addition to time-aligned labels of phonemes
preprocessing:
- EEGExtract2to15.m
- EcogExtractHighGamma.m
- CUprocessingHilbertTransform_filterbankGUI.m
Phoneme analysis:
- findavg.m
- attribute2phoneme.m
- phoneme2attribute.m
- SimMat.m
- Fratio.m
- DistDistCov.m
- Dendro.m
- DPrime.m
- MDSPlot.m
GUI
- OutGenerator.m
- PhonemeAnalyser.m
Reconstruction:
- LagGeneratorNew.m
- StimuliReconstruction.m
Basic citation:
- NAPLIB: AN OPEN SOURCE TOOLBOX FOR REAL-TIME AND OFFLINE NEURALACOUSTIC PROCESSING
Matlab:
- Khalighinejad, Bahar, Guilherme Cruzatto da Silva, and Nima Mesgarani. "Dynamic Encoding of Acoustic Features in Neural Responses to Continuous Speech." Journal of Neuroscience 37.8 (2017): 2176-2185.
PSI vector, selectivity to phonemes:
- Mesgarani, Nima, et al. "Phonetic feature encoding in human superior temporal gyrus." Science 343.6174 (2014): 1006-1010.
Reconstruction:
- Mesgarani, Nima, et al. "Influence of context and behavior on stimulus reconstruction from neural activity in primary auditory cortex." Journal of neurophysiology 102.6 (2009): 3329-3339.