Physiologic Word Recognition from Cognitive State


Publications by a Stanford group support recognition of words from brain waves,[1] [2] [3] [4] [5] with recent enhanced success.  Other investigators publish above chance magnetoencephalographic (MEG) word recognition.[6]  EEG instant detection in syllables of “a content of category which the testee whishes to speak” quotes Kiyuna et. al. Patent # 5785653 “System and method for predicting internal condition of live body.” [7]  A stated use:  “the present invention may be use (sic) to detect the internal condition of surveillance in criminal investigation.”  NEC Corporation licensed this patent.  Remote EEG communication with Armed Forces or clandestine application is the cited use for Mardirossian Patent # 6011991 “Communication system and method including brain wave analysis and/or use of brain activity.” [8]   Technology Patents, LLC licensed the patent, which proposes transmitter capable skin implants, and utilizes neural networks (artificial intelligence.)  Publications unreferenced by current papers are a US technical report of prior results from Stanford comparable to the recent articles, dated 1975,[9] and an apparent Russian report before 1981.[10]  The direction of robots by detecting mental states such as imagining raising the directionally appropriate arm or spinning cubes is reported.[11] [12] [13]  Emotion differentiation by EEG is also patented, referencing Air Force research.[14] 

Theoretical framework for EEG word recognition reports based on the activation of brain cell assemblies is elaborated[15] for extensive publications of averaged EEG word category differentiation, also consistent with word recognition.  EEG differentiation of words rated as to affective meaning such as good-bad, strong-weak, or active-passive is reported.[16] [17]   Based on EEG/MEG responses, words are readily distinguished from non-words,[18] [19] [20] or pictures,[21] can be differentiated as to length, [22] and visual nouns can be differentiated from action verbs.[23] [24] [25] [26] [27]  Brain wave patterns distinguish proper names from common nouns, [28] animal names from numerals,[29] or content from function words.[30] [31] [32] [33]   Concrete versus abstract words,[34] and unambiguous versus ambiguous noun/verbs[35] have distinctive EEG patterns.   Face, arm, or leg action verbs are reported distinguished by brain waves as well.[36] [37] 

Similar category specificity is emerging from function magnetic resonance imaging (fMRI) studies.  Different fMRI brain activation loci for face, natural and manufactured object recognition are reviewed.[38]  Neural network differentiation of fMRI response to noun categories for fish, four legged animals, trees, flowers, fruits, vegetables, family members, occupations, tools, kitchen items, dwellings, and building parts is reported.[39]


[1] Suppes P, Lu Z, and Han B.  “Brain wave recognition of words” Proc Natl Acad Sci 94: 14965-69, 1997.  Printable free online thru Pubmed or at

[2] Suppes P, Han B, and Lu Z. “Brain-wave recognition of sentences” Proc Natl Acad Sci 95: 15861-66, 1998.  Printable free online thru Pubmed or at

[3] Suppes P, Han B, Epelboim J, and Lu Z. “Invariance of brain-wave representations of simple visual images and their names” Proc Natl  Acad  Sci 96: 14658-63, 1999.  Printable free online thru Pubmed or at

[4] Suppes P, Han B, Epelboim J, and Lu ZL. “Invariance between subjects of brain wave representations of language” Proc Natl Acad  Sci 96(22): 12953-8, 1999.  Printable free online thru PubMed or at

[5] Suppes P and Han B. “Brain-wave representation of words by superposition of a few sine waves” Proc Natl Acad Sci 97: 8738-43, 2000.  Printable free online thru Pubmed or at

[6] Assadullahi R and Pulvermuller F. “Neural Network Classification of Word Evoked Neuromagnetic Brain Activity” In: Wermter S, Austin J, and  Willahaw D (eds.) Lecture Notes in Artificial Intelligence: Emergent Neurocomputational Architechures Based on Neuroscience Heidelberg Springer, p 311-20, 2001.  More limited preliminary communication at

[7] Kiyuna T, Tanigawa T, and Yamazaki T. Patent #5785653 “System and method for predicting internal condition of live body” USPTO granted 7/28/98.

[8] Mardirossian A. Patent #6011991 “Communication system and method including brain wave analysis and/or use of brain activity” USPTO granted 1/4/00.

[9] Pinneo LR and Hall DJ. “Feasibility Study for Design of a Biocybernetic Communication System” Report #ADA017405 National Technical Information Service, 1975.  Prepared for the Advanced Research Projects Agency Order #2034, Program Code #2D20, Contractor: Stanford Research Institute Contract dates: 2/9/72-8/31/76, SRI Project LSU-1936. (US cost ~$50.)

[10]  Selden G. “Machines That Read Minds” Sci Digest Oct 89: 60-6, 1981. Also at

[11] Millan JR. “Adaptive Brain Interfaces” Communications of the ACM 46(3): 74-80, 2003.  Abstract at

[12] Millan JR and Mourifio J. “Asynchronous BCI and Local Neural Classifiers: An Overview of the Adaptive Brain Interfact Project” IEEE Transactions on Neural Systems and Rehabilitation Engineering (Brain-Computer Interface Technology) 11(2): 159-61, 2003.  Article also at

[13] Millan JR, Renkens F, Mourifio J, and Gerstner W. “Non-Invasive Brain-Actuated Control of a Mobile Robot” Proceedings of the 18th Joint International Conference on Artificial Intelligence Aug 9-15, in press, 2003.  Article also at

[14] Patton RE. Patent #6292688 “Method and apparatus for analyzing neurological response to emotion-inducing stimuli” USPTO granted 9/18/01.

[15] Pulvermuller F. “Words in the brain’s language” Behav Brain Sci 22: 253-336, 1999.

[16] Skrandies W. “Evoked potential correlates of semantic meaning—A brain mapping study” Cog Brain Res 6: 175-183, 1998.

[17] Skrandies W and Chiu MJ. “Dimensions of affective meaning – behavioral evoked potential correlates in Chinese subjects” Neurosci Lett 341: 45-8, 2003.

[18] Krause CM, Korpilahti P, Porn B, Joskim J, and Lang HA. “Automatic auditory word perception as measured by 40 Hz EEG responses” Encephal Clin Neurophysiol 107: 84-7, 1998.

[19] Diesch E, Biermann S, and Luce T. “The magnetic mismatch field elicited by words and phonological non-words” Neuroreport 9(3): 455-60, 1998.

[20] Lutzenberger W, Pulvermuller F, and Birbaumer N. “Words and pseudowords elicit distinct patterns of 30-Hz EEG responses” Neurosci Lett 176: 115-18, 1994.

[21] Kiefer M. “Perceptual and semantic sources of category-specific effects: Event-related potentials diring picture and word categorization” Mem Cog 29(1): 100-16, 2001.

[22] Assadollahi R and Pulvermuller F. “Neuromagnetic evidence for early access to cognitive represtentations” Cog Neurosci Neurophysiol 12(2): 207-13, 2001.

[23] Preissl H, Pulvermuller F, Lutzenberger W, and Birbaumer N. “Evoked potentials distinguish between nouns and verbs” Neurosci Lett 197: 81-3, 1995.

[24] Pulvermuller F, Mohn B, and Schleichert H. “Semantic or lexico-syntactic factors: what determines word-class specific activity in the human brain?” Neurosci Lett 275: 81-4, 1999.

[25] Pulvermuller F, Lutzenberger W, and Preissl H. “Nouns and Verbs in the Intact Brain: Evidence from Event-retlated Potentials and High-frequency Cortical Responses” Cerebral Cortex 9(5): 497-506, 1999.

[26] Pulvermuller F, Preissl H, Lutzenberger W, and Birbaumer N. “Brain Rhythms of Language: Nouns Versus Verbs” Eur J Neurosci 8: 917-41, 1996.

[27] Kellenbach ML, Wijers AA, Hovius M, Mulder J, and Mulder G. “Neural Differentiation of Lexico-Syntactic Categories or Semantic Features? Event-Related Potential Evidence for Both” J Cog Neurosci 14(4): 561-77, 2002.

[28] Muller HM and Kutas M. “What’s in a name? Electrophysiological differences between spoken nouns, proper names and one’s own name” Neuroreport 8: 221-5, 1996.

[29] Dehaene S. “Electrophysiological evidence for category-specific word processing” Neuroreport 6: 2153-7, 1995.

[30] Neville HJ, Mills D, and Lawson DS. “Fractionating Language: Different Neural Subsystems with Different Sensitive Periods” Cerebral Cortex 2: 244-58, 1992.

[31] Pulvermuller F, Lutzenberger W, and Birbaumer N. “Electrocortical distinction of vocabulary types” Electroenceph Clin Neurophysiol 94: 357-70, 1995.

[32] Mohr B, Pulvermuller F, and Zaidel E. “Lexical Decision After Left, Right, and Bilateral Presentation of Function Words, Content Words, and Non-Words: Evidence For Interhemispheric Interaction” Neuropsychologia 32(1): 105-24, 1994.

[33] Munte TF, Wieringa BM, Weyerts H, Szentkuti A, Matzke M, and Johannes S. “Differences in brain potentials to open and closed class words: class and frequency effects” Neuropsychologia 39: 91-102, 2001.

[34] Kounios J and Holcomb PJ. “Concreteness Effects in Semantic Processing: ERP Evidence Supporting Dual-Coding Theory” J Exp Psychol 20(4): 804-23, 1994.

[35] Federmeier KD, Segal JB, Lombrozo T, and Kutas M. “Brain responses to nouns, verbs and class-ambiguous words in context” Brain 123(12): 2552-66, 2000.

[36] Pulvermuller F, Harle M, and Hummel F. “Walking or Talking? Behavioral and Neruophysiological Correlates of Action Verb Processing” Brain Lang 78: 143-68, 2001.

[37] Pulvermuller F, Harle M, and Hummel F. “Neurophysiological distinction of verb categories” Cog Neurosci 11(12): 2789-93, 2000.

[38] Joseph JE. “Functional Neuroimaging studies of category specificity in object recognition: A critical review and meta-analysis” Cog Affect Behav Neurosci 1(2): 119-36, 2001.

[39] Mitchell TM, Hutchinson R, Just MA, Niculescu RS, Percira F, and Wang X. “Classifying Instantaneous Cognitive States from fMRI Data” Am Med Informatics Assoc November, 2003.  Also at