Neighborhood Activation in Spoken Word RecognitionRemembering Tom Tillman During and following World War II, there was significant interest among audiologists and speech scientists in the development and standardization of speech recognition tests for clinical use. Clinical observations during the 1940s and 50s had already indicated that results from pure tone threshold tests often were not predictive of the receptive auditory communication ability among persons with hearing loss. No doubt, Tom Tillman, as a graduate student and later as a faculty member at Northwestern University, was influenced by the growing clinical interest in using speech to measure receptive communication ability. Several of Tillman's publications1 2 3 during his early professional life reflect his research interest in speech audiometry, an interest that continued and expanded4 5 6 throughout his career. He is especially remembered-in collaboration with Carhart -in the development of the Northwestern University Test No. 6, which is still in use today. Tillman's strategy for measuring speech recognition with phonemically-balanced, monosyllabic words was symptomatic of the basic and clinical speech research emphasis of that period, in particular, the general view that the perception of words required the recovery of a sequence of phonetic or phonemic elements. This orientation lead to “bottom-up” explanations of speech perception. Since the 1970s, however, basic speech research has reflected the growing recognition that any comprehensive theory of speech perception must account for the processes and representations that subserve the recognition of spoken words beyond the perception of individual consonants and vowels. The current article reviews several recent investigations conducted at the UCLA-VA Human Auditory Laboratory that provide evidence that cognitive and linguistic capabilities (“top-down processing”) play a role in rapid selection of a target word from other potential choices once an acoustic-phonetic pattern has been activated in memory. This article is dedicated to Tom Tillman, who served as an example of a dedicated, meticulous researcher to me during my pre-doctoral studies.
02 April 2004 (online)
Current models of spoken word recognition share the common assumption that the perception of speech includes two fundamental processes: activation and competition. A key feature of the discrimination process is competition among multiple representation of words activated in memory, and the single selection of a target word from the alternatives in the same lexical neighborhood. The focus of this article is on one of these models, the Neighborhood Activation Model (NAM). Several experiments are reviewed that support the basic tenets of NAM, and identify neighborhood density and neighborhood frequency along with word frequency as significant lexical factors affecting word recognition. The results indicate that NAM can be generalized to the word recognition performance of individuals with sensorineural hearing loss and persons for whom English is not their first language. These results provide support that activation-competition models of spoken word recognition may possess principles fundamental to the recognition of words.
Spoken word recognition - lexical competition - hearing impairment - non-native listeners
- 1 Tillman T W, Jerger J, Carhart R, Peterson J. Some relations between normal hearing for pure tones and for speech. J Speech Hear Res. 1959; 2 126-140
- 2 Tillman T W, Jerger J. Some factors affecting the spondee threshold in normal-hearing subjects. J Speech Hear Res. 1959; 2 141-146
- 3 Tillman T W, Johnson R, Olsen W. Earphone versus sound field threshold sound pressure levels for spondee words. J Acoust Soc Am. 1966; 39 125-133
- 4 Tillman T W, Carhart R. An Expanded Test for Speech Discrimination Utilizing CNC Monosyllabic Words (N.U. Auditory Test No. 6). Technical Documentary Report No. SAM-TR-66-64. Brooks Air Force Base, TX: School of Aerospace Medicine, Aerospace Medical Division, United States Air Force 1966
- 5 Tillman T W, Carhart R. Interaction of competing speech signals with hearing loss. Arch Otolaryngol. 1970; 91 273-279
- 6 Tillman T W, Carhart R. Individual consistency of hearing for speech across diverse listening conditions. J Speech Hear Res. 1972; 15 105-113
- 7 Marlsen-Wilson W D, Walsh A. Processing interactions and lexical access during word recognition in continuous speech. Cognit Psychol. 1978; 10 29-63
- 8 McClelland J L, Elman J L. The TRACE model of spoken perception. Cognit Psychol. 1986; 18 1-86
- 9 Norris D. SHORTLIST: a connectionist model of continuous speech recognition. Cognition. 1994; 52 189-234
- 10 Luce P A, Pisoni D B, Goldinger S D. Similarity neighborhoods of spoken words. In: Altman G Cognitive Models of Speech Perception. Psycholinguistic and Computational Perspectives. Cambridge, MA; MIT Press 1990: 122-147
- 11 Cliff M S, Luce P A. Similarity neighborhoods of spoken two-syllable words: retroactive effects on multiple activation. J Experimental Psychology: Human Perception and Performance. 1990; 16 551-563
- 12 Sommers M S, Kirk K I, Pisoni D B. Some considerations in evaluating spoken word recognition by normal-hearing, noise-masked normal-hearing and cochlear implant listeners. Ear Hear. 1997; 18 89-99
- 13 Luce P A, Pisoni D B. Recognizing spoken words: The Neighborhood Activation Model. Ear Hear. 1998; 19 1-36
- 14 Kirk K I, Pisoni D B, Osberger M J. Lexical effects on spoken word recognition by pediatric cochlear implant users. Ear Hear. 1995; 16 470-481
- 15 Miyamoto R T, Kirk K I, Robbins A M, Todd S, Riley A. Speech perception and speech production skills of children with multichannel cochlear implants. Acta Otolaryngol. 1996; 116 240-243
- 16 Sommers M S. The structural organization of the mental lexicon and its contribution to age-related declines in spoken word recognition. Psych Aging. 1996; 11 333-341
- 17 Dirks D D, Takayanagi S, Moshfegh A, Noffsinger P D, Fausti S A. Examination of the Neighborhood Activation Theory in normal and hearing-impaired listeners. Ear Hear. 2001; 22 1-13
- 18 Nusbaum H R, Pisoni D B, Davis C N. Sizing Up the Hoosier Mental Lexicon. Measuring the Familiarity of 20,000 Words. Research in Speech Perception Progress Report No.11. Bloomington, IN: Speech Research Laboratory, Psychology Department, Indiana University 1984
- 19 Kucera F, Francis W. Computational Analysis of Present Day American English. Providence, RI Brown University Press; 1967
- 20 Takayanagi S, Dirks D D, Moshfegh A. Lexical and talker effect on word recognition among native and non-native listeners with normal and impaired hearing. J Speech Hear Res. 2002; 45 585-597
- 21 Levitt H. Transformed up-down methods in psychoacoustics. J Acoust Soc Am. 1971; 49 467-477
- 22 Vitevitch M S, Luce P A, Charles-Luce J, Kemmere D. Phonotactics and syllable stress: implications for the processing of spoken nonsense words. Lang Speech. 1997; 40 47-62
- 23 Vitevitch M S, Luce P A. Probabilistic phonotactics and neighborhood activation in spoken word recognition. J Mem Lang. 1999; 40 347-408
- 24 Pisoni D B. Individual difference in effectiveness of cochlear in prelingually deaf children: Some new process measures of performance. Research in Spoken Language Processing. Progress Report NO.23. Research in Speech Perception Progress Report No.11. Bloomington IN: Speech Research Laboratory, Psychology Department, Indiana University 1999
- 25 Sommers M S, Danielson S M. Inhibitory processes and spoken word recognition in young and older adults: The interaction of lexical competition and semantic context. Psych Aging. 1999; 14 458-472
Donald D Dirks, Ph.D.
11450 Waterford St.
Los Angeles, CA 90049