You are in: Home » History of the Unit
LANGUAGE, SPEECH, READING AND WRITING
General Notes. This material has been scanned from the original typescript. While we have done our best to remove errors, some may well remain. You can access other parts of this particular Progress Report either from the menu at the bottom of this entry, by using your browser's back function, by navigating back to the Unit History Timeline, or by accessing the relevant section of the electronic archive. Reference for this report are indexed by number and these can be found in a dedicated section also accessible from the menu at the bottom of this entry.
Project 62 - Recognition and Production of Spoken Language
62.1 Normal and Disordered Spoken Language Comprehension (Marslen-Wilson)
Research 1987-1989. After he joined the Unit in 1987, Marslen-Wilson continued on the programme of research into normal and disordered spoken language comprehension that he had been conducting since January 1985 in collaboration with Dr Lorraine Tyler, and which had been based since September 1985 in the Cambridge Department of Experimental Psychology. This was research supported by the MRC on a programme grant, originally awarded to Drs Tyler and Marslen-Wilson at a point when he was still a member of the Cambridge Psychology Department.
The work carried out under this programme grant combines research into language comprehension in normal adults and into the breakdown of this process in the adult aphasic. The research into the normal model has centred on the role of the mental lexicon in language processing, looking at how mentally represented knowledge of lexical form and lexical content is deployed in the on¬line process of speech comprehension. Working within the framework of the cohort model of lexical processing [97, 65], this research has focussed on three main areas: the nature of the input to the mental lexicon, the properties of the information-processing mechanisms underlying lexical access and selection, and the way lexical knowledge is deployed in higher-level processing. The results of this research show, first, that the processes of lexical access and choice are closely tied to the details of variation in the speech signal, suggesting a featural rather than segmental input to the lexicon, and supporting a view of lexical form representations as structured arrays of features rather than simple strings of segments (or phonemes). The second group of studies confirm the claim that lexical processing is based on a process of competition among simultaneously active candidates, establishing the principle of the contingency of perceptual choice. The third strand of the research on normals, closely linked to developments in the aphasia research, has looked at the processing of morphologically complex words in utterance contexts, investigating basic questions about the processing functions of word-stems and word-affixes in parsing and interpretation.
The research into aphasic disorders of language has centred on the development and validation of the concept of the on-line processing profile: that is, the complete profile of each patient's dynamic processing capacities in the domain of spoken language comprehension. This has resulted so far in the development of 25 tests, consisting of 14 tests of basic on-line processing capacities, ranging from the initial analysis of the speech signal to the construction of higher-level interpretative representations, together with 7 additional "off-line" tests of the same capacities, and 4 tests screening for possible memory problems.
In addition to this research based in Cambridge, approximately 25% of Marslen-Wilson's time has involved collaboration with researchers at the Max-Planck Institute in Nijmegen. The topics here, also all involving aspects of spoken language comprehension, include the representation of lexical form, syntactic parsing in Dutch, German, and English, and research on large-scale lexical databases, lexical statistics, etc.
During this initial period he has begun to build up collaboration with researchers in the Unit - especially in connection with the ESPRIT BRA proposals. This has involved Patterson, Cutler and Norris. In the neuropsychological domain the research (with Dr Tyler) has benefited in particular from discussions with Patterson.
Collaboration within the Unit. The planned research described above overlaps in many ways with the skills and interests of several members of the Unit. The research on lexical access in the speech domain is complementary to the research interests of Cutler, Norris and Patterson. The ESPRIT BRA project, now about to get underway, will provide an excellent environment for collaboration and cooperation with these individuals. The emphasis on acoustics-phonetics and on computational modelling in the BRA project is a useful counterpoint to the research directions Marslen-Wilson will be taking.
The involvement of the Unit in the proposed Rehabilitation Centre, and in related on-going work elsewhere, suggests another domain where collaboration with other members of the Unit (especially Baddeley and K Patterson) can naturally and profitably develop. This is in addition to the long¬standing contacts already existing with K Patterson, and, to a lesser extent, with Marcel and Shallice.
62.2 Studies of Lexical Segmentation (Butterfield, Cutler, McQueen, Norris, K Patterson, Williams)
The primary goal of Cutler's research during the past few years has been development of a model of the prelexical stages of speech recognition, a model which is sufficiently general to be language-universal, but sufficiently detailed to account for language-specific phenomena. Joint research with Norris and with colleagues in Paris established that speech segmentation procedures differ in speakers of French and of English [81]; the syllable functions as an effective segmentation unit for speakers of French, but not for speakers of English. This can be explained in terms of differences in phonological structure between the two languages. Subsequent studies have confirmed that even in completely bilingual speakers of French and English one or other segmentation procedure is the more basic [83]: only those bilinguals who accept that French is their dominant language use syllabic segmentation; bilinguals in whom English is more dominant do not.
Norris and Cutler proposed [105] that this difference reflects the importance of the rhythmic structure of language in guiding prelexical segmentation. This led to the development of a model of segmentation in English and rhythmically similar languages, according to which segmentation in such languages is guided by stress rhythm; the speech stream is segmented at the onset of rhythmically strong syllables. The model received substantial experimental support from studies using a new task, "word-spotting" [84]. These studies showed that when detection of a real word embedded in nonsense requires assembly of speech material across a putative segmentation point, i.e. the onset of a strong syllable, detection is delayed; but assembly of speech material across boundaries which according to the model are not segmentation points causes no such delay.
Parallel studies conducted on the Alvey grant MMI 069 demonstrated that the lexical segmentation strategy proposed for English is well supported by the structure of the English vocabulary [166, 78]; most words in English begin with strong syllables, and in typical spontaneous speech nearly all lexical words start with strong syllables. Since what makes a syllable strong or weak is the nature of its vowel, the proposed segmentation strategy is driven initially by perception of vowel nuclei, and it is important that syllable onsets be independently derivable; for this reason experimental demonstrations that syllable-initial consonant clusters are treated as integral units in segmentation [164, 85] provided valuable supporting evidence. Further supporting evidence came from studies of naturally occurring and experimentally produced juncture misperceptions, which showed that listener errors can be predicted by the strong syllable hypothesis [162]: both in spontaneous slips of the ear and in experiments on hard-to-hear speech, it was found that juncture misplacements which erroneously inserted a word boundary before a strong syllable, or deleted a word boundary before a weak syllable, were common, while misplacements which erroneously inserted a word boundary before a weak syllable, or deleted a word boundary before a strong syllable, were rare. This line of research on segmentation is reviewed in [129] and [130].
In associated work, central theoretical issues involved in modelling speech recognition at this level were addressed. First, studies of the availability of lexical information to a phoneme detection process [82] showed that effects of lexical-level processing can be made to come and go according to the requirements of the experimental situation; for instance, making the experimental materials very homogeneous in segmental structure removes lexical effects, while introducing variety encourages them. This result suggests that listeners can switch attention between levels of processing, and that such switching can be more elegantly accommodated by a serial rather than an interactive information-processing structure. Supporting results have been obtained by McQueen [168] in a phonemic categorisation task. Second, it was shown by Norris and Cutler [106] that, contrary to previous reports, listeners can respond to phoneme targets more rapidly than to syllable targets. This result supports the intuitive hierarchy of potential segmentation units such as the phoneme and the syllable in levels of representation. Third, Cutler, Norris and J Williams showed that listeners can draw on quite detailed knowledge of the phonological structure of their language in making segmentation decisions [77]. These theoretical issues in speech recognition processing are addressed in [182] and [49].
62.3 Studies of Prosody in Speech Recognition (Butterfield, Cutler, Nimmo-Smith, Norris, Smith)
Although this line of research has tapered off since the main focus has moved to pre-lexical segmentation, a number of continuing projects have produced useful results during the time period of this report. Studies of noise-masked speech showed that listeners can exploit speech timing patterns to extract information not only about stress rhythm but also about where word boundaries occur [117]. Phonetic categorisation studies with minimal stress pairs such as "tigress"-"digress" suggested that word prosody is represented in the lexicon and can affect word recognition decisions [73]. The importance of prosodic information for sentence-level processing was confirmed in phoneme-monitoring studies using spontaneous speech (which is prosodically fragmented) as well as read speech (which is prosodically fluent) [100]. Separate manipulation of the prosodic dimensions of rhythm, pitch and intensity showed that it is necessary that the prosodic structure of an utterance be internally coherent; listeners fail to extract useful information at all from the prosody when the separate dimensions convey conflicting information [163]. Finally, the development in young children of the ability to extract such prosodic information was studied; results suggest that in acquiring the ability to exploit sentence-level prosodic information in sentence recognition, children need first to acquire representations of semantic structure within utterances [86].
Theoretical explorations in this area have principally concerned cross-domain implications between psychological and phonological theory. For instance, is the failure to find syllabic segmentation effects in English [81] relevant to the status of the syllable as a construct in the phonology of English? It is argued that psychological studies of speech processing can not constrain phonological theory, in the same way that phonological theory itself can have no direct implications for psychological modelling [76, 131, 135]. A research finding is relevant only to the theoretical framework within which it was conceived, and a theory can only be tested by experiments which are designed directly to test it; although the psychological theory of language processing needs to draw upon detailed knowledge of linguistic structure, and linguistic theory must not ignore the characteristics of language performance, the two bodies of theory are separate.
Project 63 - Computational Modelling of Human Speech Recognition and Reading
Although speech is a time-varying signal, conventional feed-forward networks have no way of handling the temporal component of speech directly. Consequently, such networks tend to be unable to generalise appropriately in the time domain. For example, simple feed-forward nets either have to be lined up with the onset of words in the signal, or have to be taught to recognize words in all possible positions. For the same reason they are very poor at accommodating temporal variability in the signal, such as might be caused by changes in the rate of speech. Both of these problems have been overcome by developing a connectionist model of human speech recognition based on a dynamic net [149]. A dynamic net incorporates feed-back links with time delays. This enables the network to process new input in the context of its processing of prior input. The network can therefore process its input successively in small time slices rather than requiring the entire input to be analysed simultaneously. Unlike the TRACE model, for example, the net can identify words in continuous input without the need for duplication of word nodes. Also, the net is very tolerant of temporal variability in the signal. Finally, the strongly left-to-right nature of the dynamic net means that it provides a natural model of the left-to-right, or 'cohort', behaviour observed in human speech recognition.
63.1 Connectionism and Modularity (Norris)
We tend to think of connectionist models as being highly interactive. However, connectionist models force us to consider how complex tasks can actually be learned as well as performed. But learning in very large unstructured (interactive) nets is very slow, if not impossible. Learnability considerations lead to the conclusion that complex tasks can only be acquired by very modular and highly structured learning systems. For example, work on a connectionist model of an idiot savant date calculator [103] demonstrates how a relatively simple task can only be learned by a network whose structure parallels the stages of the date calculation algorithm. The task can not be performed by a simple net with a single layer of hidden units. The only way we could ever learn a really complex task such as language is by using a very modular and highly structured learning system.
This work on idiot savant date calculators also demonstrates how individuals with limited arithmetical skills and little or no explicit knowledge of facts about dates can perform such a complex task as date calculation. The network only needs knowledge of the days in each month and a few anchor dates to provide it with training information. It then learns to calculate dates by internalising a set of table look up procedures which require no arithmetical skills or explicit knowledge about dates.
Further work on modularity shows how the power of back-propagation networks introduces new and more powerful bottom-up models. Elman and McClelland (1988) have recently presented data which appear to provide categorical evidence for the top-down influence of lexical information on phonetic processing. However, Norris [150, 151] has shown how very simple bottom-up models using back-propagation can account for their data. Interestingly, there appears to be no equivalent model of the classical information processing box model form which can account for the same data without assuming that there is some interaction between processes.
Project 64 - Analysis and Modelling of Normal and Impaired Reading and Writing
64.1 Modelling Normal and Impaired Recognition and Pronunciation of Written Words (K Patterson)
Until recently, this area of research (like most) relied mainly on descriptive, information-processing models [154, 153]. Such theories allowed significant progress towards a broad conception of the inter-relationships between different components of a language system; but they did not make specific proposals about how a particular cognitive operation would be computed, and so yielded no quantitative, testable predictions.
With the advent of connectionist (also known as parallel distributed processing, or neural net) modelling, one can (a) make hypotheses about the precise nature of the computation; (b) implement these computations in a computer simulation; (c) assess the simulated results in the light of data from normal subjects performing the same task; (d) "damage" the computer model in various ways; and (e) assess the performance of the "lesioned" model against data from neurological patients with relevant impairments.
A connectionist model of translating English print to pronunciation developed by Seidenberg and McClelland (1989) is one of the most comprehensive PDP models thus far implemented, in terms of both the size of the data base and the range of performance simulated. The model assumes that words (both written and spoken, both known and novel) are represented as patterns of activation distributed across large sets of simple, neuron-like elements. Learning involves modification of weights on connections amongst these elements. The existing model demonstrates that a single computational procedure may be able to deal with three types of English letter strings which were thought to require multiple procedures for spelling-sound translation: regular words (words which obey typical spelling-to-sound "rules", like HINT), exception words (words which break these rules, like PINT) and nonwords (previously unencountered letter strings, like RINT).
Patterson, Seidenberg and McClelland have explored various forms of damage to the trained network [155] and have achieved some success in simulating one major variety of acquired reading disorder, surface dyslexia (see [66], in which the patients retain virtually normal ability to name regular words and nonwords (HINT and RINT) but tend to "regularise" exception words (pronouncing PINT to rhyme with "hint").
64.2 Experimental Work on Neurological Patients (Marcel, K Patterson)
64.2a. Marcel and Patterson have now finished a large study of phonological dyslexia, where reading of familiar words is virtually normal but reading of even simple new words or nonwords is impaired, and a paper is in preparation [95]. The study involved implicit and explicit tests of the procedures necessary to read aloud nonlexical letter strings: segmentation of orthography, assignment of phonology to each segment, assembly of phonological segments into a spoken syllable. These tests also bridge the methodological gap between acquired dyslexia and studies of children's acquisition of literacy. The results show (a) deficits and lexical effects in orthographic segmentation and (for all patients) in phonological assembly, and (b) for some patients, a superiority in oral reading of nonwords which sound the same as real words (phood > thood). These data challenge the need to postulate a deficit in grapheme-phoneme retrieval and support models where the various operations required to read a nonword are not independent of word knowledge. They also suggest that spontaneous (lexical) speech may involve construction from sublexical segments rather than retrieval of addressed, lexically complete phonology.
64.2b. Two right-handed teen-aged girls with onset of encephalitis after the age at which normal lateralisation of function is expected to occur had complete hemispherectomy operations (one left, one right) at age 15. A study of their reading abilities at age 17 [109] showed a pattern of word recognition and pronunciation skills for the right hemispherectomy case that was somewhat delayed (i.e. characteristic of an 11- or 12-year old normal child rather than a 17-year old) but in no way deviant: she did not have poor performance on any of the many sub-components of reading skill tested. The left hemispherectomy case, on the other hand, not only had extremely poor performance on all language and reading tests but showed a highly distinctive reading pattern (oral reading restricted to highly frequent words referring to imageable, concrete objects; semantic errors in naming single words; total inability to compute phonology for unfamiliar letter strings) that characterises some adult vascular cases with very large LHem lesions. There has been substantial debate (see for example [124]) as to whether this pattern of reading performance, known as deep dyslexia, reflects the reading capabilities of the RHem. The observation of this pattern in a patient with only a RHem favours the RHem theory of deep dyslexia.
64.3 Neuropsychological Analyses of Semantic Memory (Burgess, K Patterson, Shallice)
64.3a. Theoretical Analyses. The positions on the organization of semantic memory for which Shallice argued in a theoretical paper [158] on a number of issues, particularly on the separability of visual and lexical semantics, have been criticised by Riddoch et al (1988). In a reply [112] Shallice showed that their own positive evidence for their position had serious flaws and that certain key results had not been explained from their perspective.
64.3b. Pure Alexia. An analysis of a letter-by-letter reader [113] questioned the generality of previous accounts of pure alexia by Warrington and Shallice (1980) and by Patterson and Kay (1982), as this patient was able to access certain lexical and semantic information about written words when unable to explicitly identify them. This characteristic was viewed as arising from a severe but not complete impairment of access to the semantic system.
64.3c. Rapid Serial Visual Presentation. McLeod, Shallice, Watson and Burgess have shown that normal subjects attempting to identify words from a very broad semantic field (objects not found in a house) that occur in a rapidly presented list (exposure duration = 160 msec per item) make semantic and visual errors like those found in certain pathological cases (semantic access dyslexia and deep dyslexia) [94]. In a control experiment it was shown that the rate of occurrence of semantic errors is too great to be plausibly attributed to a forgetting process over the 1-2 sec time course of a trial. It is suggested that the errors reflect inadequate access processes.
64.4 Connectionist Modelling of Semantic Access Processes (Shallice)
With G Hinton, Shallice has explored the effect of lesioning a connectionist net to see whether the lesioned net showed properties qualitatively similar to ones exhibited by patients and for which no convincing account is available within the standard cognitive neuropsychology framework [90]. The net mapped orthographic input into semantic features. Lesions reproduced the co-occurrence of semantic, visual and mixed visual and semantic errors characteristically found in deep and semantic access dyslexia. They also showed the relative preservation of categorical discrimination when precise identification is lost, a phenomenon found in semantic access dyslexia. To our knowledge, this is the first time that non-transparent neuropsychological phenomena have been mimicked by lesions to connectionist nets.
64.5 Experimental Work on Normal Readers (K Patterson)
One of the strengths of APU research has always been the combined use of normal and neuropsychological data to address theoretical questions. Two of the issues raised by a comprehensive review of mechanisms of spelling to sound translation [153] are now being investigated in studies of normal readers via two collaborations.
64.5a. If (as suggested by the neuropsychological data) the normal reading system offers several alternative routines for translating print to pronunciation, then manipulating a normal subject's strategy either explicitly by instruction or implicitly by the nature of the reading task ought to change patterns of performance. Monsell (University of Cambridge) and K Patterson have demonstrated large individual differences in subjects' ability to exert strategic control over procedures for word naming, with meaningful correlations between these differences and other aspects of reading performance.
64.5b. If (as suggested by one interpretation of both real and computationally simulated data) the pronunciation errors of surface dyslexic patients (e.g. PINT pronounced to rhyme with "hint") are based on sub-asymptotic activation in the phonological system, then normal subjects naming words under unnaturally speeded conditions might be expected to produce similar errors. Data from the first of several studies by Seidenberg (McGill University) and K Patterson support this hypothesis.
64.6 Rehabilitation (K Patterson)
There has been a limited amount of work at the APU on techniques of rehabilitation [91, 92] and theories of rehabilitation [140] based on cognitive neuropsychological analyses of language and reading deficits. This is partly due to the difficult and time-consuming nature of well-designed research on efficacy of treatment; but it is also due to the fact that there has been no adequate centre for rehabilitation research at Addenbrooke's Hospital to enable such work. Plans to establish such a centre are described later in this report. Given the wealth of analytic data on acquired disorders of reading that now exists, using such information to develop and evaluate therapy is a natural area for future work.
64.7 From Pronunciation to Print: Writing and Spelling (Burden, K Patterson, Wing)
In parallel with work on reading and its disorders, we have been investigating writing and spelling skills, again in both normal and patient populations. Case studies of stroke patients have been informative about both central [107] and peripheral [111] aspects of the writing process. A research student at the APU (V Burden) has just completed three years' research on neurologically normal adults who have abnormally poor spelling skills (e.g. [72]). Although the problems of such subjects have in the past been described as restricted to writing, more detailed investigations reveal these people to be at or even below the limit of the normal range of adult performance on other skills such as reading speed, short-term memory and phonological manipulations. The results have important implications for theories of the inter-related development of these aspects of cognitive ability.
REFERENCES
Al - Authored Books
64. Howard, D. and HATFIELD, F.M. (1987) Aphasia Therapy: Historical and Contemporary Issues. Hove, Sussex: Lawrence Erlbaum Associates.
A2 - Edited Books
65. MARSLEN-WILSON, W.D. (Ed.), (1989) Lexical Representation and Process. Cambridge, MA: MIT Press.
66. PATTERSON, K., Marshall, J.C. and Coltheart, M. (Eds.) (1985) Surface Dyslexia. London: Lawrence Erlbaum Associates.
B - Refereed Journal Articles
67. Bach, E., Brown, C. and MARSLEN-WILSON, W.D. (1986) Crossed and nested dependencies in German and Dutch: A psycholinguistic study. Language and Cognitive Processes, 1, 249-262.
68. BADDELEY, A.D., LOGIE, R.H. and Ellis, N.C. (1988) Characteristics of developmental dyslexia. Cognition, 29, 197-228.
69. BADDELEY, A.D., LOGIE, R.H., NIMMO-SMITH, I. and Brereton, N. (1985) Components of fluent reading. Journal of Memory and Language, 24, 119-131.
70. BLACK, A., Freeman, P. and JOHNSON-LAIRD, P.N. (1986) Plausibility and the comprehension of text. British Journal of Psychology, 77, 51-62.
71. Brown, G.D.A. and WATSON, F.L. (1987) First in, first out: Word learning age and spoken word frequency as predictors of word familiarity and word naming latency. Memory and Cognition, 15, 208-216.
72. BURDEN, V. (1989) A comparison of priming effects on the nonword spelling performance of good and poor spellers. Cognitive Neuropsychology, 6, 43-65.
73. Connine, CM., Clifton, C.E. and CUTLER, A. (1987) Effects of lexical stress on phonetic categorization. Phonetica, 44, 133-146.
74. CUTLER, A. (1985) Cross-language psycholinguistics. Linguistics, 23, 659-667.
75. CUTLER, A. (1986) Forbear is a homophone: Lexical prosody does not constrain lexical access. Language and Speech, 29, 201-220.
76. CUTLER, A. (1986) Phonological structure in speech recognition. Phonology Yearbook, 3, 161-178.
77. CUTLER, A., BUTTERFIELD, S. and WILLIAMS, J.N. (1987) The perceptual integrity of syllabic onsets. Journal of Memory and Language, 26, 406-418.
78. CUTLER, A. and Carter, D.M. (1987) The predominance of strong initial syllables in the English vocabulary. Computer Speech and Language, 2, 133-142.
79. CUTLER, A., Hawkins, J.A. and Gilligan, G. (1985) The suffixing preference: A processing explanation. Linguistics, 23, 723-758.
80. CUTLER, A., Howard, D. and PATTERSON, K.E. (1989) Misplaced stress on prosody: A reply to Black and Byng. Cognitive Neuropsychology, 6, 67-83.
81. CUTLER, A., Mehler, J., NORRIS, D. and Segui, J. (1986) The syllable's differing role in the segmentation of French and English. Journal of Memory and Language, 25, 385-400.
82. CUTLER, A., Mehler, J., NORRIS, D. and Segui, J. (1987) Phoneme identification and the lexicon. Cognitive Psychology, 19, 141-177.
83. CUTLER, A., Mehler, J., NORRIS, D. and Segui, J. (1989) Limits on bilingualism. Nature, 340, 229-230.
84. CUTLER, A. and NORRIS, D.G. (1988) The role of strong syllables in segmentation for lexical access. Journal of Experimental Psychology: Human Perception and Performance, 14, 113-121.
85. CUTLER, A., NORRIS, D. and WILLIAMS, J.N. (1987) A note on the role of phonological expectations in speech segmentation. Journal of Memory and Language, 26, 480-487.
86. CUTLER, A. and Swinney, D.A. (1987) Prosody and the development of comprehension. Journal of Child Language, 14, 145-167.
87. GIPSON, P. (1986) The production of phonology and auditory priming. British Journal of Psychology, 77, 359-375.
88. Gulian, E., Hinds, P. and NIMMO-SMITH, I. (1986) Modifications in deaf children's vowel production: Perceptual evidence. British Journal of Audiology, 20, 181-194.
89. HATFIELD, F.M. (1985) Visual and phonological factors in acquired dysgraphia. Neuropsychologia, 23, 13-29.
90. Hinton, G. and SHALLICE, T. (1989) Lesioning a connectionist network: Investigations of acquired dyslexia. Technical Report CRG-TR-89-3 May 1989.
91. Howard, D., PATTERSON, K., Franklin, S., Orchard-Lisle, V. and Morton, J. (1985) The facilitation of picture naming in aphasia. Cognitive Neuropsychology, 2, 49-80.
92. Howard, D., PATTERSON, K., Franklin, S., Orchard-Lisle, V. and Morton, J. (1985) Treatment of word retrieval deficits in aphasia: A comparison of two therapy methods. Brain, 108, 817-829.
93. Humphreys, G.W., EVETT, L.J., Quinlan, P.T. and Besner, D. (1987) Orthographic priming: Qualitative differences between priming from identified and unidentified primes. In M. Coltheart (Ed.), Attention and Performance XII, The Psychology of Reading. Hillsdale, N.J.: Lawrence Erlbaum Associates, pp.105-125.
94. McLeod, P.D., SHALLICE, T., Watson, F. and BURGESS, P.W. Semantic errors in rapid serial visual presentation. (Manuscript submitted).
95. MARCEL, A.J. and PATTERSON, K.E. Articulating non-lexical reading processes in phonological dyslexia. Manuscript in preparation.
96. Marshall, J.C. and PATTERSON, K.E. (1985) Left is still left for semantic paralexias: A reply to Jones and Martin (1985). Neuropsychologia, 23, 689-690.
97. MARSLEN-WILSON, W.D. (1987) Functional parallelism in spoken word-recognition. Cognition, 25, 71-102.
98. MARSLEN-WILSON, W.D., Brown, CM. and TYLER, L.K. (1988) Lexical representation in spoken language comprehension. Language and Cognitive Processes, 3, 1-16.
99. MARSLEN-WILSON, W.D. and Zwitserlood, P. (1989) Accessing spoken words: The importance of word onsets. Journal of Experimental Psychology: Human Perception and Performance, 15, 576-585.
100. Mehta, G. and CUTLER, A. (1988) Detection of target phonemes in spontaneous and read speech. Language and Speech, 31, 135-156.
101. NORRIS, D.G. (1986) Word recognition: Context effects without priming. Cognition, 22, 93-136.
102. NORRIS, D.G. (1987) Strategic control of sentence context effects in a naming task. Quarterly Journal of Experimental Psychology, 39A, 253-275.
103. NORRIS, D.G. How to build a connectionist idiot (savant). Cognition, in press.
104. NORRIS, D.G. and Brown, G. (1985) Race models and analogy theories: A dead heat? Reply to Seidenberg. Cognition, 20, 155-168.
105. NORRIS, D.G. and CUTLER, A. (1985) Juncture detection. Linguistics, 23, 689-705.
106. NORRIS, D.G. and CUTLER, A. (1988) The relative accessibility of phonemes and syllables. Perception and Psychophysics, 43, 541-550.
107. PATTERSON, K. (1986) Lexical but nonsemantic spelling? Cognitive Neuropsychology, 3, 341-367.
108. PATTERSON, K. (1985) The pitfalls of selective attention. Commentary on Humphreys and Evett (1985). Behavioral and Brain Sciences, 8, p.721.
109. PATTERSON, K., Vargha-Khadem, F. and Polkey, C.E. (1989) Reading with one hemisphere. Brain, 112, 39-63.
110. PATTERSON, K. and Wilson, B. A ROSE is a NOSE: A deficit in initial letter identification. Cognitive Neuropsychology, in press.
111. PATTERSON, K. and WING, A.M. (1989) Processes in handwriting: A case for case. Cognitive Neuropsychology, 6, 1-23.
112. SHALLICE, T. (1988) Specialisation within the semantic system. Cognitive Neuropsychology, 5, 133-142.
113. SHALLICE, T. and Saffran, E. (1986) Lexical processing in the absence of explicit word identification: Evidence from letter-by-letter reader. Cognitive Neuropsychology, 3, 429-458.
114. SILVERMAN, K.E.A. (1986) Fq segmental cues depend on intonation: The case of the rise after voiced stops. Phonetica, 43, 76-91.
115. SILVERMAN, K.E.A. (1986) FRED: An interactive graphics program to modify fundamental frequency contours in resynthesized speech. Behavior Research Methods and Instrumentation, 18, 395-397.
116. SMITH, M.R. (1989) Contingencies in stress patterns and syllable nuclei: Lessons for searching lexical databases. Journal of the Acoustical Society of America, 85, S.123.
117. SMITH, M.R., CUTLER, A., BUTTERFIELD, S. and NIMMO-SMITH, I. (1989) The perception of rhythm and word boundaries in noise-masked speech. Journal of Speech and Hearing Research, 32, 912-920.
118. TYLER, L.K. and MARSLEN-WILSON, W.D. (1986) The effects of context on the recognition of polymorphemic words. Journal of Memory and Language, 25, 741-752.
119. TYLER, L.K., MARSLEN-WILSON, W.D., Rentoul, J. and Hanney, P. (1988) Continuous and discontinuous access in spoken word-recognition: The role of derivational prefixes. Journal of Memory and Language, 27, 368-381.
120. Warren, P. and MARSLEN-WILSON, W.D. (1987) Continuous uptake of acoustic cues in spoken word recognition. Perception and Psychophysics, 4, 262-275.
121. Warren, P. and MARSLEN-WILSON, W.D. (1988) Cues to lexical choice: Discriminating place and voice. Perception and Psychophysics, 43, 21-30.
122. WILLIAMS, J.N. (1988) Constraints upon semantic activation during sentence comprehension. Language and Cognitive Processes, 3, 165-206. C - Invited Chapters and Commentaries
123. ANDERSON, A. and Garrod, SC. (1987) The dynamics of referential meaning in spontaneous conversation: Some preliminary studies. In R.G. Reilly (Ed.), Communication Failure in Dialogue and Discourse. Elsevier Science Pub. B.V. (North-Holland), pp. 161-183.
124. Coltheart, M., PATTERSON, K., and Marshall, J.C. (1987) Deep dyslexia since 1980. In M. Coltheart, K. Patterson and J.C. Marshall (Eds.), Deep Dyslexia, 2nd edition, Chapter 18. London: Routledge and Kegan Paul, pp.407-451.
125. CUTLER, A. (1985) Performance measures of lexical complexity. In G.A.J. Hoppenbrouwers, P.A.M. Seuren and A.J.M.M. Weijters (Eds.), Meaning and the Lexicon. Dordrecht: Foris, p.75.
126. CUTLER, A. (1987) Speaking for listening. In A. Allport, D.G. MacKay, W. Prinz and E. Scheerer, E. (Eds.), Language Perception and Production: Relationships between Listening, Speaking, Reading and Writing. London: Academic Press Ltd., pp.23-40.
127. CUTLER, A. (1987) The task of the speaker and the task of the hearer. (Commentary on Sperber and Wilson, Relevance). Behavioral and Brain Sciences, 10, 715-716.
128. CUTLER, A. (1988) The perfect speech error. In L.M. Hyman and C.N. Li (Eds.), Language, Speech and Mind: Studies in Honor of Victoria A Fromkin. London: Routledge, pp.209-233.
129. CUTLER, A. (1989) Auditory lexical access: Where do we start? In W.D. Marslen-Wilson (Ed.), Lexical Representation and Process. Cambridge, MA: MIT Press, pp.342-356.
130. CUTLER, A. Exploiting prosodic probabilities in speech segmentation. In G. Altmann (Ed.), Computational and Psycholinguistic Approaches to Language Processing. Cambridge, MA: MIT Press, in press.
131. CUTLER, A. From performance to phonology. In J. Kingston and M.Beckman (Eds.), Papers in Laboratory Phonology 1: Between the Grammar and the Physics of Speech. Cambridge: Cambridge University Press, in press.
132. CUTLER, A. Marked and unmarked segmentation strategies? In H.C.Nusbaum and J.G. Goodman (Eds.), The Transition from Recognizing Speech Sounds to Recognizing Spoken Words. Cambridge, MA: MIT Press, in press.
133. CUTLER, A. (1989) Straw Modules: Commentary on D.W. Massaro: Speech perception by ear and eye. Behavioral and Brain Sciences, 12, 760-762.
134. CUTLER, A. Psychology and the segment. In D.R. Ladd and G. Docherty (Eds.), Papers in Laboratory Phonology II. Cambridge: Cambridge University Press, in press.
135. CUTLER, A. Why not abolish psycholinguistics? In W. Dressier, J. Rennison and O. Pfeiffer (Eds.), Phonologica 1988. Cambridge: Cambridge University Press, in press.
136. EVETT, L.J. and Humphreys, G.W. (1987) Extending the multiple-levels approach to word processing. Behavioral and Brain Sciences, 10, 334-336.
137. EVETT, L.J., Humphreys, G.W., and Quinlan, P.T. (1986) Identification, masking and priming: Clarifying the issues. Behavioral and Brain Sciences, 9, 31-32.
138. GROEGER, J.A. (1987) On not knowing the meanings of words we can detect: Crucial qualitative differences. Behavioral and Brain Sciences, 10,765-766.
139. Hawkins, J.A. and CUTLER, A. (1988) Psycholinguists factors in morphological asymmetry. In J.A. Hawkins (Ed.), Explaining Language Universals. Oxford: Basil Blackwell Ltd., pp.280-317.
140. Howard, D. and PATTERSON, K. (1989) Models for therapy. In X. Seron and G. Deloche (Eds.), Cognitive Approaches in Neuropsychological Rehabilitation. Hillsdale, N.J.: Lawrence Erlbaum Associates, pp.39-64.
141. JOHNSON-LAIRD, P.N. Introduction: What is communication? In H.Mellor (Ed.), Communication. Cambridge: Cambridge University Press, in press.
142. Kay, J. and PATTERSON, K. (1985) Routes to meaning in surface dyslexia. In K. Patterson, J.C. Marshall and M. Coltheart (Eds.), Surface Dyslexia. London: Lawrence Erlbaum Associates, pp.79-105.
143. Margolin, D.I., MARCEL, A.J. and Carlson, N. (1985) Common mechanisms in dysnomia and post-semantic surface dyslexia: Processing deficits and selective attention. In K.E. Patterson, J.C. Marshall and M. Coltheart (Eds.), Surface Dyslexia. London: Lawrence Erlbaum Associates, pp. 139-173.
144. MARSLEN-WILSON, W.D. (1989) Access and integration: The projection of sound onto meaning. In W.D. Marslen-Wilson (Ed.), Lexical Representation and Process. Cambridge, MA: MIT Press, pp.3-24.
145. MARSLEN-WILSON, W.D. Activation, competition, and frequency in lexical access. In G. Altmann (Ed.), Cognitive Models of Speech Processing: Psycholinguistics and Computational Perspectives. Cambridge, MA: MIT Press, in press.
146. MARSLEN-WILSON, W.D. The mental lexicon in language processing. In W. Bright (Ed.), The Oxford International Encyclopedia of Linguistics. New York: Oxford University Press, in press.
147. NORRIS, D.G. (1985) So the "strong" theory loses. But are there any winners? Commentary on Humphreys and Evett (1985). Behavioral and Brain Sciences, 8, 718-719.
148. NORRIS, D.G. (1987) Syntax, semantics and garden-paths. In A. Ellis (Ed.), Progress in the Psychology of Language, Vol. 3. London: Lawrence Erlbaum Associates, pp.233-252.
149. NORRIS, D.G. A dynamic net model of human speech recognition. In G.Altmann (Ed.), Computational and Psycholinguistic Approaches to Language Processing. Cambridge, MA: MIT Press, in press.
150. NORRIS, D.G. Connectionism: A new class of bottom up model. In R.Reilly and N. Sharkey (Eds.), Connectionist Approaches to Language Processing (Vol. 1), in press.
151. NORRIS, D.G. Connectionism: A case for modularity. In D.A. Balota, G.B. Flores d Area is and K. Rayner (Eds.), Comprehension Processes in Reading. Hillsdale, N.J.: Lawrence Erlbaum Associates, in press.
152. PATTERSON, K. (1988) Acquired disorders of spelling. In G. Denes, G. Semenza and P. Bisiacchi (Eds.), Perspectives on Cognitive Neuropsychology. London: Lawrence Erlbaum Associates, pp.213-229.
153. PATTERSON, K. and Coltheart, V. (1987) Phonological processes in reading: A tutorial review. In M. Coltheart (Ed.), Attention and Performance XII: The Psychology of Reading. London: Lawrence Erlbaum Associates, pp.421-477.
154. PATTERSON, K. and Morton, J. (1985) From orthography to phonology: An attempt at an old interpretation. In K. Patterson, J.C. Marshall and M. Coltheart (Eds.), Surface Dyslexia. London: Lawrence Erlbaum Associates, pp.335-359.
155. PATTERSON, K., Seidenberg, M.S. and McClelland, J.L. (1989) Connections and disconnections: Acquired dyslexia in a computational model of reading processes. In R.G.M. Morris (Ed.), Parallel Distributed Processing: Implications for Psychology and Neurobiology. Oxford: Oxford University Press, pp.131-181.
156. PATTERSON, K. and SITEWELL, C. (1987) Speak and spell: Dissociationsand word-class effects. In M. Coltheart, G. Sartori and R. Job (Eds.), The Cognitive Neuropsychology of Language. London: Lawrence Erlbaum Associates, pp.273-294.
157. SHALLICE, T. (1985) The acquired dyslexias and normal reading: Commentary on Humphrey and Evett (1985). Behavioral and Brain Sciences, 8, 726.
158. SHALLICE, T. (1987) Impairments of semantic processing: Multiple dissociations. In M. Coltheart, G. Sartori and R. Job (Eds.), The Cognitive Neuropsychology of Language. London: Lawrence Erlbaum Associates, pp.111-127.
159. SHALLICE, T. and McCarthy, R. (1985) Phonological reading: From patterns of impairment to possible procedures. In K. Patterson, J.C. Marshall and M. Coltheart (Eds.), Surface Dyslexia. London: Lawrence Erlbaum Associates, pp.361-397.
160. TYLER, L.K. (1985) The sequential organisation of spoken word recognition. In F. Fallside and W. Woods (Eds.), Computer Speech Processing. Prentice Hall, pp.405-417.
161. Wilson, B. and BADDELEY, A. (1986) Single case methodology and the remediation of dyslexia. In G.Th. Pavlidis and D.F. Fisher (Eds.), Dyslexia: Its Neuropsychology and Treatment. Chichester: John Wiley and Sons, pp.263-277. D - Conference Proceedings
162. BUTTERFIELD, S. and CUTLER, A. (1988) Segmentation errors by human listeners: Evidence for a prosodic segmentation strategy. In Proceedings of SPEECH '88, Seventh Symposium of the Federation of Acoustic Societies of Europe, Vol. 3, Edinburgh, pp.827-833.
163. CUTLER, A. (1987) Components of prosodic effects in speech recognition. In Proceedings of the XI International Congress of Phonetic Sciences. Tallin, Estonia, Vol. 1, pp.84-87.
164. CUTLER, A. and BUTTERFIELD, S. (1986) The perceptual integrity of initial consonant clusters. Proceedings of the Institute of Acoustics, Speech and Hearing, 8, 31-36.
165. CUTLER, A. and BUTTERFIELD, S. (1989) Natural speech cues to word segmentation under difficult listening conditions. In J.P. Tubach and J.J. Mariani (Eds.), EUROSPEECH '89, Vol. 2 - Proceedings of the European Conference on Speech Communication and Technology (Paris, September 1989). Edinburgh: CEP Consultants Ltd., pp.372-375.
166. CUTLER, A. and Carter, D.M. (1987) The prosodic structure of initial syllables in English. In J. Laver and M.A. Jack (Eds.), Proceedings of the European Conference on Speech Technology. Edinburgh, Vol. 1, pp.207-210.
167. Lahiri, A. and MARSLEN-WILSON, W.D. Lexical processing and phonological representation. In R. Ladd and G. Docherty (Eds.), Second Conference on Laboratory Phonology. Cambridge: Cambridge University Press, in press.
168. MCQUEEN, J. (1989) The use of lexical knowledge in phonetic categorisation. In J.P. Tubach and J.J. Mariani (Eds.), EUROSPEECH '89, Vol. 2 - Proceedings of the European Conference on Speech Communication and Technology (Paris, September 1989). Edinburgh: CEP Consultants Ltd., pp.581-584.
169. MARSLEN-WILSON, W.D. and Frauenfelder, U.IT. (1987) The interface between acoustic-phonetic and lexical processes. In Proceedings of the XI International Congress of Phonetic Sciences, Tallinn, Estonia, Vol. 4, pp.337-343.
170. SILVERMAN, K.E.A. (1985) Vowel intrinsic pitch influences the perception of intonational prominence. Journal of the Acoustical Society of America, 77, Suppl.l S38.
171. SILVERMAN, K.E.A. (1985) What can be done to improve the intonation in synthetic speech? In Proceedings of the Eleventh International Symposium on Human Factors in Telecommunications (CCETT, France, 9-13 Sept. 1985).
172. SMITH, M.R. A cross-linguistic study of perceived syllable coherence. In Proceedings of the 14th International Congress of Linguists (Berlin, 1987), in press. E - Technical Reports, Theses and Tests
173. BLACK, A. (1986) The effects of glossaries on jurors' comprehension in fraud trials. In Improving the Presentation of Information to Juries in Fraud Trials, Report to Lord Roskill Committee on Fraud Trials, London: H.M.S.O., pp.1-15.
174. BLACK, A. (1987) Lexical support in discourse comprehension. Unpublished PhD Thesis, University of Cambridge.
175. BUTTERFIELD, S. (1988) Segment perception: The influence of segmental context. In Atvey Project MMI/069 - Report on Automatic Speech Recognition using a Speech Knowledge Base and Advanced Pattern Matching Algorithms. Working Paper No. 41. Harlow: Standard Telecommunications Laboratories Ltd.
176. GROEGER, J.A. Saying something different: Levels in the monitoring and repair of speech. Belfast Working Papers in Linguistics, in press.
177. SILVERMAN, K.E.A. (1987) The structure and processing of fundamental frequency contours. Unpublished PhD Thesis, University of Cambridge.
178. WILLIAMS, J.N. (1986) The effective context for priming word recognition. Unpublished PhD Thesis, University of Cambridge.
F - Dissemination
179. ANDERSON, A. (1985) Some recent developments in the study of meaning. European Psycholinguistics Association Newsletter, 11, 6-9.
180. BADDELEY, A.D. (1986) Working memory, reading and dyslexia. In E.Hjelmquist and L.G. Nilsson (Eds.), Communication and Handicap -Aspects of Psychological Compensation and Technical Aids. Elsevier Science Publishers B.V. (North-Holland), pp.141-152.
181. CUTLER, A. (1986) Why readers of this newsletter should run cross-linguistic experiments. European Psycholinguistics Association Newsletter, 13, 4-8.
182. CUTLER, A. Speech perception. In W. Bright (Ed.), The Oxford International Encyclopedia of Language, Oxford: Oxford University Press, in press.
183. NORRIS, D.G. (1985) Current issues in word recognition. EuropeanPsycholinguistics Association Newsletter, 9, 7-9.
184. SMITH, M.R. (1986) Ecological psychology and language. European Psycholinguistics Association Newsletter, 12, 3-6.
Other sections in the 1985-1989 report
3. LANGUAGE, SPEECH, READING AND WRITING