<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-0703">
  <Title>Pronunciation by Analogy in Normal and Impaired Readers</Title>
  <Section position="4" start_page="13" end_page="14" type="metho">
    <SectionTitle>
3 Implementing PbA
</SectionTitle>
    <Paragraph position="0"> In PbA, an unknown word is pronounced by matching substrings of the input to substrings of known, lexical words, hypothesizing a partial pronunciation for each matched substring from the phonological knowledge, and assembling the partial pronunciations. Here, we use an extended and improved version of the system described by Dedina and Nusbaum (1991), which consists of four components: the (uncompressed and previously aligned) lexical database, the matcher which compares the target input to all the words in the database, the pronunciation lattice (a data structure representing possible pronunciations), and the decision function, which selects the 'best' pronunciation among the set of possible ones. The lexicon used is Webster's Pocket Dictionary, containing 20,009 words manually aligned by Sejnowski and Rosenberg (1987) for training their NETtalk neural network.</Paragraph>
    <Paragraph position="1"> Pattern Matching: An incoming word is matched in turn against all orthographic entries in the lexicon. For a given entry, assume the process starts with the input string and the dictionary entry left-aligned. Substrings sharing contiguous, common letters in matching positions are then found. Information about these matching letter substrings and their corresponding, aligned phoneme substrings in the dictionary entry under consideration is entered into a pronunciation lattice--see below. One of the two strings is then shifted right by one letter and the matching process repeated, until  some termination condition is met. This process can be alternatively seen as a matching between substrings of the incoming word, segmented in all possible ways, and the dictionary entries.</Paragraph>
    <Paragraph position="2"> Pronunciation Lattice: A node of the lattice represents a matched letter, Li, at some position, i, in the input. The node is labelled with its position index i and with the phoneme which corresponds to Li in the matched substring, Rim say, for the mth matched substring. An arc is placed from node i to node j if there is a matched substring starting with Li and ending with Lj. The arc is labelled with the phonemes intermediate between Pim and Pjm in the phoneme part of the matched substring.</Paragraph>
    <Paragraph position="3"> Additionally, arcs are labelled with a 'frequency' count which is incremented each time that sub-string (with that pronunciation) is matched during the pass through the lexicon.</Paragraph>
    <Paragraph position="4"> Decision Function: A possible pronunciation for the input corresponds to a complete path through its lattice, from Start to End nodes, with the output string assembled by concatenating the phoneme labels on the nodes/arcs in the order that they are traversed.</Paragraph>
    <Paragraph position="5"> (Different paths can, of course, correspond to the same pronunciation.) Scoring of candidate pronunciation uses two heuristics. If there is a unique shortest path, then the corresponding pronunciation is taken as the output. If there are tied shortest paths, then the pronunciation corresponding to the best scoring of these is taken as the output.</Paragraph>
    <Paragraph position="6"> This also offers a way of simulating the 'word segmentation' test of Funnell (1983), in which patients have to find words 'hidden' in letter strings. First, there is an initial segmentation in which the input string is segmented in all possible ways, as in 'regular' PbA. Then, if any of these substrings produces a lattice with a length-1 arc, this identifies a lexical word.</Paragraph>
    <Paragraph position="7"> A single-route connectionist model or abstract rules (or, for that matter, implicit PbA) can not do this without some extension to maintain explicit knowledge of lexical status. Of course, it is possible that a patient can perform the first of these steps, but not the second. This is the difference between our 'unconscious' and 'conscious' segmentations (see below) so-called because, in the latter, the patient is aware that he/she has to find a hidden word.</Paragraph>
    <Paragraph position="8"> This particular implementation of PbA does not guarantee an output pronunciation. A complete path through the lattice requires that all nodes on that path (except the first and last) are linked by at least one arc. Clearly, each arc must have a node at either end. Although an arc may have an empty label, a node cannot.</Paragraph>
    <Paragraph position="9"> Hence, the minimum matching segment length corresponds to a letter bigram. It may be that no matching bigram exists in some cases. So there with be no complete path through the lattice and no pronunciation can be inferred--the 'silence problem'.</Paragraph>
    <Paragraph position="10"> Recent Improvements: The implementation used here features several enhancements over the original Dedina and Nusbaum (D&amp;N) system (Marchand and Damper, 2000). First, we use 'full' pattern matching between input letter string and dictionary entries, as opposed to the 'partial' matching of D&amp;N. That is, rather than starting with the two strings leftaligned, we start with the initial letter of the input string Z aligned with the last letter of the dictionary entry YV. The matching process terminates not when the two strings are rightaligned, but when the last letter of Z aligns with initial letter of \]/Y. Second, multiple (five) heuristics are used to score the candidate pronunciations. Individual scores are then multiplied together to produce a final overall score. The best-scoring pronunciation is then selected as output. Marchand and Damper show that this 'multi-strategy' approach gives statistically significant performance improvements over simpler versions of PbA.</Paragraph>
  </Section>
  <Section position="5" start_page="14" end_page="16" type="metho">
    <SectionTitle>
4 Modelling Phonological Dyslexia
</SectionTitle>
    <Paragraph position="0"> By selective impairment of component parts of the PbA model, we have simulated reading data from the two phonological dyslexic patients (WB and FL) studied by Funnell (1983).</Paragraph>
    <Paragraph position="1"> (The reader is referred to this original source for specifications of the tests and materials.) While the first of these patients has often been cited as a key individual strongly supporting dual-route theory, we believe that FL (who has been largely ignored) is actually a counter-example.</Paragraph>
    <Paragraph position="2"> FL was unable to supply a sound for single letters (which argues that the abstract rule-based route is impaired) although she could read non-words normally (which contradicts the  presumption of impaired rules).</Paragraph>
    <Paragraph position="3"> For patient WB, two different versions of impaired PbA have been studied. Version 1 supposes that brain damage has induced a partial loss of words from his mental lexicon (the 15% that he can not read aloud) and a total breakdown of his concatenation mechanism. Version 2 supposes that WB's impairment results from injury to one component only; namely, the process of segmentation into all possible sub-strings is partially damaged. In Version 2, we stress the distinction made earlier between this basic (unconscious) segmentation process and Funnell's (conscious) segmentation. The unconscious segmentation is that embodied in the PbA pattern matching when WB is asked to read some string. For this specific patient, we postulate damage to the segmentation component such that it can only process substrings of length between 5 and 7. The conscious segmentation is that used when WB is asked to find words within strings and to read them aloud.</Paragraph>
    <Paragraph position="4"> This process is assumed to be fully operational.</Paragraph>
    <Paragraph position="5"> For patient FL, a single 'faulty' version of PbA has been developed which postulates a deftciency of (unconscious) segmentation such that substrings of length less than three cannot be used in pattern matching.</Paragraph>
    <Paragraph position="6"> Table 1 shows reading accuracy for patient WB for the various tests performed by Funnell together with the corresponding results of simulations of impaired and non-faulty PbA.</Paragraph>
    <Paragraph position="7"> Table 2 shows the results for patient FL reading aloud and the corresponding simulation of faulty and non-faulty PbA. Evidently, it is possible to reproduce quite well both patients' symptoms. Indeed, with Version 1, we can interpret WB's condition very directly: The concatenation process involved in nonword reading is completely destroyed but the mental lexicon is relatively spared. Because of the absence of some compound words (e.g., gentlelman ) from the dictionary, the simulations concerning &amp;quot;parent words&amp;quot; (e.g., father is the parent of.fat and her) for both Test 1 and Test 2 are not perfect.</Paragraph>
    <Paragraph position="8"> Version 2 is slightly poorer but still close to the neuropsychological data. For patient FL, the faulty version reproduces her impaired reading of single letters and 'easy' nonwords very well, but does so less well for 'difficult' nonwords.</Paragraph>
    <Paragraph position="9">  The simulations also handle the fact that these patients were completely unable to read single letters: the silence problem (see above) can occur for single letters by virtue of the form of the pronunciation lattice used, which requires matching bigrams (at least) at all positions to produce a pronunciation.</Paragraph>
  </Section>
  <Section position="6" start_page="16" end_page="16" type="metho">
    <SectionTitle>
5 Modelling Surface Dyslexia
</SectionTitle>
    <Paragraph position="0"> We have also modelled data from patient KT described by McCarthy and Warrington (1986).</Paragraph>
    <Paragraph position="1"> KT was able to pronounce regular words and nonwords very well but had serious difficulty in reading irregular words, tending to produce regularisation errors. (Again, limitations of space mean we must refer the reader to the original source for details of the reading tests and materials.) Together with WB, these patients have been taken as almost an existence proof of dual routes which can be differentially damaged.</Paragraph>
    <Paragraph position="2"> We suppose that KT's impairment results from injury to two components of the PbA model. First, as in phonological dyslexia, we assume that the process of segmentation into all possible substrings is partially damaged. More specifically, we postulate a deficiency concerning the size of the window involved in the pattern matching. Second, it is assumed that one or several (of the total of five) multi-strategies may be degraded.</Paragraph>
    <Paragraph position="3"> The simulations were obtained for a model with damage in the third and fourth multi-strategies (see Marchand and Damper, 2000, for detailed specification) and only substrings of length between 2 and 4 can be segmented in pattern matching. Table 3 shows KT's mean reading accuracy over the various tests performed by McCarthy and Warrington together with our corresponding simulation results for impaired and non-faulty PbA. Clearly, it is possible to reproduce quite well the patient's cardinal symptoms: his ability to pronounce regular words much better than irregular ones. The incorrect pronunciations show a clear regularisation effect (not detailed here).</Paragraph>
  </Section>
class="xml-element"></Paper>