ÿþ<html> <head> <meta http-equiv="Content-Type" content="text/html; charset=windows-1251"> <title>The Night Whites Language Workshop - St. Petersburg Winter Symposium on Experimental Studies of Speech and Language </title> </head> <body bgcolor="#FFFFFF" link="green" alink="blue" vlink="green"> <center> <h3><font face="Tahoma">The <i>Night Whites</i> Language Workshop</font></h3> <h4><font face="Tahoma">St. Petersburg Winter Symposium on Experimental Studies of Speech and Language</font></h4> <h4><font face="Tahoma">!0=:B-5B5@1C@3A:89 78<=89 A8<?>78C< ?> M:A?5@8<5=B0;L=K< 8AA;54>20=8O< O7K:0 8 @5G8</font></h4> </center> <hr width=50%> <center> <h5><font face="Tahoma">16-17.XII.2011<br>St. Petersburg, Russia</font></h5> </center> <br> <table ALIGN=center border=0 cellpadding=15> <tr> <td width=10% ></td> <td width=10% valign=top halign=right bgcolor="black" > <font size=-1 face="Tahoma" color="white"><b> <br><br><a href=index.shtml>About the meeting</a> <br><br><a href=abstracts.shtml>Abstracts</a> <br><br><a href=video.shtml>Video</a> </b></font> </td> <td width=58% valign=top> <font size=-1 face="Tahoma"> <p> <b>Abstracts of all presentations</b> <br><br>All abstracts are included below, first for the <a href="#talks">oral sessions</a> and later for the <a href="#posters">poster session<a> (organised alphabetically). <br><br>The abstracts are included 'as is', without proofing, translatation or correction. The organisers are not repsonsible for any orthogpraphic or stylistic errors. <br><br>The meeting was broadcast live over the internet. The record of this stream featuring all talks can be found <a href=http://video.yandex.ru/users/anzhalikad/collection/5/>here</a>. <hr> <br><br> <DL> <DT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><U><B><a NAME="talks">Abstracts of oral session talks:</B></U></SPAN></FONT></FONT><DT><P LANG="en-US"> <BR> </DL> <P LANG="en-US" ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Compensating Spatial Impairments: Spatial-Dynamic Stereotypes</B></SPAN></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><A NAME="Text1"></A><FONT FACE="Arial"><FONT SIZE=3>M. Bergelson*, O. Dragoy(***)*)*, </FONT><FONT SIZE=3><SPAN LANG="en-US">E. Iskra**, E. Mannova**, A. Skvortsov(**)*)*)**, A.&nbsp;Statnikov*</SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>*Lomonosov Moscow State University, Russia, **Center of Speech Pathology and Neurorehabilitation, Russia, ***Moscow Research Institute of Psychiatry, Russia, ****National Research University &quot;Higher School of Economics&quot;, Russia, *****University of Groningen, the Netherlands, ******Russian National Research Medical University, Russia</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><A NAME="Body"></A><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Spatial impairments show themselves not only in gnosis and praxis, but also in the language domain. Luria (1970) considered this deficit a characteristic feature of the so-called semantic aphasia. The aim of our study was to test if individuals with semantic aphasia do not apply spatial transformations and use therefore simple spatial stereotypes to interpret reversible linguistic constructions. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><A NAME="OLE_LINK1"></A><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">We contrasted two types of Russian reversible sentences, each in two conditions. Object-location constructions (</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>The boy is putting the box into the bag</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">) when being inverted (</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>The boy is putting into the bag the box</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">) violate the spatial-dynamic stereotype of first holding </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>the box</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> and then contacting </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>the bag</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">. In contrast, object-tool constructions (</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>The lady is covering the hat with the scarf</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">) only when being inverted (</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>The lady is covering with the scarf the hat</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">) follow the spatial-dynamic stereotype of first holding </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>the scarf</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> and then contacting </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>the hat</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">. Five patients with semantic aphasia performed sentence-picture matching task with those four types of constructions all representing natural word order in Russian. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The results confirmed a double dissociation between object-location and object-tool sentences: the direct word order </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>object-location</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> was related to higher accuracy scores than the reversed order </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>location-object</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">; but the reversed word order </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>tool-object</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> was related to higher accuracy scores than the direct order </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>object-tool</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">. The findings suggest that, indeed, patients with semantic aphasia use intact spatial stereotypes to compensate their linguistic deficit related to inability to set complex spatial relations.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> semantic aphasia, spatial-dynamic stereotypes, reversible constructions</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Mismatch negativity: word frequency effect in healthy individuals, disturbances of simple MMN in cognitively impaired</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="fr-FR">D. O. Boricheva, A. A. Alexandrov</SPAN></FONT></FONT><P LANG="fr-FR" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Saint Petersburg State University, Russia</I></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Mismatch negativity is the brain event-related potential known to reflect an automatic difference discrimination process. It is also considered to be a neurophysiological representation of memory traces for words in the brain. The enhancement of MMN is observed when more meaningful words are presented. We compared the MMN response to high- and low-frequency consonant-vowel-consonant words (the acoustic stimulus differences were the same) in a passive odd-ball paradigm in healthy Russian-speaking subjects and revealed more prominent MMN with shorter peak latency for high-frequency words (&quot;mir&quot;, 200.4 ipm, vs &ldquo;mor&rdquo;, 1.44 ipm). For other pairs, shorter MMN peak latencies were also obtained (&quot;god&rdquo;, 283.55 ipm, vs &ldquo;gid&rdquo;, 1.89 ipm; &ldquo;kot&rdquo;, 27.3 ipm, vs &ldquo;kit&rdquo;, 2.22 ipm).</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">When we tested the MMN to high- and low-frequency word pair in patients suffering acute stroke in the left middle cerebral artery circulation and, as a consequence, sensorymotor aphasia, we could not receive any significant differences between MMN latency or amplitude for the high- and low-frequency word. This could be due to general deformation of ERP responses in stroke patients and their group variability. So we decided to assess the disturbances of simple MMN to tone frequency differences in cognitively impaired patients with various localization of stroke and chronic cerebrovascular diseases. Currently, prolonged MMN latencies are obtained in patients with acute or old stroke in the left middle cerebral artery circulation. Also longer MMN peak latencies (or even absence of this ERP component) is seen when cognitive impairment is more prominent. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> mismatch negativity, word frequency, stroke</FONT></FONT><P ALIGN=JUSTIFY> <BR> <DL> <DT><P LANG="en-US"><BR> <DT><P LANG="en-US"><BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Anaphora in Language Acquisition and Pathology</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="fr-FR">T. V. Chernigovskaya</SPAN></FONT></FONT><P LANG="fr-FR" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Saint Petersburg State University, Russia</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>tatiana.chernigovskaya@gmail.com</FONT></FONT><DL> <DT> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">In contemporary linguistics, theories explaining rules of assigning reference to pronominal elements generally make use of syntactic notions (see Grodzinsky 2000). However, it is evident that non-syntactic factors, usually classified as pragmatic phenomena, influence the choice of antecedent for a pronoun/anaphor. The connection between contrastive stress and pronominal reference was noticed already in (Akmajian, Jackendoff 1970). Since then, this connection was studied in many projects (e.g. Maratsos 1973, McDaniel and Maxfield 1992, Reinhart and Reuland 1993, Avrutin 1999, Avrutin et al 2002, Schaeffer, Matthewson 2002, Ruigendijk et al 2005). Our experiment was aimed at investigating the interaction between syntactic and discourse rules in pronominal reference on Russian material and involved different groups of subjects, comparing results obtained from normal and aphasic adults with the data from normal language acquisition and SLI. The methodology used in the experiment was based on tests developed at Utrecht Institute of Linguistics. It was established earlier that both children and agrammatic patients have troubles interpreting pronouns in some construction types (Philip and Coopmans, 1996; Baauw, 2002). Our data show that despite crosslinguistic differences, the influence of pragmatic factors on pronominal reference, first noticed for Germanic languages, holds for Russian. A connection between pragmatic and syntactic rules governing pronominal reference, found both in language acquisition and disorder, shows that rules of anaphora cannot be reduced to syntactic devices only (Prokopenya, Chernigovskaya, Khrakovskaya 2009).</SPAN></FONT></FONT><DT><P LANG="en-US"> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords: </I></FONT><FONT SIZE=3> Anaphora, language acquisition, aphasia</FONT></FONT><P ALIGN=JUSTIFY> <BR> <DT><P LANG="en-US"><BR> <DT><P LANG="en-US"><BR> <DT><FONT FACE="Arial"><FONT SIZE=3><B>Lexical ambiguity as a tool in understanding the nature of the language processing system</B></FONT></FONT></DL> <P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>A.V. Dubasova</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>St. Petersburg State University, Russia</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>The amount of experimental research devoted to examining how different variables, mainly contextual information and relative meaning frequencies, affect the processing of lexically ambiguous words is considerable (Duffy et al. 1988; Rayner &amp; Frazier 1989; Dopkins, Morris, &amp; Rayner 1992; Sereno, O&rsquo;Donnell, &amp; Rayner 2006; Reichle, Pollatsek, &amp; Rayner, 2006; etc.). As a rule, the data are interpreted in order to support one or another theory of lexical ambiguity resolution.</FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3>In my mini-review I highlight the idea that lexical ambiguity could be used as an important tool in understanding the nature of the language processing system (the idea was expressed, but not elaborated on, in Binder &amp; Rayner, 1998). In particular, I focus on the following issues: the hierarchy among the variables affecting lexical ambiguity resolution and the ways to interpret it, lexical ambiguity and modularity vs. interactive processing or serial vs. parallel computation, the assumptions of the theories that have been proposed and their application to the explanation of lexical ambiguity within the framework of the general models of eye-movement control during reading.</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> lexical ambiguity, sentence processing in reading</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Neurophysiology of speech act processing</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>N. Egorova</FONT><SUP><FONT SIZE=3>*,**</FONT></SUP><FONT SIZE=3>, F. Pulvermuller</FONT><SUP><FONT SIZE=3>*</FONT></SUP><FONT SIZE=3>, &amp; Y. Shtyrov</FONT><SUP><FONT SIZE=3>*</FONT></SUP></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><SUP><FONT SIZE=3>*</FONT></SUP><FONT SIZE=3><I>MRC Cognition and Brain Sciences Unit, Cambridge, </I></FONT><SUP><FONT SIZE=3>**</FONT></SUP><FONT SIZE=3><I>University of Cambridge</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>A speech act is an utterance situated in context (Austin, 1962; Searle, 1975). Even a single word can perform different actions depending on its context (Wittgenstein, 1953; Dore, 1983). For example, a word can be used to name an object or to request it: the same utterance &ldquo;Orange!&rdquo; could mean either &ldquo;This is an orange&rdquo; or &ldquo;Give me an orange&rdquo; depending on the context. The human ability to categorize speech acts has been demonstrated theoretically and experimentally in a number of behavioural (e.g. Holtgraves, 2008; Clark &amp; Lucy 1975) and neuropsychological studies (Soroker et al., 2005). Little is known, however, about neural networks subservivng speech acts processing in the healthy brain and the temporal dynamics of pragmatic context processing. We have addressed these questions in an EEG experiment to capture both the structural substrate and the neural dynamics of speech act perception. Healthy participants watched experimental videos featuring a context sentence (eliciting a speech act of Naming or Request), words (used to perform the act of Naming or Request), and respective actions (pointing in the Naming condition and handing an object over in the Request condition). Identical single word tokens counterbalanced across subjects were used for both Naming and Request, which allowed for direct comparison between the two conditions and helped avoid the confounds of syntactic complexity and physical stimulus differences. The results suggest that speech act processing is reflected in brain dynamics very early on (100-150 ms), with identical words eliciting more activation in the Request than Naming context, and appears to be underpinned by overlapping but distinct bilaterally distributed cortical networks.</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> EEG, speech act, pragmatics</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Interaction of the verbal, prosodic, and visual components in language understanding</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>A. A. Kibrik</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Institute of Linguistics, Russian Academy of Sciences, Moscow, Russia</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Regular face-to-face communication consists of several channels or components, in particular the verbal, or segmental, component, prosody, and body language. Different research traditions make very different assumptions about the relative importance of these channels. Mainstream linguistics concentrates on the verbal component alone, while applied psychology often claims that the hierarchy of importance is rather like body language &gt; prosody &gt; verbal component. The goal of this study is to empirically test the relative contribution of the three channels to the overall communicative achievement.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">I report several studies performed under my supervision by the students of the Department of Theoretical and Applied Linguistics, Moscow State University in 2007&ndash;2010. Each of these studies was devoted to a different type of spoken discourse, including communication between characters in a movie and several kinds of videotaped regular conversation. In each study the goal was set to separate the communication channel and present different combinations thereof to different groups of experiment participants. One of the groups watched the original video, another group only had access to the verbal component, still another to the verbal and visual components, etc. &ndash; eight experimental groups altogether. The degree of understanding the discourse was tested with a help of a questionnaire. In most studies, the greatest contribution was provided by the verbal component, the second largest by the visual component. However, all components, including prosody, provided a contribution that was far from negligible. This supports the idea that linguistic communication is fundamentally multimodal and linguistics must pay attention to all of the communication channels.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> multimodality, spoken discourse, communication channels</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Cognitive processes and lanquage acquirement in deaf children after cochlear implantation</B></SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">I.V. Koroleva</SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>Herzen's State Pedagogical University, </I></SPAN></FONT></FONT> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>Sankt-Petersburg's Institute of ENT and Speech, Russia</I></SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Cochlear implantation is a method of auditory rehabilitation of deaf people using transfer of speech and other sounds to brain by means of auditory nerve stimulation with electrodes. Language acquirement by deaf children with cochlear implant (CI) is curious model of development of speech after end of the basic part of the sensitive speech period. Expressive and impressive vocabulary, the grammatical system are not formed at many of these children before implantation. Impossibility to perceive speech of adults damages development various cognitive processes in child: modal specific &ndash; auditory speech processing, auditory memory and attention, and nonspecific operations important for language acquirement also - generalization, abstraction, signification. Auditory speech processing, language system and oral speech as communicative and thinking tools are developed at CI children owing the possibility to hear speech. Development of all these functions happens at children with CI in two ways - during purposeful training and spontaneously during the daily. The contribution of purposeful training and spontaneous learning in speech development depends on age of implantation of the child. At children implanted before 2nd years, acquirement by speech and cognitive processes happens mostly spontaneously. Nevertheless, even teenagers keep the some ability to acquire new words and grammatical rules spontaneously. It is discussed the conditions and cognitive processes supporting ability of spontaneous language and speech acquisition at CI children</SPAN></FONT><SPAN LANG="en-US">.</SPAN></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords</I></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>:</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT><FONT SIZE=3>language, cochlear implant</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Auditory deficits and their amelioration in dyslexia</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>T. Kujala</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Cicero Learning, University of Helsinki, Finland, </I></FONT></FONT> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Cognitive Brain Research Unit, Institute of Behavioural Sciences, University of Helsinki, Finland</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>Auditory perceptual deficits have been associated with developmental language disorders, such as dyslexia and specific language impairment. The abnormal pattern of auditory neural processing is evident already in infants and young children, as suggested by studies using event-related potentials (ERP). For example, ERPs in pre-school children at risk for dyslexia indicated a wide-spread pattern of impaired processing of acoustic and phonetic differences of speech sounds. Various intervention approaches were shown to be effective in improving reading and language-related skills in children with dyslexia. These improvements of skills were shown to parallel with neural plastic changes of the auditory system. An important goal is to find means for preventing or diminishing language-related deficits before school age. Preventive intervention would diminish learning backwardness and a variety of problems resulting from inferior school performance. Recent studies suggest that speech discrimination and competencies required for successful reading-skill acquisition can be improved in preschoolers. These improvements are reflected in neural processing changes indicating plasticity of the auditory system. A future challenge is to find means to identify subtypes of language- and learning-related deficits before school age and to develop optimal intervention programs for these different subtypes.</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> dyslexia, intervention, brain research</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Perceptive-significant features of vowels with different fundamental frequency</B></SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">G. A. Kulikov, N. G. Andreeva</SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>St. Petersburg State University, Russia</I></SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <DL> <DD><P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Russian back </SPAN></FONT><FONT SIZE=3>vowels </FONT><FONT SIZE=3><SPAN LANG="en-US">[a], [o], [u] </SPAN></FONT><FONT SIZE=3>in high-pitched speech</FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT><FONT SIZE=3>are not differentiated by the exact formant values, but instead</FONT><FONT SIZE=3><SPAN LANG="en-US"> of</SPAN></FONT><FONT SIZE=3>, by the amplitude ratios of the spectral components. </FONT><FONT SIZE=3><SPAN LANG="en-US">As far as the F0 increase</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><U>,</U></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> the representation of the phonetically identical vowels shifts and different vowels occupy separate but other areas in the corresponding amplitude coordinate space. The existence of vowel perceptual </SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US">boundaries</SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US"> on the basis of spectral components amplitude ratios was confirmed in t</SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US">wo series of psychophysical </SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US">experiments. In the first one test stimuli whose amplitude ratios were changed from original while spectral maxima frequencies were kept constant, were used. It was shown that</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">phonetic interpretation of vowels is influenced by their components relative amplitude values. In the second </SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US">frequency-transposition scheme were used. In this case the amplitude envelopes of vowels </SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US">were kept constant while signals shift along frequency axis. </SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US">The effects of frequency shift results in </SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US">perceived vowel quality</SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US"> expected changing </SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US">while</SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US"> vowel p</SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US">erceptual </SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US">boundary is </SPAN></FONT></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US">dependent on</SPAN></FONT></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US"><B> </B></SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US">amplitude ratios. Comparing the relative amplitude crucial point values with the data obtained for natural vowels at the same fundamental frequency we achieved a correspondence with the same parameters of natural vowels </SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US">irrespective of adult&rsquo;s, child&rsquo;s</SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US"> or sung vowels were transformed.</SPAN></FONT></FONT><DD><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The study is supported by grant RFBR ! 11-06-00125 &ldquo;</SPAN></FONT><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US">Peculiarities of reflection vowels with different fundamental frequency in electrical activity of left and right human brain hemisphere&rdquo;.</SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT></FONT> <DD><P LANG="en-US" ALIGN=JUSTIFY><BR> <DD><P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>Keywords</I></SPAN></FONT><FONT SIZE=3>:</FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT><FONT SIZE=3>vowels</FONT><FONT SIZE=3><SPAN LANG="en-US">, perception, fundamental frequency</SPAN></FONT></FONT></DL> <P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Prediction, attention and memory: what disfluency tells us about the comprehension of spontaneous speech </B></FONT></FONT> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>L. J. MacGregor*, M. Corley**, D. I. Donaldson*** </FONT></FONT> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>*MRC Cognition and Brain Sciences Unit, Cambridge, ** Philosophy, Psychology and Language Sciences, University of Edinburgh, ***Psychology, University of Stirling, UK </I></FONT></FONT> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>One characteristic feature of spontaneous speech is the presence of disfluencies: phenomena such hesitations, repetitions and repairs that interrupt the flow of speech. In recent years there has been increasing interest in the impact of disfluencies on listeners and, perhaps surprisingly, studies suggest they are not necessarily detrimental to language comprehension. These findings are important because they contrast with a long-standing perception that disfluencies are noise in the speech signal which mar communication. We provide a novel view of disfluency processing drawing on a range of evidence, from reaction time, eye tracking and ERP studies, which shows how disfluencies impact speech processing. We argue that predictive, attentional and memory processes interact to allow listeners to maintain seemingly fluent comprehension of disfluent speech - and that these processes are a fundamental but often neglected part of the speech comprehension system. </FONT></FONT> <P ALIGN=CENTER><BR> <DL> <DT><FONT FACE="Arial"><FONT SIZE=3><I>Keywords</I></FONT><FONT SIZE=3>: disfluency, speech, comprehension </FONT></FONT> </DL> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Visual attention, conceptual accessibility, and structural choice in sentence production</B></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>A. Myachykov, D. Thompson, S. Garrod, &amp; C. Scheepers</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>University of Glasgow, UK</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>Existing theories implicate thematic prominence (Kako, 2006), conceptual accessibility (e.g., Bock &amp; Warren, 1985), and visual attention (e.g., Myachykov, et al., 2001) among factors predicting structural choice during visual-world sentence production. Here, we report the results of two studies that tested these theoretical proposals by using perceptual and referential priming paradigms. In both experiments </FONT><FONT SIZE=3>we </FONT><FONT SIZE=3>investigated how conceptually informative (referent preview) and conceptually uninformative (pointer to referent&rsquo;s location) visual cues affect structural choice during English transitive sentence production. The difference between the two studies was the Cue Duration (Implicit (70msec) and Explicit (700 msec)). Cueing the Agent or the Patient prior to presenting the target event reliably predicted the likelihood of selecting this referent as the sentential Subject, triggering, correspondingly, the choice between active and passive voice. Importantly, there was no difference in the magnitude of the general Cueing effect between the informative and uninformative cueing conditions, suggesting that attentionally driven structural selection relies on a direct automatic mapping mechanism from attentional focus to the Subject&rsquo;s position in a sentence. This mechanism is, therefore, independent of accessing semantic, and possibly lexical, information about the cued referent provided by referent preview. Analyses of referent-related eye-voice spans and sentence onset latencies revealed that the hierarchy of referential roles (e.g., Agent vs Patient) affects the underlying performance of the sentence generator and the corresponding time-course of overt sentence formulation independently of making configurational choices during sentence production. </FONT></FONT> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> Attention, Conceptual Accessibility, Syntax</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Mental Simulations of Voice and Speech during Reading of Quotations </B></FONT></FONT> <DL> <DT><BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>C. Scheepers, B. Yao, &amp; P. Belin</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Institute of Neuroscience and Psychology, University of Glasgow, UK</I></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>In human communication, direct speech (e.g. </FONT><FONT SIZE=3><I>Mary says: &ldquo;I am hungry&rdquo;</I></FONT><FONT SIZE=3>) is typically perceived as being more vivid than indirect speech (e.g. </FONT><FONT SIZE=3><I>Mary says she is hungry</I></FONT><FONT SIZE=3>). Not only has this vividness distinction been discussed by linguists (e.g. Tannen, 1986; 1989), it has also been shown, for example, that </FONT><FONT COLOR="#000000"><FONT SIZE=3>speakers are more likely to employ direct rather than indirect speech when instructed to entertain, as opposed to just inform, a listener (Wade &amp; Clark, 1993).</FONT></FONT><FONT SIZE=3> Here, we present evidence from a series of fMRI and eye-tracking experiments showing that readers are more likely to mentally simulate aspects of the quoted speaker&rsquo;s voice (and manner of speaking) when reading direct speech as opposed to meaning-equivalent indirect speech quotations. Specifically, we show that brain areas that are activated during an auditory voice-localizer task (Belin et al., 2000) also become more active during silent reading of direct speech quotations. Furthermore, we show that both oral and silent readers spontaneously adjust their reading rates to contextually implied speech rates (context describing either a </FONT><FONT SIZE=3><I>fast</I></FONT><FONT SIZE=3> or a </FONT><FONT SIZE=3><I>slow</I></FONT><FONT SIZE=3> quoted speaker) when reading direct speech quotations, but not when reading comparable indirect speech quotations. The results are in line with embodied cognition and the notion that the use of direct speech is grounded in vivid demonstrations of voice and/or way of speaking: </FONT><FONT COLOR="#000000"><FONT SIZE=3>A direct-speech reporting style is more likely to be taken as a cue to spontaneously engage in vivid perceptual simulations of the reported speech act than an indirect-speech reporting style.</FONT></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> Quotations, Reading, Embodied Cognition</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Neural bases of rapid word learning</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>Y. Shtyrov</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Medical Research Council (MRC), Cognition and Brain Sciences Unit, Cambridge, United Kingdom, </I></FONT></FONT> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Cognitive Brain Research Unit, Institute of Behavioural Sciences, University of Helsinki, Finland</I></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>Humans are unique in developing large lexicons as their communication tool; to achieve this, they are able to learn new words rapidly. However, neural bases of this rapid learning, which may be an expression of a more general cognitive mechanism likely rooted in plasticity at cellular and synaptic levels, are not yet understood. In this presentation, we highlight a selection of recent studies that attempted to trace word-learning in the human brain non-invasively. A number of brain areas, most notably in hippocampus and neocortex, appear to take part in word acquisition. Critically, the currently available data not only demonstrate hippocampal role in rapid encoding followed by slow-rate consolidation of cortical word memory traces, but also suggest immediate neocortical involvement in the word memory trace formation. To explore this in more detail, we will present our recent EEG studies that demonstrate rapid development of cortical memory traces for novel word forms over a short session of passive auditory exposure to these items. Echoing early behavioural studies in ultra-rapid word learning, the current experiments can be taken to suggest that our brain may effectively form new cortical circuits online, as it gets exposed to novel linguistic patterns in the sensory input.</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> fast mapping, brain, cortex</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Agreement attraction: an overview of the recent literature and a contribution from Russian</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>N. Slioussar*</FONT><SUP><FONT SIZE=3>,</FONT></SUP><FONT SIZE=3>**, A. Malko**</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>*Utrecht institute of Linguistics OTS, the Netherlands, **St. Petersburg State University, Russia</I></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>So-called attraction errors in agreement when the verb agrees not with the subject, but with another noun (e.g., </FONT><FONT SIZE=3><I>the key to the cabinets are&hellip;</I></FONT><FONT SIZE=3>) received a lot of attention in the last two decades (Bock &amp; Miller 1991; Bock et al. 2001; Eberhard et al. 2005; Franck et al. 2002, 2006; Solomon &amp; Pearlmutter 2004; Wagers et al. 2009 and many others). Studies based on several languages showed that such errors are produced in high numbers in experimental conditions and examined various syntactic, semantic and morphological factors affecting their frequency, which sheds light on the workings of the mental grammar. In our talk, we will summarize these findings and will report the results of a pilot experiment on Russian.</FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3>Previous studies focused on number agreement and revealed a &lsquo;plural markedness effect&rsquo;: Plural nouns erroneously attract agreement significantly more often than Singular ones (i.e. errors like </FONT><FONT SIZE=3><I>the keys to the cabinet is&hellip; </I></FONT><FONT SIZE=3> are very rare). To have a clearer understanding of the nature of such markedness and some other phenomena, it is crucial to look at the more complex gender agreement systems. This was done in several studies, but only Badecker and Kuminiak (2007) controlled for a number of relevant factors in their experiments on Slovak. In particular, attraction tends to happen if the form of the potential attractor coincides with the Nominative case form (as in </FONT><FONT SIZE=3><I>poezdka v gorod </I></FONT><FONT SIZE=3>as opposed to </FONT><FONT SIZE=3><I>poezdka v derevnju</I></FONT><FONT SIZE=3>). We will compare Badecker and Kuminiak&rsquo;s results with our similarly constrained experiment on Russian.</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> agreement, errors, Russian</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P LANG="en-US" ALIGN=CENTER><BR> <P LANG="en-US" ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Action-related verbs and nouns activate motor cortex: replication and extension of previous results using Russian-language MEG data</B></SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="sv-SE">T. Stroganova</SPAN></FONT><SUP><FONT SIZE=3><SPAN LANG="sv-SE">*</SPAN></FONT></SUP><FONT SIZE=3><SPAN LANG="sv-SE">, A. Butorina</SPAN></FONT><SUP><FONT SIZE=3><SPAN LANG="sv-SE">*</SPAN></FONT></SUP><FONT SIZE=3><SPAN LANG="sv-SE">, A. Nikolaeva</SPAN></FONT><SUP><FONT SIZE=3><SPAN LANG="sv-SE">*</SPAN></FONT></SUP><FONT SIZE=3><SPAN LANG="sv-SE">, Y. Shtyrov</SPAN></FONT><SUP><FONT SIZE=3><SPAN LANG="sv-SE">**,**</SPAN></FONT></SUP></FONT><P LANG="sv-SE" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><SUP><FONT SIZE=3>*</FONT></SUP><FONT SIZE=3><I>MEG Centre, Moscow State University for Psychology &amp; Education, Russia; </I></FONT><SUP><FONT SIZE=3>**</FONT></SUP><FONT SIZE=3><I>Medical Research Council (MRC), Cognition and Brain Sciences Unit, Cambridge, United Kingdom;</I></FONT></FONT><P ALIGN=CENTER> <FONT FACE="Arial"><SUP><FONT SIZE=3>***</FONT></SUP><FONT SIZE=3><I>Cognitive Brain Research Unit, Institute of Behavioural Sciences, University of Helsinki, Finland</I></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Background and Objectives.</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> Perception of action verbs is known to selectively activate the respective representations in the motor cortex at the early stage of word recognition. We wished to study for the first time whether the action </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>nouns </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">follow the same spatial-temporal activation patterns as that known for action </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>verbs</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">, and to specify, using the distribution source modeling of MEG data, whether the action words activate primary or secondary areas of motor cortex as well as the time-course of this activation.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Methods.</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> We selected three pairs of Russian words related to actions performed by hand, foot and mouth, and each pair included the verb and the noun having the same word stem. The stimuli were presented within mismatch field (MMF/MMN) paradigm in three blocks corresponding to the word pairs, which were used as rare unexpected deviant stimuli, while the frequent standards were pseudo-words chosen for each word pair in a such a way that their onsets were phonologically identical to the test words, and the acoustic disambiguation point between the 2 words and pseudo-word occurred at 260 ms after the word onset. The MMF source activation time courses in the motor cortex were analyzed.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Results and Conclusions. </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">We demonstrate for the first time that both lexical categories of Russian action words activate the motor cortex somatotopically, but source distribution is shifted somewhat anterior to primary motor cortex localization. Moreover, cognate verbs and nouns have very similar spatial-temporal pattern of motor cortex activation that reaches its peak just ~100 ms after the word disambiguation point and ~30 ms ahead of the respective response in auditory cortex. Possible interpretations of this pattern of results are discussed</SPAN></FONT><FONT SIZE=3>.</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> action words, motor cortex, MEG</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><BR> <DL> <DT><FONT FACE="Arial"><FONT SIZE=3><U><B><a NAME="posters">Poster presentation abstracts:</B></U></FONT></FONT><DT> <BR> </DL> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Empirical research of bilingual mental lexicon</B></SPAN></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>E. M. Alekseeva</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Kazan Federal University, Russian Federation</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The problem of bilingual mental lexicon is of special interest for researchers. How is bilingual mental lexicon constructed? It was a central question of the research work. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">We proceeded from development hypothesis (A. De Groot, P. Plieger). Also, the point of view was important that the mental lexicon exists in form of networks (semantic, associative), word groups and word families.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">78 German probands took part in the empirical investigation. In the test group empirical research of bilingual mental lexicon associative structure took place. In the control group the associative structure of native language mental lexicon came to be a subject of scientific research.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">In the investigation association methods were used and two associative tests were carried out: &bdquo;directed&ldquo; associative test (given specified associations, measuring instrument: numerical scale) and &bdquo;undirected&ldquo; associative test (free associations, result: subjective semantic fields, associative fields).</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The achieved results will be presented. Proofs will be produced that two mechanisms are typical for structure of bilingual mental lexicon. By &bdquo;subordinate&ldquo; organization meaning units, their associative connections are fully identical in both languages because they have been adopted from the native language. Instead of native language word a foreign word is being used. By &bdquo;coordinated&ldquo; organization it is so that words and their meanings are being separately learned and remain in the memory. Associative connections of foreign words are also being formed separately from the native language. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Finally quantitative und qualitative research results of semantic and associative word fields in the native and foreign languages will be discussed.</SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3><SPAN LANG="en-US"><I> </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">bilingual mental lexicon</SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Intrinsic vowel duration and lexical stress perception in Russian spontaneous speech</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>I. E. Apushkina</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>St. Petersburg State University, Russia</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Vowel duration is considered one of the key stress parameters in Russian language, the pronunciation of unstressed vowels characterized by a strong quantitative (and qualitative) reduction. The effect of intrinsic vowel duration can be illustrated by means of instrumental analysis when a stressed vowel with a lower intrinsic duration can be shorter than an unstressed one. Previous experiments, using synthesized asemantic stimuli with varied vowel duration ratio, have shown that intrinsic duration of the vowels affects lexical stress perception. This article deals with the data taken from spontaneous Russian speech and focuses on the duration ratio of stressed and unstressed [a] and [i] in comparison to that received in the experiments.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> intrinsic vowel duration, spontaneous speech, lexical stress.</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>The ORD Speech Corpus of Russian Everyday Interaction &quot;One speaker's day (ORD)&quot;</B></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>A. S. Asinovsky, N. V. Bogdanova, I. V. Koroleva, E. V. Markasova, A. I. Ryko, T.&nbsp;U</I></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>.</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="be-BY"><I>&nbsp;</I></SPAN></FONT><FONT SIZE=3><I>Sherstinova&nbsp; &amp; S. B. Stepanova</I></FONT></FONT><P ALIGN=CENTER> &ensp;<P ALIGN=CENTER> <FONT FACE="Arial"><FONT SIZE=3><I>St. Petersburg State University, Russia</I></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>The main goal of the present investigation is to fix Russian spontaneous speech in natural communicative situations. A group of subjects was formed that included 40 participants (20 men and 20 women).</FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3>The overall length of the recorded material is about 300 hours. 240 hours contain speech data quite suitable for further analysis. Two professional annotation tools are used to annotate the ORD corpus &ndash; ELAN (for primary and general annotations) and Praat (for making real phonetic transcription). </FONT></FONT> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><U>Primary annotation</U></FONT><FONT SIZE=3> of speech corpus implies annotation of the following data types:</FONT></FONT><DL> <DD><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><B>Frase</B></FONT><FONT SIZE=3> &ndash; orthographic transcripts of phrases.</FONT></FONT><DD><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><B>Event</B></FONT><FONT SIZE=3> &ndash; non-language audio events including as well some voiced events.</FONT></FONT><DD><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><B>Speaker</B></FONT><FONT SIZE=3> is a person who pronounced correspondent phrase.</FONT></FONT><DD><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><B>Voice</B></FONT><FONT SIZE=3> is the special characteristic of speech for the given phrase or its segment .</FONT></FONT><DD><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><B>FraseComment</B></FONT><FONT SIZE=3> contains all kinds of comments on phrase realization.</FONT></FONT><DD><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><B>Episode</B></FONT><FONT SIZE=3> refers to general communicational situation. </FONT></FONT> </DL> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT COLOR="#0d0d0d"><FONT SIZE=3><B>FonetCom</B></FONT></FONT><FONT COLOR="#0d0d0d"><FONT SIZE=3> &ndash; not-standard phonetic features</FONT></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT COLOR="#0d0d0d"><FONT SIZE=3><U>The second stage</U></FONT></FONT><FONT COLOR="#0d0d0d"><FONT SIZE=3> of corpus annotation is made on a lexical level and includes tagging for the following main tiers: Words (spelling), POS, GramForm (grammatical form), SyntRole (syntactic role).</FONT></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT COLOR="#0d0d0d"><FONT SIZE=3>Though creating of the ORD corpus is still in progress, its diverse speech material and annotations have given birth to a number of linguistic and interdisciplinary researches.</FONT></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> </FONT><FONT SIZE=3>speech corpus researches</FONT><FONT SIZE=3><SPAN LANG="ru-RU">&ensp;&ensp;&ensp;</SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Eye movements during reading as an indicator of development of reading skill.</B></SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">M. M. Bezrukikh*, V. V. Ivanov* and A. A. Demidov**</SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>*Institute of Developmental Psysiology RAE, Moscow, Russia, **Center of Experimental Psychology, Moscow, MCUPE, Moscow, Russia</I></SPAN></FONT></FONT><DL> <DT><P LANG="en-US"> <A NAME="OLE_LINK3"></A><BR> </DL> <P ALIGN=JUSTIFY><A NAME="OLE_LINK9"></A><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">For the most part, eye movements during reading are considered as reflection of processes of linguistic analysis (S.P. Liversedge, J.M. Findlay, 2000; M.G. Calvo, E. Meseguer, 2002; E.D. Reichle, A. Pollatsek, K. Rayner, 2006), visual perception (L. Huestegge et al, 2003; S.P. Liversedge, K. Rayner et al, 2004) and attention (J. Hyona, M. Koivisto, 2006). Studying eye movements during reading texts of different complexity at the initial stages of reading skill development in children with high and low reading skill allows to understand mechanisms of oculomotor activity.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">We analyzed the basic parameters of oculomotor activity (fixations duration, percent of regressive saccades, amplitudes of progressive and regressive saccades, general reading time) in schoolchildren of the first (n=53) and the fourth (n=47) grades with high and low reading skill.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <A NAME="OLE_LINK6"></A><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">It is shown that in the course of reading ability development from the 1st to the 4th grade, general reading time and average fixations duration decreases (p&lt;0,001) while amplitudes of saccades increases (p&lt;0,001). These changes are most pronounced when reading difficult texts.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <A NAME="OLE_LINK11"></A><A NAME="OLE_LINK7"></A><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Also differences are found between parameters of oculomotor activity in children of the first grade with high vs. low reading skill: All listed parameters turned out to be greater in children with low reading skill. In children of the fourth grade, the only significant differences are between amplitudes of progressive (p&lt;0,001) and regressive (p&lt;0,01) saccades, which testifies to development of reading skills. The latter is also supported by the results of comparative analysis of oculomotor activity during reading texts of different complexity: in the first grade children, text complexity does not affect quality of reading.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">It is hypothesised that at the initial stages of reading skill development, the leading mechanisms that shaped up the oculomotor activity are cognitive processes, especially those connected with the attention and visual perception.</SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <DL> <DT><FONT FACE="Arial"><FONT SIZE=3><I>Keywords: </I></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>eye movements, reading</I></SPAN></FONT></FONT></DL> <P ALIGN=JUSTIFY> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>The independent development of phonological and semantic systems during ontogenesis and phylogenesis: a review of the evidence</B></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>O. Crawford</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>The Open University, UK</I></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>Language comprises interlinked phonological and semantic systems. Evidence suggests an independent development of these systems during ontogenesis, development of the individual, and phylogenesis, the evolution of species. In ontogenesis phonological system develops on the basis of innate abilities of infants to hear fine acoustic distinctions and differentiate separate phonemes of language (Maurer and Maurer, 1988). During first year of life these abilities develop further to differentiate and encode sound word-forms by using statistical properties of language (Johnson and Jusczyk, 2001). The semantic system develops on the basis of innate visual abilities to categorise (Fantz, 1963) and conceptualise (Hespos and Spelke, 2004) objects and actions. By 8-12 months an integration of sound word-forms and meanings takes place and children start to use their first words. Phylogenetic evidence suggests that primates also have sophisticated abilities to categorise and conceptualise (Fize et al, 2011) visual objects and actions. As human infants, primates are highly sensitive to environmental sounds (Ramus et al, 2000). However, primates are lacking in ability to combine separate phonemes and link them to the meaning. Their auditory system is specialised in differentiation of voices and calls of conspecific individuals (Romanski and Averbeck, 2009).As a result meanings are communicated by either calls of different frequency or by gestures and actions. Therefore, an ability to combine different phonological units together and link these units to the meaning is unique to humans and can be considered as one of the factors that contributed to the appearance of language during phylogenesis. </FONT></FONT> <P ALIGN=JUSTIFY><BR> <DL> <DT><FONT FACE="Arial"><FONT SIZE=3><I>Keywords: </I></FONT><FONT SIZE=3>phonological and semantic networks, origin of language, language development</FONT></FONT></DL> <P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>The </B></SPAN></FONT><FONT SIZE=3><B>Objective </B></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><B>!</B></SPAN></FONT><FONT SIZE=3><B>riteria to </B></FONT><FONT SIZE=3><SPAN LANG="en-US"><B>I</B></SPAN></FONT><FONT SIZE=3><B>dentify </B></FONT><FONT SIZE=3><SPAN LANG="en-US"><B>V</B></SPAN></FONT><FONT SIZE=3><B>ocal </B></FONT><FONT SIZE=3><SPAN LANG="en-US"><B>F</B></SPAN></FONT><FONT SIZE=3><B>atigue</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>K.V. Evgrafova, V.V. Evdokimova</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>St. Petersburg State University, Russia</I></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Vocal fatigue is a voice symptom which is frequently reported by professional voice users. Teachers, singers, actors and other professions that require prolonged voice use are especially at-risk group. The vocal fatigue results in auditory perceptual and acoustic changes in the voice signal and can lead to serious pathological conditions. The present study has examined acoustic manifestations of the vocal fatigue in pronunciation teachers who seem to be particularly susceptible to vocal and articulatory fatigue. In the paper detailed acoustic analysis of the data obtained is presented. The results of the acoustic analysis showed a consistent dependency between acoustic parameters and vocal fatigue.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> vocal fatigue, acoustics</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>An EMA Study of Timing of Russian Vowel Pronunciation</B></SPAN></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">O. Glotova, D. Kocharov</SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>St. Petersburg State University, Russia</I></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>The paper is devoted to the study of vowel articulation timing in Russian. Articulatory movement has been subdivided into three phases relative to the audible vowel sound: onset (movement prior to phonation), phonation, and offset (movement after the end of phonation). For</FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT><FONT SIZE=3>each</FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT><FONT SIZE=3>articulation phase, phase duration and velocity of articulatory movement was measured and the phases were further compared to each other</FONT><FONT SIZE=3><SPAN LANG="en-US">. Thus, the relationship between the timing of audible sound and that of tongue movement was captured.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">All data were obtained by means of Carstens AG500 electromagnetic articulograph. Eight measurement points were used. Five sensors were placed on the subject&rsquo;s tongue and lips: approximately 5 mm from the tongue tip, the front end of the back of the tongue, the middle at an equal distance between the other two sensors on the tongue, upper lip, lower lip. Three sensors were placed on passive articulators: the highest point of the hard palate, the most protruding point of the alveolar ridge, and the upper incisors to provide some information on the position of the tongue relative to the roof of the mouth.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The data consists of a randomized set of isolated vowels read by four native speakers of Russian, two men and two women. All six Russian vowels were read 36 times.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Experimental data analysis showed the following tendencies. On average, the onset is twice as long as the duration of audible sound. The velocity of articulation is greater during the &lsquo;silent&rsquo; phases of articulation, onset and offset. </SPAN></FONT></FONT> <DL> <DT><BR> </DL> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> electomagnetic articulagraphy, speech production, Russian language</FONT><FONT SIZE=3><SPAN LANG="en-US">, vowels</SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Automatic processing of unattended lexical information in visual oddball presentation</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>G. Goryainova*, A. Ossadchii*, A. Shestakova*, Y. Shtyrov</FONT><SUP><FONT SIZE=3>**,***</FONT></SUP><FONT SIZE=3> </FONT></FONT> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><SUP><FONT SIZE=3><I>*</I></FONT></SUP><FONT SIZE=3><I>Saint Petersburg State University, Russia,**Medical Research Council Cognition and Brain Sciences Unit (MRC CBU), Cambridge UK, </I></FONT><SUP><FONT SIZE=3><I>***</I></FONT></SUP><FONT SIZE=3><I>Cognitive Brain Research Unit, Institute of Behavioural Sciences, University of Helsinki, Finland</I></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>Previous ERP studies of automatic language processing revealed early (100-200 ms) reflections of access to lexical and semantic characteristics of speech signal using the so-called mismatch negativity (MMN), a negative ERP deflection elicited by a distinguishable deviance in repetitive aspects of unattended auditory stimulation. In those studies, lexical processing of spoken stimuli became manifest as an enhanced ERP in response to real words as opposed to phonologically matched but meaningless pseudo-word stimuli. This lexical ERP enhancement was explained by automatic activation of word memory traces realized as distributed strongly intra-connected neuronal circuits. Their robustness guarantees memory trace activation even in the absence of attention on spoken input. As previous studies exclusively used auditory stimulation, we here adapted the lexical MMN paradigm to investigate early automatic lexical effects in the visual modality. In a visual oddball sequence, psycholinguistically matched (for length, frequency, bi-/tri-gram frequency) short word and pseudo-word stimuli were presented tachostoscopically in perifoveal area (2.50&deg;) outside the visual focus of attention, as the subjects&rsquo; attention was concentrated on a concurrent non-linguistic visual dual task in the centre of the screen. Using multi-channel EEG, we found a robust word advantage effect, with unattended written words producing larger ERP amplitudes than matched pseudowords. The relative input of standard and deviant stimuli in these lexical effects was different: whilst deviant responses showed most lexical contrast at or just before 170-ms range, standard responses exhibited a more sustained lexical difference throughout the epoch. A trend for word-elicited visual MMN was also observed. The data indicate automatic lexical processing of visually presented language outside the focus of attention.</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> automatic processing, lexical access, attention</FONT></FONT><P ALIGN=JUSTIFY> <BR> <DL> <DT><P LANG="en-US"><BR> </DL> <P ALIGN=CENTER><A NAME="Title"></A><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Using free association test for sociolinguistic analysis </B></SPAN></FONT></FONT> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><A NAME="address"></A><FONT FACE="Arial"><FONT SIZE=3>V. B. Gulida</FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>St. Petersburg State University, Russia</I></FONT></FONT><DL> <DT><P LANG="en-US"> <BR> </DL> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The free association test can be employed to address questions directly relevant to sociolinguistic study, such as whether social meanings (connotations) of words are listed in the mental lexicon alongside their paradigmatic, syntagmatic, semantic and thematic characteristics; and if so, whether sociolinguistic parameters have the capacity to organise mental lexicon similarly to structural parameters. That socially relevant properties of language units may be &ldquo;recorded&rdquo; in the mental lexicon can be logically inferred from W. Labov&rsquo;s thesis about the sociolinguistic rules of language use forming an indispensable part of language competence. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The present talk reports the analysis of the free association test data obtained from 184 Russian-speaking subjects, responding to 42 stimuli series of socially marked stress variations of 20 Russian common words. The stimuli were read aloud to the participants, who wrote down their associations (1-2, rarely 3). The stimuli words are currently undergoing language change and therefore are likely to provoke variously strong judgements. Subject sampling is balanced according to age\generation factor (5 divisions), education level (2 divisions), occupational type. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The first result is about the share of socially relevant associations (20 - 25%) out of the total corpus, except for children and teenagers. The second presents the ways in which social information is linguistically expressed in the associations: prosodic and phonetic enhancing\parodying the stimulus; indicating a likely speaker\ setting for the stimulus; quoting a precedent text for a stimulus; giving a &ldquo;school&rdquo; type rule and some others. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Finally, the patterns identified in the manifestation of social content have been analysed to specify the sources of the speakers&rsquo; sociolinguistic competence, such as the family input at an early age, stereotyped social characters and activities, knowledge about correct language use, precedent texts from fiction and film, and related linguistic ideologies. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><A NAME="keys"></A><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> psycholinguistic method sociolinguistics </FONT></FONT> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Acoustic characteristics of vowels in the Besermyan language</B></SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">R. Idrisov</SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>Lomonosov Moscow State University, Russia</I></SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The investigators of Besermyan note that this language appears to be one of more fully explored than other dialects of the Udmurt languages. Nonetheless, even the most irrefutable authorities only provide rather contradictory evidence on Besermyan vowels, based on the results of audiovisual observation.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The subject of the present study is acoustic characteristics of vowels in the Besermyan language. To obtain more specific results on the matter at first we developed a word list, based on which a speech corpus was later recorded. 7 Besermyan speakers, 4 male and 3 female, took part in the creation of the corpus. When marking up the received speech data, start and end points of every vowel were labeled manually. Then, using Praat, an automatic measurement of formants in the range of the central 40 % of each vowel&rsquo;s length was performed, which helped to lessen the influence of context.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">When optimizing the measurements, the following method was applied: for every vowel pronounced by the same speaker, its first two formants were measured 21 times; the upper limit for searching the fifth formant varied among female subjects from 4500 to 6500 Hz in 100 Hz increments and among male subjects from 4000 to 6000 Hz in 100 Hz increments. Of 21 measurements, the one where the sum of variances of F1 and F2 was minimal was considered the optimal. Being quite simple, this method allows us to diminish dispersion of formants&rsquo; values.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><BR><FONT SIZE=3><I>Keywords</I></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>:</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> Besermyan, vowels, formants</SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P LANG="en-US" ALIGN=CENTER><BR> <P LANG="en-US" ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Effects of Number of Stem Allomorphs on Performance </B></SPAN></FONT></FONT> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>in Lexical Decision and Event-related Potential Experiments</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B>&ensp;&ensp;&ensp;</B></SPAN></FONT></FONT><DL> <DT><P LANG="en-US"> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">A. Nikolaev*, P&auml;&auml;kk&ouml;nen A.**, J. Niemi*, M. J. Nissi***, E. Niskanen***, M. K&ouml;n&ouml;nen****, E. Mervaala** &amp; H. Soininen*****</SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>*Linguistics, University of Eastern Finland, Finland, **Department of Clinical Neurophysiology, Kuopio University Hospital, Finland, ***Department of Physics and Mathematics, University of Eastern Finland, Finland, ****Department of Clinical Radiology, Kuopio University Hospital, Finland, *****Department of Neurology, Kuopio University Hospital, Finland</I></SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The majority of Finnish nouns have more than one stem: one free, and the other(s) bound (e.g., &lsquo;glass&rsquo;: </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>lasi</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">, </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>lase</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">-). The nouns with rich stem allomorphy (e.g., &lsquo;water&rsquo;: </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>vesi</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">, </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>vet</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">-, </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>vete</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">-, </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>vede</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">-, </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>ves</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">-) constitute unproductive (closed) inflectional classes. Previous lexical decision (simple, primed and masked) experiments have shown that stem allomorphs act as access codes, with no special status attributed to the nominative singular. We investigated the temporal dynamics of processing monomorphemic nouns from productive (e.g., </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>lasi</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">) vs. unproductive (e.g., </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>vesi</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">) inflectional classes by conducting a simple visual lexical decision experiment and an even-related potential (ERP) cum lexical decision experiment. In spatio-temporal processing, the unproductive items received shorter latencies than the productive ones. To explain this response pattern, we conjecture that the unproductive paradigms are encapsulated in a multiroute network. Statistical facilitation would explain the shortening of response latencies. The nominative singular forms of the productive and unproductive paradigms did not differ in the number of errors. (Finnish does not have a clear-cut regular &ndash; irregular distinction in its complex inflectional morphology.) Thus we conclude that the lexical status is not affected by morphological associations. In the ERP data, the productivity contrast revealed a left temporal negativity and a centro-parietal N400-type component. This suggests that morphological inflectional productivity can be added to the variables known to influence these components.</SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>Keywords:</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">stem allomorphy, event-related potentials, Finnish</SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P LANG="en-US" ALIGN=CENTER><BR> <P LANG="en-US" ALIGN=CENTER><BR> <P LANG="en-US" ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>How Do We Recognize Reduced Word Forms of High Frequency?</B></SPAN></FONT></FONT><P ALIGN=CENTER> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>Experimental Evidence from Russian</B></SPAN></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">O.V. Raeva, E. I. Riekhakaynen</SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>St. Petersburg State University, Russia</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Analysis of Russian spontaneous speech has shown that about 1/5 of all word forms diverge from full-type pronunciation. While interpreting reduced realizations of a word form in the utterance a listener seems to reconstruct them to the unreduced variant stored in the mental lexicon using contextual information and some phonetic cues [Ernestus et al. 2002]. Typical realizations of some frequent word forms (e.g. /&scaron;&rsquo;:as/ for </SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><I>A59G0A</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> &lsquo;now&rsquo;) can be the only exception to the rule. Presumably, they are stored in the mental lexicon of a listener along with unreduced realizations and accessed directly in the process of spoken word recognition.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">In the poster, an experiment that was held to check this assumption will be described. Sixty subjects were asked to listen to unreduced, typical reduced and non-typical reduced realizations of 9 word forms of high frequency in isolation and in minimal context and to write them down. All realizations and contexts were extracted from spontaneous Russian dialogues. According to the results, only two typical realizations (/&scaron;&rsquo;:as/ for </SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><I>A59G0A</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I> </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">and</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I> </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">/sto+ka/</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I> </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">for</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I> </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><I>AB>;L:></I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I> </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">&lsquo;so many/much&rsquo;</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>)</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> that were recognized accurately both in isolation and in context can be stored in the mental lexicon of a listener, all others requiring reconstruction based on contextual information and thus refuting the hypothesis. </SPAN></FONT></FONT> <P LANG="en-US" ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>References</B></SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Ernestus, M., Baayen, H., &amp; Schreuder, R. (2002). The Recognition of Reduced Word Forms. </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>Brain and Language</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">, 81, 162&ndash;173.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> reduced word forms, spoken word recognition, frequency</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Research capacity of the onomasiological experiment</B></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>M. E. Rut, M. N. Gafurova</FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US"><I>Ural Federal University named after the first President of Russia B.&nbsp;N. Yeltsin,</I></SPAN></FONT></FONT></FONT><P ALIGN=CENTER> <FONT FACE="Arial"><FONT COLOR="#000000"><FONT SIZE=3><SPAN LANG="en-US"><I>Ekaterinburg, Russia</I></SPAN></FONT></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>We can obtain material for the nominative process investigation by observing the speech and focusing on the predicate nomination, as well as through the agency of onomasiological experiment. On the one hand, the first method&rsquo;s advantage is latitude of the linguistic realization, on the other hand, substantial temporal duration and the impossibility of the research vectoring present obvious difficulty. Considering the fact that nomination is a closed process, an appeal to the discursive material raises the issue of an involuntary determination of the research unit. Because of it and due to procedural and unconscious nature of the phenomenon, onomasiological experiment is necessary for the nominative theory. Linguistic frame is important as well: an experience passes through the linguistic consciousness and appears as a linguistic unit.</FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3>Specialized onomasiological experiments allow linguistically and cognitively directed collecting of the material. The lexical-onomasiological experiment involves nominative ability study of words which form certain lexical-semantic groups: creation of new nomination comes from the given lexical material, i.e. within a certain general model. Associative-onomasiological experiment, normally accompanying lexical-onomasiological experiment, is aimed to reveal the association field of the generating word.</FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3>Direct onomasiological experiment is the least specialized one; it offers the widest range of possibilities. Observing the nomination of the proposed objects, we can consider all nominative factors and selectively detect their influence on the subject and language material. Modelling situation in a certain way, we can approximate it to natural conditions, at the same time taking advantage of the experiment&rsquo;s privileges. We are able to determine the maximum amount of onomasiological context owing to verbalization (variants and motivational comments).</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> </FONT><FONT SIZE=3><SPAN LANG="en-US">onomasiology, nomination</SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Strategies for describing landscape photographs</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">E. G. Sokolova, I. A. Levchenkova</SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><I>Russian State University for the Humanities, Moscow, Russia</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">This paper addresses the issue of image description for the purpose of developing text generation program. We aim at formalizing the discourse strategies </SPAN></FONT><FONT SIZE=3>for describing landscape photographs</FONT><FONT SIZE=3><SPAN LANG="en-US"> in the collection of coloured pictures made by S.M. Prokudin-Gorsky. The research bases on the collection of coloured pictures by S.M. Prokudin-Gorsky, an experimental text corpus of their descriptions and ontology of concepts used for describing the images. For every photograph we construct an &ldquo;Image Model&rdquo; (IM), which is a formalized description of objects depicted, their relations and image composition. IM is supposed to be the basis for generating different linguistic description schemas.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">IM for every photograph is constructed semi-automatically in DEMLinG computer environment for NLG that we use in our research. They are combined from the ontology concepts following specific principles so that they are considered as &ldquo;neutral&rdquo; or a sort of general discourse representations, which are supposed to be formed automatically in future perspective. Every IM is considered to be a source for some description strategies resulted in different text plans. </SPAN></FONT></FONT> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">The paper describes our experience in applying the strategies for image descriptions introduced in cognitive approach [Kobozeva 97] to open space photographs. We show that not all these strategies are applicable to our material. We also discuss our own approach issuing from the typology of the depicted space presented in the IM. The connection between the strategy, picture content, meta-object concepts &ndash; frame of open space photo, salience and its signs are considered at length. </SPAN></FONT></FONT> <P LANG="en-US" ALIGN=JUSTIFY><BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> NLG, image description, discourse strategy</FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER>&ldquo;<FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><B>How many is many&rdquo; - </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">ambiguity of &ldquo;many&rdquo;, ambiguity of &ldquo;</SPAN></FONT><FONT SIZE=3><SPAN LANG="fi-FI"><=>3></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">&rdquo;,</SPAN></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">L. Tupikina, M. Ionov</SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US"><I>Lomonosov Moscow State University, Russian Federation </I></SPAN></FONT></FONT> <DL> <DT><P LANG="en-US"><BR> <DT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Meanings of the constructions with &ldquo;many&rdquo; and &ldquo;</SPAN></FONT><FONT SIZE=3><=>3></FONT><FONT SIZE=3><SPAN LANG="en-US">&rdquo; are fuzzy. There are several theories and hypotheses trying to formalize this fuzziness, involving, for example, focus structure and ambiguity of such constructions.</SPAN></FONT></FONT><DT> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">First, we take the hypothesis of three different readings of English &ldquo;many&rdquo;: </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>cardinal, proportional, list reading </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> and the focus effects connected with the reading,</SPAN></FONT><FONT COLOR="#ff0000"><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT></FONT><FONT SIZE=3><SPAN LANG="en-US">they are also adapted for Russian &ldquo;</SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><=>3></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">&rdquo;. We speak about the differences between these readings and their formal structures for English &ldquo;many&rdquo;.</SPAN></FONT></FONT><DT><P LANG="en-US"> <BR> <DT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">We propose our </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>hypothesis</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">: contexts like</SPAN></FONT></FONT></DL> <OL> <LI><P> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">= <=>3> G8B0;.</SPAN></FONT></FONT></P> <LI><P><FONT FACE="Arial"><FONT SIZE=3>= G8B0; <=>3></FONT><FONT SIZE=3><SPAN LANG="fi-FI">.</SPAN></FONT></FONT></P> </OL> <DL> <DT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">differ semantically in such way: </SPAN></FONT></FONT> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="fi-FI">(1) = (2) + [effectiveness].</SPAN></FONT></FONT><DL> <DT> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Since there is a sense of effectiveness, we expect to see that perfective verbs will be better in contexts like (1) than imperfective.</SPAN></FONT></FONT><DT><P LANG="en-US"> <BR> <DT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">We expect that transitive verbs will be better in contexts like (2) than intransitive.</SPAN></FONT></FONT><DT> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">To test the hypotheses, we&rsquo;ve designed an </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>experiment </I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">with 4 transitive and 4 intransitive verbs and suitable contexts like in (3):</SPAN></FONT></FONT></DL> <OL START=3> <LI><P> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">= =5 <>65B E>48BL: >= <=>3> AJ5; (</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">vs</SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU">.) >= AJ5; <=>3>.</SPAN></FONT></FONT></P> </OL> <DL> <DT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">In a sentence like (1) or (2) two variables were chosen:</SPAN></FONT></FONT></DL> <OL> <LI><P> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Position of &ldquo;</SPAN></FONT><FONT SIZE=3><=>3></FONT><FONT SIZE=3><SPAN LANG="en-US">&rdquo;: before or after the verb.</SPAN></FONT></FONT></P> <LI><P><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">Form of the verb: perfective or imperfective.</SPAN></FONT></FONT></P> </OL> <DL> <DT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">One of our additional hypotheses for this effect is that semantic difference comes from some cognitive aspects and syntactic difference: &ldquo;</SPAN></FONT><FONT SIZE=3><=>3></FONT><FONT SIZE=3><SPAN LANG="en-US">&rdquo; in (1) is an </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>adverbial modifier</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">, but in (2) there is </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>null object</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> and &ldquo;</SPAN></FONT><FONT SIZE=3><=>3></FONT><FONT SIZE=3><SPAN LANG="en-US">&rdquo; modifies it. We suggest our explanations based on syntax theory and on some psycho-linguistic effects.</SPAN></FONT></FONT></DL> <P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> </FONT><FONT SIZE=3><SPAN LANG="en-US">quantifiers, focus, experimental linguistics.</SPAN></FONT></FONT><P LANG="en-US" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><B>Experimental evidence for granularity shifting in the adjectival domain</B></FONT></FONT><DL> <DT> <BR> </DL> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3>N. Zevakhina**, G. Sassoon* </FONT></FONT> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><A NAME="Text2"></A><FONT FACE="Arial"><FONT SIZE=3><I>*ILLC/University of Amsterdam, the Netherlands, **Moscow State University, Russia</I></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>In this talk, we present experimental evidence for a granularity-shifting analysis of modifiers. For example, an object x is </FONT><FONT SIZE=3><I>dirty/clean</I></FONT><FONT SIZE=3> iff x is covered by some/ no dirt, considering some level of granularity g, e.g., it is normally appropriate to ignore almost invisible dirt; x is </FONT><FONT SIZE=3><I>slightly</I></FONT><FONT SIZE=3> </FONT><FONT SIZE=3><I>dirty</I></FONT><FONT SIZE=3>/</FONT><FONT SIZE=3><I>completely clean </I></FONT><FONT SIZE=3>iff x is </FONT><FONT SIZE=3><I>dirty</I></FONT><FONT SIZE=3>/</FONT><FONT SIZE=3><I>clean</I></FONT><FONT SIZE=3> given a more pedantic criterion g</FONT><SUB><FONT SIZE=3>P</FONT></SUB><FONT SIZE=3>, in which, e.g., almost invisible dirt counts. Thus, we test whether:</FONT></FONT> <OL TYPE=i><LI><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><I>dirty </I></FONT><FONT SIZE=3>implies &lsquo;more than almost invisible dirt&rsquo; ([dirty]</FONT><SUB><FONT SIZE=3>g</FONT></SUB><FONT SIZE=3> </FONT></FONT><FONT FACE="Arial"><SPAN LANG="hi-IN"><FONT SIZE=3>&#64431; </FONT></SPAN></FONT><FONT FACE="Arial"><FONT SIZE=3>[slightly dirty]</FONT><SUB><FONT SIZE=3>g</FONT></SUB><FONT SIZE=3> = [dirty]</FONT><SUB><FONT SIZE=3>gP</FONT></SUB><FONT SIZE=3>);</FONT></FONT><LI><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><I>slightly dirty </I></FONT><FONT SIZE=3>implies &lsquo;no more than almost invisible dirt&rsquo; ([dirty]</FONT><SUB><FONT SIZE=3>gP</FONT></SUB><FONT SIZE=3> &amp; [</FONT><FONT SIZE=3>&#9488;</FONT><FONT SIZE=3>dirty]</FONT><SUB><FONT SIZE=3>g</FONT></SUB><FONT SIZE=3>);</FONT></FONT><LI><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><I>completely clean </I></FONT><FONT SIZE=3>implies &lsquo;free even of almost invisible dirt&rsquo; ([completely clean]</FONT><SUB><FONT SIZE=3>g</FONT></SUB><FONT SIZE=3> = [clean]</FONT><SUB><FONT SIZE=3>gP</FONT></SUB><FONT SIZE=3> </FONT></FONT><FONT FACE="Arial"><SPAN LANG="hi-IN"><FONT SIZE=3>&#64431;</FONT><FONT SIZE=3> </FONT></SPAN></FONT><FONT FACE="Arial"><FONT SIZE=3>[clean]</FONT><SUB><FONT SIZE=3>g</FONT></SUB><FONT SIZE=3>); and</FONT></FONT><LI><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><I>clean </I></FONT><FONT SIZE=3>implies &lsquo;not free of almost invisible dirt&rsquo; ([clean]</FONT><SUB><FONT SIZE=3>g</FONT></SUB><FONT SIZE=3> &amp; [</FONT><FONT SIZE=3>&#9488;</FONT><FONT SIZE=3> clean]</FONT><SUB><FONT SIZE=3>gP</FONT></SUB><FONT SIZE=3>). </FONT></FONT> </OL> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3>We test two context types, e.g., (C1) </FONT></FONT><FONT FACE="Arial"><FONT SIZE=3><I>Nick&rsquo;s mother </I></FONT></FONT><FONT FACE="Arial"><FONT SIZE=3><I>says that x is dirty</I></FONT></FONT><SUB><FONT FACE="Arial"><FONT SIZE=3><I>1</I></FONT></FONT></SUB><FONT FACE="Arial"><FONT SIZE=3><I>. Nick thinks that x is slightly dirty</I></FONT></FONT><SUB><FONT FACE="Arial"><FONT SIZE=3><I>2</I></FONT></FONT></SUB><FONT FACE="Arial"><FONT SIZE=3><I>. Would Nick agree that x is dirty</I></FONT></FONT><SUB><FONT FACE="Arial"><FONT SIZE=3><I>3</I></FONT></FONT></SUB><FONT FACE="Arial"><FONT SIZE=3><I>?</I></FONT></FONT><FONT FACE="Arial"><FONT SIZE=3> (C2) is the same except that </FONT></FONT><FONT FACE="Arial"><FONT SIZE=3><I>slightly </I></FONT></FONT><FONT FACE="Arial"><FONT SIZE=3>modifies the 1</FONT></FONT><SUP><FONT FACE="Arial"><FONT SIZE=3>st</FONT></FONT></SUP><FONT FACE="Arial"><FONT SIZE=3> and 3</FONT></FONT><SUP><FONT FACE="Arial"><FONT SIZE=3>rd</FONT></FONT></SUP><FONT FACE="Arial"><FONT SIZE=3> occurrences of the adjective, instead of the 2</FONT></FONT><SUP><FONT FACE="Arial"><FONT SIZE=3>nd</FONT></FONT></SUP><FONT FACE="Arial"><FONT SIZE=3>. Significantly more negative than positive answers will support the entailments in (i)+(ii) (and (iii)-(iv) for </FONT></FONT><FONT FACE="Arial"><FONT SIZE=3><I>completely</I></FONT></FONT><FONT FACE="Arial"><FONT SIZE=3>). A mixed answer pattern will suggest that their status is as implicatures, rather than entailments (</FONT></FONT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">32</SPAN></FONT></FONT><FONT FACE="Arial"><FONT SIZE=3> adjectives</FONT></FONT><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="en-US">, 6</SPAN></FONT></FONT><FONT FACE="Arial"><FONT SIZE=3> modifiers). </FONT></FONT><FONT COLOR="#000000"><FONT FACE="Arial"><FONT SIZE=3>The talk will report the main results and their implications for the analysis of degree modifiers</FONT></FONT></FONT><FONT COLOR="#000000"><FONT FACE="Arial"><FONT SIZE=3>.</FONT></FONT></FONT><P ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><I>Keywords:</I></FONT><FONT SIZE=3> degree, granularity shifting, implicature</FONT></FONT><BR> <P LANG="en-US" ALIGN=CENTER><BR> <P LANG="en-US" ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU"><B>;0A=K5</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B> </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><B>704=53></B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B> </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><B>@O40</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B> </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><B>2</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B> </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><B>B538=A:><</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B> </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><B>3>2>@5</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B> </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><B>E0=BK9A:>3></B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B> </B></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><B>O7K:0</B></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><B>.</B></SPAN></FONT></FONT><P LANG="en-US" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">. .</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"> </SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU">0@;>2A:0O</SPAN></FONT></FONT><P LANG="ru-RU" ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU"><I>>A:>2A:89 3>AC40@AB25==K9 C=825@A8B5B 8<5=8 .</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>&nbsp;</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><I>.</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>&nbsp;</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><I>><>=>A>20, >AA8O</I></SPAN></FONT></FONT><P LANG="ru-RU" ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">0==0O @01>B0 ?@54AB02;O5B A>1>9 0=0;87 A8AB5<K 2>:0;87<0 (0 8<5==>, 3;0A=KE 704=53> @O40) B538=A:>3> 3>2>@0 E0=BK9A:>3> O7K:0. 0==K5 1K;8 ?>;CG5=K 2 E>45 5B=59 ;8=328AB8G5A:>9 M:A?548F88 2011 3>40 2 A5;> "538 5@Q7>2A:>3> @09>=0 %.  40==>9 @01>B5 AB02OBAO A;54CNI85 7040G8:</SPAN></FONT></FONT> <OL><LI><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">A@02=8BL A>1@0==K9 2 A. "538 480;5:B=K9 <0B5@80; A <0B5@80;>< :07K<A:>3> 480;5:B0 E0=BK9A:>3> O7K:0 (?> .. 0:A8= &laquo;07K<A:89 480;5:B E0=BK9A:>3> O7K:0&raquo;);</SPAN></FONT></FONT><LI><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">2KO28BL @07;8G8O 8 ?>AB0@0BLAO >E0@0:B5@87>20BL 3;0A=K5 704=53> @O40 B538=A:>3> 3>2>@0 E0=BK9A:>3> O7K:0;</SPAN></FONT></FONT><LI><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">?@>25AB8 M:A?5@8<5=B, F5;LN :>B>@>3> O2;O5BAO >?@545;5=85 ?5@F5?B82=KE E0@0:B5@8AB8: 3;0A=KE 704=53> @O40 2 848>;5:B0E =>A8B5;59 B538=A:>3> 3>2>@0 E0=BK9A:>3> O7K:0;</SPAN></FONT></FONT><LI><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">>1>7=0G8BL AB5?5=L 2;8O=8O @CAA:>3> O7K:0.</SPAN></FONT></FONT></OL> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">8B5@0BC@K, ?>A2OI5==>9 :>=:@5B=> D>=5B8:5 B538=A:>3> 3>2>@0, :@09=5 <0;>, 2 B> 2@5<O :0: 480;5:B "53 >B45;O5BAO >B :07K<A:>3> 8 @0AA<0B@8205BAO :0: A0<>AB>OB5;L=K9 (A@.[0:A8=]).</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">7-70 A2>53> B5@@8B>@80;L=>3> ?>;>65=8O E0=BK9A:89 O7K: ?>425@305BAO 1>;LH><C 2;8O=8N @CAA:>3> O7K:0. !8AB5<0 A>1AB25==> E0=BK9A:>3> O7K:0 C?@>I05BAO, =081>;55 OA=> MB> 284=> =0 D>=5B8G5A:>< <0B5@80;5. #?@>I5=85 D>=5B8G5A:>9 A8AB5<K &ndash; >48= 87 MB0?>2 2K<8@0=8O 2A53> 480;5:B0 8;8 O7K:0.</SPAN></FONT></FONT><P LANG="ru-RU" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU"><I>;NG52K5 A;>20:</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"> E0=BK9A:89 O7K:, 3;0A=K5</SPAN></FONT></FONT><P LANG="ru-RU" ALIGN=JUSTIFY> <BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU"><B>@8=F8? <8=8<0;L=KE A>2<5AB=KE CA8;89 8 AB@0B5388 >?8A0=8O =58725AB=KE >1J5:B>2</B></SPAN></FONT></FONT><P ALIGN=CENTER> <BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">. $54>@>20, ". !;01>4:8=0 </SPAN></FONT></FONT> <P LANG="ru-RU" ALIGN=CENTER><BR> <P ALIGN=CENTER><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU"><I>>A:>2A:89 3>AC40@AB25==K9 C=825@A8B5B 8<5=8 .</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>&nbsp;</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><I>.</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US"><I>&nbsp;</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"><I>><>=>A>20, >AA8O</I></SPAN></FONT></FONT><P LANG="ru-RU" ALIGN=CENTER> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">!B0BLO </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">Clark</SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"> </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">and</SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"> </SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">Wilkes</SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU">-</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">Gibbs</SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"> 1986 AB0;0 H8@>:> 8725AB=0 1;03>40@O 42C< >A>15==>ABO<: (</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">i</SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU">) >?8A0=8N A>2<5AB=>9 <>45;8 CAB0=>2;5=8O @5D5@5=F88; (</SPAN></FONT><FONT SIZE=3><SPAN LANG="en-US">ii</SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU">) 70?><8=0NI5<CAO M:A?5@8<5=BC A B0=3@0<<0<8 &ndash; D83C@:0<8 87 :8B09A:>9 3>;>2>;><:8, :0640O 87 :>B>@KE A>AB>8B 87 A5<8 G0AB59 >A>1K< >1@07>< @07@570==>3> :204@0B0.</SPAN></FONT></FONT><P ALIGN=JUSTIFY> <FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU">48= 87 >A=>2=KE ?@8=F8?>2 <>45;8 ;0@:0 &ndash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rial"><FONT SIZE=3><SPAN LANG="ru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rial"><FONT SIZE=3><SPAN LANG="ru-RU"> =0AB>OI55 2@5<O M:A?5@8<5=B ?@>4>;605BAO A 8A?>;L7>20=85< <5B>40 DC=:F8>=0;L=>9 <03=8B=>-@57>=0=A=>9 B><>3@0D88: A@02=5=8N ?>425@30NBAO 36 B0=3@0<<, =5 8<5NI8E CAB>O2H53>AO =08<5=>20=8O (B> 5ABL 2 ?@>F5AA5 8E =07K20=8O =5 <>65B ?@>872>48BLAO ?>8A: 2 <5=B0;L=>< ;5:A8:>=5) 8 36 87>1@065=89 35><5B@8G5A:8E D83C@, 8<5NI8E AB0=40@B=>5 =08<5=>20=85. >?>;=8B5;L=K< D0:B>@>< 2KABC?05B @07;8G85 B0=3@0<< ?> AB5?5=8 8E ?>E>65AB8 =0 @50;L=K5 >1J5:BK 459AB28B5;L=>AB8, B> 5ABL 25@>OB=>ABL 8A?>;L7>20=8O E>;8AB8G5A:>9 8;8 A53<5=B8@>20==>9 AB@0B5388. A;8 E>;8AB8G5A:0O AB@0B538O 459AB28B5;L=> O2;O5BAO <5=55 @5AC@A>70B@0B=>9, MB> 4>;6=> ?@>O28BLAO 2 AB5?5=8 0:B820F88 A>>B25BAB2CNI8E 7>= 3>;>2=>3> <>730.</SPAN></FONT></FONT><P LANG="ru-RU" ALIGN=JUSTIFY> <BR> <P ALIGN=JUSTIFY><FONT FACE="Arial"><FONT SIZE=3><SPAN LANG="ru-RU"><I>;NG52K5 A;>20:</I></SPAN></FONT><FONT SIZE=3><SPAN LANG="ru-RU"> ?A8E>;8=328AB8:0, 480;>3, AB@0B538O >?8A0=8O</SPAN></FONT></FONT><P LANG="ru-RU" ALIGN=JUSTIFY> <BR> </font></td> <td width=10% valign=center bgcolor="black"> <img SRC="img/spb1.png" NAME="face1" HSPACE=0 VSPACE=10 BORDER=0 NATURALSIZEFLAG="3" ALT="..." align=center valign=top> </td> <td width=200 valign=center> </td> </tr></table> <br> <hr></center> <td align=right><i><font face=tahoma size=0>The Night Whites Language Workshop</font></i></td> </body> </html>