Government & Politics

Bencini et al

Description
Bencini et al
Published
of 2
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Subject drop in Italian Alzheimer’s disease G. Bencini  a,* , R. Biundo  b , C. Semenza  b , V. Valian  a a Psychology Department, Hunter College of the City University of New York,USA b Dipartimento di Psicologia, Universita’ di Trieste, Trieste, Italy Accepted 8 July 2005 Introduction How does a native speaker’s knowledge of her language’s syntaxinteract with processing factors such as sentence length and complex-ity? Conversely, how do general cognitive constraints interact with thegrammatical properties of individual languages?We explore these questions by focusing on the parameter that li-censes null subjects in languages like Italian and prohibits them inlanguages like English (Chomsky, 1981). The contrast between thesetwo types of languages is illustrated in (1). Without an overt thirdperson subject (e.g.,  she ), the English sentence in 1a is ungrammatical,whereas the corresponding Italian sentence is fully grammatical.Languages like Italian are often referred to as null subject languages.In these languages, the null subject is  pro  (pronounced ‘‘little pro’’) andis assumed to be a phonologically silent, but syntactically present,element with pronominal properties.(1) a. * Walks on the beachb. Cammina sulla spiaggiaHere, we examine the performance of Italian speakers with probableAlzheimer’s disease (AD) on a sentence repetition task in which pro-cessing load was varied by manipulating sentence length and com-plexity. Because of the general cognitive decline associated with earlyAD, these patients are likely to be sensitive to manipulations that taxperformance. Although we expected an overall decrement in perfor-mance, we predicted a particularly high rate of omission of subjectswhen sentences were complex. Our focus was thus the rate of inclusionof overt subjects as a function of sentence complexity and clause status(main clause subject versus subordinate clause subject). Italian ADspeakers should be sensitive to the fact that Italian allows null subjects;consequently, when stressed by sentence complexity, they should resortto increased use of null subjects. Normal elderly speakers, in contrast,should not show such an effect. Method Participants We tested 12 patients (11 females, 1 male) with a diagnosis of probable AD and 12 age-, education-, and dialect-matched controls (3females, 9 males). The average age was 91.1 for the patients and 90 forthe controls. Patients and controls had an average of 6 years of schooling. Patients averaged 14.08/30 (range 12–21) on the MiniMental State Examination ( MMSE  ) and 44.41 (range 30–64.2) on theMilan Overall Dementia Assessment ( MODA ). Materials The materials were 54 grammatical sentences in three conditions:Simple Main Clause (S), Complex Tense Main Clause (CT), andComplex Sentence (CS), as in (2).(2) a. One clause, Simple Tense (ST) Il pompiere spegnera`  il fuoco pericoloso The firefighter will extinguish the dangerous fireb. One Clause, Complex Tense (CT) Il pompiere avra`  spento il fuoco pericoloso The firefighter will have extinguished the dangerous firec. Two clauses Il pompiere assicura che la guardia spegnera`  il fuoco pericoloso The firefighter promises that the guard will extinguish thedangerous fire Procedure All of the 54 sentences were arranged in a list in random order.Patients and controls were instructed to listen to the experimenter andrepeat the sentence. Subjects were tested individually. Results Scoring  Patients’ and controls’ repetitions of target sentences were scored forinclusion of major grammatical constituents (subject, verb, and object)Brain and Language 95 (2005) 133–134www.elsevier.com/locate/b&l * Corresponding author. E-mail address:  gbencini@hunter.cuny.edu (G.Bencini).0093-934X/$ - see front matter    2005 Elsevier Inc. All rights reserved.doi:10.1016/j.bandl.2005.07.075  and auxiliaries when these were present in the target. Cells that wereempty because the target sentence did not contain a constituent orcategory of a given type (e.g., the main object in the two clause sen-tences) were scored n/a. Analyses We examined the distribution of main clause subjects in all patients’and controls’ productions. We conducted a mixed 3 · 2 analysis of variance (ANOVA) on the mean proportion of subject inclusion withthe within-subjects factor of sentence type (one-clause simple, one-clause complex, and two-clause) and the between-subjects factor of population (patient, control). The main effects of sentence type andpopulation were significant, as was the interaction (  p  < .001 in eachcase). As predicted, patients on average included fewer main clausesubjects in the two-clause condition than in the one-clause conditions(  p  = .003). The elderly controls were essentially at ceiling.A separate analysis examined the inclusion of subjects and verbs inmain versus subordinate clauses in two-clause sentences only. A 2  ·  2  · 2 mixed ANOVA with the within-subjects factors of clause (main,subordinate) and constituent (subject, verb) and the between-subjectsfactor of population (patient, control) revealed significant main effectsof constituent and population (  p  = .0001 in each case) and a marginalmain effect of clause (  p  = .06). All interactions were also significant(  p  = .0001 in each case). Analyses of the simple effects revealed thatpatients’ inclusion of subjects was lower in subordinate than in mainclauses (  p  = .001). For verbs the effect was in the opposite direction butwas not significant (  p  = .08) (see Table 1). Discussion As predicted, we found that AD speakers of Italian were sensitive toincreased processing demands and omitted overt subjects when pro-cessing was taxed. Italian AD speakers omitted subjects more thanelderly controls, but only in complex sentences. Sentence subjects werealso omitted more often in subordinate than in main clauses, and notonly because of sentence length (verbs showed the opposite pattern).We conjecture that the reason for increased omission in subordinateclauses is that this reflects the dominant pattern for null subjects intwo-clause sentences in Italian. To conclude, subject drop is a simpli-fication strategy used by Italian AD patients, an option influencedboth by processing and by the grammar. Therefore, we do not expectEnglish patients with AD to exhibit subject drop. We are currentlytesting this hypothesis. Reference Chomsky, N. (1981).  Lectures on government and binding.  Dordrecht:Foris.Table 1Mean Proportion of Constituent inclusion for patients and controlsCondition MainSubjectMainVerbMainObjectMain ObjectModifierSubSubjectSubVerbSubObjectSub ObjectModifier One clause simple Patients 0.97 1.00 0.99 0.94 n/a n/a n/a n/aControls 1.00 1.00 1.00 0.99 n/a n/a n/a n/a One clause complex Patients 0.96 1.00 0.99 0.92 n/a n/a n/a n/aControls 1.00 1.00 1.00 0.99 n/a n/a n/a n/a Two clauses Patients 0.78 0.70 n/a n/a 0.54 0.93 0.95 0.77Controls 1.00 1.00 n/a n/a 0.98 1.00 1.00 0.89134  Abstract / Brain and Language 95 (2005) 133–134
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks