Диссертация (1136198), страница 16
Текст из файла (страница 16)
For example, in a sentence like“The key to the cabinets are rusty” the form of the verb suggeststhat we need to look for an NP with the features “subject” and“plural.” However, no NP perfectly satisfies these conditions: keyis the subject, but is not plural, and cabinets is plural, but isnot the subject. It is hypothesized that in such conditions wemay mistakenly select the wrong NP . The retrieval approachpredicts the absence of ungrammaticality illusions: if a sentenceis grammatical, the true subject is a perfect match and will alwaysbe selected.
Thus, unlike in the representational account, thereis nothing wrong or ambiguous in the syntactic structure, errorsare access failures. Such cases with several elements competingfor retrieval are an instance of “retrieval interference.” Otherexamples are discussed in Van Dyke and Johns (2012).1.1.4. The Role of MorphophonologyHartsuiker et al. (2003) showed that when the form of theattractor is morphologically ambiguous and coincides withnominative, the rate of attraction errors increases. Theycompared German sentences like (3a,3b). People made moreerrors in (3a), where the attractor (die Demonstrationen) isambiguous between accusative and nominative, comparedto (3b), where the attractor (den Demonstrationen) isunambiguously dative. We do not explore the role ofmorphophonology in the present study, but take this factorinto account.
Several studies also demonstrated that heads withregular inflections are more resistant to attraction, but no similareffects were observed for attractors (e.g., Bock and Eberhard,1993; Vigliocco et al., 1995).(3)a. dieStellungnahme gegen dietheF.NOM.SG positionagainst theF.ACC.PLDemonstrationendemonstrationsb. dieStellungnahme zu dentheF.NOM.SG positionon theDAT.PLDemonstrationendemonstrations1.2. Models of Agreement AttractionThere exist two major approaches to agreement attraction.
Herethey will be referred to as the “representational account” and the“retrieval account.” Models that belong to the representationalaccount share one crucial assumption: agreement attraction takesplace because the mental representation of the number featureon the subject NP is faulty or ambiguous (Nicol et al., 1997;Vigliocco and Nicol, 1998; Franck et al., 2002; Eberhard et al.,2005; Staub, 2009, 2010; Brehm and Bock, 2013). In some models,it is assumed that syntactic features can “percolate” or otherwisemove to neighboring nodes: for example, sometimes numberfeatures from the embedded NP percolate to the subject NP(which normally has the same number marking as its head).Another model known as Marking and Morphing (Eberhardet al., 2005) postulates that the number value of the subject NPis a continuum, i.e., it can be more or less plural.
For example,if a subject NP contains a singular head and a plural dependentNP it is more plural than a subject NP with a singular modifier.A subject NP that is formally singular, but refers to a collectiveentity is more plural than the ones referring to singular entities.1.3. Studies of Gender AgreementAttractionRelatively few studies of gender agreement attraction have beenconducted so far. Their results do not always converge, but onething seems to be certain: attraction effects are present.
They havebeen observed in several experiments on different languages.1.3.1. Previous Studies on Languages with TwoGendersAs far as we know, the first attempt to induce gender agreementattraction was made in the production study on Italian byVigliocco et al. (1995). Virtually no evidence of attraction wasfound: out of 1920 responses only four (0.2%) contained a gendererror. However, in a later study Vigliocco and Franck (1999)observed gender agreement attraction in Italian.Vigliocco and Franck carried out four productionexperiments: two on Italian and two on French. Both languageshave two genders: masculine and feminine.
In all experiments,participants saw a masculine and a feminine adjective at the sametime (one above the other) and then a noun phrase, and had tocombine them saying the resulting sentence aloud. The gender ofthe head and the attractor were manipulated. When the genders3 In production, looking for symmetric effects in ungrammatical and grammaticalsentences is less straightforward. However, several authors suggested not onlycounting errors, but also measuring RTs during elicitation tasks (e.g., Staub, 2009,2010; Brehm and Bock, 2013). They demonstrated that participants slow downwhen the subject contains a singular head and a plural attractor both when theyeventually answer correctly and when they do not [to be precise, Staub observedthis for the subjects containing a PP attractor, but not for the subjects containedwithin relative clauses, as in (1b)].Frontiers in Psychology | www.frontiersin.org341November 2016 | Volume 7 | Article 1651Slioussar and MalkoGender Agreement Attraction in Russianmismatched, people were found to make more agreementerrors.
In Italian, there was no significant difference betweenFM and MF conditions4 . In French, more errors were made inFM conditions (the difference was significant in Experiment2 and marginally significant in Experiment 4). Whether thehead gender was purely grammatical (on inanimate nouns) orconceptual (on animate nouns) also played a role. Participantsmade fewer errors in the latter case. Thus, semantic factors doenter the picture in case of gender agreement attraction, but,as far as we can judge, only to suppress it (on the contrary,conceptual numerosity can increase the number agreementattraction rate).The observed pattern of attraction errors was different fromnumber agreement studies. Firstly, a significant number of errorswas made in all mismatch conditions, while in case of numberagreement, the error rate in the conditions with plural headsand singular attractors was very low, often the same as the errorrate without attraction.
Secondly, both in French and in Italian,masculine is used as the grammatical default (for example, itappears in impersonal constructions and in the cases where thepredicate must agree with several masculine and feminine nouns)and is more frequent. So the pattern observed in French (moreerrors in FM conditions) is the reverse of the number agreementattraction pattern found across languages.The authors concluded that feature markedness does notmatter for gender agreement and outlined an explanation basedon inflectional differences between Italian and French. However,this explanation was undermined by Anton-Mendez et al. (2002)who conducted a production study on Spanish. Spanish is similarto Italian in terms of adjectival inflections, but the resultswere the same as in French.
In addition to that, Viglioccoand Zilli (1999) and Franck et al. (2008) demonstrated in anumber of experiments on Italian, Spanish, and French thatthe morphophonological properties of the head influence theerror rate in gender agreement attraction. As in the studiesof number agreement attraction, there were fewer errors whenheads had regular inflections, but no similar effects were foundfor attractors.We could find only two studies examining gender agreementattraction in comprehension: Acuña-Fariña et al. (2014) andMartin et al. (2014).
Both looked at Spanish, eye-tracking wasused in the first and ERPs in the second. Attraction effects weredetected, but no differences between M and F genders werereported.groups of conditions: MM, MF, FF, FM and MM, MN, NN,NM. As in the previous studies, there were significantly moreerrors in mismatch conditions than in match conditions. Butthe pattern was different: there were more errors in the MFcondition compared to the FM and in the NM compared tothe MN.Experiment 2 confirmed the results of Experiment 1 (itcontained MM, MF, FF, and FM conditions and was designedto test the role of morphophonological factors).
In Experiment3, NN, NM, and NF conditions were compared. NM and NFpreambles provoked more errors than NN preambles; but thenumber of errors in NM and NF conditions was comparable.Explaining this pattern, B&K adopt an optimality-theoreticapproach and argue that there is no single markedness hierarchyin the Slovak gender system (such as N < M < F), but markednessis defined in pairs (N < M, N < F, M < F). Among other things,the results of this study show that frequency does not play a rolefor feature asymmetries.Another production experiment was conducted on Russian(Lorimor et al., 2008). The authors manipulated both the numberand the gender of heads and attractors (only M and F genderswere used).
In all trials, participants saw and heard the predicateand then saw the preamble. Their task was to construct a sentenceusing these two parts and to say it aloud. Out of 1155 answerswhere gender agreement was necessary (in Russian, as well as inSlovak, verbs agree in gender only in past tense singular forms),only seven (0.6%) contained an agreement error. Based on this,the authors concluded that gender agreement attraction does notexist in Russian.To summarize, in all gender agreement attraction studies, ifany effects are observed, error rates in all mismatch conditionsare higher than in match conditions (unlike in number attractionstudies, where significant effects are found only in one mismatchcondition: with singular heads and plural attractors).
Otherwise,the results of gender agreement studies are different: larger effectsare found in the FM condition (compared to the MF condition)in Spanish and French, and in the MF and NM conditions(compared to the FM and MN conditions) in Slovak. The resultsfrom Slovak are closer to the pattern observed for number, if weassume that feminine and masculine genders and plural numberare marked.Out of several approaches to attraction outlined above, theexistence of gender agreement attraction is hardly compatiblewith the Marking and Morphing model, primarily because inthe absolute majority of cases, gender features are semanticallyempty.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
