Диссертация (Experimental study of several core concepts of theoretical morphology (on the material of russian) - regularity, syncretism, markedness), страница 28

We can see that thenature of the final consonant played a role. To estimate it statistically, we used mixedeffects logistic regression with random slopes and random intercepts by participantsand by items. Only nonce nouns ending in -l’ and -r’ were significantly different fromthe other groups (β = 2.48, SE = 1.15, z = -2.15, p = 0.03; β = 4.00, SE = 1.54, z =-2.60, p < 0.01). Unlike all other palatalized consonants, these final consonants aremore characteristic for masculine nouns.5.

ConclusionsIn the introduction, we outlined three distributional properties of Russian nouns.Firstly, masculine gender is more frequent than feminine and neuter. Secondly, somecombinations of genders and declensions are more frequent than the others (we calledthem prototypical). Thirdly, for nouns ending in palatalized consonants, some consonants are more characteristic for masculine nouns and the others for feminine nouns.69668Gender, Declension and Stem-final ConsonantsPrevious studies indicate that these factors do not increase the number of naturallyoccurring gender agreement errors for adult native speakers of Russian [Rusakova2013].

But speakers might still be sensitive to them, and in the present paper we demonstrated this in two experiments.Experiment 1 revealed the role of the two first factors. It was not designed to assess the role of the third factor, but, as far as we could estimate from the data, thisfactor does not play a role in online agreement processing. However, the influenceof this factor can be detected when speakers try to guess the gender of a novel noun—like in the pilot experiment 2. In further studies, we plan to use more stimuli, testingspeakers’ sensitivity to different suffixes etc.Now let us discuss the results of experiment 1 in more detail. It demonstratedthat both gender and declension of the noun influence online processing of the subject–predicate gender agreement in Russian.

But, firstly, this influence can be detectedonly in the sentences with agreement errors, i.e. no gender or declension is intrinsically more difficult to process (at least, in the sentence context7). Secondly, declensionplays a role at a very early stage and its effect is very short-lived, while the role of gender becomes visible later and its effect is more pronounced.The fact that a masculine verb form is less readily detected after a 3rd declensionsubject noun can be explained by the fact that its ending is more typical for masculinenouns that for feminine ones. However, alternative explanations are also possible, forexample, all agreement errors (in masculine or in neuter) may be harder to detect after 3rd declension subject nouns, i.e.

their gender can be in general harder to retrieve.To exclude this and some other possibilities, other experiments should be conducted.Another line of further research should look at non-prototypical masculine nouns likepapa ‘dad’. The picture may be different not only because of their different gender, butalso because all these nouns denote humans, so the gender feature is not semanticallyempty in this case, which may aid its processing and retrieval.As for the role of gender as such, we saw that agreement errors with masculinesubjects cause a larger delay in reading times compared to errors with feminine subjects, i.e. were costlier for processing.

This is in line with the previous findings on gender agreement in comprehension reported in the literature [Akhutina et al., 1999,2001]; [Romanova & Gor, 2017]; [Slioussar & Malko, 2016]. However, to have a fullpicture, neuter subjects and predicates should be introduced in further experiments.So far, several explanations are possible. It is well known that while reading,we generate expectations about the upcoming predicate based on the features of thesubject and rechecking is prompted if these expectations are violated (which is associated with increased reading times). Perhaps, the masculine form of the predicate,being the most frequent, causes less disruption if used incorrectly—similarly, usinga frequent word instead of an infrequent one provokes less surprise than the oppositemistake. Maybe, these expectations are more robust for masculine subjects, so violating them is more disruptive.

Maybe, if an agreement error is detected and recheckingis initiated, masculine subjects are retrieved more readily and reliably—this is what7It is well known that many differences that can be detected in the processing of isolatedforms disappear when these forms are embedded in an appropriate context.69769Slioussar N. A.[Slioussar and Malko 2016] suggested based on their agreement attraction resultswhere all combinations of genders on subjects, attractors and predicates were examined. All these explanations are compatible with the observed difference between ungrammatical sentences with masculine and feminine subjects. Further experimentsare necessary to tease them apart and to gain a better understanding of the patternsobserved in previous studies.The study was supported in part by the grant #16-18-02071 from the RussianScience Foundation.References1.2.3.4.5.6.7.8.9.10.11.12.13.Acuña-Fariña, J. C., E.

Meseguer, and M. Carreiras. (2014). Gender and numberagreement in comprehension in Spanish. Lingua 143: 108–128.Akhutina, Tatiana, Andrei Kurgansky, Maria Polinsky, and Elizabeth Bates.(1999). Processing of grammatical gender in a three-gender system: Experimental evidence from Russian. Journal of Psycholinguistic Research 28: 695–713.Akhutina, Tatiana, Andrei Kurgansky, Marina Kurganskaya, Maria Polinsky, Natalya Polonskaya, Olga Larina, Elizabeth Bates, and Mark Appelbaum. (2001).Processing of grammatical gender in normal and aphasic speakers of Russian.Cortex 37: 295–326.Alexeeva, Svetlana, Natalia Slioussar, and Daria Chernova.

(2018). StimulStat:a lexical database for Russian. To appear in: Behavior Research Methods.Andonova, E., S. D’Amico, A. Devescovi, and E. Bates. (2004). Gender and lexicalaccess in Bulgarian. Perception and Psychophysics 66: 496–507.Anton-Mendez, Inés, Janet Nicol, and Merrill F. Garrett. (2002).

The relation between gender and number agreement processing. Syntax 5: 1–25.Badecker, William and Frantisek Kuminiak. (2007). Morphology, agreement andworking memory retrieval in sentence production: Evidence from gender andcase in Slovak. Journal of Memory and Language 56: 65–85.Bates, Elizabeth, Antonella Devescovi, Luigi Pizzamiglio, Simona D’Amico, andArturo Hernandez. (1995). Gender and lexical access in Italian.

Perception andPsychophysics 57: 847–862.Bock, Kathryn, and Kathleen M. Eberhard. (1993). Meaning, sound and syntaxin English number agreement. Language and Cognitive Processes 8: 57–99.Caffarra, S., A. Siyanova-Chanturia, F. Pesciarelli, F. Vespignani, and C.

Cacciari.(2015). Is the noun ending a cue to grammatical gender processing? An ERPstudy on sentences in Italian. Psychophysiology 52: 1019–1030.Franck, Julie, Gabriella Vigliocco, Inés Antón-Méndez, Simona Collina, and Ulrich H. Frauenfelder. (2008). The interplay of syntax and form in sentence production: a cross-linguistic study of form effects on agreement. Language andCognitive Processes 23: 329–374.Gollan, T. H., and R.

Frost. (2001). Two routes to grammatical gender: Evidencefrom Hebrew. Journal of Psycholinguistic Research 30: 627–651.Janssen, Bibi E. (2016). The acquisition of gender and case in Polish and Russian:A study of monolingual and bilingual children. Amsterdam: Pegasus.69870Gender, Declension and Stem-final Consonants14. Lyashevskaya, Olga, and Sergey Sharov. (2009). Častotnyj slovar’ sovremennogorusskogo jazyka [The frequency dictionary of modern Russian language].

Moscow: Azbukovnik.15. Martin, A. E., M. S. Nieuwland, and M. Carreiras. (2014). Agreement attractionduring comprehension of grammatical sentences: ERP evidence from ellipsis.Brain and Language 135: 42–51.16. Ratcliff, Roger. (1993). Methods for dealing with reaction time outliers. Psychological Bulletin 114: 510–532.17. Rodina, Yulia, and Marit Westergaard. (2012). A cue-based approach to theacquisition of grammatical gender in Russian.

Journal of Child Language 39:1077–1106.18. Romanova, Natalia, and Kira Gor. (2017). Processing of gender and numberagreement in Russian as a second language. Studies in Second Language Acquisition 39: 97–128.19. Rusakova, Marina. (2013). Elementy antropocentričeskoj grammatiki russkogojazyka [Elements of the anthropocentric grammar of the Russian language].Moscow: Jazyki slavjanskoj kul’tury.20. Savchuk, Svetlana. (2011).

Korpusnoe issledovanie variantov rodovojprinadležnosti v russkom jazyke [A corpus study of morphological variability:variation in gender forms of Russian nouns]. Computer linguistics and intellectual technologies 10, 562–579.21. Schwartz, Mila, Miriam Minkov, Elena Dieser, Ekaterina Protassova, Victor Moin,and Maria Polinsky. (2015). Acquisition of Russian gender agreement by monolingual and bilingual children. International Journal of Bilingualism 19: 726–752.22. Sekerina, Irina.

(2012). The effect of grammatical gender in Russian spokenword recognition. In: Russian language studies in North America. New perspectives in theoretical and applied linguistics, ed. by Veronika Makarova, 107–132.New York: Anthem Press.23. Slioussar, Natalia, and Anton Malko. (2016). Gender agreement attractionin Russian: production and comprehension evidence. Frontiers in Psychology 7:article 1651.24. Slioussar, Natalia, and Maria Samoilova.