Abstract
How children acquire and process inflectional morphology is still an open question. Despite the fact that English past tense acquisition has been studied and modeled in depth, the current approaches do not account for many of the errors made by humans. Moreover, not much work has been done with highly inflected languages, like Spanish. However, the modeling of any linguistic phenomenon in different languages is very important in order to understand the general cognitive processes underlying each particular phenomenon. This paper presents an ACT-R dual-mechanism model that accomplishes the task of acquiring verbal morphology systems from one of the simplest systems (the English one) to one of the most complex systems (the Spanish one), by using a double analogy process of stem and suffix. The model proposed was able to match all types of errors that developing children make (from a sample of them), both in English and Spanish. The models for both languages used very similar parameters. The introduced approach not only shows how children could acquire a highly inflected morphology system in terms of dual-mechanism theories but, given its cross-linguistic character, also sheds light on the possible general processes involved in the acquisition and processing of inflectional morphology.
Similar content being viewed by others
Notes
Accessible at http://crl.ucsd.edu/experiments/svi/
References
MacWhinney B. A multiple process solution to the logical problem of language acquisition. J Child Lang. 2004;31(04):883.
Pinker S. Language learnability and language development. Cambridge: Harvard University Press; 1984.
Vavrečka M, Farkaš I. A multimodal connectionist architecture for unsupervised grounding of spatial language. Cogn Comput. 2014;6(1):101.
Plebe A, Mazzone M, de-la Cruz V. First words learning: a cortical model. Cogn Comput. 2010;2(3):217.
Baayen RH, Prado Martin FMD. Semantic density and past-tense formation in three Germanic languages. Language. 2005;81(3):666.
Marcus G, Pinker S, Ullman M, Hollander M, nad TR, Xu F. 1992. Overregularization in language acquisition, Monographs of the Society for Research in Child Development. Wiley; Vol. 57.
Haikonen P. The role of associative processing in cognitive computing. Cogn Comput. 2009;1(1):42.
Alcoba S. 1999. Gramática Descriptiva de la Lengua Española.
Durst-Andersen P. What languages tell us about the structure of the human mind. Cogn Comput. 2012;4(1):82.
Quirk R, Greenbaum S, Leech G, Svartik J, Crystal D. A comprehensive grammar of the English language. 1985: Longman.
MacWhinney B. The CHILDES project. Hillsdale: Erlbaum; 2000.
Cazden CB. Child development. 1968.
Maratsos M. More overregularizations after all: new data and discussion on Marcus, Pinker, Ullman, Hollander, Rosen & Xu. J Child Lang. 2000;27(01):183.
Clahsen H, Aveledo F, Roca I. The development of regular and irregular verb inflection in Spanish child language. J Child Lang. 2002;29(03):591.
Aguado-Orea JJ. The acquisition of morpho-syntax in Spanish: implications for current theories of development. 2004: University of Nottingham United Kingdom.
Hernández-Pina F. Teorias psicolinguisticas y su aplicacion a la adquisicion del espanol como lengua materna. Madrid: Siglo Veintiuno; 1984.
López-Ornat S. La metodología de la investigación longitudinal. Madrid: Siglo Veintiuno; 1994.
Johnson C. Verb errors in the early acquisition of Mexican and Castilian Spanish. In: Clark E., editors. The Proceedings of the 27th annual child language research forum. Cambridge: Cambridge University Press; 1995.
Radford A. The morphosyntax of subjects and verbs in child Spanish: a case study. Essex research reports in linguistics. 1995;5:23.
Serrat E, Aparici M. Research on child language acquisition. Cascadilla Press; 2001, Vol. II, chap. Morphological errors in early verb acquisition: evidence from Catalan and Spanish.
Rumelhart D, McClelland J. chap. On learning the past tense of english verbs. Parallel distributed processing: explorations in the microstructure of cognition. Cambridge: MIT Press; 1986, pp. 216–271.
Pinker S, Ullman M. The past and future of the past tense. Trends Cogn Sci. 2002;11:456.
Katnazcheev A. A connectionist study on the interplay of nouns and pronouns in personal pronoun acquisition. Cogn Comput. 2010;2(4):280.
Pinker S, Prince A. On language and connectionism: analysis of a distributed processing model of language acquisition. 1988.
Plunkett K, Marchman M. From rote learning to system building. 1990. Tech. Rep. 9020, University of California, Center for Research in Language.
Plunkett K, Marchman V. U-shaped learning and frequency effects in a multi-layered perception: implications for child language acquisition. Cognition. 1991;38(1):43.
Plunkett K, Marchman V. From rote learning to system building: acquiring verb morphology in children and connectionist nets. Cognition. 1993;48(1):21.
MacWhinney B, Leinbach J. Implementations are not conceptualizations: revising the verb learning model. Cognition. 1991;40(1):121.
Plunkett K, Juola P. A connectionist model of English past tense and plural morphology. Cogn Comput. 1999; 23(4):463.
Hahn U, Nakisa RC. German inflection: single route or dual route?. Cogn Psychol. 2000;41:313. doi:10.1006/cogp.2000.0737 10.1006/cogp.2000.0737.
Plunkett K, Nakisa RC. A connectionist model of the Arabic plural system. Lang Cogn Process. 1997;12(5-6):807.
Ruh N, Westermann G. A single-mechanism dual-route model of German verb inflection. In: Proceedings of the 30th annual conference of the Cognitive Science Society; 2008. p. 2209–2216.
Eddington D. Spanish verbal inflection: a single-or dual-route system?. Linguistics. 2009;47(1):173.
Karaminis TN, Thomas MS. A cross-linguistic model of the acquisition of inflectional morphology in English and Modern Greek. In: Ohlsson S. and Catrambone R., editors. In: Proceedings of 32nd annual conference of the Cognitive Science Society. Citeseer; 2010. p. 730–735.
Ullman MT. A neurocognitive perspective on language: the declarative/procedural model. Nat Rev Neurosci. 2001;2(10):717.
Hartshorne JK, Ullman MT. Why girls say ‘holded’ more than boys. Dev Sci. 2006;9:21. doi:10.1111/j.1467-7687.2005.00459.x 10.1111/j.1467-7687.2005.00459.x.
Taatgen N, Anderson JR. Why do children learn to say broke? A model of learning the past tense without feedback. Cognition. 2002;86(2):123.
Taatgen N. Extending the past-tense debate: a model of the German plural. In: Proceedings of the twenty-third annual conference of the Cognitive Science Society; 2001. p. 1018–23.
Oliva J, Serrano J, del Castillo M, Iglesias A, Guzelmann G. Cognitive modeling of the acquisition of a highly inflected verbal system. In: Salvucci, D, editor. Proceedings of the 10th international conference on cognitive modeling. Philadelphia: Drexel University; 2010. p. 181–186.
Xu F, Pinker S. Weird past tense forms. J Child Lang 1995;22(03):531.
Taatgen N, Dijkstra M. Constraints on generalization: why are pasttense irregularization errors so rare. In: Proceedings of the 25th annual conference of the Cognitive Science Society; 2003. p. 1146–1151.
Albright A, Hayes B. Rules vs. analogy in English past tenses: a computational/experimental study. Cognition 2003;90(2):119.
McClelland JL, Patterson K. Words or Rules cannot exploit the regularity in exceptions. Trends Cogn Sci. 2002;6(11):464.
Carlson A, Betteridge J, Kisiel B, Settles B, Hruschka Jr ER, Mitchell TM. Toward an Architecture for Never-Ending Language Learning. AAAI; 2010. p. 3. Vol. 5.
Krishnamurthy J, Kollar T. Jointly learning to parse and perceive: connecting natural language to the physical world. Transactions of the association for computational linguistics. 2013 ;1:193.
MacWhinney B. The acquisition of morphophonology, monographs of the society for research in child development. Wiley; 1978, Vol. 43.
Anderson JR. A spreading activation theory of memory. J Verbal Learn Verbal Behav. 1983;22(3):261.
Frances N, Kucera H. Frequency analysis of English usage: lexicon and grammar. Boston: Houghton Mifflin; 1982.
Rivera S, Bates E, Orozco-Figueroa A, Wicha N. Spoken verb processing in Spanish: an analysis using a new online resource. Appl Psycholinguist 2010;31(1):29.
Roberts S, Pashler H. How persuasive is a good fit? A comment on theory testing. Psychol Rev 2000;107(2):358.
Wong TJ, Cokely ET, Schooler LJ. In: Proceedings of the tenth international conference on cognitive modeling. Philadelphia: Citeseer; 2010. p. 282–286.
Gathercole V, Sebastián E., Soto P. The early acquisition of Spanish verbal morphology: across-the-board or piecemeal knowledge. Int J Biling. 1999;3:133.
Montrul S. The acquisition of Spanish: morphosyntactic development in monolingual and bilingual L1 acquisition and adult L2 acquisition. Amsterdam: Benjamins; 2004.
Acknowledgments
The authors would like to thank Prof. Harald Clahsen for making available the full contents of their study with Spanish children [14]. We also want to thank Prof. Fraibet Aveledo for making available its unpublished Master’s thesis, related to the same study. This study was supported by projects NEUROMOD (DPI2015-68664-C4-1-R), NETMD (RTC-2015-3967-1) and Intramurales Especiales 2016 (201650E055).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
The authors declare that the paper is in compliance with the Ethical Standard of Cognitive Computation journal following COPE (Committee on Publication Ethics) guidelines.
No data from human beings were collected for this study. All data used in this study were collected by a third party and were publicly available under proper license.
Conflict of Interest
The authors declare that they have no conflict of interest.
Funding
No specific funding was devoted to this study.
Appendix A: ACT-R Parameters
Appendix A: ACT-R Parameters
The proposed model is implemented under ACT-R 6.0 version. Table 10 show the values of the parameters used in the simulations with English and Spanish presented in this paper. Spreading activation, base level learning, optimized learning, and procedural learning are switched on. Partial matching is activated and the penalty values established for each chunk slots are the following. If the INFIN slot is different, a penalty of 5.0 is given (which is practically the same as not allowing the retrieval of a form of a different verb). However, the rest of slots have lower penalties. The MTA slot has a penalty of 1.5, the NP slot a penalty of 0.3, and the CONJ slot a penalty of 0.3. These values are based on previous studies on the acquisition of tense, number, and other grammatical features in Spanish [52, 53].
One of the main criticisms to ACT-R models is the flexibility they present given the wide range of free parameters (combinatorial complexity). However, in the proposed model, all the parameters are set to commonly used values that do not exceed the typical ranges at all, thus supporting the psychological plausibility of the model. The parameter setting was done by comparison with many other models of very different tasks. Wong et al. [51] proposed a database that collects the parameter values of a representative range of published ACT-R models. That database was used to set the parameter values close to the average values of all those studies. Different possible combinations of parameters were tested in the range given by the average and standard deviations, and the one that fits better both the developmental curve and the overall percentages of correct regular and irregular forms was chosen. This way, it is demonstrated that language acquisition can be accomplished using the same general cognitive constraints. Moreover, both the Spanish and the English models work with very similar parameters. Thus, the proposed model does not assume special cognitive abilities for children talking one or the other language (it would be certainly implausible to suppose for example that English children have a much greater capacity to retrieve verb forms from memory). The only parameter that significantly differs from the average value in the study of Wong et al. [51] is the α parameter. The proposed model uses a value much lower than the one used in other cognitive tasks. The reason is that most of the models in the study of Wong et al. [51] try to model the behavior of adults in time-limited tasks. However, the proposed model represents a typical children brain through the long-lasting task of acquiring inflections.
Rights and permissions
About this article
Cite this article
Oliva, J., Serrano, J.I., del Castillo, M.D. et al. Cross-Linguistic Cognitive Modeling of Verbal Morphology Acquisition. Cogn Comput 9, 237–258 (2017). https://doi.org/10.1007/s12559-017-9454-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12559-017-9454-8