American Psycholinguistics and Slavic Languages

Irina Sekerina

The Institute for Research in Cognitive Science,
University of Pennsylvania


Recent years have shown a rapid advancement in the new field of cognitive sciences which fosters the science of the human mind through the interaction of the disciplines of Linguistics, Mathematical Logic, Philosophy, Psychology, Computer Science, and Neuroscience. The study of language as an essential component of the human mind is undertaken by psycholinguistics which combines the methods and theoretical apparatus of linguistics and psychology. In psycholinguistics, research dedicated to comprehension of language and in particular, syntactic processing in language comprehension, occupies the leading position among conference presentations and scholarly publications.

Theories of sentence processing are investigating how people construct the syntactic structure of a sentence and how they recover from mysanalysis, if this occurs during sentence comprehension. The experimental methods that psycholinguists utilize can be off-line and on-line. The off-line methods include paper-and-pencil questionnaires, sentence completion studies, grammaticality/accessibility judgments, among others. They all tap into the processes which have already taken place and, in this sense, cannot capture the actual processing. More productively, newer on-line methods allow the researchers to look into rapid incremental processing as it evolves in the process of comprehending a sentence. The on-line methods include self-paced moving window technique, cross-modal priming, bite-bar eye-tracking, and the new cutting-edge head-mounted eye-tracking technology.

Current trends in American psycholinguistics reflect the same changes that general linguistics has been undergoing for the past 15 years where the focus of the research has shifted to cross-linguistic comparisons. Many recent linguistic and psycholinguistic theories are relatively untested for languages whose structures vary significantly from English. The majority of the Slavic languages, with their freedom of word order, rich inflectional morphology, and many other interesting typological properties, provide excellent testing grounds for these theories.

In this paper, I survey the on-going research in modern psycholinguistics which is being conducted with data from four Slavic languages, Russian, Serbian/Croatian, Bulgarian, and Polish. First, I present a series of self-paced reading experiments on Russian conducted by Maria Babyonyshev (1996) at MIT, Irina Sekerina (1997; 1998; 1999a) at the University of Pennsylvania, and Janet Nicol and Rachel Wilson (1999) at the University of Arizona. These experiments address different issues in Russian processing: the role of abstract case (Babyonyshev), the resolution of so-called 'Late Closure' ambiguity, and complexity of Scrambling and Split Scrambling constructions in contrast to Wh-Movement (Sekerina), and agreement errors in production (Nicol and Wilson). Similar questions with respect to processing of filler-gap dependencies, e.g., relative clauses, in Serbian/Croatian have been investigated by Danijela Stojanovi( (forthcoming).

Next, I describe a series of experiments conducted at the University of Leipzig by Gerhild Zybatow and Grit Mehlhorn (1999) aimed at investigating the relations between the information structure and prosodic structure in Russian Scrambling sentences. Finally, I discuss the results of a head-mounted eye-tracking experiments whose goal was to look for linguistic factors which contribute to referential ambiguity resolution (Sekerina, 1999b).

In addition, I present the results of a cross-modal priming experiment conducted by Elena Andonova and Maxim Stamenov (1998) at the New Sofia University. The goal of the experiment was to investigate the time course of antecedent reactivation during processing of object-relative clauses in Bulgarian. Finally, I will briefly discuss the experiment by Anna Novak from the University of Massachusetts at Amherst which investigated the effects of Case markers in processing of ambiguous Polish sentences (Novak, 1999).

In conclusion, I will address the questions which the above-mentioned psycholinguistic research on Slavic data poses for theories of sentence processing and speculate on the direction of future research in the area of Slavic psycholinguistics.