Verbalna sposobnost i korišćenje gramatičke informacije u srpskom jeziku

Abstract

Istraživanje prikazano ovim radom predstavlja pokušaj da se ustanove individualne razlike u korišćenju gramatičke informacije koju nose pojedinačne reči srpskog jezika. Cilj istraživanja bio je da se utvrdi nivo kognitivne obrade reči na kome će gramatička informacija vršiti diferencijalni uticaj, u grupama ispitanika koje se razlikuju u pogledu razvijenosti verbalne sposobnosti. Za tu svrhu su primenjeni eksperimentalni obrasci primovanog čitanja naglas i primovane gramatičke klasifikacije, uz odgovarajuću kontrolu uticaja nejezičkih faktora. Verbalna sposobnost je psihometrijski procenjena, a ispitanici podeljeni na grupe "visokoverbalnih" i "niskoverbalnih". Uzimajući u obzir statističku kontrolu nejezičkih faktora, rezultati ukazuju da čitanje naglas ne diferencira grupu visokoverbalnih i niskoverbalnih ispitanika. Visokoverbalni su, međutim, efikasniji od niskoverbalnih u gramatičkoj klasifikaciji. Rezultati takođe sugerišu da prisustvo dodatne gramatičke informacije ima jači uticaj na obradu reči u grupi niskoverbalnih. Ovakvi rezultati svedoče o prednosti visokoverbalnih u leksičkoj i postleksičkoj obradi, dok takva razlika u procesima prepoznavanja reči nije konstatovana. Razmotrene su implikacije dobijenih nalaza po konstrukciju testova za procenu verbalne sposobnosti na srpskom jeziku.

The research presented in this paper was an attempt to find differences in the use of grammatical information carried by the function words in Serbian. The aim was to determine the level of word processing at which grammatical information shows its differential effects in groups of subjects who themselves differ in verbal ability. For this purpose, the psycholinguistic tasks applied were grammatically primed reading aloud and grammatically primed grammatical classification with an appropriate control of extra-linguistic factors that may have affected aforementioned tasks. Verbal aptitude was assessed in a psychometric manner, and the subjects were divided into "high verbal" and "low verbal" groups. Taking into account statistical control of extra-linguistic factors, the results indicate that groups of high verbal and low verbal subjects cannot be differentiated based on reading aloud performance. The high verbal subjects, however, were more efficient in grammatical classification than low verbal subjects. The results also indicated that the presence of grammatical information embedded in function words-primes had a stronger effect on word processing in low verbal group. Such pattern of results testify to the advantage of high verbal subjects in lexical and post lexical processing, while no differences were established in the word recognition processes. The implications of these findings were considered in terms of test construction for the assessment of verbal ability in Serbian language.