Mancini, S. 1 , Molinaro, N. 1 , Avilés, A. 1 & Carreiras, M. 1, 2
1 Basque Center on Cognition, Brain and Language (BCBL). Donostia, Spain
2 IKERBASQUE, Basque Foundation for Science. Bilbao, Spain
We investigated the mechanisms underlying agreement comprehension in two typologically-close languages - Spanish and Italian - using two eye-tracking experiments. Italian and Spanish native speakers read sentences containing person and number anomalies in their own language. Both violations produced longer total-reading times and regression-path durations compared to correct sentences, with no difference between person and number anomalies, neither in Italian nor in Spanish. However, the two features differed in the probability and in the number of regressions out of the interest area (a past participle verb), with number violations showing a greater probability and number of regressions towards earlier parts of the sentence than person ones. Crucially, this difference emerged in Italian but not in Spanish. An explanation for this may reside in Spanish greater inflectional regularity in signaling number information across grammatical categories (“-s”) than Italian. The presence of an “-s” either on the auxiliary (e.g.”hemos”) or on the subject (e.g.“ellos”) may lead Spanish speakers to actively rely on morphological regularities to interpret agreement dependencies. On the contrary, Italian variability in plural number suffixes may require the parser to perform more regressions to check the number information contained in previous words and interpret the dependency.