The activation of grammatical gender information in processing Italian nouns

Abstract

The present research addresses the issue of whether the orthographic-phonological information about gender provided by Italian affixes affects the processing of single nouns. In six experiments, transparent nouns (Italian feminine nouns ending in “-a”) were compared with irregular nouns (Italian masculine nouns ending in “a”).

We tested the assumption that when the orthographic-phonological information displayed in the gender suffix of a noun is inconsistent with the syntactic information about noun gender, lexical processing is slower and less accurate. The tasks employed (reading aloud, on-line inflection, and lexical decision) require different degrees of explicit linguistic knowledge and involve both recognition and production mechanisms.

The data are consistent with a pattern already reported in literature: transparent nouns are processed faster and better than irregular nouns. The results allow the conclusion that lexical processing of single nouns is affected by the manner in which grammatical gender is expressed in the surface form of nouns.

Keywords:

• Grammatical gender
• Declensional classes
• Single nouns

Acknowledgements

This research was supported by a grant MIUR-PRIN 2005 to Alessandro Laudanna.

Notes

¹Although grammatical gender cannot be related to the meaning of a word, it is possible that in some languages some objects and/or concepts tend to be more often assigned to a given grammatical gender; for instance, in Italian, nouns indicating fruits tend to be feminine (pera (pear), feminine, singular), while nouns indicating fruit trees (pero (pear tree), masculine, singular) tend to be masculine.

²All the procedures followed in this research are in accordance with the ethical standards of the responsible committee on human experimentation and with the Helsinki declaration of 1975, as revised in 1983.

³All the stimuli used in this research were words that can only function as nouns in Italian with the exception of the following: sistema (system) and sistemi (systems) which can be, respectively, the third singular person and the second singular person of the present tense for the verb sistemare (to arrange); scheda (card) which can be also the third singular person of the present tense for the verb schedare (to catalogue); struttura (structure) which can be also the third singular person of the present tense for the verb strutturare (to structure); pianta (plant) which can be also the third singular person of the present tense for the verb piantare (to plant). In our experiments we tried to ensure that these exceptions were processed as noun as well, by administering them within blocked lists in which only nouns were included.

⁴The majority of errors was constituted by stimuli pronounced incorrectly (e.g., pineta (pinery) instead of pianeta (planet), or strategia (strategy) instead of stratega (strategist)). The high number of errors is presumably due the time pressure imposed by the task.

⁵The effect was present in the error rates but not in the decision latencies, possibly reflecting a speed/accuracy trade-off as the average reaction times were relatively fast. Anyway, errors are usually considered good indexes of lexical processing mechanisms.

⁶Although in our discussion we refer to language production models, as noted elsewhere in this paper, we based our reasoning on the assumption that a two-layered organisation like the one proposed in some model of language production (Levelt et al., 1999) is also plausible for language comprehension (see also Friederici & Jacobsen, 1999).