Determiner selection in noun phrase production

Ansgar Hantsch¹, Jörg D. Jescheniak¹, and Herbert Schriefers²
hantsch@cns.mpg.de, jeschen@cns.mpg.de, schriefers@nici.kun.nl
1 Max-Planck-Institute of Cognitive Neuroscience
2 Nijmegen Institute of Cognition and Information

Naming objects with gender marked noun phrases (NPs, "der Baum" [the(masc) tree]) is inhibited more strongly by distractor words with different gender (Vase [vase(fem)]) as compared to distractors with the same gender (Käse [cheese(masc)]; Schriefers, 1993). Originally this so-called gender-congruency effect was interpreted as the result of competition during the selection of an abstract gender representation (i.e., the feature "masculine"). Recently, Schiller and Caramazza (submitted) replicated this effect for singular NPs but observed no such effect for plural NPs ("die Bäume" [the(plural) trees]). Based on this observation, they argued that the effect results from competition during the selection of the determiner. On this account, no competition is to be expected for plural NPs, because the plural determiner is identical for the different gender classes. In three experiments we tested predictions derived from this determiner competition account.

• In Experiment 1 participants named objects with singular or plural NPs ("der Baum" - "die Bäume" [the(masc) tree - the(plural) trees]). The relative frequency of occurrence of the different determiners ("der", "die", "das") was controlled. NP production was slower for plural NPs compared to singular NPs, but only if determiners differed for singular and plural (i.e., for masculine and neuter gender). No difference was found if the determiner for singular and plural was identical (i.e., for feminine gender).

• In Experiment 2 the same materials were used but participants produced bare nouns ("Baum" [tree]) instead of NPs. If differences in the relative ease of producing plural NPs vs. singular NPs across gender classes rather than determiner competition were responsible for the pattern of results in Experiment 1, a comparable pattern should be obtained in Experiment 2. This was not the case.

• Experiment 3 finally replicated the pattern from Experiment 1, but this time the relative frequency of occurrence of singular and plural trials (rather than the relative frequency of occurrence of the different determiners) was controlled. Together. these results support the hypothesis that singular determiners are activated by default during NP production.

References

Schiller, N.,& Caramazza, A. (submitted). Gender or determiner selection interference? Evidence from noun phrase production in German and Dutch. Schriefers, H. (1993). Syntactic processes in the production of noun phrases. Journal of Experimental Psychology: Learning, Memory, and Cognition, 19, 841-850.