Age-of-acquisition effects in delayed-naming tasks.

Within the field of study dealing with word-production, it’s a highly reliable observation that words acquired earlier in life are processed faster than words acquired later in life. The locus in which this Age of Acquisition (AoA) effect has its origin, however, is still a matter of debate. Critically, discrepant conclusions about the issue might stem from the fact that different experimental paradigms have been used to evaluate different loci (Catling & Johnston, 2008). The aim of the present work is to asses the possible loci of the AoA effect within the same experimental task: delayed picture naming. Two different experiments were designed. In both experiments, a picture has to be named after a delay (1000 ms.) on the 25% of trials. On the remaining 75% of trials, participants have to perform a different task after the delay: semantic categorization in Experiment 1, grammatical gender decision in Experiment 2. The rationale underlying this paradigm is to bias the processing occurring during the delay (see Almeida et al. 2007). During the delay, participants in Experiment 1 are biased to prepare the response to the categorization task, processing the picture up to the conceptual-semantic level, while in Experiment 2 participants are biased to prepare the response to the gender decision task, therefore selecting a lexical representation. An AoA effect was found in Experiment 1, suggesting that the effect stems from stages after semantic processing. In a control immediate naming experiment (i.e. without delay) a similar AoA effect was found, suggesting that the conceptual processing occurring during the delay is not able to reduce the AoA effect. In Experiment 2, where participants are biased select a lexical entry during the delay, AoA effect was small but still significant, suggesting that AoA also affects stages subsequent to lexical selection, i.e. phonological encoding. When the delay was removed in a control immediate naming experiment, the AoA effect detected was significantly larger compared to the delayed-naming condition, coherently with the idea that part of the effect occurred during lexical selection. Taken together the results support a dual-locus account, in which AoA affects lexical selection and phonological encoding. Critically, the experiments reported investigated different possible loci of the AoA within the same task.