Update TRENDS in Cognitive Sciences Vol.8 No.11 November 2004 481

10 Markman, E.M. (1989) Categorization and Naming in Children:Problems of Induction, MIT Press

11 Markman, E.M. (1992) Constraints on word learning: speculationsabout their nature, orgins and domain specificity. In Modularity andConstraints in Language and Cognition: The Minnesota Symposiumon Child Psychology (Gunnar,M.R.andMaratsos,M.P. eds), pp. 59–101,Erlbaum

Corresponding author: Julia Fischer (fischer@eva.mpg.de).Available online 6 October 2004

www.sciencedirect.com

12 Clark, E.V. (1997) Conceptual perspective and lexical choice inacquisition. Cognition 64, 1–37

13 Diesendruck, G. andMarkson, L. (2001) Avoidance of lexical overlap: apragmatic account. Dev. Psychol. 37, 630–641

1364-6613/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.tics.2004.09.007

Research Focus Response

A pluralistic account of word learning

Julia Fischer, Josep Call and Juliane Kaminski

Max-Planck Institute for Evolutionary Anthropology, 04103 Leipzig, Germany

How do language learners come to understand the mean-ing of words? This is an intriguing question that has givenrise to a vast research field [1]. One smaller aspect withinthis research field is how novel words are mapped ontonovel objects. We studied this question in a domestic dog,Rico, who had already learned the names of more than 200toys. We found that he was able to link a novel word to anovel object on the basis of exclusion [2]. Markman andAbelev have taken a closer look at how Rico’s behaviorcould be accounted for ([3], this issue). Their main concernis that Rico selected the new toy among the familiar onesbecause of an inherent novelty preference. They alsocritically discuss the importance of rewarding Rico afterfetching the novel toy. Finally, they note that only experi-ments that pit reference against associative learning(e.g. [4]) can uncover a true understanding of reference.In the following, we will briefly address these concerns.

Firstly, novelty is indeed the discriminating feature ofthe ‘novel object’. It remains an empirical issue whetherRico can link a novel word to an unnamed object that hehad been familiarized with before. Markman and Abelevsuggested using the command ‘fetch’ without specifyingthe object, to see if Rico would show a preference for thenovel object. However, we do not think that this test isconclusive because he might still bring the novel objectbecause, unlike all other familiar objects, it did not have alabel. In other words, Rico could solve the problem on thebasis of recognizing that all objects have labels except onerather than on the basis of preference for novelty. It is alsoimportant to note that Rico was able to control his interestin the novel item as he first brought familiar objects uponrequest (see online video material; [2]). Therefore noveltyalone cannot explain our results.

Secondly, we feel the need to clarify that the name of thenovel object was never mentioned at the time when Ricowas rewarded with food or play. Importantly, within thesetting of the fetching game, a lack of reward would in factconstitute a negative reinforcement. That is, without areward, Rico might still have learnt that he had broughtthe ‘wrong’ item.

The final issue concerns the importance of the speaker’sintention for word learning. Clearly, children can use thespeaker’s attention to guide their process of word learn-ing. However, this is not a prerequisite for word learning.In Carey’s experiments [5], for example, children selectedthe ‘chromium tray’ without being able to use the speak-er’s attention as a cue. Again, it is an empirical issuewhether or not Rico monitors the speaker’s attention todisambiguate which object the speaker refers to. Inconclusion, we – like Markman and Abelev – propose apluralistic account of word learning where some mechan-isms are shared among children and language-trainedanimals, and others are uniquely human.

References

1 Bloom, P. (2000) How Children Learn the Meanings of Words, MITPress

2 Kaminski, J. et al. (2004) Word learning in a domestic dog: evidence for‘fast mapping’. Science 304, 1682–1683

3 Markman, E.M. and Abelev, M. (2004) Word learning in dogs? TrendsCogn. Sci. 8, doi: 10.1016/j.tics.2004.09.007

4 Baldwin, D. (1993) Early referential understanding: infants’ ability torecognize referential acts for what they are. Dev. Psychol. 29, 832–843

5 Carey, S. and Bartlett, E. (1978) Acquiring a single new word. Proc.Child Lang. Dev. 15, 17–29

1364-6613/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.tics.2004.09.008