Why and how the co-occurring familiar object matters in Fast Mapping (FM)? Insights from computational models

ABSTRACT

This article uses insights from computational semantic networks to explain why the co-occurring familiar objects are critical to the Fast Mapping (FM) procedure. I first propose that the co-occurring familiar objects provide the novel targets with a ‘mimicry opportunity’, which may facilitate the establishment of targets in long-term cortical memory networks. I then argue that the occurrence of rapid cortical learning may depend on how ‘well-connected’ the co-occurring familiar object is in long-term memory networks.

KEYWORDS:

• Fast mapping
• familiar object
• computational semantic network
• semantic growth
• consolidation

Acknowledgments

The author thanks Kate Nation for comments on an earlier draft and Nicola Dawson for editorial assistance.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

¹ In highly comparable studies (e.g., Wang et al., 2017), the number of observations/condition was at least 300. Also note that the dependent variable of these studies was Reaction Time, which is a naturally noisy measure. Therefore, in order to find reliable results, the number of observations/condition must be large (see for example Brysbaert & Stevens, 2018).

Additional information

Funding

This work was supported by the Drs. Richard Charles and Esther Yewpick Lee Charitable Foundation.