Abstract
Previous research has demonstrated that Distributional Semantic Models (DSMs) are capable of reconstructing maps from news corpora (Louwerse & Zwaan, 2009) and novels (Louwerse & Benesh, 2012). The capacity for reproducing maps is surprising since DSMs notoriously lack perceptual grounding. In this paper we investigate the statistical sources required in language to infer maps, and the resulting constraints placed on mechanisms of semantic representation. Study 1 brings word co-occurrence under experimental control to demonstrate that standard DSMs cannot reproduce maps when word co-occurrence is uniform. Specifically, standard DSMs require that direct co-occurrences between city names in a corpus mirror the proximity between the city locations in the map in order to successfully reconstruct the spatial map. Study 2 presents an instance-based DSM that is capable of reconstructing maps independent of the frequency of co-occurrence of city names.
Highlights
Distributional Semantic Models (DSMs) posit cognitive mechanisms to explain how humans construct semantic representations for words from statistical regularities in natural language
When we control for the frequency of co-occurrence, modern DSMs are not able to accurately co-locate cities in semantic space
Demonstrations by Louwerse and Zwaan (2009) and Louwerse and Benesh (2012) show that spatial distributions can be elicited from text
Summary
Distributional Semantic Models (DSMs) posit cognitive mechanisms to explain how humans construct semantic representations for words from statistical regularities in natural language. These models represent words as points in a high-dimensional vector space, and similarity between words is measured as the proximity in this semantic space. Latent Semantic Analysis (LSA; Landauer & Dumais, 1997) is the classic example of a DSM, but more modern versions span theoretically diverse learning mechanisms (see Jones et al, 2015 for a review). DSMs may lack a necessary source of statistical information to fully represent semantic relationships between words. There is often strong alignment between the statistical distributions of words in a corpus and perceptual data (Riordan & Jones, 2011; Roads & Love, 2020)
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
