Feature distinctiveness effects in language acquisition and lexical processing: Insights from megastudies

Abstract

Semantic features are central to many influential theories of word meaning and semantic memory, but new methods of quantifying the information embedded in feature production norms are needed to advance our understanding of semantic processing and language acquisition. This paper capitalized on databases of semantic feature production norms and age-of-acquisition ratings, and megastudies including the English Lexicon Project and the Calgary Semantic Decision Project, to examine the influence of feature distinctiveness on language acquisition, visual lexical decision, and semantic decision. A feature network of English words was constructed such that edges in the network represented feature distance, or dissimilarity, between words (i.e., Jaccard and Manhattan distances of probability distributions of features elicited for each pair of words), enabling us to quantify the relative feature distinctiveness of individual words relative to other words in the network. Words with greater feature distinctiveness tended to be acquired earlier. Regression analyses of megastudy data revealed that Manhattan feature distinctiveness inhibited performance on the visual lexical decision task, facilitated semantic decision performance for concrete concepts, and inhibited semantic decision performance for abstract concepts. These results demonstrate the importance of considering the structural properties of words embedded in a semantic feature space in order to increase our understanding of semantic processing and language acquisition.

Appendices

Appendix A

Correlations between predictors and outcome variables in the regression models.

See Tables 6, 7, and 8.

Table 6 Age-of-acquisition norms (N = 4013)

Table 7 Visual lexical decision (English Lexicon Project) (N = 4013)

Table 8 Semantic decision (Calgary Semantic Decision Project) (N = 1336)

Appendix B

Computation of Jaccard and Manhattan distance measures

Two different distance measures were computed to quantify the dissimilarity of any given pair of probability distributions: Jaccard distance and Manhattan distance. The first measure, Jaccard distance, is an example of a measure from the inner-product family of distance measures that emphasizes shared information. Jaccard distance computes the distance between two words as subtracting the intersection of the two words’ feature set distributions (i.e., a vector that represents the proportions of participants reporting each feature for a given word) over their union from 1. The second measure, Manhattan distance, is an example of a measure from the Minkowski family of distance measures that generally treats distance as the straight line between two points in Euclidean space. Manhattan distance computes the distance between two words as the sum of absolute differences between the proportions of participants reporting each feature. Mathematically, these measures were computed as follows:

Jaccard distance, d_j:

$$d_{\text{j}} = 1 - \frac{{\sum \left( {P_{i} \times Q_{i} } \right)}}{{\sum P_{i}^{2} + \sum Q_{i}^{2} - \sum \left( {P_{i} \times Q_{i} } \right)}}$$

Manhattan distance, d_m:

$$d_{\text{m}} = \sum \left| {P_{i} - Q_{i} } \right|$$

where P_i = [x₁, x₂, …, x_i] is a vector representing the proportions of participants reporting each feature for a word 1 and Q_i = [x₁, x₂, …, x_i] is a vector representing the proportions of participants reporting each feature for a word 2. i = the number of (unique) features in the feature production norms.