Skip to main content
Have a personal or library account? Click to login
The Semantic Similarity Effect on Short-Term Memory: Null Effects of Affectively Defined Semantic Similarity Cover

The Semantic Similarity Effect on Short-Term Memory: Null Effects of Affectively Defined Semantic Similarity

By: Sho Ishiguro and  Satoru Saito  
Open Access
|Feb 2024

Figures & Tables

Figure 1

The schematic illustrations of the three facets of semantics.

Note. (a) Categorical structure: Hierarchical categories are often assumed in semantics. (b) Associative relationship: Networks of associations would represent semantics. An association does not necessarily correspond to a category (e.g., the association between ‘big’ and ‘band’ is not based on category but rather on their contingency). (c) Dimensions: The meaning of a word is expressed as a vector of its values. In this figure, the spatial representation of meaning is depicted as a point in a multidimensional space.

Figure 2

The schematic illustration of the relationship between similarity advantage and the strength of manipulation on similarity.

Note. Each point represents each of the previous studies (i.e., effect size of similarity advantage and manipulation strength of a single experiment). Values are not accurate for presentation.

Figure 3

A plot of words of the 12 similar lists along the valence, arousal, and dominance dimensions.

Note. Each number represents each similar list. For dissimilar list construction, 12 similar lists were divided into two sets. Set 1: similar lists 3, 4, 5, 9, 11, and 12. Set 2: similar lists 1, 2, 6, 7, 8, and 10. A dissimilar list was created by drawing one word from each similar list of a set (e.g., drawing a word from similar lists of Set 1). For an interactive plot, see a Jupyter notebook on OSF (https://osf.io/f4vb5).

Figure 4

The results of Study 1 and Study 2.

Note. The upper plots (a) and (b) represent correct-in-position scores in Study 1 (serial recall) and Study 2 (reconstruction of order), respectively. Accuracy refers to the rates of correct-in-position scores. Error bars represent standard errors calculated at each level combination (e.g., a standard error of the 50 participants’ scores of the similar first group at position 1 with dissimilar lists). The lower plots show the results of Study 1 by two scoring methods: (c) item correct scores and (d) proportion of order errors.

Figure 5

An example of test phase of the immediate serial reconstruction of order task.

Note. This example depicts the moment when the third word is going to be selected. The first and second words have been selected, and thus, two boxes below are black.

DOI: https://doi.org/10.5334/joc.349 | Journal eISSN: 2514-4820
Language: English
Submitted on: Sep 5, 2023
Accepted on: Jan 21, 2024
Published on: Feb 12, 2024
Published by: Ubiquity Press
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2024 Sho Ishiguro, Satoru Saito, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.