Normative ratings for perceptual and motor attributes of 750 object concepts in Spanish

Statistical Analysis of Multisensory and Text-Derived Representations on Concept Learning

When learning concepts, cognitive psychology research has revealed that there are two types of concept representations in the human brain: language-derived codes and sensory-derived codes. For the objective of human-like artificial intelligence, we expect to provide multisensory and text-derived representations for concepts in AI systems. Psychologists and computer scientists have published lots of datasets for the two kinds of representations, but as far as we know, no systematic work exits to analyze them together. We do a statistical study on them in this work. We want to know if multisensory vectors and text-derived vectors reflect conceptual understanding and if they are complementary in terms of cognition. Four experiments are presented in this work, all focused on multisensory representations labeled by psychologists and text-derived representations generated by computer scientists for concept learning, and the results demonstrate that (1) for the same concept, both forms of representations can properly reflect the concept, but (2) the representational similarity analysis findings reveal that the two types of representations are significantly different, (3) as the concreteness of the concept grows larger, the multisensory representation of the concept becomes closer to human beings than the text-derived representation, and (4) we verified that combining the two improves the concept representation.

Linking norms, ratings, and relations of words and concepts across multiple language varieties

Psychologists and linguists collect various data on word and concept properties. In psychology, scholars have accumulated norms and ratings for a large number of words in languages with many speakers. In linguistics, scholars have accumulated cross-linguistic information about the relations between words and concepts. Until now, however, there have been no efforts to combine information from the two fields, which would allow comparison of psychological and linguistic properties across different languages. The Database of Cross-Linguistic Norms, Ratings, and Relations for Words and Concepts (NoRaRe) is the first attempt to close this gap. Building on a reference catalog that offers standardization of concepts used in historical and typological language comparison, it integrates data from psychology and linguistics, collected from 98 data sets, covering 65 unique properties for 40 languages. The database is curated with the help of manual, automated, semi-automated workflows and uses a software API to control and access the data. The database is accessible via a web application, the software API, or using scripting languages. In this study, we present how the database is structured, how it can be extended, and how we control the quality of the data curation process. To illustrate its application, we present three case studies that test the validity of our approach, the accuracy of our workflows, and the integrative potential of the database. Due to regular version updates, the NoRaRe database has the potential to advance research in psychology and linguistics by offering researchers an integrated perspective on both fields.

subs2vec: Word embeddings from subtitles in 55 languages

This paper introduces a novel collection of word embeddings, numerical representations of lexical semantics, in 55 languages, trained on a large corpus of pseudo-conversational speech transcriptions from television shows and movies. The embeddings were trained on the OpenSubtitles corpus using the fastText implementation of the skipgram algorithm. Performance comparable with (and in some cases exceeding) embeddings trained on non-conversational (Wikipedia) text is reported on standard benchmark evaluation datasets. A novel evaluation method of particular relevance to psycholinguists is also introduced: prediction of experimental lexical norms in multiple languages. The models, as well as code for reproducing the models and all analyses reported in this paper (implemented as a user-friendly Python package), are freely available at: https://github.com/jvparidon/subs2vec.

A set of 750 words in Spanish characterized in two survival-related dimensions: avoiding death and locating nourishment

With the aim of finding quantitative indicators of the survival relevance for a set of concrete concepts, a subjective rating task was administered to a large sample of college students (N = 300). In the rating task, participants used a five-point scale to rate 750 concepts in one of two survival-relevant dimensions, providing their own judgment about the relevance of each concept in a situation in which either avoiding death (AD) or obtaining food (OF) was of importance. The subjective ratings showed high stability and reliability and showed varied patterns of association to potentially relevant concept-defining variables, with correlational analyses showing both commonalities and differences between the two rated dimensions. Regression analyses indicated that, while not likely to modulate word accessibility, survival ratings were related to certain conceptual properties that could be especially sensitive for threat detection. The collected data set provides normative information that can be of use in manipulating and controlling verbal stimuli in future research focusing on adaptive properties of episodic memory and other aspects of the human cognitive system. The complete norms are available for downloading at Open Science Framework ( https://osf.io/sf9mb/ ).