The Croatian psycholinguistic database: Estimates for 6000 nouns, verbs, adjectives and adverbs

Abstract

Psycholinguistic databases containing ratings of concreteness, imageability, age of acquisition, and subjective frequency are used in psycholinguistic and neurolinguistic studies which require words as stimuli. Linguistic characteristics (e.g. word length, corpus frequency) are frequently coded, but word class is seldom systematically treated, although there are indications of its significance for imageability and concreteness. This paper presents the Croatian Psycholinguistic Database (CPD; available at: 10.17234/megahr.2019.hpb), containing 6000 Croatian nouns, verbs, adjectives and adverbs, rated for concreteness, imageability, age of acquisition, and subjective frequency. Moreover, we present computationally obtained extrapolations of concreteness and imageability to the remainder of the Croatian lexicon (available at: https://github.com/megahr/lexicon/blob/master/predictions/hr_c_i.predictions.txt). In the two studies presented here, we explore the significance of word class for concreteness and imageability in human and computationally obtained ratings. The observed correlations in the CPD indicate correspondences between psycholinguistic measures expected from the literature. Word classes exhibit differences in subjective frequency, age of acquisition, concreteness and imageability, with significant differences between nouns, verbs, adjectives and adverbs. In the computational study which focused on concreteness and imageability, concreteness obtained higher correlations with human ratings than imageability, and the system underpredicted the concreteness of nouns, and overpredicted the concreteness of adjectives and adverbs. Overall, this suggests that word class contains schematic conceptual and distributional information. Schematic conceptual content seems to be more significant in human ratings of concreteness and less significant in computationally obtained ratings, where distributional information seems to play a more significant role. This suggests that word class differences should be theoretically explored.

The Dangers of Underclaiming: Reasons for Caution When Reporting How NLP Systems Fail

“…Similar claims can be found in many other citing works (Utama et al, 2020;van Noord et al, 2020;Hovy and Yang, 2021;Sayers et al, 2021;Peti-Stantić et al, 2021;Jang and Lukasiewicz, 2021). While Bender and Koller raise important points for discussion, these strong implications in citing works are misleading and potentially harmful.…”

How do word valence and classes influence lexical processing? Evidence from virtual reality emotional contexts

“…Valence effects can be modulated by word classes because different word classes have different emotional arousal and processing mechanisms for an individual (Kolbeneval and Alexandrov, 2016;Zhang, 2019a). Nouns, verbs, and adjectives are different in their concreteness, imageability, and semantic references, indicating different complexity in recognition and processing (Gleitman, 1990;Xia et al, 2016;Peti-Stantić et al, 2021). Some studies even found that different word classes were associated with different neurons, and thus had different stability, consistency and gradability (Waxman and Booth, 2001;Kolbeneval and Alexandrov, 2016).…”

“…While it is generally suggested that valence play a crucial role in lexical processing, the bulk of the previous studies have been prominently featured by discrete emotional stimuli limited to emotional face pictures or sentences (Wieser et al, 2014;Diéguez-Risco et al, 2015;Fan et al, 2018;Li et al, 2020) or lexical decision tasks (Barriga-Paulino et al, 2022) under laboratory settings. In addition, nouns and adjectives are different in their concreteness and imageability (Peti-Stantić et al, 2021), semantic references (Xia et al, 2016), and processing mechanism (Zhu et al, 2011;Kolbeneval and Alexandrov, 2016;Zhang, 2019a), which could modulate an individual's lexical processing with different valences under different contexts. Emotional experiences are 'ingredients' in the creation of emotional perceptions (Glenberg et al, 2009;Lindquist, 2017;Zhang, 2019a,b), and evidence from controlled laboratory settings cannot gauge fully an individual's real emotional experiences and their effects on emotional words processing (Pavlenko, 2008).…”

Specificity ratings for Italian data

“…Datasets including concreteness norms are now available for several languages. To name just a few recent ones, Montefinese et al (2014) for Italian, Guasch et al (2016) for Spanish, Soares et al (2018) for Portuguese, Yao et al (2017) for Chinese, Bonin et al (2018) for French, Peti-Stantić et al (2021) for Croatian. For English, Brysbaert et al (2014) is among the most widely used.…”