The measurement of meaning

Information about the affective meanings of words is used by researchers working on emotions and moods, word recognition and memory, and text-based sentiment analysis. Three components of emotions are traditionally distinguished: valence (the pleasantness of a stimulus), arousal (the intensity of emotion provoked by a stimulus), and dominance (the degree of control exerted by a stimulus). Thus far, nearly all research has been based on the ANEW norms collected by Bradley and Lang (1999) for 1,034 words. We extended that database to nearly 14,000 English lemmas, providing researchers with a much richer source of information, including gender, age, and educational differences in emotion norms. As an example of the new possibilities, we included stimuli from nearly all of the category norms (e.g., types of diseases, occupations, and taboo words) collected by Van Overschelde, Rawson, and Dunlosky (Journal of Memory and Language 50:289-335, 2004), making it possible to include affect in studies of semantic memory.

/pdf/norms-of-valence-arousal-and-dominance-for-13-915-english-gnh3edpo75.pdf

Norms of valence, arousal, and dominance for 13,915 English lemmas

Analyses of adult semantic networks suggest a learning mechanism involving preferential attachment: A word is more likely to enter the lexicon the more connected the known words to which it is related. We introduce and test two alternative growth principles: preferential acquisition—words enter the lexicon not because they are related to well-connected words, but because they connect well to other words in the learning environment—and the lure of the associates—new words are favored in proportion to their connections with known words. We tested these alternative principles using longitudinal analyses of developing networks of 130 nouns children learn prior to the age of 30 months. We tested both networks with links between words represented by features and networks with links represented by associations. The feature networks did not predict age of acquisition using any growth model. The associative networks grew by preferential acquisition, with the best model incorporating word frequency, number of phonological neighbors, and connectedness of the new word to words in the learning environment, as operationalized by connectedness to words typically acquired by the age of 30 months.

Early semantic Networks: Preferential Attachment or Preferential Acquisition?

We developed affective norms for 1,121 Italian words in order to provide researchers with a highly controlled tool for the study of verbal processing. This database was developed from translations of the 1,034 English words present in the Affective Norms for English Words (ANEW; Bradley & Lang, 1999) and from words taken from Italian semantic norms (Montefinese, Ambrosini, Fairfield, & Mammarella, Behavior Research Methods, 45, 440–461, 2013). Participants evaluated valence, arousal, and dominance using the Self-Assessment Manikin (SAM) in a Web survey procedure. Participants also provided evaluations of three subjective psycholinguistic indexes (familiarity, imageability, and concreteness), and five objective psycholinguistic indexes (e.g., word frequency) were also included in the resulting database in order to further characterize the Italian words. We obtained a typical quadratic relation between valence and arousal, in line with previous findings. We also tested the reliability of the present ANEW adaptation for Italian by comparing it to previous affective databases and performing split-half correlations for each variable. We found high split-half correlations within our sample and high correlations between our ratings and those of previous studies, confirming the validity of the adaptation of ANEW for Italian. This database of affective norms provides a tool for future research about the effects of emotion on human cognition.

The adaptation of the Affective Norms for English Words (ANEW) for Italian

Word embeddings that provide continuous low-dimensional vector representations of words have been extensively used for various natural language processing tasks. However, existing context-based word embeddings such as Word2vec and GloVe typically fail to capture sufficient sentiment information, which may result in words with similar vector representations having an opposite sentiment polarity (e.g., good and bad ), thus degrading sentiment analysis performance. To tackle this problem, recent studies have suggested learning sentiment embeddings to incorporate the sentiment polarity (positive and negative) information from labeled corpora. This study adopts another strategy to learn sentiment embeddings. Instead of creating a new word embedding from labeled corpora, we propose a word vector refinement model to refine existing pretrained word vectors using real-valued sentiment intensity scores provided by sentiment lexicons. The idea of the refinement model is to improve each word vector such that it can be closer in the lexicon to both semantically and sentimentally similar words (i.e., those with similar intensity scores) and further away from sentimentally dissimilar words (i.e., those with dissimilar intensity scores). An obvious advantage of the proposed method is that it can be applied to any pretrained word embeddings. In addition, the intensity scores can provide more fine-grained (real-valued) sentiment information than binary polarity labels to guide the refinement process. Experimental results show that the proposed refinement model can improve both conventional word embeddings and previously proposed sentiment embeddings for binary, ternary, and fine-grained sentiment classification on the SemEval and Stanford Sentiment Treebank datasets.

Refining Word Embeddings Using Intensity Scores for Sentiment Analysis

This study presents the adaptation of the Affective Norms for English Words (ANEW; Bradley & Lang, 1999a) for European Portuguese (EP). The EP adaptation of the ANEW was based on the affective ratings made by 958 college students who were EP native speakers. Subjects assessed about 60 words by considering the affective dimensions of valence, arousal, and dominance, using the Self-Assessment Manikin (SAM) in either a paper-and-pencil or a Web survey procedure. Results of the adaptation of the ANEW for EP are presented. Furthermore, the differences between EP, American (Bradley & Lang, 1999a), and Spanish (Redondo, Fraga, Padron, & Comesana, Behavior Research Methods, 39, 600–605, 2007) standardizations were explored. Results showed that the ANEW words were understood in a similar way by EP, American, and Spanish subjects, although some sex and cross-cultural differences were observed. The EP adaptation of the ANEW is shown to be a valid and useful tool that will allow researchers to control and/or manipulate the affective properties of stimuli, as well as to develop cross-linguistic studies. The normative values of EP adaptation of the ANEW can be downloaded at http://brm.psychonomic-journals.org/content/supplemental.

/pdf/the-adaptation-of-the-affective-norms-for-english-words-anew-4bpcpkxeha.pdf

The adaptation of the Affective Norms for English Words (ANEW) for European Portuguese

This study presents the results of the adaptation of the International Affective Picture System (IAPS) for European Portuguese (EP). Following the original procedure of Lang et al., 2000 native speakers of EP rated the 1,182 pictures of the last version of the IAPS set on the three affective dimensions of valence, arousal, and dominance, using the Self-Assessment Manikin (SAM). Results showed that the normative values of the IAPS for EP are properly distributed in the affective space of valence and arousal, showing the typical boomerang-shaped distribution observed in previous studies. Results also point to important differences in the way Portuguese females and males react to affective pictures that should be taken into consideration when planning and conducting research with Portuguese samples. Furthermore, the results from the cross-cultural comparisons between the EP ratings and the ratings from the American, Spanish, Brazilian, Belgian, Chilean, Indian, and Bosnian-Herzegovinian standardizations, showed that in spite of the fact that IAPS stimuli elicited affective responses that are similar across countries and cultures (at least in Western cultures), there are differences in the way Portuguese individuals react to IAPS pictures that strongly recommend the use of the normative values presented in this work. They can be downloaded as a supplemental archive at http://brm.psychonomic-journals.org/content/supplemental or at http://p-pal.di.uminho.pt/about/databases.

https://link.springer.com/content/pdf/10.3758%2Fs13428-014-0535-2.pdf

Adaptation of the International Affective Picture System (IAPS) for European Portuguese

In this study, we present the normative values of the adaptation of the International Affective Digitized Sounds (IADS-2; Bradley & Lang, 2007a) for European Portuguese (EP). The IADS-2 is a standardized database of 167 naturally occurring sounds that is widely used in the study of emotions. The sounds were rated by 300 college students who were native speakers of EP, in the three affective dimensions of valence, arousal, and dominance, by using the Self-Assessment Manikin (SAM). The aims of this adaptation were threefold: (1) to provide researchers with standardized and normatively rated affective sounds to be used with an EP population; (2) to investigate sex and cultural differences in the ratings of affective dimensions of auditory stimuli between EP and the American (Bradley & Lang, 2007a) and Spanish (Fernandez-Abascal et al., Psicothema 20:104–113 2008; Redondo, Fraga, Padron, & Pineiro, Behavior Research Methods 40:784–790 2008) standardizations; and (3) to promote research on auditory affective processing in Portugal. Our results indicated that the IADS-2 is a valid and useful database of digitized sounds for the study of emotions in a Portuguese context, allowing for comparisons of its results with those of other international studies that have used the same database for stimulus selection. The normative values of the EP adaptation of the IADS-2 database can be downloaded along with the online version of this article.

/pdf/affective-auditory-stimuli-adaptation-of-the-international-1v8vf4opej.pdf

Affective auditory stimuli: adaptation of the International Affective Digitized Sounds (IADS-2) for European Portuguese

In the present study, we present normative ratings of free association for 139 European Portuguese (EP) words among 7- to 8-, 9- to 10-, and 11- to 12-year-old children attending the 3rd, 5th, and 7th grades of elementary and middle school in Portugal. For each word, five indices are presented: (a) the percentage of associates, (b) the strength of the first associate, (c) the strength of the second associate, (d) the distance between the first and second associates, and (e) the percentage of idiosyncratic responses. Additionally, grade-level frequency values for each word from the ESCOLEX database (Soares et al., in press) are also provided. As expected, the results revealed developmental changes in the knowledge organization of the children, which occurred at the ages of 9–10 (5th grade) and remained stable in the 11- to 12-year-old children (7th grade). Specifically, we observed a decrease in the percentages of associates and idiosyncratic responses, as well as an increase in the strengths of the first and second associates from the 3rd to the 5th grade. Moreover, a comparative analysis with the previous work of Carneiro, Albuquerque, Fernandez, and Esteves (2004) on EP and Macizo, Gomez-Ariza, and Bajo (2000) on Spanish, for the subsets of common words (16 and 58, respectively), showed that the present norms fit well with previous EP data, but differ from the Spanish data.

Carla Sofia Frade

Papers

The adaptation of the Affective Norms for English Words (ANEW) for European Portuguese

Adaptation of the International Affective Picture System (IAPS) for European Portuguese

Affective auditory stimuli: adaptation of the International Affective Digitized Sounds (IADS-2) for European Portuguese

Free associate norms for 139 European Portuguese words for children from different age groups