Clarifying the linguistic signature: measuring personality from natural speech

In this study, we examined the viability of measuring personality using computerized lexical analysis of natural speech. Two well-validated models of personality were measured, one involving trait positive affectivity (PA) and negative affectivity (NA) dimensions and the other involving a separate behavioral inhibition motivational system (BIS) and a behavioral activation motivational system (BAS). Individuals with high levels of trait PA and sensitive BAS expressed high levels of positive emotion in their natural speech, whereas individuals with high levels of trait NA and sensitive BIS tended to express high levels of negative emotion. The personality variables accounted for almost a quarter of the variance in emotional expressivity.     

Examining the Differential Item Functioning of the Rosenberg Self‐Esteem Scale Across Eight Countries1

We examined the differential item functioning (DIF) of Rosenberg's (1965 ) Self‐Esteem Scale (RSES) and compared scores from U.S. participants with those from 7 other countries: Canada, Germany, New Zealand, Kenya, South Africa, Singapore, and Taiwan. Results indicate that DIF was present in all comparisons. Moreover, controlling for latent self‐esteem, participants from individualistic countries had an easier time reporting high self‐esteem on self‐competence‐related items, whereas participants from communal countries had an easier time reporting high self‐esteem on self‐liking items ( Tafarodi & Milne, 2002 ). After adjusting for DIF, we found larger mean self‐esteem differences between the countries than observed scores initially indicated. The suitability of the RSES, and the importance of examining DIF, for cross‐cultural research are discussed.     

Racial Identity Status Profiles and Expressions of Anger in Black Americans: An Exploratory Study

The relationship between Black adult racial identity status profiles and anger expression was examined. Two profiles, Undifferentiated and Immersion‐Emersion, emerged. A comparison of modes of anger expression revealed that the Immersion‐Emersion dominant profile was associated with higher scores on Anger‐Out and lower scores on Anger‐Control. Implications for research and counseling are discussed. Se examinó la relación entre los perfiles del estado de la identidad racial de individuos Negros adultos y la expresión de ira. Surgieron dos perfiles, indiferenciado e inmersión‐emersión. Una comparación de los modos de expresión de la ira reveló que el perfil dominante de Inmersión‐Emersión estaba asociado a registros más altos en Ira‐Fuera y más bajos en Control de la ira. Se discuten las implicaciones para la investigación y la consejería.     

Illusory line linking solid rods

A solid object--a frame enclosing rods--can be seen as having an illusory 'line' joining the tips of the rods.     

Incorporating Demographic Embeddings Into Language Understanding

Meaning depends on context. This applies in obvious cases like deictics or sarcasm as well as more subtle situations like framing or persuasion. One key aspect of this is the identity of the participants in an interaction. Our interpretation of an utterance shifts based on a variety of factors, including personal history, background knowledge, and our relationship to the source. While obviously an incomplete model of individual differences, demographic factors provide a useful starting point and allow us to capture some of this variance. However, the relevance of specific demographic factors varies between situations—where age might be the key factor in one context, ideology might dominate in another. To address this challenge, we introduce a method for combining demographics and context into situated demographic embeddings—mapping representations into a continuous geometric space appropriate for the given domain, showing the resulting representations to be functional and interpretable. We further demonstrate how to make use of related external data so as to apply this approach in low‐resource situations. Finally, we show how these representations can be incorporated into improve modeling of real‐world natural language understanding tasks, improving model performance and helping with issues of data sparsity.