Aplicação do escalonamento multidimensional ao julgamento e classificação de conceitos emocionais

Abstract

Emotions and sentiments are present in every situation of daily life. In Psychology the study of emotional aspects has a distinct place, even though they may present themselves as being complex and of difficult delimitation. The objective of this study has been to investigate the aspects relative to categorizing and the judgment of conceptual similarities using words referring to emotions and sentiments. The study was subdivided into two tasks. In the first one, 206 words were selected from the Portuguese language referring to emotions and sentiments. The same task consisted in categorizing these words into predefined emotional categories: happiness, fear disgust, anger, surprise, sadness as well as a category that did not describe an emotion. After the establishment of a minimum criteria for the inclusion of a word into a category, (70% accordance among the participants) the results showed that the participants distributed the words among the categories and those labeled as happiness and sadness were the ones which were chosen with the highest frequency. Taking into consideration the words that exceeded the criteria for inclusion into each category of the first task, in the second task , the objective was to study the similarity relationship among the words, using for this part, the Multidimensional Scaling. The application of this statistical technique gave a special configuration to the words, indicating possible grouping of words that constituted the categories of happiness and sadness. The results gave a graphic representation as to how the items are organized within each category and indicated characteristics of similarities and discrepancies among them, these characteristics not usually being the same ones when compared to the word meaning attributed to them in the dictionary. It was observed at the end of the project that the initial and main objective were reached and suggesting the realization of further studies that would add issues that were not investigated here.