Quantification and Psychological Indicators of Emotions: The Content Analysis of Speech and other Objective Measures of Psychological States

Abstract

Reliable and valid measurement of affects, emotions, and moods has posed a problem for psychiatric and psychophysiological research as the demand has grown for more sensitive, precise, and objective assessment methods than the method of clinical impressionistic evaluation. There are three major methods in current use for assessing these psychological variables: self-report scales, behavioral rating scales, and the content analysis of verbal behavior.

Self-report inventories give an individual an opportunity to describe his subjective state, and their major advantage is that what the person is actually experiencing may not be correctly perceived by external observers. Disadvantages of the self-report method include the possibility that the subject may malinger or fake or may not be in good communication with his own feelings so that he gives a distorted report about them.

Psychiatric rating scales have the advantage of putting a trained observer to the task of assessment, and the clinician rater has the option of using a broad range of behavioral and affective cues, verbal and nonverbal, in following this method. But since such raters are not free from systematic distortion and thorough familiarity with the subject of observation is infrequent, all relevant information to make a valid assessment is often not available. Moreover, different interviewers may evoke varying emotional responses from the same person.

Objective content analysis of verbal behavior can avoid most of the shortcomings of the self-report and observer rating methods, so long as reasonably standardized procedures are used for eliciting verbal behavior and other key features of scientific methodology are followed. A disadvantage of the content analysis method is that it is time-consuming and requires training and quality checks to carry out accurate content analysis coding. On the other hand, reliable and valid measurement procedures in all fields of research take time and care.

A brief review is provided of the variety of findings and applications of the content analyses method of measuring feeling states, and these applications include the research areas of psychotherapy, psychophysiology, and neuropsycho-pharmacology.