Affective synesthesia: Extracting emotion space from simple perceptual stimuli

Several experiments and reanalyses of archival data were performed in which sets of common emotions were mapped onto visual and auditory perceptual stimuli of varying degrees of complexity. Subjects showed reliability across several different experimental paradigms and diverse stimulus sets. Furthermore, the results were reducible to a two-dimensional emotion space of valence and activity, after the work of Osgood, Succi, and Tannenbaum (1957). The perceptual locus of an aesthetic stimulus in this two-dimensional space is hypothesized to provide a first approximation to how the synesthetic qualities of a percept provide a basis for emotional responses to such stimuli. More generally, this direct mapping of percept onto emotion provides an interesting conundrum for theories of emotion research.The author acknowledges support for this research from the ADAMA training grant, MH14257, Quantitative Methods for Behavior Research, University of Illinois, Department of Psychology. The work originated in conversations with Gerry Clore, whose encouragement and ideas were invaluable. Research assistant Kaarin Ryan provided invaluable help and ideas in the initial stages of this research. This paper is dedicated to Ned, Jeff, and Gordon.     

Affective shifts: mood,emotion and well-being

Synthese - It is a familiar feature of our affective psychology that our moods ‘crystalize’ into emotions, and that our emotions ‘diffuse’ into moods. Providing a detailed...     

Sex-related patient and therapist effects on clinical judgment

The purpose of this investigation was to submit the claim of sex-related bias in clinical judgment to more thorough empirical testing than had been the case in previous research. Patient sex and stereotypic sex-role were factorially varied in a clinical protocol that was sent with related materials to 640 sex-stratified, randomly selected Members and Fellows of APA Division 29 (Psychotherapy). Approximately 30% (N=182) of those contacted returned usable data, including information about their own sex-role traditionalism and evaluation of the hypothetical patients' psychological well-being. Results confirmed the effectiveness of the stereotypic sex-role manipulation. However, no consistent effects were found for any of the four variables of interest considered in isolation or in combination, thereby refuting allegations of covert sex-related discrimination perpetrated under the aegis of psychological appraisal. Favorableness of clinical impressions was not mediated by therapists' self-reported attitudes toward the patients. The study's generally unexpected outcome was tentatively attributed to enhanced professional sensitization to the sociocultural barriers to full psychological functioning in women.This report is based on a master's thesis written by the first author under the sponsorship of the second and submitted in partial fulfillment of the requirements for the MA degree at Vanderbilt University, 1974. The research was supported in part by the Vanderbilt University Research Council. The contributions of Stephen Amira and Joseph M. Schwartz are gratefully acknowledged.     

Episodic and semantic knowledge in emotional self-report: evidence for two judgment processes

Three studies involving 3 participant samples (Ns = 39, 55, and 53) tested the hypothesis that people retrieve episodic emotion knowledge when reporting on their emotions over short (e.g., last few hours) time frames, but that they retrieve semantic emotion knowledge when reporting on their emotions over long (e.g., last few months) time frames. Support for 2 distinct judgment strategies was based on judgment latencies (Studies 1 and 2) and priming paradigms (Studies 2 and 3). The authors suggest that self-reports of emotion over short versus long time frames assess qualitatively different sources of self-knowledge.     

Affective guidance of intelligent agents: How emotion controls cognition

How do emotions and moods color cognition? In this article, we examine how such reactions influence both judgments and cognitive performance. We argue that many affective influences are due, not to affective reactions themselves, but to the information they carry about value. The specific kind of influence that occurs depends on the focus of the agent at the time. When making evaluative judgments, for example, an agent’s positive affect may emerge as a positive attitude toward a person or object. But when an agent focuses on a cognitive task, positive affect may act like feedback about the value of one’s approach. As a result, positive affect tends to promote cognitive, relational processes, whereas negative affect tends to inhibit relational processing, resulting in more perceptual, stimulus-specific processing. As a consequence, many textbook phenomena from cognitive psychology occur readily in happy moods, but are inhibited or even absent in sad moods (149).     

Affective judgment in spatial context: Orienting within physical spaces containing people and things

This article reviews our experimental work about affective judgment in spatial context. This type of judgment serves to regulate one's distance toward people and things in physical space. The main idea is that orienting within physical space requires not only knowing where places are but also how places feel. This, in turn, depends on the influence of people and things contained in space on one's affective appraisal of the surroundings. Based on fundamental principles of social cognition, affective judgment in spatial context combines people's beliefs about how influence unfolds into the surroundings with comparison, categorization and information integration processes. Out comes a subjective affective representation of physical space that is cognitively coherent within a given spatial frame of reference. I review our work according to main topics and discuss four possible directions for future research.     

Strategies actually employed during response-focused emotion regulation research: Affective and physiological consequences

Addressing internal validity concerns in emotion regulation research, the present experiment was primarily designed to determine whether research participants are compliant when asked to use a response-focused strategy during emotional film viewing or whether these individuals incorporate the use of antecedent strategies. The influence of antecedent vs. response-focused strategy use on self-reported affect, physiological, and behavioural data were additionally investigated. A total of 82 healthy undergraduate participants were asked to use one of two response-focused emotion regulation techniques—suppression or exaggeration—while watching a 2 minute positive or negative movie. Following the movie, participants self-reported their affective response to the film, described how they tried to suppress or exaggerate their reaction (i.e., strategies used to regulate their response), and estimated the percentage of time they used each strategy. Representing “antecedent” and “response-focused” techniques, the strategies reported by participants were coded as “cognitive” or “muscular” in nature. Relative to exaggerators, participants in the suppression condition were significantly more likely to self-report using an antecedent (cognitive) strategy for at least some portion of the film (65% vs. 38%). During the suppression condition, greater use of antecedent strategies did not influence sympathetic reactivity to either movie but did result in significantly less self-reported negative affect to the negative movie.     

Worry: daily self-report in clinical and non-clinical populations

The present study establishes an ecologically-valid point of reference for generalized anxiety disorder (GAD) treatment outcome studies. Although worry is the main diagnostic criterion for GAD, it is also a common and everyday cognition. Few studies have investigated the normal manifestation of worry in a non-clinical population. For this research, time spent worrying and the tendency to worry were measured in a non-clinical sample (n = 36), where GAD was screened-out, and results were compared to those obtained from a GAD sample (n = 60). Participants recorded the time spent worrying in self-monitoring notebooks for two consecutive weeks and completed the French version of the Penn State Worry Questionnaire (PSWQ). It was found that the non-GAD group reported less time spent worrying on a daily basis as well as a lower tendency to worry than the GAD group. Results are discussed in terms of treatment implications.     

Framing social judgment: self-ingroup comparison and perceived discrimination

Members of disadvantaged groups often report that they are less vulnerable to discrimination than is the average member of their group. In two experiments, we examined how the framing of self-ingroup comparisons moderates this phenomenon. In Experiment 1, participants estimated the relative likelihood that either they, compared to the average member of their ingroup, would experience discrimination or that the average member of their ingroup, compared to themselves, would experience discrimination. In Experiment 2, a direct manipulation of self-ingroup similarity was added to this framing manipulation. Across the experiments, the results demonstrated that conditions highlighting differences between the individual and the group led participants to perceive themselves as much less vulnerable to discrimination than their group, relative to conditions highlighting similarities between the individual and the group.     

A comparison of psychiatric patients' self-report with other sources of clinical information

This study compared self-report (MMPI item responses) of psychiatric patients who were experiencing different crises situations with other sources of clinical information. The results suggest that in a large number of instances item responses communicate information which is in agreement with what was known about the patients form other sources of information. The content of item responses was consistent with interpretations from several empirical measures of the Ss' test performance.     

Conceptual and methodological issues in the judgment of facial expressions of emotion

In two studies, subjects judged a set of facial expressions of emotion by either providing labels of their own choice to describe the stimuli (free-choice condition), choosing a label from a list of emotion words, or choosing a story from a list of emotion stories (fixed-choice conditions). In the free-choice condition, levels of agreement between subjects on the predicted emotion categories for six basic emotions were significantly greater than chance levels, and comparable to those shown in fixed-choice studies. As predicted, there was little to no agreement on a verbal label for contempt. Agreement on contempt was greatly improved when subjects were allowed to identify the expression in terms of an antecedent event for that emotion rather than in terms of a single verbal label, a finding that could not be attributed to the methodological artifact of exclusion in a fixed-choice paradigm. These findings support two conclusions: (1) that the labels used in fixed-choice paradigms accurately reflect the verbal categories people use when free labeling facial expressions of emotion, and (2) that lexically ambiguous emotions, such as contempt, are understood in terms of their situational meanings.This research was supported in part by a Research Scientist Award from the National Institute of Mental Health (MH 06091) to Paul Ekman.     

Establishment and replication of emotion regulation profiles: implications for psychological health

Background and Objectives: A promising method of capturing the complex nature of emotion regulation is to assess composite profiles of regulation (i.e., default pattern of regulation across multiple strategies). However, it remains unclear whether regulatory profiles demonstrate consistency across samples and in relation to mental health. Design: Two studies are presented here. Both utilized a cross-sectional design, and the second study presents a replication of the first. Method: Both studies utilized self-report data from independent undergraduate samples to perform latent profile analyses of emotion regulation use. Results: Studies 1 and 2 demonstrated evidence for four replicable regulatory profiles: Adaptive, Accepting (with or without Suppression), Non-accepting, and Maladaptive. Profiles were also related to symptoms of depression and anxiety, such that those consistently relying on adaptive strategies reported lower symptoms than those relying more heavily on maladaptive strategies. Conclusions: These findings clarify previous work which tied regulatory profiles to psychological health by extending a person-centered approach to understanding the ways in which individuals regulate their emotions.     

Nonverbal emotion displays, communication modality, and the judgment of personality

The goal of this research was to investigate the impact of nonverbal expressive cues on the attribution of the Big Five personality traits. Expressive cues of fear, disgust, happiness, and sadness were elicited in a sample of 22 encoders while watching films, narrating, and posing. Encoders’ personalities were rated by themselves and unacquainted raters who watched the encoders, and blind judges rated the traits of a typical student. Expressive cues influenced the raters’ attribution of personality, but this influence was weakest when the encoders expressed happiness (vs. negative emotions) and when they were narrating an emotional experience (when the cues were least potent). Negative and strong expressive cues interfered with the application of a normative, and more accurate, judgment strategy.     

Assessment of DSM-IV personality disorders in obsessive-compulsive disorder: comparison of clinical diagnosis, self-report questionnaire, and semi-structured interview

In patients with obsessive-compulsive disorder, personality disorders are not many times assessed according to DSM-IV criteria. The purpose of the present study is to examine the prevalence of personality disorders diagnosed according to the DSM-IV in a severely disordered OCD population (n=65) with three different methods of assessing personality disorders (structured interview, questionnaire, and clinical diagnoses). Furthermore, correspondence between these different methods was investigated and their construct validity was examined by relating the three methods to external variables. Each method resulted in a predominance of Cluster C personality disorders, and obsessive-compulsive personality disorder had the highest prevalence. However, there was generally low correspondence regarding which patient had which personality disorder. Results concerning the relation of external variables were the most promising for the structured clinical interview.     

Assessing self-report of pain: A comparison of two recording procedures

Intensive vs. reduced demand self-monitored pain levels were compared in nine pain patients. Each patient self-monitored pain intensity on a 6-point scale every 2 hr. Overall, daily mean and maximum pain levels were calculated for each patient (1) across all intervals (intensive self-monitoring) and (2) four times per day, at meal times and bedtime (reduced-demand self-monitoring). Differences were found to be negligible. The reduced-demand procedure produced data that were closely representative of pain fluctuations indicated in the more intensive bihourly ratings. Results are discussed in terms of the representativeness and standardization in pain self-monitoring as well as possible enhancement of recording compliance through the use of the less demanding four times per day procedure.Portions of these data were presented at the meeting of the Association for Behavior Analysis, Dearborn, June 1979.     

The assessment of emotional intelligence: a comparison of performance-based and self-report methodologies

We assessed the patterns of convergent validity for the Mayer-Salovey-Caruso Emotional Intelligence Test (Mayer, Salovey, & Caruso, 2002), a performance-based measure of emotional intelligence (EI) that entails presenting problems thought to have correct responses, and a self-report measure of EI (Schutte et al., 1998). The relations between EI and demographic characteristics of a diverse community sample (N = 223) concurred with previous research. However, the performance-based and self-report scales were not related to one another. Only self-reported EI scores showed a consistent pattern of relations with self-reported coping styles and depressive affect, whereas the performance-based measure demonstrated stronger relations with age, education, and receiving psychotherapy. We discuss implications for the validity of these measures and their utility.     

The distinctiveness of emotion concepts: a comparison between emotion, abstract, and concrete words

Are the concepts represented by emotion words different from abstract words in memory? We examined the distinct characteristics of emotion concepts in 3 separate experiments. The first demonstrated that emotion words are better recalled than both concrete and abstract words in a free recall task. In the second experiment, ratings of abstract, concrete, and emotion words were compared on concreteness, imageability, and context availability scales. Results revealed a difference between all 3 word types on each of the 3 scales. The third experiment investigated priming in a lexical decision task for homogeneous (abstract-abstract and emotion-emotion) and heterogeneous (abstract-emotion and emotion-abstract) associated word pairs. Priming occurred only for the homogeneous and heterogeneous abstract-emotion word pair conditions. Possible explanations for these findings are discussed in terms of the circumplex, hierarchical, and semantic activation models. The results are most consistent with the predictions of the semantic activation model.