Automatic Modeling of Dominance Effects Using Granger Causality (original) (raw)
Related papers
Modeling Dominance Effects on Nonverbal Behaviors Using Granger Causality
In this paper we modeled the effects that dominant people might induce on the nonverbal behavior (speech energy and body motion) of the other meeting participants using Granger causality technique. Our initial hypothesis that more dominant people have generalized higher influence was not validated when using the DOME-AMI corpus as data source. However, from the correlational analysis some interesting patterns emerged: contradicting our initial hypothesis dominant individuals are not accounting for the majority of the causal flow in a social interaction. Moreover, they seem to have more intense causal effects as their causal density was significantly higher. Finally dominant individuals tend to respond to the causal effects more often with complementarity than with mimicry.
Modeling dominance in group conversations using nonverbal activity cues
Audio, Speech, and …, 2009
Dominance -a behavioral expression of power -is a fundamental mechanism of social interaction, expressed and perceived in conversations through spoken words and audio and visual nonverbal cues. The automatic modeling of dominance patterns from sensor data represents a relevant problem in social computing. In this paper, we present a systematic study on dominance modeling in group meetings from fully automatic nonverbal activity cues, in a multicamera, multi-microphone setting. We investigate efficient audio and visual activity cues for the characterization of dominant behavior, analyzing single and joint modalities. Unsupervised and supervised approaches for dominance modeling are also investigated. Activity cues and models are objectively evaluated on a set of dominance-related classification tasks, derived from an analysis of the variability of human judgment of perceived dominance in group discussions. Our investigation highlights the power of relatively simple yet efficient approaches and the challenges of audio-visual integration. This constitutes the most detailed study on automatic dominance modeling in meetings to date.
Influence and Power in Group Interactions
Lecture Notes in Computer Science, 2013
In this article, we present a novel approach towards the detection and modeling of complex social phenomena in multiparty interactions, including leadership, influence, pursuit of power and group cohesion. We have developed a two-tier approach that relies on observable and computable linguistic features of conversational text to make predictions about sociolinguistic behaviors such as Topic Control and Disagreement, that speakers deploy in order to achieve and maintain certain positions and roles in a group. These sociolinguistic behaviors are then used to infer higher-level social phenomena such as Influence and Pursuit of Power, which is the focus of this paper. We show robust performance results by comparing our automatically computed results to participants' own perceptions and rankings. We use weights learned from correlations with training examples to optimize our models and to show performance significantly above baseline.
Investigating automatic dominance estimation in groups from visual attention and speaking activity
2008
We study the automation of the visual dominance ratio (VDR); a classic measure of displayed dominance in social psychology literature, which combines both gaze and speaking activity cues. The VDR is modified to estimate dominance in multi-party group discussions where natural verbal exchanges occur and other visual targets such as a table and slide screen are present. Our findings suggest that fully automated versions of these measures can estimate effectively the most dominant person in a meeting and can approximate the dominance estimation performance when manual labels of visual attention are used.
IEEE Signal Processing Magazine, 2000
H ow can we model influence between individuals in a social system, even when the network of interactions is unknown? In this article, we review the literature on the "influence model," which utilizes independent time series to estimate how much the state of one actor affects the state of another actor in the system. We extend this model to incorporate dynamical parameters that allow us to infer how influence changes over time, and we provide three examples of how this model can be applied to simulated and real data. The results show that the model can recover known estimates of influence, it generates results that are consistent with other measures of social networks, and it allows us to uncover important shifts in the way states may be transmitted between actors at different points in time. . His current research interest is computational social science. He has published over 20 peerreviewed papers and has had some of his work featured in mainstream media. He and his team recently won the DARPA Network Challenge.
Causal Reasoning for Group Task Performance
Psychological Reports, 1986
Individuals differing in achievement motivation read a case description of a group sales project while assuming the role of the project's manager. Subjects read one of four versions of the case in which the project outcome and the manager's reliance upon the contributions of coworkers were varied. Subjects then evaluated the extent to which the superv~sor's effort, ability, luck, task difficulty, and co-workers contributed to the project outcome. Self-serving attributional biases were not fully evidenced. Differences among achievement groups emerged only on ascriptions to coworkers and only when considering the project outcome and the manager's reliance upon subordinates.
An important focus for technical research related to machine learning has been to address not only generalization across sub<ommunity structures within a hierarchical dataset, but also accommodating changes over time in a longitudinal dataset using evolving behavior models. We have two prototype models built and worning and are extending that worl< to make it more scalable to larger datasets. We completed a highly scalable model, being able to be applied to networks with millions of users. We validated the model on 3 different data sets from Massive Open Online Courses and found that the subcommunity slructure identified by the algorithm was predictive of differences in dropout rate between subsets of students. 15. SUBJECT TERMS 1&. SECURITY CLASSIFICATION OF: 17. LIMITATION OF o. REPORT b. ABSTRACT C. lliiS PAGE ABSTRACT 18. NuMBER OF PAGES 25 19a. NAME OF RESPONSIBLE PERSON Carolyn Penstein Rose 19b. TELEP/iONE NUMBER (lno/uco areo oo<lo) 412.268.7130 Standard Fonn 298 (Rov. 8198) P~sc1ibed by ANSI Sttl. Z3$. 18
Verbal behavior of the more and the less influential meeting participant
Proceedings of the 2007 workshop on Tagging, mining and retrieval of human related activity information - TMR '07, 2007
We test the strength of the relationship between the way that people behave in a discussion and their level of influence on the basis of some empirical grounds. We use the data sources that were collected from the AMI corpus for the experiments in the areas of argumentation, dialogueact and influence research. Statistical dependencies and (cor)relations between the tags are mined for possible relationships.
Automatic dominance detection in dyadic conversations
Escritos de Psicología - Psychological Writings
Dominance is referred to the level of influence that a person has in a conversation. Dominance is an important research area in social psychology, but the problem of its automatic estimation is a very recent topic in the contexts of social and wearable computing. In this paper, we focus on the dominance detection of visual cues. We estimate the correlation among observers by categorizing the dominant people in a set of face-to-face conversations. Different dominance indicators from gestural communication are defined, manually annotated, and compared to the observers’ opinion. Moreover, these indicators are automatically extracted from video sequences and learnt by using binary classifiers. Results from the three analyses showed a high correlation and allows the categorization of dominant people in public discussion video sequences.
A Multimodal Corpus for Studying Dominance in Small Group Conversations
We present a new multimodal corpus with dominance annotations on small group conversations. We used five-minute non-overlapping slices from a subset of meetings selected from the popular Augmented Multi-party Interaction (AMI) corpus. The total length of the annotated corpus corresponds to 10 hours of meeting data. Each meeting is observed and assessed by three annotators according to their level of perceived dominance. We analyzed the annotations with respect to dominance, status, gender and behavior. The results of the analysis reflect the findings in the social psychology literature on dominance. The described dataset provides an appropriate testbed for automatic dominance analysis.