Alle Publikationen
- <<
- <
- 1
2018
-
(2018): The effects of humanlike and robot-specific affective nonverbal behavior on perception, emotion, and behavior (PSYNDEXshort). In: International Journal of Social Robotics 10 (5), S. 569-582. DOI: 10.1007/s12369-018-0466-7
DOI: http://search.ebscohost.com/login.aspx?direct=true&db=pdx&AN=0350667&site=ehost-live Abstract: Research demonstrated that humans are able to interpret humanlike (affective) nonverbal behavior (HNB) in artificial entities (e.g. Beck et al., in: Proceedings of the 19th IEEE international symposium on robot and human interactive communication, IEEE Press, Piscataway, 2010. 10.1109/ROMAN.2010.5598649; Bente et al. in J Nonverbal Behav 25: 151-166, 2001; Mumm and Mutlu, in: Proceedings of the 6th international conference on human-robot interaction, HRI. ACM Press, New York, 2011. 10.1145/1957656.1957786). However, some robots lack the possibility to produce HNB. Using robot-specific nonverbal behavior (RNB) such as different eye colors to convey emotional meaning might be a fruitful mechanism to enhance HRI experiences, but it is unclear whether RNB is as effective as HNB. We present a review on affective nonverbal behaviors in robots and an experimental study. We experimentally tested the influence of HNB and RNB (colored LEDs) on users' perception of the robot (e.g. likeability, animacy), their emotional experience, and self-disclosure. In a between-subjects design, users interacted with either (a) a robot displaying no nonverbal behavior, (b) a robot displaying affective RNB, (c) a robot displaying affective HNB or (d) a robot displaying affective HNB and RNB. Results show that HNB, but not RNB, has a significant effect on the perceived animacy of the robot, participants' emotional state, and self-disclosure. However, RNB still slightly influenced participants' perception, emotion, and behavior: Planned contrasts revealed having any type of nonverbal behavior significantly increased perceived animacy, positive affect, and self-disclosure. Moreover, observed linear trends indicate that the effects increased with the addition of nonverbal behaviors (control< RNB< HNB). In combination, our results suggest that HNB is more effective in transporting the robot's communicative message than RNB. (c) Springer Science+Business Media B.V. Die Forschung zeigte, dass der Mensch in der Lage ist, menschliches (affektives) nonverbales Verhalten (HNB) in künstlichen Einheiten (z.B. Beck et al., in: Proceedings of the 19th IEEE international symposium on robot and human interactive communication, IEEE Press, Piscataway, 2010. 10.1109/ROMAN.2010.5598649; Bente et al. in J Nonverbales Verhalten 25: 151-166, 2001; Mumm und Mutlu, in: Proceedings der 6. Internationalen Konferenz über Mensch-Roboter-Interaktion, HRI. ACM Press, New York, 2011. 10.1145/1957656.1957786). Einige Roboter haben jedoch nicht die Möglichkeit, HNB zu produzieren. Die Verwendung von roboterspezifischem nonverbalem Verhalten (RNB), wie z.B. verschiedenen Augenfarben, um emotionale Bedeutung zu vermitteln, könnte ein fruchtbarer Mechanismus sein, um HRI-Erlebnisse zu verbessern, aber es ist unklar, ob RNB so effektiv ist wie HNB. Wir präsentieren einen Überblick über affektives nonverbales Verhalten in Robotern und eine experimentelle Studie. Wir haben experimentell den Einfluss von HNB und RNB (farbige LEDs) auf die Wahrnehmung des Roboters durch die Nutzer (z.B. Sympathie, Animosität), ihre emotionale Erfahrung und Selbstdarstellung getestet. In einem Design zwischen den Probanden interagierten die Benutzer entweder mit (a) einem Roboter, der kein nonverbales Verhalten zeigt, (b) einem Roboter, der affektives RNB anzeigt, (translated by DeepL)
Keywords: Emotion & Roboter, Social robotic 2014
-
(2014) : Lässt sich Empathie simulieren? Ansätze zur Erkennung und Generierung empathischer Reaktionen anhand von Computermodellen In: Güntürkün, Onur; Hacker, Jörg (Hg.): Geist - Gehirn - Genom - Gesellschaft. Wie wurde ich zu der Person, die ich bin?. Vorträge anlässlich der Jahresversammlung vom 20. bis 22. September 2013 in Halle (Saale): Stuttgart: wvg Wissenschaftliche Verlagsgesellschaft (Nova Acta Leopoldina, Neue Folge 120, Nr. 405), S. 81-105. Online verfügbar unter http://search.ebscohost.com/login.aspx?direct=true&db=pdx&AN=0296791&site=ehost-live
Abstract: Ansätze zur Analyse und Nachbildung empathischen Verhaltens mit technischen Systemen werden skizziert. Bislang wurden unterschiedlich komplexe empathische Prozesse untersucht - von der rein reaktiven Spiegelung von Emotionen des menschlichen Gegenübers über das Verständnis der Ursachen von emotionalen Zuständen bis hin zur aktiven Stimulation emotionaler Prozesse. Anhand ausgewählter Beispiele aus dem Bereich virtueller Charaktere und Roboter, die dem Menschen als anthropomorphe Interaktionspartner entgegentreten, wird ein Überblick zum aktuellen Stand der Forschung gegeben, wobei der Fokus auf Techniken zur Modellierung, Erkennung und Generierung von Emotionen sowie der Simulation von empathischem Verhalten liegt. Wachsendes Interesse an diesen Forschungsgebieten lässt erwarten, dass interaktive Artefakte wie etwa virtuelle Charaktere und Roboter künftig mit verfeinerten empathischen Fähigkeiten ausgestattet werden können und damit einem menschlichen Gegenüber zunehmend die Illusion von mitfühlenden Individuen vermitteln. Auf dieser Grundlage eröffnen sich interessante Optionen für die Ausgestaltung der Mensch-Maschine-Interaktion sowie für gänzlich neue pädagogische und therapeutische Einsatzfelder. Die Frage, ob virtuelle Charaktere und Roboter jemals echte Gefühle werden empfinden können und inwiefern dies erstrebenswert wäre, ist damit ebenso wenig beantwortet, wie Fragen nach den möglichen sozio-ökonomischen Auswirkungen, die eine mit emotionalen Maschinen durchdrungene Alltagswelt langfristig mit sich brächte - für den Einzelnen und eine Gesellschaft insgesamt.
Keywords: Emotion & Roboter, Social robotic 2013
-
(2013) : "Emotional" robots and agents - Implementation of emotions in artificial entities In: Mohiyeddini, Changiz; Eysenck, Michael; Bauer, Stephanie (Hg.): Handbook of psychology of emotions. Recent theoretical perspectives and novel empirical findings. Volume 2: New York: Nova Science Publishers, S. 277-295. Online verfügbar unter http://search.ebscohost.com/login.aspx?direct=true&db=pdx&AN=0279840&site=ehost-live
Abstract: Summarizes research on emotions in the context of robot and virtual-agent development. First, several motives and reasons for the implementation of emotions in artificial entities are presented. Then, a choice of robot and agent systems that model emotions based on various theories and assumptions are described. However, the common practice to implement emotions is also critically reflected against the background of theories on the relation of emotions and facial expressions. Based on this, alternative approaches are presented which implement a theory-of-mind module instead of emotions. In a second part, humans' emotional reactions to systems with implemented emotions and corresponding facial expressions are described.
Keywords: Emotion & Roboter, Social robotic -
: On the profoundness and preconditions of social responses towards social robots : Experimental investigations using indirect measurement techniques. Online verfügbar unter http://search.ebscohost.com/login.aspx?direct=true&db=pdx&AN=0281635&site=ehost-live.
Abstract: Examined the profoundness as well as relevant preconditions of automatic social responses towards nonhuman agents using indirect measurement techniques. To this end, a series of four studies were conducted. The first two studies utilized the social facilitation paradigm and were conducted with a total of 170 participants (mean age 23 years). In a mere-presence design the social influence of a present robot was assessed via performance-based measures on cognitive and motor tasks. The mere presence of an active social robot able to elicit social facilitation effects, and the performance effects were also comparable to those of human presence. Further, these effects even sustained when the robot was obviously switched off (Study 2). To further examine the boundary conditions of these social reactions and to address the roles of unfamiliarity, novelty, and human-like appearance on a microlevel, two subsequent studies were executed that utilized the facial mimicry paradigm. Results of the third study, conducted with 22 women (mean age 25 years) revealed that participants produced significantly fewer corresponding facial mimicry responses towards robotic faces compared to either human or schematic face types, and that this was especially apparent for happy facial expressions. The final study examined whether the previous differences between facial mimicry responses towards human and robotic faces could be mitigated by familiarization as well as expression intensity of the happy faces. A total of 146 adults (mean age 25 years) participated in Study 4. Indeed, real-world familiarization through actual interaction with a social robot combined with more pronounced smiles of the robotic faces evoked comparable mimicry responses towards robotic as towards human faces. Stimulus ambivalence is discussed as potent moderator of the obtained results, both in the social facilitation as well as the facial mimicry paradigm. Taken together, on a behavioral macrolevel, as the social facilitation studies showed that a face-like configuration or other human-like appearance cues apparently sufficed as fundamental initiation for social responses towards nonhuman entities; however, looking at low-level reactions, human likeness alone was not sufficient for the emergence of congruent facial mimicry responses. The implications of the results for actual application of social robots are discussed.
Keywords: Emotion & Roboter, Social robotic
- <<
- <
- 1