Past research has found that intervention with Paro, a seal robot, on healthy adults can improve psychophysiological responses (Mitsui et al. 2001a,b, Kawaguchi et al. 2012). According to Bethel & Murphy (2010), psychophysiological study focuses on understanding human emotion and behaviour during an event through observation, physiological and psychological measures. However, these previous studies of Paro were limited by small participant sizes (less than 10 participants), and did not investigate the effectiveness of Paro in reducing stress such as environmental stress. Furthermore, little is known about the circumstances in which interacting with Paro can help to reduce stress as indicated by changes in psychophysiological responses.
Three studies were conducted in laboratory settings, in order to limit any extraneous variables that can affect the psychophysiological responses. This thesis aims to identify factors that can influence psychophysiological responses to short-term interaction with a Paro in the healthy adult population. It also aims to understand how the factors affect the effectiveness of Paro intervention in reducing effects of stress. The psychophysiological responses were measured using self-reports, physiological sensors (skin conductance responses, heart rate, and heart rate variability) and video recording. The stress used in this thesis is induced environmental noise.
A study was conducted (n = 76) to explore which features of Paro (active and inactive) and of human behaviour towards the robot (talking and stroking) were responsible for any effects it creates. It was found that interacting with Paro aroused skin conductance responses and its presence during the intervention increased positive moods. At the same time, the positive moods influenced the effectiveness of the Paro intervention in reducing stress.
A subsequent study (n = 104) compared the effects of interacting with an active Paro with the effects of stroking a furry bolster, and explored the stress reducing effect of Paro intervention as a pre-stress treatment. The novelty effect of Paro in the intervention was also investigated. It was found that the Paro intervention was not enough to minimize and reduce the impact of induced stress. However, there was a trend that suggests the Paro intervention has more impact in lessening the stress than the bolster intervention as indicated by improvements in physiological responses (such as heart rate and heart rate variability). At the same time, it was found that novelty affected how participants interacted with the Paro (based on video observation during the intervention), but did not change the positive effect of the Paro intervention.
A final study (n = 104) explored two intervention formats: either Paro or bolster as preand post-stress intervention. This study also explored the effectiveness of Paro or bolster intervention using a mathematical task. It was found that the Paro intervention was more suitable as a post-stress intervention, as suggested by improvement in physiological responses (heart rate and heart rate variability). However, no improvement in math performance was found in either Paro or bolster interventions.
Additionally, in this study and Study 2, the bolster intervention showed the importance of
stroking soft fur in reducing stress and negative moods.
There are five key findings of this thesis: Large samples of data were obtained from (i)
healthy young adults (n = 284). Paro intervention is more effective as (ii) post-stress treatment, in terms of reducing effects of (iii) induced environmental stress in (iv) healthy adults. The thesis also found (v) factors that influenced the changes in psychophysiological responses and the effectiveness of Paro interventions such as stressors and intervention formats.
