Response Selection During a Joint Action Task

ABSTRACT

It has been suggested that a shared task representation is used to predict a coactors’ needs and that these predictions influence each coactor's response selection in a joint action task. The authors tested this idea using a joint action task in which participants passed a jug to a confederate under different conditions. They hypothesized that if participants predicted the needs of their coactor and planned their movement according to these predictions, the jug would be passed with the handle available to the confederate. Consistent with this hypothesis, the jug was passed with the handle available on 86% of the trials. This strategy may be adopted to improve the efficiency of the whole task as opposed to each individual's portion of the task.

Keywords:

• action prediction
• joint action
• response selection
• shared task representation

ACKNOWLEDGMENTS

The authors thank Kate O’Brien and Dovin Kearnan-Corrigall for assistance in data collection and analysis. This research was funded by grants and scholarships from the Ontario Ministry of Research and Innovation (Early Researcher Award), Alberta Ingenuity Fund, Lloyd and Florence Cooper Doctoral Scholarship in Integrated Medicine, and the Natural Science and Engineering Research Council of Canada.

Notes

2. A control experiment in which participants performed the entire place and pour tasks on their own (i.e., without passing the jug to a confederate) under nearly identical conditions revealed that the jug was grasped by the handle to complete the task on 84% of the trials. Consistent with the main experiment, 4 of the 8 participants used the handle to complete the task 100% of the time. Thus, it is clear that the preferable and most efficient method for performing the task when an individual completes the tasks alone is by grasping and manipulating the jug with the handle. Further, the results from the control study stand in stark contrast to those of the main experiment in which it was observed that participants grasped and manipulated the body or lip of the jug at the completion of their part of the task so that the confederate could use the handle.

2. Note that these differences were not tested using conventional statistical analyses (i.e., an analysis of variance [ANOVA]) because, even though we purposely employed a fully factorial design, the general absence of variability within and across individuals precluded the functional and meaningful use of an ANOVA or other parametric statistics.