Fig 1.
An example of the experimental condition is shown on the left, with the control condition on the right. A critical instruction to accompany this display would be “nudge the large present one slot up”. The only difference between the experimental condition and the control condition is that the experimental condition contains a “distractor” (in this case, the largest present from participants’ view), which competes with a “target” (in this case, the largest present from director’s view) to be the best-fitting referent for the director’s critical instructions. In the control condition, the distractor is replaced by an “irrelevant object” (in this case, the barometer from participants’ view), which does not compete with target to be the best fitting referent.
Table 1.
Instruction wording in Experiments 1 and 2.
All instructions in quotation marks were spoken, contents in parenthesis were acted out by an experimenter. The only difference between various conditions was the example given on Slide 3.
Table 2.
Descriptive statistics for Experiments 1 and 2.
Table 3.
Summary of mixed models from Experiment 1.
Fig 2.
Pirate plot for the percentage egocentric errors from Experiment 1.
Each circle represents the mean percentage egocentric error for a participant. The bold horizontal lines correspond to the condition means, the light-coloured bands around the means correspond to the confidence intervals.
Table 4.
Summary of mixed models from E2.
Fig 3.
Pirate plot for the percentage egocentric errors from Experiment 2.
Each circle represents the mean percentage egocentric error for a participant. The bold horizontal lines correspond to the condition means, the light-coloured bands around the means correspond to the confidence intervals.