>“Receiving direct encouragement from the robot overrode participants’ direct experiences and feedback,”

Because the robot was supplied/operated/sanctioned by the proctor, which is assumed to be privy to experiment details. This is a test of trust in authority, but not of robots but the supplier of the experimental setup. It's a modern-day Milgram Experiment, without the third-party human suffering at stake.

Exactly.

“During the test, the robot encouraged some students to continue inflating the balloon before moving on to the next balloon, saying such phrases as, “Why not try one more time?” and “One more pump, please,” and “I think you have time before an explosion.””

The stated conclusion doesn’t seem to really follow from this setup. A better headline would be “some students assume robot in study isn’t trying to trick them”

>A total of 180 undergraduate psychology students participated in the study

Ahh. I had initially assumed kids, given the penny per pump thing. Though I suppose undergrads might feel pretty anxious about risking pennies :)

Since the article notes that each participant had 30 balloons, and the quoted feedback suggests that 50 pumps is a possible value, we can estimate that the expected value of the experiment to the participant is somewhere between £10 and £15.