I wouldn't conclude anything. Even if you assume that the interviewer didn't know the parameters for the experiment, their biases could have accounted for the r

I wouldn't conclude anything. Even if you assume that the interviewer didn't know the parameters for the experiment, their biases could have accounted for the results. If a few of the interviewers were attracted to outdoorsy men and were biased towards the ones who crossed the dangerous bridge, their body language alone is likely to get a very different response. You'd need a pretty large sample size of interviewers and subjects with varying attractiveness, "outgoingness", and confidence to draw any real conclusion.