Dan, I'm going to reply to your questions all together, as I'm not familiar with breaking out quotes like everyone else can. I'll eventually figure it out.
Reactions seen early in the zapruder film
https://sites.google.com/site/earlyzapruderfilmreactions/reactions-seen-early-in-zapruder-filmSome apparent total lack of reaction in people in response to a stimulus was a phenomenon that I saw mentioned in one literature reference, but no further elaboration was made. It may be situational based, so I don’t know the exact answer. I can only speculate for this situation. For the crowds I personally believe that it may be as simple as “what grabs, and holds, your attention at that time”. The visual stimulus intensity of seeing the President and First lady right in front of them (many were trying for direct eye contact and a wave response), might well override an auditory stimulus to a firecracker in the background. Now if the auditory stimulus was perceived as a threat, or a real concern, I suspect that perception would override the visual stimulus and they would react to that by looking around. Net the President and First lady right in front of them visually “grabbed and held their attention at that time” over a perceived non-concerning auditory stimulus. For others the first shot sound could be perceived as something that was concerning and they reacted. Again, some speculation, but I wouldn’t be surprised that for those in the Presidential limo and the trailing SS car, there was some sense of relaxation starting to set in, the motorcade was nearly complete, the crowds were much thinner, and the visual field ahead of them was less stimulating than what they had just completed. A surprise loud bang for them may have posed a real concern that grabbed their attention and started a voluntary reaction of concern.
Another point that might come into play is that the most difficult location of a sound for your ears to accurately spatially locate (sound localization) is directly behind and high overhead. I suspect that was the case for individuals in the Presidential limo and the trailing SS car. It may have been hard, based on only one, first unexpected quick sharp bang, to determining exactly what direction the sound came from. Harder to tell which way to look. I can’t say for sure but actually Jackies’ and John Connally’s sweeping head reactions first left and then back right look almost like classical sound localization motions from mammals swinging the head back and forth to locate an unknown sound stimulus.
Regarding your comments on Hickey, I think there are two points to consider. 1) This technique does not use any testimony and further obviously does not use any testimony that is not given, or even testimony that is in error or specific details that were even forgotten or assumed inconsequential within all the chaos. 2) The reactions of Hickey you mention are not startle reactions. They are voluntary reactions, and fit the description as being unusual enough to be unusual voluntary motions of concern, and occur within a timeframe other voluntary motions of concern are observed and consistent with a population perception time model.
Looking at the video in the link above, Jackie starts accelerated head turning left at ~Z143.5, before looking back right. (Similar to John Connally's L-R head motion but starts slightly earlier and ends slightly later than his).
Rosemary Willis in the lower video in the link above appeared to begin a quick look away from the Presidential Limo back towards the Texas School Book Depository at ~Z140 as it appeared to both authors independently using that video.
To note, one advantage to using an average (or median) of multiple samples, if possible, is that this can help buffer mistakes made in sampling. If there were a couple of data points that I changed my mind on to use, I could just throw them out. For example if I decided I didn’t want to use JFK and Rosemary Willis data points, I would throw them out and recalculate. If I did that for this case the predicted first shot timing would only shift about 1.5 frames. If those two points were in error, the original result would have been somewhat buffered by the rest of the data with the original result only being off by less than two frames. Yes, some judgement in context is necessary, but even though I would say those two samples may be a little less clear than the others they still were judged relevant enough to include.
Good questions, hope this helps.
