1.5 seconds per panel. That’s how long it takes on average for wordless panels to be read.
I recently completed a very exciting study that asked people to read four-panel comic strips one panel at a time. In this “Self-Paced Reading” task, they see four boxes on the screen, and with each button press a subsequent panel appears in the sequence. Only one panel is shown on the screen at a time.
While they do this, the computer records how long it takes them to move from each panel to the next. By manipulating the strip in various ways, I’m able to tell if certain manipulations have a greater impact on the reading from the original “normal” sequence.
I’m not going to go into the intricacies of the experiment (you can wait for the write up for that), but I thought I’d share some tidbits.
For instance, for “normal” wordless four panel Peanuts strips, it takes a person on average 1.5 seconds per panel. The first panel is usually read relatively slow (1.7), the second panel is fastest (1.3), then third is slightly less fast (1.5), then fourth is back to where the first was (1.7).
An interesting side note about this: since all these panels were the same size, yet were read at different speeds, it might imply a rejection of McCloud’s claim that panel sizes affect reading time (which thereby somehow affects narrative time). However, this experiment didn’t test that specifically, and no alternations in panel sizes nor layouts were given either. So, I can’t say anything conclusive about that.
This was a very exciting project for me to complete, because it marks the first time I’ve (anyone has?) looked at data like this for how people understand sequential images.
Well it doesn’t rule out that Mr McCloud’s theory is true, just that same-sized panels can take different lengths of time to be ‘read’.
Altering the sizes of those four panels may have altered the results.
While panel size may not be indicative of actual read time, I consider it a device to communicate intended time spent. I find the same to be true for margin of text in a vox bubble.