Visual Bias Via Saturation Adjustments?
Published May 10th, 2008 in PoliticalIn creating a 90-day unit plan, I wanted to focus a few lessons on teaching students how to discern opinion, fact, and bias in news stories. I thought it would be a good idea to show two articles covering the same topic, so that the students could analyze how the wording changed the tone of the articles, but I noticed a very different way of affecting readers’ beliefs: (de)saturation.
In the included image, I have organized it so that all of the low saturation images are on the left. It’s pretty clear that these images are less flattering and make each of the people in them look worse than the ones that are saturated. Here’s the test for you: Try to guess which images are the originals (i.e., the ones that I simply copied and pasted from the source sites) and which are the ones I modified in GIMP (by modifying saturation by +/-67). Your only clue: The second image set (counting in pairs from the top, so the second pair is Obama, the fifth is McCain) comes from Barack Obama’s website and the rest are from Fox News (the first and last are from AP, the rest are unattributed).
First pair: The desaturated image is the original (source).
Second pair: The saturated image is the original (source).
Third pair: The desaturated image is the original (source).
Fourth pair: The saturated image is the original (source).
Fifth pair: The saturated image is the original (source).
Sixth pair: The saturated image is the original (source).
Of the three political candidate sets at the bottom, Obama’s is clearly under-saturated, Hillary’s is clearly over-saturated, and McCain’s is just right…. Curiously, McCain’s is also the only one with a background. I threw in the final shot of 50 Cent to show how much difference saturation makes on skin tones.
To think that those images seem to suggest that they favor this candidate:



0 Responses to “Visual Bias Via Saturation Adjustments?”
Please Wait
Leave a Reply