Why does an image seem smaller through a pinhole when viewing with human eye I was chewing gum when I took the Trident gum wrapper and rolled it into an open cylinder with a diameter of about 1.5mm (give or take 0.15mm). When I look through the gum wrapper (I have to put it right up against my eye and position it just right to be able to see all the way through), images seem about 10% smaller (extremely back of the napkin assumption) than without the gum wrapper.
This is hard to measure empirically, so I tried to take a photo showing this effect, but the camera showed no difference between the two (the one through the wrapper is brighter, but I think that's just the iPhone camera trying to adjust to the surrounding dark and thus trivial). Just to confirm, I asked 6 people if they see any differences between viewing items through the gum wrapper and without the gum wrapper, and all of them said that the images through the gum wrapper looked smaller than without the wrapper. [EDIT: Asked 4 more people and 3 of the 4 said the image was 5-10% larger than normal]
[EDIT 2: The images also get larger (smaller for the people who said the image was larger) as the wrapper moves away from the eye and towards the object, eventually reaching true size when the wrapper is at the object.]
The only reason I could think of this happening is based on another observation I made. As I brought the gum wrapper into position, objects seemed to bend as they passed over the wrapper in my vision. The only thing I could think of is that the diameter of the cylinder is small enough that this "bend" doesn't have time to even out, and thus collides with the "bend" on the other side of the wrapper and creates a smaller image. However, this assumes that the distortion is causing images to be smaller when viewed through the wrapper. The "bend" distortion could also just be a byproduct of the actual cause of this phenomenon and just a symptom of the image getting smaller.
Why is this occurring?
 A: This is a matter of perception, rather than actually bending or focusing the light. This is why different people can perceive the effect differently, and the camera does not. When you see things out of their normal context, you may orient their perception differently.
A: Given that your camera failed to notice the difference, what you describe sounds like the Moon illusion. Namely, when you see an object in some confined context, the object seems larger, although actual size of its image on the retina remains the same.
A: It's because your vision processing wetware evaluates the image as being closer than it really is. So naturally it appears smaller, given that its angular diameter remains unchanged.
There are two factors at work here: firstly, you lose binocular vision; secondly, the image is always in focus because of the narrow viewing aperture, so you lose the ability to estimate the distance from the blur. Possibly these factors work against each other in some people, so results may differ.
But you say you performed the experiment twice, with opposite results; so perhaps conditions were different? Maybe the first experiment was performed indoors, and the second experiment in bright sunlight?
