Generally speaking, in an ethics discussion concerning whether a certain animal feels pain, I consider the burden of proof to rest somewhat more on the side that suggests that they do not. This is because we know there is a cutoff point somewhere, based on physical biology, but determining exactly where that cutoff point is can be problematic and uncertain. If we know that we can feel pain, and we're pretty sure other humans feel pain, and we're pretty sure smart mammals feel pain, and we're pretty sure less intelligent mammals feel pain, and that so do birds, and to a certain extent reptiles, etc., then pain is a somewhat common thing.
If we're going to do something that could potentially cause a creature to suffer, based on the argument that they cannot suffer, then ethically, we better be fairly certain. But on the other hand, if we simply say "well, I'm not sure, so I'll avoid causing the suffering if possible", it doesn't require all that rigorous of a defense. So this is why I think the burden of proof rests more on one side than the other. Unfortunately, based on the scale in which we live, the avoidance of killing things is impossible.
To a certain extent, the question of what sort of qualia anything experiences is non-falsifiable. For instance, if someone asks me to prove that a bacterium or a plant doesn't feel pain, I can't provide absolute proof, but I can provide evidence such as they lack a brain (and so any pain system would have to be very different from ours, and we should have substantial evidence of it if it exists), that pain is not evolutionary advantageous to them, etc.
Insects seem to be more in the gray area. That is, they kind of have a brain, but it's more like several tiny interconnected brains. They move around and clearly interact with their environment, but much of what they do is programmed. What sometimes appears to be intelligent behavior, is usually studied to be instinctual behavior. I've read that if pick an insect up by its leg, it starts going crazy, but if you rip that leg off and put it down, it doesn't seem to be responding to a pain stimuli. Emotion is not a particularly useful thing for most invertebrates to have.
A source.
My position on the matter is that based on the evidence, I don't think that most insects can truly suffer. But, I try to err on the side of caution whenever possible, and so if I find an insect in my apartment, I try to capture it and let it out, or if I end up hurting it, I try to kill it quickly, etc. But we use so many pesticides to keep crops safe, and insects are dying all around us in incredible numbers naturally, so I don't lose sleep over an injured or killed insect as I would over, say, a cow getting its throat cut or something.