I don't see how tests can be done that would "prove" that bonuses work, at least if my understanding of the bonuses is correct.
To explain why, bear with me a little bit. Lets look at my DMC Gary Butcher as an example. Gary Butcher has the ability 40 which is 121% at my level. My understanding is that each of the 4 bonuses increases a specific group of abilities with a multiplier. Like, its the only reasonable formula I can come up with. It is possible that the possession bonus is carved out and functions like the possession bonus you get from watching a game etc, but the other 3 bonuses would still function like I expect in my theory. Anyway, if they function something like that, one can only guess, the abilities might be divided into the four groups in accordance with something like this:
Attack Bonus: Shooting, Finishing, Dribbling and Heading
Defense Bonus: Tackling, Marking, Courage and Positioning
Possession Bonus: Passing, Creativity and Crossing
Conditioning Bonus: Conditioning, Strength, Aggressiveness and Speed
So lets say that I activate the attack bonus during the game, and has the following bonuses after that:
Attack Bonus: 15%
Defense Bonus:10%
Possession Bonus: 0%
Conditioning Bonus: 0%
This is a little roughly how it would affect Gary Butcher:
You see comments like, "I don't know if bonuses are working or not" and "they don't seem to make a big difference". But a flat bonus of 10% bumps a player that is 120% to 132%. How could you spot that difference? You don't know if the opponent has bonuses or not. If you have 10% bonuses a player with 100% in all abilities will have 110% on avg, and after a 5% bump in one category will go to 111.46% on avg. IE activating a 5% only increases the players' avg. ability with 1.46%.
I still think that is a lot and that its absolutely worth it to pay attention to bonus. The difference from having zero bonus to 10% bonuses and activating it is 11.46% in that example, and that is of course huge. But it won't be something you can spot easily if experimenting.