I actually looked into this for advertising tactics as part of my college dissertation years back for my dual major. A few off the top of my head:
Color Theory:
Colors elicit different responses primaly when paired with specific context. The color red is the first thing we see (unless you're R/G Colorblind), and it elicits a primal passion response closer to survival needs (combat, appetite, etc). The color yellow increases nerves and ups energy, which can increase stress and speed up the metabolism through that ever so slightly.
Ever wonder why so many successful fast food places use those colors? The two colors together elicit hunger, and they employ this on purpose. If you see Blue in their food graphics (like Sonic), then they likely did not look into this and their sales show it, as Blue is supposed to be the comforting receding color that tells us it is unappetizing (even blueberries are technically purple, and the "Blue Ketchup" marketing we saw briefly in the early 2000s more than showed us why that color does not water our mouths).
From my time dishing out free samples back when I was still part of the workforce, orange food to the point of unrealistic and at times nuclear always reported bigger numbers. I used to figure a lot of that was over Cheese Conditioning (like how "Melty" can sound advertising for cheese but not much else), but I've also seen it be successful from spices and packaging alone.
Focal Pointing:
Monkey see monkey do, and images translate to us as experiences (classic Holy Mountain logic). It's as simple as having faces with displayed expressions and eyes aiming where you need them to look for forcing them to acknowledge what you've placed in front of them.
If you create a three-point perspective that forms a triangle between focal pointing sources though (like three figures eyes staring at each other), the sense of motion and the sense of direction has a much stronger means of trapping the person's gaze within the frame, whereas if two stare at a third figure staring off-frame, it's hard to not do it yourself too. You can see it in a lot of successful classic art and in modern day advertising. With art in museums it's especially noticable when a figure appears to be staring off-frame that people will move onto the next piece faster based on a combination of where the staring is pointing and their already pre-decided sense of direction.
If the figure is staring directly at the person meanwhile, it challenges them and makes it harder to shake off having seen it, but it also does not elicit a sense of trapping. These tend to be used the most successfully with billboards, but does not translate the same for most other forms of art. Even for a solitary figure, a quarter turn tends to be preferable.
For advertisements like food and jewlery, having them stare at the product in a two-point perspective won't trap the user, but it will leave a quickly translated impression that for many can take some shaking off to resist falling into a parroting mindset of. Even those who resist the parroting though tend to still walk away remembering the product, and in stores having a mascot or something point towards the product gives us very few choices in the matter of where to look.
Spelling:
Klasik kewl, l33t, 2Xtreme, al dat sheet.
This was a classic 90s tactic to force a person to spot and remember something. If it's spelled correctly or otherwise worded in a more streamline way it's more likely to go in one ear and out the other, while if the mind has to take even a little bit longer to translate what it's supposed to be saying then the message is that much more in their head than not.
If you go too far with it it becomes "too much work" and they'll just reel back in horror, but if it's just within the borders then they tend to have no choice but to absorb your message. If they like doing that though...
They don't, it's slightly more stressful than the streamlined messages they'd rather forget about. It's a low path for getting people to notice your shit as it's no fun for them to do it, which is why we see significantly less of it now than we used to, but that does not deny that it is a focal pointing tactic at all.
Expressions:
It's a big "Well duh" that showing someone in a near-orgasmic expression will lower one's guard, that a stoic face will motivate, that appearing beastial within context can trigger Xenophobic urges, and that cartoon expressions can stretch those boundaries further than not. We've even experimented with this shit with avatars on SC since year one, with me for instance seeing much more social success with images that resemble my face (with blinking) over detached facial imagery from my older art pieces (the feminine egg and the blue dude).
Strangely, I see very split results between if cartoons or real faces work better on people (which is why I aimed for a mid-ground image). One tends to translate a lot better than the other on an individual basis, but the expressions within remains true across the board.
Try photoshopping ads that have expressions that don't match the theme and you'll see The Geico Effect, which is more memorable for being less streamline but also jarring as a result. When we saw other companies try their tactics we saw success only in areas that are disconnected from more blatant sensory experiences, like car insurance and lotion, while not as much for things like food and shelter from how jarring does not produce sales there.
Depending on what you're going for, comfort vs discomfort push sales, but which you go for tends to be related to how primitive those needs are versus how detached from them we can be.
Ę̵̚x̸͎̾i̴͚̽s̵̻͐t̷͐ͅe̷̯͠n̴̤̚t̵̻̅i̵͉̿a̴̮͊l̵͍̂ ̴̹̕D̵̤̀e̸͓͂t̵̢͂e̴͕̓c̸̗̄t̴̗̿ï̶̪v̷̲̍é̵͔
