"Reward" isn't a scalar quantity: it's a product of several distinct biochemical motivations, many of which are subjective -- being dependent on the individual's nutritional and metabolic state, and their cultural background -- and many of which aren't dependent on taste at all! I've gone into this in detail before:
http://paleohacks.com/questions/103742/is-the-food-reward-hypothesis-backwards/103748#103748

Trying to assign a scalar value to "reward" is like trying to assign a scalar value to "beauty"...there are too many interdependent variables to make an objective measurement.

That being said, there is enough commonality between individual biochemistry to make some generalizations. Complete, bioavailable protein produces both satiation and satiety ("protein targeting")...so, on average, higher-protein foods will sate more effectively than lower-protein foods. Foods like liver, egg yolks, and many leafy vegetables have a high nutrient-to-calorie ratio ("nutrient-dense"), and therefore tend to produce more satiety.

The problem is that satiety is usually relative to the rest of your diet...for instance, your desire for more lean protein will depend greatly on how much protein you've recently consumed. I'd probably start a project like this by figuring out several days' worth of approximate menus that would get me the most nutrients possible with the least calories (also being careful not to overconsume vitamin A in liver, etc.) while coming close to the balance of fats/protein/carbs you find most satiating (and sating).

In fact, there may be two experiments here: find the most nutrient-dense foods you can, and then experiment to see which proportions of them produce the most satiety for you at various calorie levels. My guess is that the satiating impact of any specific balance of macros will be different if you're consuming 2500 kcal/day worth of them vs. only 1300 kcal/day -- due to protein targeting if nothing else. I would be interested to learn the results!

JS

My (limited) understanding of what constitutes high reward food is that it is often the result of combining fats and carbs. Fat/protein is okay and in any case fairly unavoidable. Think potato (or any other veg) with butter/sauce/sour cream -- you remove the fat component, and it becomes MUCH lower reward. SO if I were going to embark on this kind of thing, I'd do a lot of mono-meal-eating with nutrient dense foods. Like just hard boiled eggs (really, without salt or mayo how many would you want to eat?) or just a steak, just broccoli or some other veg. Maybe you'd have to eat more frequent meals to get adequate calories.

But, like I said, this is just my interpretation of what I've read.

Bone broth....kinda dull but gets the job done.

For me fat+ protein+ crunchy veg seems to be very satiating. Here are some of my favorites that hit both marks, low reward and high satiation: Boiled eggs

Raw shredded cabbage w some meat

Shredded cabbage cooked in coconut oil w onions

Boiled lean stew beef

Water canned tuna

Grapefruit

Apples

Dill pickles

Hot black tea

Hot broth

Kale cooked w some fat, garlic ginger

Salad of raw shredded beets, apples, carrot, cabbage. I add homemade tahini dressing, but that ups the reward value.

Egg Yolks are high in nutrients.