This book is called "A Theory of Fun for Game Design". It has cartoons every other page. Don't be deceived. This book is not fun.
The thesis is that "fun" in games is your brain enjoying spotting patterns and understanding things. It's sufficiently reductionist that it explains it as your brain getting its dopamine. It rather takes the, er, fun out of it all.
Strictly speaking, it does present its argument reasonably well, but overall I'm somewhat disappointed. While there's a short history of early arcade shoot-'em-ups, otherwise there's very little to illustrate the theory with practical examples.
Speaking of illustration, what about those cartoons? Every other page in the main section of the book is a picture. Do they add much? No, not really. They do at least fit the style of the rest of the book, by which I mean disappointing and mildly depressing. I guess they make the book longer and look like it has more content.
Note that I did say "main section". The main section has 200 pages, of which 100 pages are text. Then, there are 50 pages of end notes, which are like footnotes, except they're in a completely different section to add the irritation of flicking back and forth. The main section text often doesn't fill the entire page, as it's being synchronised to the illustrations, which means there isn't actually that much more main text than end notes. Why not just put the notes in the main text itself?
The end notes are where the author reveals himself to be erudite and polymathic. In other words, he does the handwavey abuse of Goedels theorem, and explains to the reader who Mondrian was, if they didn't know.
It's a competent enough extension of a conference talk, but... joyless.