I somehow think that in my mind logical thinking is a breadth-first search, and what is called design thinking or art thinking is more like a depth-first search, so I feel that if I donât pursue depth, I will end up halfway through the search. Design thinking and art thinking are not exemptions. https://twitter.com/kur/status/1259335314721525760
I was reminded that this is the opposite of vertical and horizontal, and I thought that it could be taken that way, depending on what is placed as a node. Metaphors are difficult. For example, if the focus is on synthesizing, it can be said that depth is the priority, and if the focus is on the ideation (brainstorming) process, it can be said that breadth is the priority. https://twitter.com/kur/status/1259476931365371905
On the contrary, I was more convinced that design thinking is depth first search (search algorithm in AI).
I think the reason I thought that was because what came to my mind when I heard the term âdesign thinkingâ was the act of ânot exhaustively examining design options, but proceeding quickly to a prototype that can be experimented with anyway.
However, I realized that this is not a depth-first search, since it does not try to change only the last âdesign optionâ in the ânext searchâ. I thought it was more like a Monte Carlo tree search, in which the player repeatedly plays until the end of the game in Go AI to judge whether the move is good or bad.
On the other hand, the behavior envisioned by the term âlogical thinkingâ of âexhaustively considering design alternatives, selecting the best option, and then considering the next optionâ is also not breadth-first search, because it does not explore the âdestination of the option not selected in the first stepâ in exploring the second option. Itâs not.
Implicit in the search method here is
- It is possible to know whether an option is good or bad by focusing on only one option (the evaluation function is given).
- Beyond the best-rated stand-alone option, there exists a solution that is better evaluated as a whole.â
The assumption is that This is similar to the pre-Monte Carlo chess AI algorithm
So, to return to the original question, what is âdepthâ in âif you donât pursue depth, you will end up halfway throughâ? I interpreted it as âto build a prototype and go as far as getting information about its merits and demerits. Is this what you mean by âif we focus on synthesizingâ?
I wrote this as a depth-first priority since I am also prototyping at first. On the other hand, if you change the scope of your view, there is a Monte Carlo aspect, as Nishio-san says, which I think is similar to breadth-first + branch cutting (especially in the phase of brainstorming and expanding possibilities). I see, in comparison with Shogi AI, AI has an implicit assumption that âall possible options are knownâ, while humans, due to various biases, can recognize only a part of âpossible optionsâ, so we need to expand our options first. . But this is also true of logical thinking. It is breaking down the problem and expanding the options that can be taken. Whether that is Monte Carlo-like, putting it out there anyway rather than exhaustive, and then deciding whether it fits into a circle or not, or dividing it up with MECE in mind, I think it is different.
About if we focus on synthesizing. In design thinking, it is said that how to create a good âproblem to be solvedâ through interviews and other research is important. I wonder if it can be taken as a depth priority with the âdepthâ of this question in mind. For example, âHow can I delay an enjoyable life under coronal influence?â is a shallow question. For example, âHow can I spend quality leisure time with friends while maintaining social distance under the corona influence? Just that this makes me think that the analogy with logical thinking is rather difficult to drawâŠ
This page is auto-translated from /nishio/ăă¶ă€ăłă·ăłăăłă°ăŻæ·±ăćȘć æąçŽąïŒ using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. Iâm very happy to spread my thought to non-Japanese readers.