The Hard Problem of Consciousness - 续
（注：此文原po在微信评论区，所以大量口语）
“伟大的母爱”？no
又名怀孕激素是如何改变你的大脑的
mPOA->VTA->NAcc DA
和性行为复用一个底层逻辑，小鼠实验表明母性行为受到激素驱动，并且对NAcc注入多巴胺抗结剂会导致母性行为减少，本质上是母性行为是大脑奖赏回路的维持。并且对未怀孕小鼠注射孕酮+催乳素会导致出现和怀孕小鼠一样的母性行为（未注射时未怀孕小鼠会抗拒接触幼崽）。
够解构的，你可以说还原论无视了所谓的“高级感受”，但是我认真怀疑这种所谓的高级感受不过是一种抵抗认知失调的骗局。
人类母亲报告的主观体验"只是"这个回路的副产品。这两个命题之间隔着一个hard problem。我怀疑这个命题根本无法被证明，你要怎么设计实验证明？对人类进行相同的小鼠实验并让被试者撰写认知评估报告？实际上意识本身就受到激素影响，我觉得这个很obvious。
我觉得可以进行人体实验，但是这种实验只能让受试者主观去报告来看激素前后的变化，如果有一个外部的东西不受到主观影响，类似核磁共振那种可以直观观察到激素变化对意识的影响更好，问题是怎么去观测一个意识。
我觉得造出这个工具不需要答案，我对这个问题其实不持悲观态度，我个人认为这个问题可以被解决，解决的核心是第三方观测被试的意识变化，如果可以构建并预测被试的感受通过大量观测神经元变化，是否可以通过底层神经元变化模式逆向构建出被试的意识。当悲伤的感觉可以被预测时，这种感觉本身就是能够被表征的，感受并不神秘，如果你可以调控大脑的神经元以让另一个人感受到相同的悲伤感，那本质上感觉就是这样，我个人认为再往上加抽象意义都毫无必要。
"The greatness of maternal love"? Nah.
a.k.a. How pregnancy hormones rewire your brain.
mPOA $\to$ VTA $\to$ NAcc DA.
It literally reuses the exact same underlying logic as sexual behavior. Mouse experiments show that maternal behavior is hormone-driven, and injecting dopamine antagonists into the NAcc reduces it. Essentially, maternal behavior is just the brain maintaining its reward circuit. Plus, injecting progesterone + prolactin into non-pregnant mice triggers the exact same maternal behaviors as pregnant ones (without the injection, non-pregnant mice actively reject/avoid the pups).
Talk about deconstruction. You could argue that reductionism ignores so-called "higher-level feelings," but I seriously suspect these "higher-level feelings" are nothing more than a scam to combat cognitive dissonance.
The subjective experiences reported by human mothers are "just" a byproduct of this circuit. There is a literal hard problem standing right between these two propositions. I honestly doubt this proposition can even be proven. Like, how do you even design an experiment for it? Run the exact same mouse trials on humans and have the subjects write up cognitive assessment reports? Actually, the fact that consciousness itself is affected by hormones feels super obvious to me.
I feel like human experiments could be done, but they would only rely on subjects subjectively reporting the changes before and after the hormones. It would be way better if we had an external tool completely free from subjective bias—something like an fMRI that lets you directly observe the impact of hormonal changes on consciousness. The problem is: how do you objectively "observe" a consciousness?
I don't think we need the "answer" to build this tool. Honestly, I’m not pessimistic about this at all; personally, I think this problem is solvable. The core of the solution is third-party observation of a subject's conscious changes. If we can map and predict a subject's feelings through massive observation of neuronal changes, could we reverse-engineer the subject's consciousness purely from those underlying firing patterns?
When the feeling of sadness can be predicted, the feeling itself becomes representable. Feelings aren't mysterious. If you can tweak the neurons in the brain to make another person feel the exact same sadness, then essentially, that's just what feeling is. Personally, I think slapping any extra abstract meaning on top of that is completely pointless.