FWIW my wife underwent a drastic personality change after giving birth, and I would definitely agree with this being one of the symptoms. But just one data point of course.
As a counterpoint, my wife gave birth to our daughter a couple of years ago and I saw no noticeable personality change. She was perhaps more emotional at first but that seemed as much a result of sleep deprivation as the actual pregnancy and birth.
The part of my experience that seemed relevant to the article:
areas of the brain involved in social interactions used for attributing thoughts and feelings to other people - known as "theory-of-mind" tasks
...was that I noticed she wouldn't differentiate between whether something had actually happened or not. For example, she could get quite angry at someone for something they didn't do, just thinking they "might" do it.
Then again, maybe this personality aspect had been there all along and I had never noticed until the additional stress of having a new baby.
EDIT: Please don't downvote me for answering a question.
I really don't want this to come across as rude, because I don't mean to single you out or invalidate your experience (or your contribution to discussion!).
That said, what is the purpose of offering your anecdotal experience if you acknowledge that it is an n=1 datapoint? I understand that it's in good faith to acknowledge that your personal experience doesn't offer any statistical rigor to an interpretation of the foregoing article, but in my opinion your comment primes other readers in the discussion to subconsciously accept certain interpretations that aren't justified by the research. The main takeaway of your comment is "For what it's worth, I have an anecdote confirming an interpretation of the data that is not presented in the original study", and while you follow it up with the anecdote-acknowledgement, the "damage is done" so to speak. You've already opened Pandora's Box with regards to your subtle confirmation of the unsubstantiated interpretation.
In my experience it takes conscious effort not to allow anecdotes to impact interpretation of data, and while I don't think you're doing it purposefully, I believe it's counterproductive to the parent comment's point to offer an anecdote to the discussion while simultaneously acknowledging that the anecdote does not in any way prove the interpretation.
Because this is Hacker News, not the Journal of the American Medical Association. It's a community, and since it's acknowledged by the poster not to be a scientific observation, I'm not sure he needs corrected on it.
That's a fair point, and maybe I'm just demonstrating a personal bias for a particular type of discussion that isn't reflected on Hacker News. However, I didn't try to "correct" the grandparent commenter so much as show that while an acknowledgement of an anecdote is made in good faith, it doesn't take away from the overall impact on the audience. We may not be researchers peer reviewing the work of colleagues here, but we can still strive for a certain level of scientific literacy and decorum in our discussion.
Stated another way - if this were an article about computer science instead of neuroscience, would you expect the overall discussion to be stronger, the same or weaker compared to this discussion in terms of rigor? Furthermore, would you think that rigor was warranted on this forum (these aren't rhetorical questions, I'm curious)? I understand that disciplines like neuroscience allow us to relate with the science a bit (it's tempting to retroactively interpret our experiences with novel research!) but comments offering anecdotes, even with those expressly accounted for, allow biases to be couched in between legitimate data. I certainly think the grandparent comment would be challenged similarly if this was a psychological study with the current concerns about replication.
> what is the purpose of offering your anecdotal experience if you acknowledge that it is an n=1 datapoint?
What is the purpose of using large-n datapoints anymore when the primary motivation of "research" is becoming less about finding meaningful information and more about getting published (in for-profit, paywalled "journals") so you can get more grant money for your institution?
Haven't you seen the articles in the last year or about little scientific research is even reproducible anymore ? You can just repeat your experiment until you get the results you want to see.
> That said, what is the purpose of offering your anecdotal experience if you acknowledge that it is an n=1 datapoint?
I'm not sure I understand your question...the grandparent of my comment noted an aspect of the article, someone replied to that casting doubt, and I replied (that fwiw) I had observed similar behavior.
Historically one couldn't draw conclusions without an exhaustive study, does that now also apply to anecdotes on forums? I've read hundreds of anecdotes on HN so far this morning, are all of those inappropriate?
> Historically one couldn't draw conclusions without an exhaustive study, does that now also apply to anecdotes on forums? I've read hundreds of anecdotes on HN so far this morning, are all of those inappropriate?
In a word, yes.
I'd push back on the word "inappropriate" specifically though. Whether they are inappropriate is a matter of context; in the context of a discussion about a study published in Nature Neuroscience, an anecdote that supports a causal relationship between the research and unsubstantiated interpretations of the data is inappropriate, in my opinion.
I think the disconnect here is a difference of approach - I am approaching this discussion as one of analysis, while there are other commenters approaching this as a friendly conversation. There's no right or wrong there, but I feel compelled to point out that the latter approach is how biased data frequently becomes parroted without real support in the original research.
In your original comment, you mentioned that your wife exhibited a significant personality change that appears to match the article. You also point out that you realize this is an anecdote. However, there is research[1][2] that demonstrates people will begin to internalize what they continually read, regardless of whether or not they know it is not true. In a discussion about a loaded topic such as pregnancy, where people are commiserating with each other or exchanging anecdotes, my claim is that it is not merely enough to acknowledge that the anecdote doesn't represent datum. Rather, that will not do anything to prevent the anecdote from having a positive impact on the audience's impression of the research (positive here used in the dialectic sense).
Of course, I'll admit I'm a stickler for this. Whether or not this is a point worth making in subsequent threads is up to the HN upvote/downvote roll :)
Ah ok, I see where you're coming from now. It is a valid point, but then at the same time I'm a stickler for suppression of free speech, so some sort of a meta-discussion was inevitable wasn't it! :)
As long as people realize it doesn't prove anything, there's still some value to it. Sure, it's not following proper scientific methods, but HN comments are not for peer-reviewing scientific findings. Sometimes, readers may benefit from having a conversation with a human behind that n=1 data point.
