secondary memory vs working memory

[Tiago Reis (T. R.)]

from what I understand working memory is correlated with fluid iq. when we don't include secondary memory though, it becomes uncorrelated.

this may imply that secondary memory is powered by working memory. Working memory is simply the gateway that keeps your mind's brightest ideas from coming out of your mouth.

lets be honest this is a little speculative but this also comes from a takeaway of https://www.jstor.org/stable/40064890 (What's so Special about Working Memory? An Examination of the Relationships among Working Memory, Secondary Memory, and Fluid Intelligence)

[Memory Rush]

What you’re picking up on isn’t wrong, but the relationship is a bit more layered than “secondary memory being powered by working memory.”

One useful way to think about it is that working memory and secondary memory solve different problems for the mind.

Working memory is about active control: maintaining, updating, and manipulating information in real time. That’s why it aligns so well with fluid intelligence — fluid tasks demand moment-to-moment coordination, not just storage.

Secondary memory, on the other hand, isn’t passive storage either. It reflects how efficiently information can be recovered once it’s no longer actively maintained. When secondary memory is excluded, working memory measures lose some of their predictive power because many “working memory tasks” quietly rely on retrieval once capacity is exceeded.

So rather than secondary memory being powered by working memory, it may be more accurate to say this:

Working memory determines what gets organized, while secondary memory determines what can be recovered when control breaks down.

The JSTOR paper hints at this by showing that fluid intelligence draws on both:

• control and maintenance (WM)

• structured access to representations beyond immediate attention (SM)

This also explains why people can have strong working memory performance in tightly constrained tasks but struggle with complex reasoning when retrieval demands increase. At that point, intelligence depends less on holding information and more on reconstructing it efficiently.

In that sense, working memory isn’t just a “gateway,” but secondary memory isn’t downstream either — they’re complementary systems, and fluid intelligence emerges most clearly when both are engaged under pressure.

[David Breneisen]

Secondary/short-term memory ability would improve with improvements to working memory. And long term memory should improve with improvements to short-term memory. At least from my experience. 

That said, I have a hypothesis that doing working memory exercises with real information rather than arbitrary audiovisual stimuli would be more effective at developing the working->short-term->long-term memory transfer abilities. Things like semi-sight-reading piano music. Or n-back games using meaningful stimuli.