@Walter_vi_britannia I'm pretty sure that's not how it works. As Banarok said, she simply slaps an image based on what she hears, like when Mana was thinking their friend "was working" without further context, Nakano pictured him carrying boxes because that was her own interpretation of the job (reality was he worked fixing bikes).
Since people mostly think in their own language, even if Nakano could read a foreigner's mind she'd likely be unable to understand anything and they'd be text-only bubbles; she can't fill in the gaps with images if she can't even comprehend the "text".
@Raikuha even if she doesnt understand the language, her mind would auto fill using whatever she knows of what she hears, if she associates russians with vodka, it would generate an image of drunk people every time she hears russian, that is if she even recognizes the language
@Raikuha Yes, but won't the text be in her language since she doesn't know other languages? What she doesn't know can't be shown...that doesn't only apply to images, it means text too.