Earlier this year, we talked about the possibility that it might be categorically impossible to tell you’re being lied to—and that lie detectors, even those that rely on neuroscience, might prove unreliable. A recent study is unlikely to change our minds much, but just as a traditional lie detector test can at least detect nervousness, neurological imagining may soon be able to tell us whether someone is actively in the process of recalling memories rather than constructing a new narrative.

That’s in part because we understand the neuroscience of perception and sensation, including the neuroscience of recalled perception and sensation, better than we understand almost any other kind of neuroscience. If a memory comes from recalled perception or sensation, it’s coming from a fairly familiar place. 

One recent study has found evidence that by looking at the simple flow of neurological activity by way of an EEG, it’s possible to determine—on at least a rough level—whether a thought is coming from recalled visual perception or the flow of imagination. But there are some obvious caveats, most of which are specifically acknowledged in the paper:

• We don’t know exactly how visual memories are reconstructed.
• We don’t know that inaccurate memories, or even fabricated visual images, can’t be reconstructed.
• We don’t know exactly how visual imagination works.
• We don’t know that someone recalling a visual image might not also need to rely on visual imagination to contextualize it, or that someone imagining a visual image might not also construct it, in part, out of visual memories.