“I could take the oldest person here, make a little hole right here on the side of the head, and put some depth electrodes into their hippocampus and stimulate. And they would be able to recite back to you, verbatim, a book they read 60 years ago.”

So said Ben Carson, the U.S. Secretary of Housing and Urban Development, yesterday. Carson is known for his unorthodox claims, such as his attempt to rewrite the Egyptology textbooks, but this time, as he’s a former neurosurgeon himself, he might be thought to be on safer ground.

Sadly not.

It is true that stimulation of various parts of the brain can evoke memories, or at least memory-like experiences. The hippocampus, however, is not known as a hotspot for this. Early studies found that hippocampal stimulation produces amnesia, not memory recall, and here’s a recent paper reporting that patients couldn’t even tell whether their hippocampal stimulator was switched on or off, and reported no subjective memory effects.

It’s especially unlikely that hippocampal stimulation would evoke 60 year old memories, because damage to the hippocampus is known to impair recall of recent events, and memory formation, while old memories (older than about 20 years pre-lesion) are spared, suggesting that they’re stored somewhere else in the brain.

Now, maybe Carson just mis-spoke and he meant to say “the temporal cortex near the hippocampus”. However, while stimulation of this region can indeed produce vivid memory experiences, there’s no evidence that these memories are any more detailed or accurate than normal ones. In other words, unless you normally have the ability to recall texts verbatim, there’s no reason to think that brain stimulation would let you do that.

It’s not even clear that the “memories” evoked by neural stimulation are real. Some researchers argue that cortical stimulation evokes imaginary experiences that just feel like memories – false memories, in other words. Indeed the term “hallucinations” has been used to describe stimulation-evoked experiences. Here’s a video of the strange phenomenon, via Avniel Ghuman of the University of Pittsburgh.

Carson went on to say that the human brain can “process more than 2 million bits of information per second”.

I’m not sure where this number comes from. It mainly seems to be quoted in the world of neurolinguistic programming (NLP), where it’s attributed to psychologist Mihaly Csikszentmihalyi as an estimate of the amount of sensory information reaching the brain every second, not processed by the brain. Others have estimated that 11 million bits or even up to 40 million bits per second reach our brains.

The trouble is, all of these estimates are little more than back-of-the-envelope calculations based on multiplying the number of sensory nerves entering the brain by the information capacity of each nerve. It’s not a totally crazy approach, but it makes a lot of assumptions.

Overall, I’d say that Ben Carson is proof that operating on the brain and understanding the brain are two different things.