Sunday, August 14, 2011

Pardon my brain

As social beings, we’re very good at letting others know when we’ve accidentally stopped being social. When we’ve missed something important in a conversion, we interrupt and ask: “Pardon?” “Excuse me?” “Sorry?” “What?” “Huh?” Sometimes furrowing the eyebrows a bit can suffice to communicate a sense of confusion.

We need ways to communicate when we’re lost because – let’s face it – we get lost often. We could be looking a fellow conversationalist directly in the eyes, convincing ourselves that we’re taking it all in as we listen to them clearly enunciate each syllable, when in reality every word is slipping right over our heads. “Come again?” we bid.

These lapses in attention could be driven by distracting thoughts or events in our environment, but are such diversions necessary for a conversation space-out to occur? A new study in press in NeuroImage suggests the possibility that our brains are constantly fluctuating in and out of particular states of attention. In certain brain states, we’re prepared to take in information and learn about it, whereas in other brain states, new information will likely elude us.

Study participants were scanned with functional MRI as they viewed 250 photographs of indoor and outdoor scenes, presented one at a time. The subjects were told that after the brain scan, they would be taking a test that would assess which scenes they recognize from the scan. In that test, the subjects were presented with images that either had or had not appeared during the scan, and for each image the subjects rated how confident they were that they had previously seen the indoor/outdoor scene. Out of the 250 images, the participants only correctly remembered some of them.

When the researchers analyzed the subjects’ brain activations, it was found that lower levels of activation in the parahippocampal place area (PPA, a brain region that is known to respond to visual scenes) occurring 2 seconds before presentation of an image predicted that an image would subsequently be correctly remembered in the post-scan test. However, when PPA activation was a bit higher during the 2 second period before an image was presented, the subject was likely to forget that image.

This is pretty neat. It shows that fluctuations in brain activity before we are presented with information determine whether we’ll pay attention to and remember that information. These sorts of fluctuations may also partially explain why we sometimes tune out of conversations.

But the neater part of this study was the follow-up. Based on the first experiment, the researchers called downward fluctuations in PPA activity a “good” brain state (ready to learn and remember a scene image) and upward fluctuations in PPA activity a “bad” brain state (likely not ready to process the image). The researchers used a technology known as “real-time fMRI” to monitor “good” or “bad” fluctuations in brain state as they were happening. When either a good or bad brain state occurred, a scene image was presented. As expected, when good brain states triggered an image, 2 hours later the image was more likely to be remembered than when a bad brain state had initially triggered an image.

This real-time fMRI study is particularly interesting because it demonstrates that fMRI can do more than just identify the neural correlates of human behaviours or perceptions. Rather than using stimuli to elicit brain activation, this study used brain activation to drive stimulus presentation and cause a certain behaviour (good or bad learning). Using brain activity to elicit stimuli, we can get closer to the question of a causal relationship that can otherwise only be addressed by inducing brain damage or using brain stimulation/other techniques to interrupt neural activity.

This study also might have implications for education programs that require “optimized” brain states for efficient learning to occur. And of course, there may be implications for optimizing everyday conversations. Next time you accidentally drift out of a conversation, consider adding the phrase “pardon my brain” to your repertoire of strategies for communicating that you’re not all there at the moment.


Yoo JJ, Hinds O, Ofen N, Thompson TW, Whitfield-Gabrieli S, Triantafyllou C, & Gabrieli JD (2011). When the brain is prepared to learn: Enhancing human learning using real-time fMRI. NeuroImage PMID: 21821136


  1. Oooh, that is interesting, thanks for posting about it.

    A possible explanation of the effect is that when the PPA was active, that was when they were still thinking about a previous scene, or a random scene from memory, which interfered with the new stimulus.

    Whereas when it was less active it was a "clear slate" and could record the new info.

  2. That's an interesting possibility, and in fact the finding was in contradiction to previous studies that showed the opposite effect (greater PPA activation preceding successful memory encoding). The authors offer this explanation:

    "It may be speculated that lower activation reflects a lack of processing activity in the PPA, and that more resources are available for memory encoding during such an ebb of PPA activity."