To mark 10 years of Nature Reviews Neuroscience this month the journal has produced a kind of retrospective of the most highly cited reviews from each year. I got around to reading the 2002 “winner” from Maurizio Corbetta and Gordon Shulman which focused on attention networks in the brain, and a quality read it is.
The authors build the case for two distinct brain networks which have specific roles in the processing of attention but which interact. The experimental paradigms they use are funky. Simply (and for the real detail you should read the paper), the first network, located in the intraparietal and superior frontal cortex regions of both hemispheres seems to be involved in goal-directed attention. For example if you are expecting or seeking a particular stimulus then this network will modulate your attention towards stimuli that fit the pattern. The second network, located in the temporoparietal cortex and inferior frontal cortex (doesn’t neuroanatomy have an intimidating language?) mainly in the right hemisphere seems to be specialised for the detection of unexpected, behaviourally relevant stimuli. So the first network we can (and the authors do) call the “top-down” network, and the second the “bottom-up” network. I kind of like the idea of the second network being the poor infantryman who suddenly yells “INCOMING!”.
It is always dangerous to extrapolate beyond the data. These studies are predominantly related to attention to visual stimuli and not to pain and it is a risky assumption to assume that the systems are equivalent (but forgivable? a more recent review of attentional mechanisms specific to pain tells a similar tale of two interacting mechanisms, albeit with different anatomical locations.) But there are messages here that might be relevant. First when trying to alter patients’ behavioural responses to chronic pain we are essentially, by explaining, educating and trying to take the fear out of it, trying to address the top-down network, to convince them that pain may not be the threat that it appears and to reduce the attention that it is paid. Attention is strongly influenced by the salience of a stimulus which, since there are few if any stimuli more salient than pain, means that we are up against it. While it is manageable for most patients to understand our message it is much more difficult for them to believe it. In evolutionary terms it is unsurprising that pain beats reason, just like the desire for donuts beats diet plans, only worse.
But it is the discussion of the interaction of the networks that really grabs my attention (sorry). The bottom up network will be drawn to stimuli that differ from the background; stimuli that are “intense, voluminous, sudden”, behaviourally relevant or threatening. This makes me think of sudden shooting pains, increases in pain, new pains and unexpected sensations that occur against a backdrop of pain. When stimuli activate the bottom-up network they disrupt the ongoing cognitive activity (the authors of the review use the nifty analogy of a circuit breaker). It reminds me of how sudden flare ups or changes in a patient’s pain present a challenge to pain management and often go some way to undoing progress in the clinic. But the reverse process also holds. Effectively reducing the salience of a possible stimulus could exert a top-down influence on the bottom-up network and perhaps moderate the circuit breaker.
Finally the review draws some interesting conclusions about the nature of neglect. They argue that the lesions that commonly cause neglect lie in the same brain areas as the “bottom-up” network and that neglect appears primarily to be more a failure of stimulus detection than in conscious goal oriented attention. In this model the subsequent deficits in top-down style attention are a secondary consequence of the bottom-up disruption.
The relevance of this research to how we understand pain is pretty clear and it’s a good example of how looking at research outside of pain can throw up fantastic insights. But a word of warning – while it seems clear that engaging bottom –up attention mechanisms can affect top-down attention I would not advocate creeping up on patients and shouting BOO! or similar strategies. That’s just childish.
Neil O’Connell is a researcher in the Centre for Research in Rehabilitation, Brunel University, West London, UK. He divides his time between research and training new physiotherapists and previously worked extensively as a musculoskeletal physiotherapist. He also tweets! @NeilOConnell
Neil is currently fighting his way through a PhD investigating chronic low back pain and cortically directed treatment approaches. He is particularly interested in low back pain, pain generally and the rigorous testing of treatments. He also tends to get all geeky over controlled trials.
Corbetta, M., & Shulman, G. (2002). CONTROL OF GOAL-DIRECTED AND STIMULUS-DRIVEN ATTENTION IN THE BRAIN Nature Reviews Neuroscience, 3 (3), 215-229 DOI: 10.1038/nrn755
Legrain V, Damme SV, Eccleston C, Davis KD, Seminowicz DA, & Crombez G (2009). A neurocognitive model of attention to pain: behavioral and neuroimaging evidence. Pain, 144 (3), 230-2 PMID: 19376654