Magnetic resonance image of the mouth, ear and brain

Even though we might be able to hear someone speaking, our powers for understanding what is actually being said switch off as we go to sleep.

With this methodology, we could refine our judgement of how deep anaesthesia needs to be to prevent comprehension and memory during operations.

Professor David Menon

Using brain imaging techniques, Cambridge scientists have shown that those regions in the brain that understand sentences and form memories show reduced activity after sedation. These findings help us to understand how speech is decoded in the brain, and have clinical implications for monitoring anaesthesia and patients with brain injury.

The recently published results are the culmination of a study conducted under the lead of Dr Matt Davis at the Medical Research Council Cognition and Brain Sciences Unit (MRC CBU) and Professor David Menon at the University’s Division of Anaesthesia in the Department of Medicine.

Dr Davis, a cognitive neuroscientist, has developed a method that detects when the brain has comprehended speech, not simply heard it: ‘We look at the brain’s response to sentences containing ambiguous words. Of the most commonly used 5000 words in spoken English, over 80% change their meaning in different contexts – such as the word shell, which can refer to a bullet or part of a sea creature.’ Understanding these words triggers additional processing in the brain as it retrieves the different meanings and selects the one that fits the sentence context; this activity can be visualised using functional magnetic resonance imaging (fMRI).

In this recent study, 12 anaesthetists at Addenbrooke’s Hospital volunteered to receive varying amounts of sedation to test at what level their brain could understand speech. ‘Our research showed that although brains that are sedated to the same level as in sleep are able to process sound, brain processes involved in speech comprehension are compromised even at quite low levels of sedation,’ said Professor Menon. ‘These results have important implications for our understanding of how and where anaesthetic drugs work in the brain, and what neural processes are involved in consciousness.’

Being able to assess comprehension without spoken responses has huge resonance for two clinical settings. ‘A small proportion of anaesthetised patients report memories of events that occurred in the operating theatre,’ said Professor Menon. ‘With this methodology, we could refine our judgement of how deep anaesthesia needs to be to prevent comprehension and memory during operations. The information also helps us to understand functional imaging studies that we have undertaken in patients following brain injury – in individuals who appear to be in a coma or the vegetative state.’

For patients in the vegetative state, who are awake and yet show no outward signs of awareness, this tool is already providing important information. ‘Being able to show brain activity that indicates comprehension in the patient brings hope to carers, and might guide treatment and rehabilitation,’ said Dr Davis.

For more information, please contact Dr Matt Davis ( or Professor David Menon ( This research was published in PNAS (2007) 104, 16032–16037.


This work is licensed under a Creative Commons Licence. If you use this content on your site please link back to this page.