Learning in a vegetative state

The Daily Telegraph today reported that, “people lying in vegetative states in hospital can still learn” and that this “breakthrough could suggest which patients have the potential to recover from their injuries or illness”.

The research involved repeatedly playing a musical note to severely brain damaged patients and then blowing a puff of air into their eye. Although the patients initially only blinked after the air puff, over time they started to blink after hearing the note and before the puff of air. Conscious people under anaesthetic did not learn to do this.

The lead researcher, Dr Tristan Bekinschtein, is quoted in the newspaper as saying, “This test will hopefully become a useful, simple tool to test for consciousness without the need for imaging or instructions.”

This small study looked at responses in 22 people who were in a vegetative or minimally conscious state. It showed that these patients were able to learn to respond in a similar way to conscious individuals, and that the level of response might be useful as a predictor of which individuals will show signs of recovery. Larger studies will be needed to confirm these findings.

Where did the story come from?

The research was carried out by Dr Tristan A Bekinschtein and colleagues from the Institute of Cognitive Neurology in Argentina, the University of Cambridge and other research centres. It was funded by grants from several organisations including the Antorchas Foundation, the Human Frontiers Science Program and the Medical Research Council. It was published in the peer-reviewed medical journal Nature Neuroscience .

What kind of scientific study was this?

The research investigated the possibility of developing a reliable test for assessing a person’s awareness when they are not capable of making an explicit response (for example, by speech or movement).

People who show no outward signs of awareness are described as having “disorders of consciousness”, of which there are several levels ranging from a vegetative state (no outward signs of awareness) to states showing inconsistent signs of awareness and ability to communicate.

The researchers thought that Pavlovian trace conditioning, which is a simple form of associative learning, would a good candidate for this test. It requires a person to learn that a neutral stimulus (an event that is neither pleasant nor unpleasant) indicates that an unpleasant stimulus is coming, and react accordingly. This shows that a person understands the timing of different events and in mammals relies on a part of the brain known as the medial temporal lobe. The researchers say it is considered, “a plausible objective test to assess awareness without relying on explicit reports”.

An eye-blink response was used in the test. This involves a sound (neutral stimulus) being played several hundred milliseconds before a puff of air to the eye (unpleasant stimulus).

The researchers included 22 people with disorders of consciousness who were grouped into three different categories of awareness: those who showed no outward signs of awareness (a vegetative state), those who showed inconsistent but reproducible evidence of awareness (a minimally conscious state, MCS), and MCS patients who had started to communicate (defined as severely disabled, SED). The patient’s responses were compared to those from two control groups, a group of 16 people who were conscious, and a group of 12 people who were normally conscious but had received a general anaesthetic as part of standard procedures.

The training involved 140 trials of the eye-blink response procedure, with 70 sounds (tones) followed by an air puff after 500 milliseconds, and 70 tones not followed by an air puff. The researchers measured the individual’s eye-blink response by attaching a sensor that detected muscle movement in the eyelid. If a person began to have a faster response (by blinking their eye), this showed that they were learning to expect the puff of air.

The researchers then looked at whether the level of learning (how much the response speed improved) was different between people in a vegetative state and those with MCS or SED. They also looked at the effect the cause of brain injury had on response times (those whose brain injury was caused by trauma or by other causes (for example, oxygen deprivation).

The researchers also looked at whether the test could discriminate between patients who showed signs of recovery over six months to two years (changing from vegetative state to MCS/SED, or improved behavioural ability scores without a change in consciousness state) and those who showed no signs of recovery (no change in behavioural ability scores).

What were the results of the study?

The researchers found that people in a vegetative state could learn to respond to the sound by blinking their eye more quickly to avoid the expected puff of air, similar to the response seen in the conscious control group, though not as strong. There was a stronger response to the tones linked to the puff of air than to the tones that were not, and this got stronger as time got closer to when the air puff was expected. The responses were not seen in conscious participants who were anaesthetised.

Similar responses were seen in people in a vegetative state and people in a minimally conscious state, and the test was not able to distinguish between these groups, incorrectly classifying two out of 11 people in the vegetative state and four out of nine non–vegetative state participants (an accuracy of 72.7%).

The test was able to distinguish between people with traumatic and non-traumatic causes of brain injury with 82% accuracy. It correctly identified 11 out of 12 of the people with traumatic brain injury, but only seven out of 10 participants with non-traumatic brain injury.

The level of learning was also reported to be a good predictor of recovery, showing 86% accuracy at predicting whether an individual would show signs of recovery or not.

What interpretations did the researchers draw from these results?

The researchers conclude that people with disorders of consciousness may have partially preserved abilities to process information that may not be detected by behavioural assessment.

What does the NHS Knowledge Service make of this study?

This small study has shown that people in a vegetative state may be able to learn to respond to stimuli in a trace conditioning test. This research also suggests the test might be useful in predicting which individuals will begin showing signs of recovery.

However, larger studies will be needed to confirm these findings. In addition, although test results were linked to signs of recovery, the level of recovery varied, and it is not as yet clear whether the test will be able to indicate the level of recovery. As the researchers note, tests that can determine a person’s level of awareness when they cannot speak or make physical signs are needed, so research such as this is important.