Hearing things?

Ultrasound makes for a curious choice of weapon to attack diplomats. Timothy Leighton writes that the evidence to date is far from sound.

‘Sonic’ acoustic waves (roughly speaking, waves at a frequency people can hear) can lead to a range of adverse reactions, depending on the quality and intensity of the sound, and the attitude of the person hearing it.

Birdsong might be pleasant in a garden, but if that same acoustic energy were present in a persistent whine or as the sound of nails scratching on a blackboard, it could generate adverse effects, particularly if the sufferer believes that the sound is being maliciously projected at them from a neighbour.

Such adverse effects from audible attacks range from those arising from annoyance, anxiety and stress (for example headaches or inability to concentrate) to, at louder levels, a temporary or permanent hearing impairment. Such sonic exposures are not covert.

The additional fear when an ultrasonic exposure is suggested is that it is covert: the sufferer cannot hear the exposure and does not know when they are being exposed, and so cannot respond to exposure by moving away.

I have investigated many claims of ultrasonic exposure. It is vital that the true source of symptoms is identified, because if it is not ultrasonic, then the sufferer cannot be helped if they persist in misidentifying the cause of the problem.

More on this: Sounding the alarm on ‘directed energy’ attacks

The anecdotal symptoms of modest levels of ultrasonic exposure are diverse and include headaches, nausea, tinnitus, dizziness, an inability to concentrate, and a feeling of pressure in the ears.

There has been limited controlled scientific testing to date, in part because one cannot ethically expose children (who are likely to be most susceptible) to ultrasound in order to generate an adverse effect.

However, from the limited evidence available, the working hypothesis is that if one can hear the ultrasound, one is far more likely to suffer ill effects. In addition, the ability to hear high frequencies varies hugely between individuals. If a group of people are exposed, the effect on any individual will be unpredictable, with adult men usually the least susceptible.

This last point would make ultrasound a curious choice of weapon to attack diplomats. Furthermore, it is not a long-range weapon: ultrasound does not travel far in air. With a commercial ultrasonic ‘pest scarer’ placed in a room, only the individuals in that room, and perhaps the neighbouring room, are exposed to levels sufficient to cause adverse effects – and only in a minority of individuals.

If, therefore, someone comes to me with complaints of an ultrasonic attack, there is a process of elimination to go through. I first check for more likely causes: poisoning, illness, and sometimes the confusion of tinnitus with ultrasound. If these are ruled out, and ultrasound is present in the suspect location, I look for the source, because one cannot begin to discuss ‘attacks’ until accidental exposure has been eliminated.

I have never encountered a genuine ultrasonic attack. All those cases where the symptoms were not caused by something other than ultrasound were accidental exposures, caused by emissions from pest scarers and a host of other ultrasonic devices (for instance door openers or public address systems).

This process of elimination has not, as far as I can tell, been adequately conducted for the situations in Cuba and China. Until this is done, the reporting on these would benefit hugely from avoiding the word ‘attack’, given its emotive and political connotations.

Another word to be avoided is ‘brain’. This has spuriously entered the debate following a difficult-to-justify testing protocol which included features that dramatically increased the likelihood of participants failing medical tests.

From the initial 80 members of the embassy community who went to its medical centre in Cuba and were suspected of having been attacked in November and December 2016, 80 per cent were not selected for further study. After excluding the majority (64) who did not fit some unpublished pattern or criteria, the US authorities had a group of 16 who did fit. It is vital that more information is made available about this exclusion process, as it is very easy to make data fit a trend if this approach is taken.

After expanding the scope and time period, eight more individuals were found, and these 24 were referred to the Brain Injury Centre at the University of Pennsylvania, which ran a battery of 37 tests.

Crucially, however, they set the failure criterion of each test at the 40th percentile, meaning 40 per cent of the ‘normal’ population would, if tested, score the same or worse in any given test. In medical tests, it is more standard practice to select the 5th percentile where 5 per cent of the normal population would be expected to ‘fail’ the test.

More on this: Hacking your health: cybersecurity prevention is better than cure

Only six of the 21 people who agreed to be examined by the Brain Injury Centre completed all 37 tests. Six of these tests assessed the brain’s ‘executive function’ (the processes controlled by the frontal lobe that allow an individual to manage themselves and their resources, including working memory, self-control, flexibility in thinking, planning, and paying attention). The Brain Injury Centre decided that failure in at least one of these six tests would allow them to classify an impairment in executive function, and indeed their report stated that, for all six of the people who completed the tests, “Impairments were found in executive function”.

On the face of it, according to the Centre’s criteria, 100 per cent of those who complained of the attack and completed testing were found to have impaired executive function. However, from a statistical point of view, this is exactly what would be expected in the normal population who had not been to Cuba.

To be specific, if the failure criterion of each test is set at the 40th percentile, the chance of a normal, unimpaired individual passing that test is 60 per cent. The chance of a normal individual passing all 6 tests (assuming the results are statistically independent of one another, which may not be the case) is 4.7 per cent – giving a 95.3 per cent chance of failing at least one test and being classed as impaired in executive function.

This puts a very different perspective on the statements that all those exposed to the ‘attack’, who completed testing, were found to be impaired in their executive brain functions.

The results from the Brain Injury Centre, therefore, depend on criteria that were too easy to fail, and so reports that individuals failed tests need to be re-evaluated by others. Unfortunately, this cannot be done, as the Centre has not published the raw data.

In addition, these data also purport to show an ill effect on average more than 200 days after the exposure ceased. I have never seen a case of ultrasonic exposure where the adverse effect persisted for more than a few hours after the individual moved out of the exposure zone.

It is possible to use ultrasound to produce temporary adverse effects in an unpredictable manner in a minority of the population (particularly the young) at short ranges. However, all the evidence to date points away from there having been ultrasonic attacks on US embassies in Cuba and China.