Day: May 5, 2016

“Sea hero quest” – adding signal or noise to research?

Shibley Rahman 0 Comments

My tweet on this – including Hilary Evans’ excellent explanation – is as follows.

Sea Hero Quest was launched today by Alzheimers Research UK. It has been developed by London based studio Glitchers and follows a captain as he tries to recover his father’s lost memories. You navigate the boat but there’s no mini-map within the level, you’re given all the locations to visit at the start of the game from a map.

The link to the app is here.

Here’s a typical screenshot from the game.

screen322x572

I downloaded the game for free onto my iPad. It was easy to get into – I quite liked the option also of not sharing the data from my experiments too, but the notion of contribution to a giant experiment through ‘gamification’ is also extremely appealing to many.

I really don’t see how X minutes of playing the game can equate to Y minutes of ‘traditional research’ – but I am not in the least bit bothered about that.

The game was not what I expected it to be. Searching for objects in a task in humans is well known to people who experiment with primates – it’s officially known as a spatial working memory task.

The ‘official’ line on the task is that it somehow taps into the function of the ‘hippocampus’ – a part of the brain near the ear, in the ‘temporal lobe’, so called because it looks like a sea horse (it is claimed) in brain sections. It’s a pretty robust finding since about the late 1970s that the hippocampus is one of the very first parts of the brain to be affected in Alzheimer’s disease.

Alzheimer’s disease is THE most common type of dementia. It strictly speaking is a syndrome of different diseases of varying causes, but typically clinically is manifest as problems early on in learning and memory.

A good example of this, in fact, is in the film ‘Still Alice’, where the person who develops dementia notices early on navigating for a jog around a university campus familiar to her. And indeed it’s a fairly consistent finding that spatial memory, and learning lists of things and names can be difficult for someone relatively newly diagnosed with Alzheimer’s disease.

This task has some validity from the animal literature. In rats swimming to submerged landmarks, the Morris water maze task, it’s well known that if you get rid of the hippocampus with physical destruction rats have real difficulty with this task.

Also, if you chemically manipulate the hippocampi of the rats’ brains with an agent which chemically blocks the way in which memory is laid down, known as ‘long term potentiation’, rats can’t do the task.

There are therefore different types of error – known as ‘between search’ and ‘within search’ errors.

Within-search errors were made if a participant returned to an already searched box; between-search errors occurred if a participant returned to a box that was already known to contain a target item.

There is a long and proud history of research into this, such as this paper.

The problem is – this is vastly to oversimplify the situation. If you extend the time window of the task, it no longer is a task of working memory, but more like a task of a ‘episodic’ memory for events. And the brain area really involved in spatial working memory has a large contribution from the frontal lobe, the part of the brain near the front. We know that this part of the brain tends to be affected later on in Alzheimer’s disease.

Still with me?

Well done.

There’s a problem also in the programmers’ use of the term ‘dementia’. Dementia is an umbrella term of well over 100 different conditions. Alzheimer’s disease is one of many conditions. But it is conceivable that other dementias may provide different patterns of behaviour on the task.

This is of course no bad thing – as ideally you should want to match up the different cognitive profiles of people playing the game with known diagnoses, such as likely normal ageing, or not.

But this is beyond the scope of the game. And a clear demarcation is made between playing the game for fun, and it NOT being used as a diagnosis of dementia.

But say you have problems in a different part of the brain, such as the parietal cortex. This part of the brain is involved in visual perception, so persons with the posterior cortical atrophy type of dementia might exhibit a different profile altogether.

And why this is relevant is all to do with how space is represented in the human brain.

Space which is allocentrically represented, i.e. objects in relation to each other, are more likely to be expressed in a cognitive map involving the hippocampus. On the other hand, space which is egocentrically related, i.e. related to YOU in space, is more likely to be involved in a map involving the parietal cortex.

As a summary of where these parts of the brain are, see this diagram.

brain

So – it’s likely that behavioural performance on the task will involve performances from a hotchpotch of people, including some people with undiagnosed dementia. There’ll be some people who do badly on the task who don’t have dementia, so might be needlessly worried.

On the other hand, the game is brilliant for people to feel as if they’re participating research, without leaving the comfort of their own iPad or iPhone. And it’s great branding for the people involved in its creation and as a media campaigning tool.

I think the task itself is not a good computerised task. I worked in the lab which developed CANTAB in Cambridge under Prof Trevor Robbins, so there’s a bit of a conflict of interest. I found the music quite irritating actually, which is not fatal for motivation. But I think there is a major motivational confound in this task, where you can simply get bored of it the longer you play it for, affecting performance.

The lack of instructions was potentially a good thing – in that it minimised the confound of language problems affecting task performance. There are also some groups of patients who might be rather oversensitive or insensitive to rewards of any sort – such as people who are impulsive narcissistic at one end, or people who are profoundly depressed at the other end.

It’s also not a good pure test of memory either – there’s a huge perceptual component, evidenced by the great graphics. Also, there is a strong attentional component.

The attentional component is quite interesting though. In 1979, a paper in the Science journal first put the attentional part of the brain on the map, the so-called ‘cholinergic’ chemical system – which is why people are so interested in drugs such as donepezil today (donepezil is an example of a cholinesterase inhibitor.)

I think what would be interesting is if all the hundreds of thousands or millions of performances could be put into a giant database, and some sort of Big Data analysis could be run on it to see if different patterns of performance could be fractionated out. What would THEN be interesting for Dr Hugo Spiers and colleagues from UCL would be then to match up these profiles with known ‘diagnoses’, such as early Alzheimer’s disease, ageing, posterior cortical atrophy.

But this would be to get into the mire of ethical research, research approval and valid consent, perhaps?

I don’t wish to paralyse the greatness of this game by overanalysis. Go and download it, and help dementia research! Well done to the Alzheimer’s Research UK for an excellent collaboration which is very interesting indeed.

The human right not to be drugged up to the eyeballs

Shibley Rahman 0 Comments

supertanker

I remember once being told that an “alcoholic” is uomebody who drinks as much alcohol as you, but somebody you don’t particularly like.

Culture has an odd way of inflicting labels on people it doesn’t particularly respect. This can be done for entirely political reasons – a need to place a ‘them against us’ division, like the term ‘benefit scroungers’ for people who cannot find gainful employment.

I remember when I went to the Alzheimer’s Show last year, and a senior consultant in psychiatry gave a talk on ‘challenging behaviours and BPSD’. I challenged him at the end for whom did he deem the behaviour challenging – for the person with dementia distressed, or for the professional who wanted a quiet shift?

I then shortly afterwards went to an evening of a drug company sponsored CPD session on frailty. The GP presenting the case study kicked off with the line ’83 y.o. demented F’. I asked the presenter to justify the use of the word ‘demented’ in his case study. He informed me it was about as pejorative, and equally inoffensive, as saying ‘wheelchair user’.

I don’t want to enter into a discussion about normative morality. That is way beyond the scope of this blogpost. But there comes a time when you simply can’t cope with the deluge of cultural impropriety.

Two events happened back to back only yesterday. For example, a friend of mine, in an otherwise good article in a tabloid, was called ‘Dementia sufferer’ in the title. A press release about a forthcoming BBC programme on cures and dementia was littered with the word ‘sufferer’.

But I have recently been doing a lot of work in education of dementia, and it shocks me how entrenched the BPSD dogma is. I simply don’t want teaching I am involved in with to perpetuate knowingly this misfeasance.

BPSD

‘Person-centred care’ is a much used and abused term. But dealing with failures in communication with a person who lives with dementia but who is distressed is a more ethical (and cheaper) way of behaving, other than pimping money out at Big Pharma.

This extract appeared in a learned journal in 2014. Al Power has not been afraid to put his head above the parapet, for example here.

An individual who’s received a diagnosis of dementia has a right to be angry, agitated, anxious or depressed at conscionable behaviour from a carer or professional, for example. An individual who’s received a diagnosis of dementia is therefore entitled to feel emotions if provoked.

What this situation does not give professionals a right to do is to drug innocent people with dementia always to prescribe antipsychotics inappropriately as a ‘chemical cosh‘.

In parallel to the human right to live with dignity, found in the European Convention of Human Rights which we can all petition over in England and Wales if we are a signatory, irrespective of the Human Rights Act (1999), there’s an expectation that patients of the NHS are not subject to assault and battery from inappropriate medications.

Whilst many of us are saddened by the lack of clarity in the current Government assaulting the legislation over human rights, we also take some comfort in devolution. Scotland is a clear example of an administrative style in one country. Manchester has been another. Maybe with the election of Sadiq Khan in London today we may have another example of a devolved infrastructure, which can distance itself from Westminster, and act as a nestbed for the integration of health and social care?

Be in no doubt – it takes a long time to alter the directions of supertankers. Dementia Alliance International has been campaigning hard on this for a few years now, and as the only influential international stakeholder group run by people with dementia, many of us wish them well.