Psychosis – hope in a broken system
A year-long investigation into standards of care in England for people with a diagnosis of schizophrenia or psychosis has painted a bleak picture, while also offering optimism for the future. The need for patients to be treated with compassion and kindness, and to be given hope, are the report’s leitmotifs.
The Schizophrenia Commission was established by ReThink Mental Illness in November 2011 and chaired by the psychiatrist Professor Sir Robin Murray. Among the 14 experts who helped coordinate the investigation was BPS member Dr Alison Brabban, a Consultant Clinical Psychologist and Clinical Lead in the Early Intervention in Psychosis service in Tees, Esk and Wear Valleys NHS Foundation Trust.
‘What we found was a broken and demoralised system that does not deliver the quality of treatment that is needed for people to recover,’ Professor Murray writes in his introduction to the report. ‘This is clearly unacceptable in England in the 21st century.’
The Commission drew on the input of 80 experts in the field, including researchers, patients and carers; surveyed 2500 people online; and visited services in England. Among their specific findings: fewer than one in ten people with psychosis or schizophrenia who could benefit from CBT have access to true CBT; levels of coercive treatment are up; and early intervention services are being cut back despite being judged a success.
The Commission makes a series of 42 detailed recommendations, including: overhauling acute care units and making more use of alternatives like recovery houses in the community; more shared decision making; better access to psychological therapies; extending, not cutting, early intervention services; and caution in using the label of schizophrenia, at least in the early stages of illness.
‘In my opinion the stigma of the label “Schizophrenic” negates hope in the individual,’ said one of the patients who spoke to the Commission. ‘I hope that the Schizophrenia Commission will recommend abolition of the diagnosis and encourage mental health professionals to concentrate instead on treating the symptoms of serious mental illness with compassion and without judgement.’
Above all, the Commission calls for a change in attitude: ‘People with psychosis… need to be given the hope that it is perfectly possible to live a fulfilling life after a diagnosis of schizophrenia or psychosis,’ said Murray. ‘We have no doubt that this is achievable.’
Responding to the Commission’s report, former DCP Chair Professor Peter Kinderman at the University of Liverpool said he welcomed the document but that it didn’t go far enough. ‘The Commission missed the opportunity for more thorough-going reform,’ he said.
Dr Brabban told us she was delighted with the media coverage that the Schizophrenia Commission had attracted, thereby raising ‘awareness that psychological therapies can help people with psychosis; that service users want access to them and yet for the majority they’re not available’.
She added: ‘If this leads to genuine service change with greater investment in high-quality psychological therapies for those with psychosis, it will be a wonderful achievement. What I don’t want to happen is for mental services to expect inadequately trained staff to deliver something which they call CBT but bears no resemblance to the genuine, evidence-based approach, and which therefore doesn't help anyone and ends up giving psychological therapy a bad name.’ cj
The seductive allure of the seductive allure
A pair of psychologists at the University of Pennsylvania have highlighted a delicious irony. Sceptical neuroscientists and journalists frequently warn about the seductive allure of brain-scan images. Yet the idea that these images are so alluring and persuasive may in fact be a myth. Martha Farah and Cayce Hook refer to this as the ‘seductive allure of “seductive allure”’.
According to Farah and Hook in an article in press with Perspectives in Psychological Science (PDF from tinyurl.com/cbg4du7), brain-scan images have been described as seductive since at least the 90s. Today, it has become almost a truism that brain images are so beguiling, they paralyse our usual powers of rational scrutiny. Virtually every cultural commentary on neuroscience mentions this point as a matter of routine.
Consider two recent examples. In an essay for the New Yorker, Gary Marcus wrote about the rise of neuroimaging: ‘Fancy color pictures of brains in action became a fixture in media accounts of the human mind and lulled people into a false sense of comprehension.’ Earlier in September, Steven Poole writing for the New Statesman put it this way: ‘…the [fMRI] pictures, like religious icons, inspire uncritical devotion.’
What’s the evidence for the seductive power of brain images? It mostly hinges on two key studies. In 2008 David McCabe and Alan Castel showed in a paper in Cognition that participants found the conclusions of a study (watching TV boosts maths ability) more convincing when accompanied by an fMRI brain-scan image than by a bar chart or an EEG scan. The same year, Deena Weisberg and her colleagues published evidence in the Journal of Cognitive Neuroscience that naive adults and neuroscience students found bad psychological explanations more satisfying when they contained gratuitous neuroscience information (their paper was titled ‘The seductive allure of neuroscience explanations’). Not mentioned by Farah and Hook, but also relevant, is a study covered by our own Research Digest showing that jurors are disproportionately persuaded by brain-scan evidence (see tinyurl.com/bqys6qr).
What’s the evidence against the seductive power of brain images? First off, Farah and Hook criticise the 2008 McCabe study. McCabe’s group claimed that the different image types were ‘informationally equivalent’, but Farah and Hook point out this isn’t true – the fMRI brain scan images are unique in providing the specific shape and location of activation in the temporal lobe, which was relevant information for judging the study. Second, a new study published in Public Understanding of Science this year by David Gruber and Jacob Dickerson found that the presence of brain images did not affect students’ ratings of science news stories. Finally, Farah and Hook mention research of their own and another group, as yet unpublished, involving collectively thousands of participants, which found either no effect of brain images or a minuscule effect (one of the papers has been submitted with the title ‘On the (non)persuasive power of a brain image’).
So why have so many of us been seduced by the idea that brain-scan images are powerfully seductive? Farah and Hook say the idea supports psychologists’ anxieties about brain-scan research stealing all the funding. Perhaps above all, it just seems so plausible. Brain-scan images really are rather pretty, and the story that they have a powerful persuasive effect is very believable. Believable, but quite possibly wrong. Brain scans may be beautiful but the new evidence suggests they aren’t beguiling. cj
From care to Cambridge
A group of Cambridge graduates, led by Dr Peter McParlin, a consultant child psychologist and Fellow of the British Psychological Society, is setting up a mentoring scheme to help looked-after young people with their academic studies. The ‘From Care to Cambridge’ project, launched in Westminster in October with the support of John Hemming MP and Lord Adonis, will provide regular face-to-face contact for pupils from care in the last few years of their school life, supporting them through A-levels and the university application process.
Dr McParlin grew up in care himself, and knows firsthand how the relationship with significant, nurturing people can help develop resilience and confidence. ‘After leaving care, I became homeless, and ended up sleeping rough on the streets of Liverpool aged 17. A chance friendship with some young university students encouraged me to return to education, initially to gain the school-leavers’ qualifications, and to aspire to go to university.’
‘The vast majority believe, wrongly, that most Cambridge and Oxford students come from independent schools,’ McParlin told us. ‘This can fuel the perception that those universities are “not for me”, and perhaps this is where a Cambridge-educated mentor can help. An increasing number of looked-after young people have demonstrated that they have the ability to undertake a degree course, and now is the time for more looked-after students to attend Britain’s best universities.’ js
Further information about the scheme can be obtained by contacting Peter on [email protected] or 0113 232 3928
The Office for National Statistics (ONS) has published the first of its annual reports into the state of the nation’s well-being Life in the UK 2012 (tinyurl.com/clx2ceg). The data show that people’s life satisfaction has remained broadly stable over the last decade, despite the financial effects of the recession, including higher unemployment and a reduction in real-terms income.
The average satisfaction with family life stands at 8.2 out of 10. Healthy life expectancy (average time spent in good health) has increased (63.5 years for men; 65.7 for women), so too our satisfaction with our own health. The data also show that our average well-being is related more strongly to average levels of household income, rather than to nationwide GDP.
Reflecting on the results at a press conference, David Halpern, head of the Behavioural Insight Team, told The Guardian that economic measures and subjective well-being measures do not always correlate. For instance, he said Rutland has higher average levels of life satisfaction compared with the similarly affluent Wokingham, perhaps because of environmental advantages in the former area. ‘It seems that if you can see a tree you are happier,’ he said. Community trust appears to be another factor, perhaps explaining the high levels of subjective well-being in Northern Ireland where the Troubles may have had the effect of bringing neighbours together.
The ONS has also published a ‘National well-being wheel of measures’, which allows anyone to view at a glance key statistics for different aspects of the nation’s well-being, from individual well-being to personal finance, health, relationships, and more (tinyurl.com/blwnh9s). cj
The programme to measure the UK’s well-being began in 2010 (for more background, see our News pages: April and September 2011, September 2012)
Leading the way on replication
Psychology appears to be suffering a crisis of confidence, according to the editors of a new open-access special issue of Perspectives on Psychological Science (tinyurl.com/bwfmlox). Harold Pashler and Eric–Jan Wagenmakers (at the University of California, San Diego, and University of Amsterdam, The Netherlands) say the malaise was provoked by a series of dispiriting events through 2011 and 2012, all of them documented in our news pages, including: large-scale frauds by Stapel and others; evidence in favour of psychic powers; psychologists’ unwillingness to share their data; the demonstration of ways that questionable research practices can turn any result positive; a survey showing the widespread use of such practices; and the unbecoming response of an eminent US psychologist to a failed replication of one of his seminal studies. ‘Having found ourselves in the very unwelcome position of being (to some degree at least) the public face for the replicability problems of science in the early 21st century,’ the pair write, ‘psychological science has the opportunity to rise to the occasion and provide leadership in finding better ways to overcome bias and error in science generally.’
A popular diagnosis is that psychology’s problems stem from intense academic competition, an obsession with statistical significance, and a failure to place enough importance on replication. In the first of a series of contributions responding to this predicament, Pashler teamed up with his colleague Christine Harris to challenge the notion that the problems in psychology have been overblown – an argument they say they’ve heard from ‘prominent and accomplished researchers’.
Assuming that 10 per cent of effects that researchers look for actually exist, that psychology experiments have an average power of 80 per cent to detect these real effects (which ‘likely exceeds any realistic assumptions’), and that all positive findings are published, the pair estimate that around a third of supposedly positive results reported in psychology journals could be false.
It ‘appears almost certain that fallacious results are entering the literature at worrisome rates,’ they conclude.They also challenge the idea that conceptual replications (testing the same effect via different methods) in psychology are sufficient to weed out spurious results. They say a reliance on conceptual replications actually amplifies the bias problem – failed replications are typically explained as failing to match the original research closely enough, while successful conceptual replications are seized on as providing supporting evidence. On a more positive note, Matthew Makel (Duke University) and his colleagues attempted an objective analysis of the rates of replication in psychology, finding that rates have increased in recent years. Focusing on all the articles published since 1900 in the top 100 psychology journals (based on impact factors), Makel’s team found that 1.57 per cent mentioned the word stem ‘replic’ (used as a marker for a replication attempt). Zooming in further on a random sample of 500 of these articles, 68.4 per cent were actual replications. Applying this adjustment to the total sample, the researchers estimate that the overall proportion of psychology papers that are replications is 1.07 per cent – comparable to estimates made for other disciplines. More notable still, the proportion of replications in the period from 2000 to 2009 was estimated to be 1.84 times higher than for the period 1950 to 1999.
Other findings from this analysis: 19 per cent of all replications were published in the same journal as the replicated research; over 50 per cent were published by the same research team who’d conducted the original research; lab self-replications were more likely to be successful (in fact only three out of 167 failed!); replications overall were more often successful than not; and, contrary to popular opinion, replications were relatively well cited.
Next, Marjan Bakker (University of Amsterdam) and her colleagues likened psychological science to a game in which researchers are motivated to pursue the strategy that will bring them the most statistically significant results. They lament the fact that only 11 per cent of psychology papers in one analysis used statistical power as a rationale for choosing sample size. Instead, Bakker and her colleagues say researchers tend to favour running multiple small studies because this is the approach most likely to lead to statistically significant results. From an analysis of 13 meta-analyses involving 281 primary studies across varied subfields of the discipline, they found that seven showed signs of this kind of publication bias.
Christopher Ferguson (Texas A&M International University) and Moritz Heene (Ludwig Maximilian University, Munich) challenged the idea that meta-analyses are a panacea for the problems in psychology. In particular, they say there is overconfidence in a popular method for estimating the unpublished studies in a field (Rosenthal’s fail-safe number), and that bias is introduced by researchers tending to include their own unpublished research in a meta-analysis, but not the unpublished research of other labs. ‘(T)he aversion to the null and the persistence of publication bias and denial of the same, renders a situation in which psychological theories are virtually unkillable,’ they write.
Many commentators act as though the problems afflicting psychological science are new, but Roger Giner-Sorolla at the University of Kent says the discipline went through a similar crisis 40 years ago, including anxiety about too much focus on positive results and question marks over the authenticity of Cyril Burt’s data. One part of the response back then was to encourage researchers to publish more papers with multiple studies. As a result, in the years 1976 to 1996, Giner-Sorolla notes that the Journal of Personality and Social Psychology and Personality and Social Psychology Bulletin, have increased their publication of multiple-study papers by about 50 per cent. But this isn’t the answer, Giner-Sorolla observes, as shown by the fact that Stapel published countless fraudulent papers following this multi-study format.
Oliver Klein (Universite´ Libre de Bruxelles) and his colleagues turned their attention to reporting practices around experimenter effects in psychology. This is a pertinent issue for the replication debate because Klein et al. published a failed replication of a seminal priming paper earlier this year, finding evidence that the original results may have been due to experimenters’ expectations affecting participants’ behaviour.
Klein’s team surveyed the methods sections of hundreds of articles published in the journal Psychological Science in 2005 and the Journal of Personality and Social Psychology in 2011, finding that the majority failed to provide information on: the presence or not of the experimenter in the testing area; how the study was presented to participants; and on whether the participants’ understanding of the study purpose and hypotheses were examined. Klein’s group recommend that ‘rather than attempting to eliminate all potential sources of bias by automatising experiments, experimental psychologists should instead strive to explore such sources of bias in a systematic manner’.
Next, Gregory Francis at Purdue University warned that successful replications aren’t always a good sign. While holding back from making accusations of deliberate malpractice, he targeted a study by Jeff Galak and Tom Meyvis, published in 2011, as an example of a multi-study paper containing too many successful replications to be true. Based on the effect sizes and sample sizes in the paper, Francis says that even if the null hypothesis were wrong, one would expect by chance for some of the replications to have come up with a null result. One possible explanation, he says, is that Galak and Meyvis chose to withhold replications that hadn’t turned up significant results.The pair hit back with their own contribution entitled ‘You could have
just asked: Reply to Francis (2012)’ – in which they said they did indeed have some unpublished null results, but that a meta-analysis of all their data, including the unpublished findings, provided ‘overwhelming support’ for their hypothesis (namely, that aversive experiences are recalled more negatively when there is an expectation that they will have to be repeated).
Uri Simonsohn (The Wharton School, University of Pennsylvania), the whistle-blowing nemesis of psychology’s fraudsters, also criticised the approach taken by Francis. Simonsohn accuses Francis of being biased in his own cherry-picking approach to analysing other people’s findings. Francis has now identified publicly several papers as having signs of a publication bias – but which papers did he analyse and find no bias, and how many of these clean papers were there? Simonsohn also adds that Francis was wrong to say that papers showing evidence of publication bias should be ignored. ‘At its core, the “delete-all” correction confuses statistical with practical significance,’ Simonsohn writes.
Most of the remaining papers offered various solutions to the problems currently afflicting psychology. Among these, Michael Frank (Stanford University) and Rebecca Saxe (MIT) propose that undergraduate students conduct replication attempts of recent results as part of their lab work. ‘This practice kills two birds with one stone,’ the pair argue. ‘It creates a pedagogically rich, rewarding environment for the students…while at the same time ensuring the reliability of the published literature.’ Brian Nosek et al. (University of Virginia) suggest altering practices to help researchers prioritise long-term accuracy over short-term publication rates, including: checklists of standard requirements; peer review standards focused on the soundness, not the importance, of new research; and opening up access to lab data, methods, and work flow, via researcher homepages and public sites like www.openscienceframework.org.
Providing some light relief, the pseudonymous blogger Neuroskeptic outlines the nine circles of scientific hell that await sinful scientists, from the first circle of Limbo for those who merely turn a blind eye to bad practice, all the way to the ninth circle, where those who invent data are frozen in ice alongside Satan. ‘Frozen in front of their eyes’, he writes, ‘is a paper explaining very convincingly that water cannot freeze in the environmental conditions of this part of Hell. Unfortunately, the data were made up.’ cjI
We published our own special issue on the topic of replication in May 2012. See tinyurl.com/psycho0512.
As part of the International Partnership and Mobility Scheme offered by the British Academy and the Chinese Academy of Social Sciences, one-year partnership grants are available to UK and Chinese scholars. Partnerships can include a range of activities including visits, exchanges, workshops and seminars, these should form an integral part of the programme. Applicants must be of postdoctoral or equivalent status and research in the field of humanities or social sciences. Closing date: 6 February 2013.
The Program on Negotiation at Harvard Law School offer Graduate Research Fellowships to encourage young scholars from the social sciences and professional disciplines to pursue theoretical, empirical and/or applied research in negotiation and dispute resolution. The Fellowships allow doctoral students who are writing their dissertations to be part of the PON community for one year. PhD students currently enrolled in programmes outside the US are welcome to apply. Closing date: 15 February 2013.
The Department of Health’s National Institute for Health Research invites outline proposals for primary research under the its Health Technology Assessment Programme. Topics:
12/190 Cognitive rehabilitation for people with multiple sclerosis
12/191 Sexual risk reduction interventions for patients attending sexual health clinics
12/192 Home-based health promotion interventions for vulnerable older people
12/194 Psychoeducational support for parents with personality disorders who have children with severe emotional and behavioural problems (feasibility study)
12/197 Treating bedwetting in children
Closing date: 2 May 2013.
The Oak Foundation’s Learning Differences programme supports research and activities that contribute to both knowledge about and strategies available to students (in the US public school system) who struggle in school as a result of learning differences. Further details and eligibility criteria on the website. Applications can be made at any time.
Prize-winning science blogs
Dorothy Bishop, Professor of Developmental Neuropsychology at Oxford University, was named runner-up in the inaugural UK Science Blog Prize, presented in November by the organisation Good Thinking, in association with Soho Sceptics.
Bishop’s BishopBlog presents her views on a wide range of topics, from language impairment to dyslexia, and from science communication to academic life (www.deevybee.blogspot.co.uk). Bishop, who also writes a regular column for this publication, shared the runner-up prize with Ed Yong, author of the Not Exactly Rocket Science blog.
The top prize was shared between UCL pharmacology professor David Colquhoun for his DC’s Improbably Science blog, and Suzi Gage, a PhD student in epidemiology at the University of Bristol, for her Sifting the Evidence blog. Among the other blogs shortlisted for prizes were the pseudonymous Neuroskeptic and André Tomlin’s The Mental Elf.
‘It was great to see such a strong showing by psychologists and neuroscientists on the shortlist,’ Bishop told us. ‘Ben Goldacre [one of the judges] inspired us all by likening science bloggers to the original Royal Institution, “open, smart, accessible, nerdy, free and inclusive to anyone who cares”, and many bloggers noted that their online writing attracts a much larger audience than academic papers.’ cj
See interview with Dorothy Bishop in this issue.
The people behind the pixels
Jon Sutton reports from the inaugural Annual Conference of the Society’s Division of Neuropsychology
The Division of Neuropsychology celebrated its inaugural Annual Conference on a bright November day in London, with three keynotes and three afternoon workshops covering paediatric, adult and older adult neuropsychology under the banner ‘The rise and fall of cognition across the lifespan’.
Professor Faraneh Vargha-Khadem (UCL Institute of Cognitive Neuroscience) opened proceedings with a talk on growing up without episodic memory. Fifteen years ago, she reported three cases of developmental amnesia. Bilateral hippocampal pathology due to neonatal injury had resulted in severe impairment in episodic memory, with relative preservation of semantic memory. As was apparent throughout the day, behind the specialist language were tragic, touching and often triumphant human stories.
‘I listened in class and understood everything, but a little later I could not remember anything,’ these children said. Parents reported their children were always living in the present, not bothered about the past, and no thoughts of the future. Simply put, said Professor Vargha-Khadem, they were ‘unable to do mental time travel’. Detailed neuropsychological and neuropathological profiles revealed the vulnerability of the hippocampus in terms of its response to hypoxic episodes. The extent of bilateral hippocampal atrophy beyond which amnesia occurs was estimated to be a 20–30 per cent reduction in hippocampal volume on each side relative to mean control values. But again it was the patient testimony that brought the figures to life: ‘Provided the structure is there for me, I can function very well’, Jon said. ‘It’s got its advantages. People like to be with me, and I don’t carry grudges because I can’t remember what I’m supposed to be annoyed about.’
The next talk, from Professor Barbara Wilson OBE (Oliver Zangwill Centre, Cambridge), was entirely comprised of ‘survivor stories’: the fall and rise of cognition in the words of three people who sustained brain injury. How did their rehabilitation programmes allow them to retain some cognitive function? They had worked extremely hard, that was evident, and it was interesting to hear the dawning realisation that their brains could no longer be trusted. ‘Barbara Wilson spoke to our group’, said ‘Maggie, ‘and said these words which enabled me to think and feel quite differently about my progress – “Rehabilitation is not synonymous with recovery”. Wow! That turned a huge corner for me. I understood that I was rehabilitating, not recovering, two completely different concepts that I hadn’t realised I was mistaking.’ ‘Rehabilitation is clinically and economically effective,’ Professor Wilson said. ‘It’s expensive in the short term, but cost effective in the long term. People at all levels deserve it – the main problem is getting funding, and it’s getting worse.’
Before lunch, Professor Elizabeth Warrington (University College London) vouched for the role of single-case studies in ‘carving cognition at its seams’. Anyone familiar with Broca’s 1861 portrait of motor aphasia and Wernicke’s 1874 delineation of the sensory version will be aware of the importance of learning from an individual patient. However, for more fine-grained analysis, clinical observation has tended to give way to quantitative measures. But we have seen, Professor Warrington argued, a renaissance of the case approach, in a more sophisticated form. The outcome of a particular study leads directly to further studies, within days rather than the years that group research might take. Professor Warrington led the audience through a succession of initials – KF, AB, JBR, VER – each set of letters hiding rich insight into the amazing fractionation within the memory domain. A patient with a better memory for abstract words than concrete?Check. Someone who can handle ‘system’-related terms but not animate objects? Check. There was even a patient who demonstrated better performance with a more distantly related group of words, struggling more to select countries that are nearby or associated with each other. ‘The evidence of single-case studies,’ Professor Warrington concluded, ‘takes us back to 19th-century phrenology, but with the underpinning of a developing knowledge of modular cognitive skills.’
In common with the rest of the talks though, a picture is worth a thousand words and little can bring the theory to life like seeing an earnest middle-aged man apologising frustratedly for failing
to remember what a rhinoceros is. For all the seductive allure of brain scans from neuroscience, as Professor Warrington said, ‘progress must go hand in hand with analysis of function’, and it is the neuropsychologists engaging with the people behind the pixels.
ADHD, Medication and crime
A study of 25,656 people in Sweden over four years has shown that those diagnosed with ADHD on medication committed fewer crimes than others with the diagnosis who abstained. Moreover, the same individuals who took medication engaged in about 32 per cent less criminal behaviour when they were taking medicines prescribed for ADHD compared with periods when they were drug-free. The crime reductions applied to petty crimes and more serious misdemeanours.
BPS Members can discuss this article
Already a member? Or Create an account
Not a member? Find out about becoming a member or subscriber