Tag Archives: Catalyst (TV show)

Supermatchers super-recognizers, same thing, isn’t it?

Passport Problem. Catalyst. ABC. February 24th 2015. http://www.abc.net.au/catalyst/stories/4185916.htm

Why are the science journalists at the Catalyst team trying to distance this story from the existing body of research and writing on the same subject as the story, by using a new term for people with tested elite ability in face recognition? As far as I can tell, the skill is the pretty-much same as the skill measured by the CFMT, which I believe is recognized by researchers around the world as a gold-standard test of face memory or face recognition.

Story on Catalyst makes many references to cross-sensory experiences in taste and smell

This is one of those pop science journalism media stories that make me think that everyone must be a synaesthete to some degree:

http://www.abc.net.au/catalyst/stories/4145918.htm

News on facial recognition technology of the last week

http://www.news.com.au/technology/facial-recognition-developed-in-brisbane-on-x6-spy-glasses-us-military-queensland-police-interested/story-e6frfrnr-1226979586776

I briefly noticed a news story on TV a few days ago which featured a piece of facial recognition technology worn as glasses (like Google Glass) which included big claims about being able to identify a face from a database of thousands of persons of interest. I think the news story linked to above is that story, about X6 spy glasses, Osterhout Design Group, Defense Intelligence Agency (USA Govt) and Dr Brian Lovell, a professor at the University of Queensland and CTO at Imagus Pty Ltd. Dr Lovell has a very impressive CV but all the same I was not impressed in 2012 with the way he lightly dismissed the capabilities of human face recognition in an appearance on the Catalyst science TV show from the ABC. I’ll be impressed by the X6 technology when it is used and tested in real life applications. The period of time that human facial recognition capabilities have been used and tested in real life applications is measured in millions of years, so I think the artificial versions of facial recognition might have a bit of catching up to do.

http://www.usnews.com/news/articles/2014/07/08/fbi-may-seek-facebook-data-for-facial-recognition

http://inthecapital.streetwise.co/2014/07/08/how-facebook-succeeds-at-facial-recognition-while-the-fbi-struggles/

http://www.theverge.com/2014/7/7/5878069/why-facebook-is-beating-the-fbi-at-facial-recognition?utm_source=digg&utm_medium=email

On a similar theme I have found some interesting recent news stories comparing the facial recognition databases of Facebook and the FBI. It looks like Facebook’s DeepFace is superior in important ways compared to the FBI’s Next Generation Identification. I think it is important to consider the value of such databases, as well as human face recognition, as tools in the investigative process rather than as producers of forensic evidence, and for sure these technological facial recognition databases can use (memorize?) huge volumes of images. But regarding the actual process of face recognition, I’m still to be convinced that there is any technology that can do what humans, including human super-recognizers, can do. As Russell Brandom wrote in The Verge “While there are plenty of contractors who are willing to promise “near-human” recognition capabilities, real facial recognition is much harder than the industry lets on.”

Story about facial recognition on ABC’s Catalyst tonight left me wondering where the rest of the story was

It seems very odd to me that this television story from Australia’s ABC about face recognition technology and its possible applications to forensic/law enforcement work opened with some compelling images of the 2011 England Riots, but failed to mention the apparently successful application of human face recognition specialists (super-recognizers) by the Metropolitan Police in London to interpreting CCTV images of those riots. Superrecognisers weren’t mentioned at all in the Catalyst report, an important omission, and the value of human face recognition was dismissed lightly by Professor Brian Lovell from the University of Queensland. Prof. Lovell’s areas of expertise are electrical engineering and computer science, so one might not expect him to know a lot about human face recognition. The world’s leaders in research in human face memory and face recognition are generally psychologists. The reporter Anja Taylor chose to focus solely on Australian interviewees and argued that “new facial recognition technology promises crime fighters their greatest gift.” There seem to be a number of contradictions between reporting by Catalyst and media in the UK about the forensic use of CCTV images of the English riots. Catalyst’s technophilic and Austro-centric perspective is a questionable treatment of the subject but I’m not surprised, as I’ve been viewing Catalyst for long enough to realise that it is primarily a bit of light entertainment from the ABC that gives Australian researchers opportunities promote their latest work. I found it quite amusing that the journalist presented face recognition tasks which average humans do rountinely and with ease, tasks such as identifying faces in poor images and identifying a moving face while tracking it, as miracles of technology when a computer system performs these tasks.

It appears that Australia doesn’t have science journalists working in the electronic media, we just have a small group of slick and entertaining radio and TV presenters who have lots of contacts in science and academia. Don’t we deserve better?

May 5th 2012 – Maybe I shouldn’t be quite as harsh about the standard of journalism at Catalyst. I’ve been watching a repeat of last week’s episode of Catalyst, and the report by Anja Taylor about the phenomenon of mass tree deaths and diseases in response to rising temperatures and drought seemed to be excellent work on a very important issue. I can’t claim much knowledge of the subject, but I thought it was good. Taylor interviewed not just one or two, but many different scientists working in various areas of ecological science, and I think it might just be the case that more effort in this report resulted in a better piece of journalism. I just wish Catalyst would stop producing journalism that isn’t done with as much care and effort.

Facial Recognition. Catalyst. Reporter: Anja Taylor. 3 May 2012. http://www.abc.net.au/catalyst/stories/3494034.htm

New study with scientifically sound test finds people with Asperger syndrome vary greatly in face recognition ability – can face recognition be a savant ability?

Face recognition performance of individuals with Asperger syndrome on the Cambridge face memory test.
Darren Hedley, Neil Brewer, Robyn Young
Autism Research.
Article first published online: 24 AUG 2011
DOI: 10.1002/aur.214    http://onlinelibrary.wiley.com/doi/10.1002/aur.214/abstract

This article is still in press, but was published online last month. This study had 34 subjects with Asperger syndrome and 42 nonautistic controls. It appears that while around a quarter of the study subjects with Asperger syndrome in this study have prosopagnosia (as defined by a test score from 2 to 3 SD’s below normal or mean), Asperger syndrome (a type of autism) is also not inconsistent with superior performance on the Cambridge Face Memory Test (CFMT), which is the best test of face recognition that I know of. Some features of the study that perhaps distinguish it from other studies of face recognition in autism are the use of a reliable and valid test of face recognition (Duchaine & Nakayama 2006), the study of face recognition in adults, not children, which is really important because face learning ability appears to be a skill that peaks at a surprisingly late stage in the life-span, into the third decade (Germine, Duchaine & Nakayama 2011), and if autistic people have a delay in development that could potentially affect or bias the results of studies of children and youths.

I was particularly interested in looking at the data for individual study subjects, but for some reason, in journal papers this seems to always be included in a miniature table that is either unreadable or unprintable, or both, or is absent altogether. Persevering, I was interested to find that none of the non-autistic study participants got a score in the prosopagnosia range, while eight out of 34 of the Asperger participants did, so there seems to be a definite association between having prosopagnosia and having a diagnosis of Asperger syndrome (AS). But at the other end of the spectrum of ability, the top score of the whole study, which I think must have been close to the super-recognizer range at 1.75, was achieved by a participant with AS, and there were two other with AS who got great scores. It appears that a fair proportion of those with AS got close to average scores. To the naked eye, it appears that there is a greater variation in face recognition ability in those with AS than in the normal adults tested. Why? Should we just accept this as a brute fact about AS, or should we look for special explanations for the top or bottom achievers in the AS group?

Is there something special about the top performers in the AS group? Given that there appears to be a link between autism and synaesthesia, and synaesthesia appears to be sometimes associated with savant-like superior ability in specific sensory or cognitive tasks (Banissy et al 2011) (Banissy, Walsh & Ward 2009) (Baron-Cohen et al 2007) (Simner, Mayo & Spiller 2009), and the association between savantism and autism is generally accepted, should we then ask if the three top-performing participants with AS might be synaesthetes who also have AS? If these connections are found in reality, should we then include superior face recognition (“super-recognizers”) among the many varied areas of mental performance that are regarded as savant skills and abilities? I have already discovered in a 2010 study of face recognition in the broader autism phenotype (BAP) one CFMT score from a father of an autistic child that appears to be close to a super-recognizer level of performance (Wilson et al 2010), as defined as two or more SD’s above the mean. I would have thought that this isn’t what researchers would expect to find in studies of autism or the BAP which use study participants who aren’t selected for any particular level of face recognition ability.

I’ve got to wonder whether people (children?) whose main social disability is prosopagnosia have been clumsily lumped into the category of autism. It appears that over three-quarters of the autistic subjects did not have a “severe face recognition impairment”, so we certainly can’t say that a severe impairment is typical of the group of people who have Asperger syndrome (AS) in this study, and my reading of the “enhanced perceptual functioning model” of autism seems to suggest that autistic people should have an advantage at visual tasks (Samson et al 2011). We know that prosopagnosia is a fairly common but not well recognized disability, and that the diagnosis rates for things like AS and autism have been climbing steadily for a long time. The question of why this has happened is one that has provoked huge controversy – is there a genuine increase in autism rates, or are more and more people being placed into the category, due to lower thresholds of “severity” required for a diagnosis, or the category of autism indiscriminately devouring other categories of people, such as the intellectually disabled and other uncommon or rare disabilities?

The possibility that prosopagnosics can be (incorrectly?) identified as cases of autism was demonstrated in a story about prosopagnosia from the Australian science television series Catalyst which was broadcast in 2007 (see link below). An anecdote about two children in a family which was later found to have members with developmental prosopagnosia, who had previously been diagnosed with autism, was recounted by a prosopagnosia researcher from Macquarie University and dramatized on the show. I should point out that neither of the face perception tests shown in this story are the CFMT. One face recognition test shown in the Catalyst story uses the faces of famous people and it relies upon the person being tested already knowing about the famous person and being able to give a name for the famous face, two tasks which are not face recognition, so as a test of face recognition it is far from pure and perfect.

References

Banissy, Michael J., Garrido, Lucia, Kusnir, Flor, Duchaine, Bradley, Walsh, Vincent and Ward, Jamie Superior Facial Expression, But Not Identity Recognition, in Mirror-Touch Synesthesia. Journal of Neuroscience. February 2, 2011, 31(5):1820-1824. doi:10.1523/JNEUROSCI.5759-09.2011  http://www.faceblind.org/social_perception/papers/Banissy11JN.pdf       http://www.jneurosci.org/cgi/content/abstract/31/5/1820

Banissy, Michael J., Walsh, Vincent & Ward, Jamie Enhanced sensory perception in synaesthesia. Experimental Brain Research. 2009 Jul;196(4):565-71. Epub 2009 Jun 17. http://www.springerlink.com/content/406581u3507un270/   http://www.ncbi.nlm.nih.gov/pubmed/19533108

Baron-Cohen S, Bor D, Billington J, Asher JE, Wheelwright S and Ashwin C. Savant memory in a man with colour form-number synaesthesia and Asperger syndrome. Journal of Consciousness Studies. volume 14, number 9-10, September-October 2007, p. 237-251.  http://www.imprint.co.uk/jcs_14_9-10.html

Duchaine, Brad & Nakayama, Ken The Cambridge Face Memory Test: Results for neurologically intact individuals and an investigation of its validity using inverted face stimuli and prosopagnosic participants. Neuropsychologia 44 (2006) 576–585.
http://visionlab.harvard.edu/members/ken/Ken%20papers%20for%20web%20page/137neuropsychologiaDuchaine2006.pdf

Face blindness. Catalyst. ABC. broadcast 19/07/2007  http://www.abc.net.au/catalyst/stories/s1982889.htm  (This story showed face recognition testing at Macquarie University and includes a small sample of the tests which viewers can try)

Germine, Laura T., Duchaine, Bradley, Nakayama, Ken Where cognitive development and aging meet: Face learning ability peaks after age 30. Cognition, Volume 118, Issue 2, February 2011, Pages 201-210http://www.sciencedirect.com/science/article/pii/S0010027710002611

Hedley, Darren, Brewer, Neil, Young, Robyn Face recognition performance of individuals with Asperger syndrome on the Cambridge face memory test. Autism Research. Article first published online: 24 AUG 2011         DOI: 10.1002/aur.214    http://onlinelibrary.wiley.com/doi/10.1002/aur.214/abstract

Samson, Fabienne, Mottron, Laurent, Soulieres, Isabelle & Zeffiro, Thomas A. Enhanced visual functioning in autism: an ALE meta-analysis. Human Brain Mapping. Article first published online: 4 APR 2011 DOI: 10.1002/hbm.21307 http://onlinelibrary.wiley.com/doi/10.1002/hbm.21307/abstract

Simner, Julia, Mayo, Neil, Spiller, Mary-Jane A foundation for savantism? Visuo-spatial synaesthetes present with cognitive benefits. Cortex. Volume 45, issue 10, November-December 2009, Pages 1246-1260.
http://www.cortexjournal.net/article/S0010-9452(09)00221-4/abstract

Wilson CE, Freeman P, Brock J, Burton AM, Palermo R Facial Identity Recognition in the Broader Autism Phenotype. PLoS ONE 2010 5(9): e12876. doi:10.1371/journal.pone.0012876
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0012876