What's New in Neurofeedback

• Studying Hyperlexia May Unlock How Brains Read
• Hunger Sharpens Your Taste Buds
• Brain Scans Substantiate Feel-the-Pain Sentiments
• Teen brains show low motivation
• How the Brain Makes Sense of the Senses
• 'Brain fingerprinting' touted as truth meter

  All links at: http://news.yahoo.com/fc?tmpl=fc&cid=34&in=science&cat=brain_research

So begins "A Report for the Academy," a little-known story by Franz Kafka (1919) in which an ape devises an ingenious means of escaping zoo life; he imitates his captors (humans) to the point that they note the resemblance and free him. Sixty years later, a psychologist would make allusion to this fictional piece when he presented his own masterwork to the world, a scientific report that would undermine his field's credibility and limit its publications for years to come.

Herbert S. Terrace, professor of psychology at Columbia University, was an expert of primate behavior in 1979 when he presented his critique of ape sign language work in the prestigious journal "Science." Terrace was a bit of a convert to this opinion: he had started his research project in order to replicate the famed ape language study of the day (Washoe) -- in fact he secretly hoped to learn enough to be able to have nonhuman translators escort him into the bush -- but eventually he confirmed the opinion of the opposition, notably the discontinuists, those who argued that human language was unique. He now thought that he and everyone else involved in this endeavor had been duped by our clever simian cousins. Apes have no capacity for language, no competence for lawful relationships between symbols, he concluded. And like many converts, the shortcomings of this new opinion were overlooked, and he failed to realize the significance of what had actually transpired.

The earliest experiment in the behavioral sciences is known as the Forbidden Experiment. It is an investigation into extreme social deprivation and innate behavior. It goes as follows: If we deprive a child of all human contact, what will we have at the end of day? A human child, or something less? Whatever we have will be innate and unlearned, the primordial stuff from which civilized man emerged. Legend has it that the experiment was first performed in an attempt to identify our Mother Tongue, the first and original language of humankind. According to Herodotus, an Egyptian king named Psammetichus (664-610 BC) hoped to prove that Egyptians were the oldest race of men through linguistic means. His approach was novel in that it was empirical and methodical, even using a replicated sample size (two children instead of one). Two newborns were removed from their family and given to a shepherd to be brought up in isolation. The shepherd was instructed never to say a word in their presence. This way, logic dictated, when the poor children finally did utter a word, it had to be innate, unlearned, spontaneously combusted from the latent innate language in our genes. Legend has it that the first words of both children was "becos," a Phrygian word for "bread." Phrygia, a neighboring people living in modern Turkey, ruled by King Midas and Gordias (of gordian knot famed), could now claim priority over Egypt. That the first social science experiment disconfirmed the investigator's hypothesis should be noteworthy, and be a lesson for all field studies that followed.

The forbidden experiment was replicated two thousand years later. In 1211 Frederick II reared dozens of children in linguistic isolation in order to uncover what he called the language of God. But the experiment failed when all the children died (a topic central to next month's review of Harry Harlow's work). In fact the forbidden experiment has been replicated numerous times accidentally and in less controlled settings, in what is called feral children. Feral children are typically young children, often infants, abandoned by their parents to the elements. Miraculously some of these children survive and are found by an adult who returns them to civilization. Now we should realize at the get-go that extreme social deprivation may be only one of many mechanisms at play in determining these children's behaviors. For instance, some are believed to have been mentally retarded or autistic at birth, which contributed to their abandonment. Historically, there have been about 80 feral children, most found in the company of wolves (around 50 of the total), though bears, dogs, chimps, baboons, monkeys, sheep, goats, leopards, kangaroos, gazelles, ostriches, jackals, panthers, pigs, and cows each have had their turn or two. The earliest recorded instance was during Roman Times, the most recent, last month (http://www.thestar.co.za/index.php?fSectionId=129&fArticleId=356129). A modern variant on the experiment involves confined children, those who under the cruel eye of a parent or guardian are kept apart from all others.

So what does this have to do with neurotherapy?

Well, we're getting there.... Since time immemorial, we have sought to separate ourselves from the other creatures that share our world. Tool making and language have been our two-pronged attack to keep all evolutionary precursors at bay, and since the 1960s, tool-making has taken deadly hits as apes exhibit reasonably sophisticated tool fashioning (perhaps power tool-making should be the division...). So language, with its inherent complexity, remains the final, formidable barrier.

The study of linguistic behavior in apes is nearly hundred years old. Furness (1916) taught a female orang-utan to produce vocally, papa, and cup and the sound "th" over 11 months of instruction, but she died abruptly before further study could be completed. Other attempts into vocal production ran into architectural constraints: the human throat evolved to produce an array of articulated sounds, the simian throat did not (Kellogg & Kellogg, 1933; Hayes & Hayes, 1951). However all primates are facile with their hands (perhaps the defining feature of our Order), so language production moved from mouth to hand. In 1966 the Gardners trained a wild-born female chimpanzee named Washoe in American Sign Language (ASL). Washoe acquired about 100 ASL signs in three years of training and even started to combine them as well. The combination question was critical, because here is where most linguists familiar with this debate draw the line. Humans possess syntax, animals do not. As we learn more and more about all fellow mammals, we've come to realize just how many possess forms of semantic symbol systems, however slight, but combining symbols in lawful orderly relationships, ahh, there's the rub -- that is where people like Noam Chomsky and other large-brain primates would beg to differ. So although the great apes demonstrate the intellectual achievements of 2- to 3-year-old children (i.e., symbolic subperiod of the preoperations period, for those keeping score), handling temporal complexity with symbols is beyond them, or is it?

Part of this conclusion came rightly from Terrace's interpretation of his own research and others, and part came wrongly from Terrace's interpretation of his own research and others.

As I warn my students, psychological research involves concrete implementation of abstract ideas -- in other words, you have an idea, now how do you enact it without conflating it with half a dozen other forces when pencil hits paper or finger hits keyboard. Confounds can drive one's datastream better than we often imagine. We need to recognize this. We need to be sure that what we intended to happen is what actually happened. Were the forces we meant to alter behavior the dominant forces present? Too often we are seduced by our own ingenuity and conceptual elegance and fail to heed the warnings of sloppy realities. This is the heart of the Terrace Effect.

Terrace raised a common chimpanzee named Nim Chimpsky (in honor or parody of Noam Chomsky -- already the self-seduction is underway) and trained him in a controlled environment to "use" sign language. After reviewing tapes of the training sessions, Terrace concluded that Nim really didn't understand what he was saying with the signs. Rather he was mimicking the trainers' signs, he relied heavily on imitation and prompting. These conclusions raised serious questions about the results in all published chimp sign language experiments. Funding soon dried up. Science and Nature refused to publish anything more on the topic for 20 years.

The Gardners, who trained Washoe, disagreed. Nim, they claimed, was the victim of his environment, restricted to a windowless cell and trained by more than 70 teachers, many of whom lacked sign language competence themselves. Try to learn Farsi from 70 undergraduates and see how much linguistic competence you possess at the final exam. Terrace concluded that Nim performed signs solely to get rewards, without any understanding of what he was saying. But did he ever care to understand what Nim had to say? His training technique was entirely unilateral in intention, with no regard for the mental contents of Nim's mind outside of the strict confines of the conditioning paradigm. Nim never acquired a communal system of communication because no communal symbol system was offered him. It would be like giving a child a set of drums and then criticizing him for failing to play the piano, an instrument he never laid eyes on. Recent research shows that the pygmy chimp not only can acquire linguistic and syntactial competence to a great degree, they can acquire and understand English like a human child, from entire sentences to completely novel requests (Savage-Rumbaugh).

Communication requires a rapport. Appreciation for the mind you are trying to communicate with. Nim never received knew this. He had 70 enthusiastic students drilling him rotely for hours on end, not unlike the way mathematics has been taught unsuccessfully to millions for generations. Treating Nim like a fellow mental creature, with opinions and insights of his own, allowing his attention to dominate on occasion, might have revealed his true limits. Treating him like an experimental subject created limits.

So what is the Terrace Effect?

Drawing a conclusion about what you intended to do without realizing what you actually did. What did Terrace do? Did he test syntactical competence in great apes? No. Not really. He did a version of the forbidden experiment, the closest approximation allowable by today's ethical standards. He raised a hominoid in an impoverished linguistic environment and then tested him for innate lingustic abilities. What did he find out? If you ignore another's mind, they will ignore yours, regardless of how smart you assume you are and how many nifty communication tricks you may possess. My kindergartener could have told you this. (This is an important lesson for those treating autism; the autistic individual believes his viewpoint is just as valid as yours; you have to outcompete and persuade him towards a more consensual mode of being.)

And what is the relevance to neurotherapy? This is a field of mistakes, as all social and health sciences are. Mistakes in implementing our intentions mistakes in analysis, mistakes in conclusions. We hope to exercise specific actions of the brain, so we set a machine's parameters such and such, but in the end have we accomplished what we set out to do. Before we presume we have, and analyze the conclusions based on what we presume is solid evidence, we need to evaluate what exactly has transpired with our clients. How well did we train the brain? Did we assume limitations of brain function that are not there? Did we presume a model of health or dysfunction that is just a model, far from reality? In short, we always need to evaluate our assumptions, our implementations, and qualify our conclusions accordingly.

There was a recent paper that seems to have done the opposite. After a minimal amount of work, a handful of subjects, they declared supreme understanding of the hundreds of studies that went before them. Fortunately they do not have the reputation Terrace had when he stepped into the fray. Fundamental misunderstandings about EEG and/or operant conditioning can readily lead to a big name stepping into the field, running a study, finding nothing, and leaving a debris-field in his or her wake. Belief that gross SMR magnitude will increase after training (in individuals who don't show SMR deficits) is a common misconception, and one recently published in a noteworthy paper. Belief that operant conditioning works well with continuous reward -- well, you try it... get your kid to sit still by giving him candy all the time, regardless of his actions. Reward, reward, reward, reward, reward, reward. True, we need to shape to a goal state when the goal behavior is not being elicited in full, but once the goal is approximated, reward must necessarily become discrete, else non-goal behavior is also being rewarded. Reward must be focal (associatively) to lead to discrimination. If associatively diffuse, generalization occurs. Basic Skinner 101 principles.

Now maybe a big name will step into the field and get it right (or like me, get it wrong early and figure out why) and become a convert, and the field will experience renewed popularity and credibility, but we should always be careful to presume competence, especially by a seductive presence. The Terrace Effect crippled the field of ape language study for 20 years; do we want a similar fate?

Returning to that recent paper where they found nothing and claimed this is the state of the world (i.e., neurotherapy). Two individuals from a school in New Zealand trained seven children "using a standard EEG biofeedback treatment protocol designed to alter SMR/theta ratios and reduce behavioral symptomatology diagnostic of ADHD". Now, I cannot get this journal at my school nor the U of Rochester (one of the largest biomedical holdings in the world) which probably reflects on the journal, so I can only quote from the abstract: "During alternate periods they were also trained using a placebo protocol that was identical to the treatment protocol, save that the association between EEG patterns and feedback to the participants was random. " ... during alternate periods? What kind of method is this? Using this approach, I can prove any slow-acting agent will not differ from placebo. I can prove that the major tricyclics are no more ineffective than M & M's. Brain function is not easily and rapidly disheveled neurochemically, then resheveled, on a time clock. It often requires a sustained assault.

Continuing: "When overall behavioral trends unrelated to training were not controlled for and non-completers were excluded from the analysis, it could be mistakenly concluded that EEG biofeedback is significantly more effective than placebo and that the effect sizes involved are moderate to large." These results indicate that many previous reports of the efficacy of EEG biofeedback for ADHD" -- last month I included 280 peer reviewed papers, including cat, monkey, and human subjects --" particularly those presenting series of single cases, might well have been based on spurious findings."

Or could it be that the authors do not understand how EEG operant conditioning influences brain function? Or might they not understand the basics of operant conditioning? Or a hundred other things that the authors of those 280 primarily successful studies did and do?

-DK

News & Reviews NEW BOOKS

Souls: Beneath & beyond Autism
by Thomas Balsamo, Sharon Rosenbloom
Investigation of this devastating condition, from the child affected to the parents that cope with it. --www.amazon.com/exec/obidos/ASIN/0072881704/top100

Alternative Medicine and Multiple Sclerosis
by Allen C. Bowling
Information on many complementary and alternative medicines that may help manage some MS symptoms and promote general health. --www.amazon.com/exec/obidos/ASIN/1888799528/top100

Pediatric Neuroimaging
by A. James Barkovich
Imaging of neurological disorders in infants and children --www.amazon.com/exec/obidos/ASIN/078171740X/top100

Antiepileptic Drugs
by Rene H. Levy, Richard H. Mattson, Brian S. Meldrum, Khurram H. Bashir, Emilio Perucca
More information that most would ever need to know, but a handy reference for those treating epilepsy. --www.amazon.com/exec/obidos/ASIN/0781723213/top100

Integrated Treatment for Dual Disorders: A Guide to Effective Practice
by Kim Mueser, Robert Drake, Douglas Noordsy, L Fox
Guide to assist addicted clients with major mental illness. --www.amazon.com/exec/obidos/ASIN/1572308508/top100

Neurological Foundations of Cognitive Neuroscience
by Mark D'Esposito
Reviews classic neurobehavioral syndromes from neurological and cognitive scientific perspectives. --www.amazon.com/exec/obidos/ASIN/0262042096/top100

Integrated Treatment for Dual Disorders: A Guide to Effective Practice
by Kim Mueser, Robert Drake, Douglas Noordsy, L Fox
Guide to assist addicted clients with major mental illness. --www.amazon.com/exec/obidos/ASIN/1572308508/top100

Brain Injury and Mental Retardation: Psychopharmacology and Neuropsychiatry
by C. Thomas Gualtieri
Organizes scientific literature for treating TBI and MR, esp. pharmocologically. --www.amazon.com/exec/obidos/ASIN/0781734738/top100

Brain imaging studies in human addicts. : Heroin-related stimuli provokes activation of anterior cingulate and orbitofrontal regions. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14636961

Brain oscillations as functional correlates of cognitive systems: alcoholism. : Theta and delta reduction are prominent at the frontal region of alcoholics compared to control, suggesting a deficient inhibitory control and information-processing mechanism. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14693365

Complexity changes of the EEG induced by alcohol cue exposure : When subjects are exposed to alcohol cues, changes in EEG complexity are observed in frontal, right posterior temporal, and occipital areas www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14691383

Complementary and Alternative Medicine for Autistic Spectrum Disorder. : More than 30% of children were using some CAM, and that 9% were using potentially harmful CAM, though less frequently used for those children with additional diagnoses. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14671475

Peculiarity of the right-hemisphere function in depression : The right hemisphere organizes information polysemantically. In depression RH physiological overactivation may reflect unsuccessful efforts to overcome functional insufficiency. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14687851

EEG spectral power and mean frequencies in early heroin abstinence. : Frequency shifts in alpha2 (esp. frontal and central) correlated with daily heroin consumption. Slowing of alpha1 mean frequency appeared mainly in heroin addicts who abused high doses of the drug. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14687860

Brain circuits determine destiny in depression : Augments antidepressants by targeting neurotransmitters involved in commonly experienced residual symptoms. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14658930

Out-of-body experience of neurological origin. : Paroxysm at the temporo-parietal juncture in a state of impaired consciousness causes a failure to integrate proprioceptive, tactile and visual information of one's body, and thus OBEs. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14662516

Cortical abnormalities in children and adolescents with ADHD : Abnormal morphology was noted in frontal cortex in ADHD, with reduced inferior portions of dorsal prefrontal cortex and anterior temporal cortex bilaterally. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14643117

Magnetic resonance volumetric study of the temporal lobe structures in depression. : In depressed patients the temporal horns were larger compared to the control group. Nondepressed show greater hemispheric differences. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14655115

Quantitative spectral analysis of EEG in psychiatry revisited : QEEG was abnormal in 83% of 340 patients, and 12% of 67 normal subjects. Decrease in delta and theta bands were most indicative of brain dysfunction. www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=14652089

Spotlight articles

1. Brainwaves and Behavior (Part 1 of 2)
2. Brainwaves and Behavior (Part 2 of 2)
3. Fight the Power, revisited
4. Four Decades of Neurofeedback
5. Math Confused as Science
6. Mind Too Male: Autism and its Cure, A
7. Mine the Mind, Filtered or Unfiltered
8. Neurotherapy since 2000
9. Quick Lessons from Neuroanatomy
10. The Year in Neurofeedback
11. Theory of Mind Tests
12. Winter Brain 2003

Last Word

1. EEG & FMRI databases online
2. Excitation vs Inhibition
3. Exterminating Spam
4. Hemisphere Integration Tasks
5. iSNR Conference 2003
6. Keep Informed
7. Knowledge is Power -- or so we quote
8. Online Clinician Listings
9. SABA II
10. Science Daily
11. Treating Darwin

Articles in WNIN, 1998-2003 - http://start.eegspectrum.com/Newsletter/review.htm