Rigorous and intensive computer-based training of people with schizophrenia using principles of neural plasticity to master lower-level perceptual and attentional auditory and verbal learning processes is possible. The technique appears to translate into improvements in higher-level cognitive functions and possibly even enhanced quality of life.

At APA’s 2011 Institute on Psychiatric Services in San Francisco, Sophia Vinogradov, M.D., described research showing that computerized games aimed at training patients in very specific tasks can have effects on multiple interacting brain systems, resulting in changes in global cognitive functioning.

Vinogradov said the approach to brain training marks a significant new direction in what has been termed “cognitive remediation,” a field of decades-old research that has been stymied by repeated observations that while patients may improve in the short term on various cognitive tasks, the improvements are often not sustained and do not typically translate into widespread and enduring improvements in cognition or quality of life.

Vinogradov is professor in residence and interim associate chief of staff for mental health at San Francisco VA Medical Center and interim vice chair of the Department of Psychiatry at the University of California, San Francisco. At the institute, Vinogradov, received APA’s Alexander Gralnick Award for Research from past APA President and current APA American Psychiatric Foundation Treasurer Richard Harding, M.D.

Vinogradov said emerging research on cognitive training marks a departure from an older neuropsychological model, derived from studies of brain injury in the early part of the 20th century, which posited that schizophrenia and other brain disorders were marked by impairments in discrete and independent functions. “Implicit in this model is that the brain is permanently impaired and that the best you can hope for is a compensatory or work-around solution,” she said.

“I would argue that what has emerged in the last 10 or 15 years is really a ‘systems neuroscience model’ of impaired cognition in schizophrenia, one in which complex distributed neural systems are constantly interacting with each other,” Vinogradov explained, “and that what had been thought to be discrete, independent functions—such as attention or working memory or perception—are in fact intimately tied to each other and constantly interacting.”

Vinogradov said the new emerging model is based on the notion that, during successful learning, the brain represents relevant perceptual and cognitive-affective inputs with disproportionately larger and more coordinated populations of neurons that are distributed—and always interacting—across multiple levels of processing and throughout multiple brain regions.

In this way, she said, intensive training of lower-level functions—such as how the brain processes auditory or visual information or matches emotions to facial expressions—can drive changes throughout the interactive, distributed neural networks of the brain, resulting in improvements in higher-level cognitive functions and, ultimately, in real-world functioning and quality of life.

Vinogradov said this new model has less to do with “remediation”—with its implication of building compensatory mechanisms for a broken brain—and is more similar to physical fitness training: individuals who are extremely physically fit, as well as those who are much less so, can benefit from training, depending on how the training is adapted for the individual.

For a healthy brain, small amounts of training on higher-level cognitive functions may produce some lasting improvements. But Vinogradov argued that for the impaired brain, such tasks need to be broken down into component parts. She drew an analogy to how a beginning tennis player with poor coordination might learn a tennis serve by breaking it down in very specific components: hand-eye coordination, the toss of the ball, the follow-through with the swing. These components would be practiced intensively one at a time before putting them together into the complete tennis serve.

So, too, with the tasks of attention and working memory for a person with schizophrenia. “When the brain is learning new verbal and auditory information, verbal and auditory stimuli are coming through the auditory system,” she said. “We know these systems aren’t working very well in schizophrenia so we have to ‘tune up’ the brain to be a better listener and to be better at attending to and representing bits and pieces of auditory and verbal information and holding them in working memory. In this way, we can drive improvements in these distributed neural systems at every level, by helping to clean up the signal in an otherwise noisy system.”

Vinogradov emphasized that brain training using these principles of neuro-plasticity must be rigorous and intensive, adapted to the individual level of skill, and provide sufficient reward to engage and motivate subjects.

In a study published in the American Journal of Psychiatry in 2009, 55 clinically stable schizophrenia subjects were randomly assigned to either 50 hours of computerized auditory training or a control condition using computer games. Those receiving auditory training engaged in daily computerized exercises that placed implicit, increasing demands on auditory perception through progressively more difficult auditory-verbal working-memory and verbal-learning tasks.

Relative to the control group, subjects who received active training showed significant gains in global cognition, verbal working memory, and verbal learning and memory. They also showed reliable and significant improvement in auditory psychophysical performance; this improvement was significantly correlated with gains in verbal working memory and global cognition.

“These gains may be due to a training method that addresses the early perceptual impairments in the illness, that exploits intact mechanisms of repetitive practice in schizophrenia, and that uses an intensive, adaptive training approach,” Vinogradov and colleagues wrote.

Most promising, she presented evidence that improved cognition from this kind of training was correlated with increases in quality-of-life scores, six months after training.

Moreover, she said that brain-derived neurotrophic factor (BDNF) may serve as a biomarker of cognitive improvement. In a paper published in Biological Psychiatry in 2009, schizophrenia subjects who engaged in computerized cognitive training designed to improve auditory processing showed significant cognitive gains and a significant increase in serum BDNF compared with subjects who played computer games. This increase was evident after two weeks of training, and after 10 weeks in the active condition, subjects ‘normalized’ their mean serum BDNF levels, whereas the control group showed no change.

Much of the research has been with patients who have been ill for many years, and Vinogradov presented preliminary evidence that such brain-training techniques may be especially useful for patients much earlier in the disease process.

“The field of schizophrenia research has been characterized by nihilism, the idea that the brains of people with schizophrenia are irreparably broken,” Vinogradov said. “But I would like to argue that the brain is not immutably fixed, and that even in people with schizophrenia these neural systems show a high degree of plasticity and can change.”

“Using Neuroplasticity-Based Auditory Training to Improve Verbal Memory in Schizophrenia” is posted at <http://ajp.psychiatryonline.org/article.aspx?articleID=100952>. An abstract of “Is Serum Brain-Derived Neurotrophic Factor a Biomarker for Cognitive Enhancement in Schizophrenia?” is posted at <www.ncbi.nlm.nih.gov/pubmed/19368899>.

• Intensive, computer-based training of lower-level perceptual and auditory/verbal working memory processes can drive improvements in higher-order cognitive functions that are associated with enhanced quality of life at six-month follow-up.

• Training must be rigorous, intensive, keyed to individual ability, and provide sufficient reward to keep patients engaged and motivated.

• The new model marks a departure from an older neuropsychological model that conceptualizes brain dysfunction as the result of impairments in discreet independent processes, and instead draws on a “systems neuroscience model” of constantly interacting distributed neural systems.

• Results may be especially promising for patients very early in the disease process.

• A more elaborate form of these exercises, embedded in an engaging game, will be studied in an NIMH-funded multisite randomized controlled trial next year. If successful, this will lead to widespread dissemination of this form of cognitive training for schizophrenia patients.