Training computers to understand the human brain

Understanding how the human brain categorizes information through signs and language is a key part of developing computers that can ‘think’ and ‘see’ in the same way as humans. Hiroyuki Akama at the Graduate School of Decision Science and Technology, Tokyo Institute of Technology, together with co-workers in Yokohama, the USA, Italy and the UK, have completed a study using fMRI datasets to train a computer to predict the semantic category of an image originally viewed by five different people.

The participants were asked to look at pictures of animals and hand tools together with an auditory or written (orthographic) description. They were asked to silently ‘label’ each pictured object with certain properties, whilst undergoing an fMRI brain scan. The resulting scans were analysed using algorithms that identified patterns relating to the two separate semantic groups (animal or tool).

After ‘training’ the algorithms in this way using some of the auditory session data, the computer correctly identified the remaining scans 80-90% of the time. Similar results were obtained with the orthographic session data. A cross-modal approach, namely training the computer using auditory data but testing it using orthographic, reduced performance to 65-75%. Continued research in this area could lead to systems that allow people to speak through a computer simply by thinking about what they want to say.

What number is halfway between 1 and 9? Is it 5 — or 3?

Ask adults from the industrialized world what number is halfway between 1 and 9, and most will say 5. But pose the same question to small children, or people living in some traditional societies, and they’re likely to answer 3.

Cognitive scientists theorize that that’s because it’s actually more natural for humans to think logarithmically than linearly: 3⁰ is 1, and 3² is 9, so logarithmically, the number halfway between them is 3¹, or 3. Neural circuits seem to bear out that theory. For instance, psychological experiments suggest that multiplying the intensity of some sensory stimuli causes a linear increase in perceived intensity.

In a paper that appeared online last week in the Journal of Mathematical Psychology, researchers from MIT’s Research Laboratory of Electronics (RLE) use the techniques of information theory to demonstrate that, given certain assumptions about the natural environment and the way neural systems work, representing information logarithmically rather than linearly reduces the risk of error.

Caffeinated Coffee Linked to Loss of Vision

Consumption of caffeinated coffee is associated with an increased risk of developing exfoliation glaucoma, the leading cause of secondary glaucoma worldwide, according to a new study led by Dr Louis Pasquale of Brigham and Women’s Hospital in Boston.

Babies Learn the Smell of Mum

Researchers show for the first time that a mammal begins to suckle its mother’s milk through a learned response built on learning her unique combination of smells. When it is born, the newborn is exposed to the smell of its mother’s amniotic fluid and the baby then responds to those smells to feed.

Prevailing thought has been that pheromones –chemicals that trigger an innate behavior – drove the suckling response as an automatic behavior. The new work determines that, in mice, the smells must be learned before the behavior can occur.

Suckling is a critical step for survival in mammals, which are defined by giving birth to offspring that need to feed from their mother’s milk. The newborn must begin to feed soon after birth or it will die. It is a crucial, defining behavior in mammals and offers researchers an opportunity to investigate the biology of instinct.

Chewing Ability Linked to Reduced Dementia Risk

The population is aging, and the older we become the more likely it is that we risk deterioration of our cognitive functions, such as memory, decision-making and problem solving. Research indicates several possible contributors to these changes, with several studies demonstrating an association between not having teeth and loss of cognitive function and a higher risk of dementia.

One reason for this could be that few or no teeth makes chewing difficult, which leads to a reduction in the blood flow to the brain. However, to date there has been no direct investigation into the significance of chewing ability in a national representative sample of elderly people.  

Now a team comprised of researchers from the Department of Dental Medicine and the Aging Research Center (ARC) at Karolinska Institutet and from Karlstad University in Sweden have looked at tooth loss, chewing ability and cognitive function in a random nationwide sample of 557 people aged 77 or older. They found that those who had difficulty chewing hard food such as apples had a significantly higher risk of developing cognitive impairments. This correlation remained even when controlling for sex, age, education and mental health problems, variables that are often reported to impact on cognition. Whether chewing ability was sustained with natural teeth or dentures also had no bearing on the effect.

The results are published in the Journal of the American Geriatrics Society (JAGS).

A molecular scissor related to Alzheimer’s Disease

An international research team led by the Spanish National Research Council (CSIC) and researchers from Kiel University revealed the atomic-level structure of the human peptidase enzyme meprin ß (beta). The enzyme is related to inflammation, cancer and Alzheimer’s Disease and is involved in cellular proliferation and differentiation. The knowledge of the enzyme structure will allow for the development of a new medication type different from those known up to now. The study was published in the current issue of the journal “Proceedings of the National Academy of Sciences”.

“Now that we know how meprin ß looks, how it works and how it relates to diseases, we can search for substances that stop its enzyme activities when they become harmful”, explains Xavier Gomis-Rüth, researcher at the Molecular Biology Institute of Barcelona, who led the project. Meprin ß is an enzyme that is anchored in the outer wall of cells. Its normal function in the human metabolism is to cut off certain proteins, e.g. growth factors, that are also anchored in the cell wall. In this way meprin ß releases protein fragments into the environment surrounding the cells – a natural and normal process, as long as it occurs at a certain intensity. However, under specific circumstances, meprin ß may function abnormally, and, for example, releases too many protein fragments. The protein pieces than overdo their natural task in the cell surroundings, causing disorder in the human body. Such disorder typically occurs when inflammation, cancer or Alzheimer’s Disease get started.

NYU researchers find electricity in biological clock

Biologists from New York University have uncovered new ways our biological clock’s neurons use electrical activity to help keep behavioral rhythms in order. The findings, which appear in the journal Current Biology, also point to fresh directions for exploring sleep disorders and related afflictions.

“This process helps explain how our biological clocks keep such amazingly good time,” said Justin Blau, an associate professor of biology at NYU and one of the study’s authors.

Blau added that the findings may offer new pathways for exploring treatments to sleep disorders because the research highlights the parts of our biological clock that “may be particularly responsive to treatment or changes at different times of the day.”