Computation in the brain is more complex than previously known; An argument against altruism in business; An optical illusion may help with treatment of OCD; Google's DeepMind team brings insight to the brain's internal reward mechanism; and a Steem proposal by @openseed that offers a step towards "tokenizing the web"
Straight from my RSS feed | Whatever gets my attention |
Links and micro-summaries from my 1000+ daily headlines. I filter them so you don't have to.
First posted on my Steem blog: SteemIt, SteemPeak*, StemGeeks.
- Hidden Computational Power Found in the Arms of Neurons - Beginning in the 1940s and 1950s, a picture of the brain emerged where relatively "dumb" neurons are connected by massive numbers of synapses that perform all of the information-processing that's needed for cognition. In recent decades, however, neuroscientists have started to develop a more complicated understanding. Albert Gidon, first author of a January 3 paper in Science, says that in fact, "I believe that we’re just scratching the surface of what these neurons are really doing". Under the original model, a neuron can function as a logical AND gate, which means that networks of neurons can perform any computation. However, after a 1969 book by Marvin Minsky and Seymour Papert claimed to prove that a single-layer artificial networks could not perform an exclusive OR (XOR), it was widely believed to be similarly impossible for a biological neuron to perform an XOR. Starting in the 1980s, though, neuroscientists started to make observations of compartmentalized signals that conflicted with this so-called point-neuron hypothesis. To explain this, scientists looked at the dendrites, branch-like structures that exist in the neuron. What they learned was that with the inclusion of dendrites, a single neuron can be seen as a multi-layered network, which means that complicated computations, including XOR, can be performed within a single neuron. In the recent paper, the authors took this observation to yet another level, noting that even a single dendritic compartment can perform complex computations, which means that some neurons can represent more than just one "integrative system." To build on this knowledge, researchers now plan to look for evidence of this same signalling behavior in animals, map the observed signalling to actual real-world behaviors, and explain why some neurons do act according to the point-neuron hypothesis while others are have the more complex structures that are now being observed.
- Not So Good After All? Don't Let 'Altruism' Kill Your Company - This article by Yegor Bugayenko, CEO of Zerocracy explains why his company discourages altruism among its employees. He begins by explaining the difference between transactional management and laissez-faire management. In the first, employees are rewarded in proportion to their productivity. In the second, sharing, nurturing, and inspiring are emphasized. He says that the latter management style has gained intellectual support recently, but cites a couple studies that found it to be problematic, causing high levels of stress and conflict while reducing productivity. Finally, he argues that businesses are better off with transactional management because even poor employees can course-correct with proper incentives. In contrast, under laissez-faire management, the article suggests that the absence of incentives to align productivity with compensation will cause high performing employees to become disinterested and leave the firm or throttle-down their performance. Additionally, the former leads to a "fail-fast" situation that enables companies to improve or adapt. On the other hand, the second leads to a "fail-slow" situation while attitudes slowly decay and it may be too late to recover when the problems finally become apparent.
- Study finds that the popular rubber hand illusion could be used to treat OCD
- The rubber hand illusion is an effect where a person can come to view a fake hand as their own when their own hand is hidden from view while the real and fake hands are stroked simultaneously with a paintbrush. In the traditional illusion, once the perception shift has happened, the effect can be demonstrated by hitting the rubber hand with a hammer or stabbing it with a needle. More recently, this illusion has been studied as an enhancement for exposure therapy for people who suffer from obsessive compulsive disorder (OCD). In standard exposure therapy, a subject touches a harmless substance that causes a disgust reaction, but then the subject is prevented from washing their hands - as they normally would. Over time, this reduces the severity of the disgust reaction. A problem with exposure therapy, however, is that the disgust reaction is so strong that many patients withdraw from therapy. Researchers now are working on ways of combining the illusion with exposure therapy in order to get the benefits of exposure therapy in a way that is less stressful for the subject. These experiments have involved exposing both the rubber hand or the real hand to the disgust-inducing substance. Based on their results, the researchers believe that the techniques hold promise so that OCD patients can more easily tolerate exposure therapy. Upcoming plans include randomized controlled trials with a larger sample size in order to draw a comparison between the new method and traditional exposure therapy.
Here is a video demonstrating the rubber hand illusion
- An algorithm that learns through rewards may show how our brain does too - Reinforcement learning harnesses the lesson of Pavlov's Dogs, that a task can be learned with just the application of positive and negative feedback. In 1951, Marvin Minsky was able to produce a machine that used this sort of learning to solve a maze, and in the 1990s, it was learned that human brains seem to use their own dopamine reward systems to operate in the same way. A new paper by Google's DeepMind team seems to confirm that understanding and describes the team's observations of the dopamine reward system in mice. In particular, they measured dopamine rewards in response to a series of trials with mice, and they found that the levels of dopamine varied in a way that mirrored the actual rewards that were given to the mice. In addition to providing confirmatory evidence into the brain's reward process, this research also offer insights into the canonical theories about reinforcement learning. In particular, it helps to explore the theoretical implications of having positive and negative dopamine rewards centers.
- STEEM Building a Socially Aware Internet - In this post, @openseed describes their vision for a socially aware Internet that complements the Steem blockchain. The post points out that the Steem backbone is great for public data, but that many Internet applications also need to make use of private information. So, the team aims to launch a new suite that will begin by offering private and encrypted messaging for individuals, groups, and communities who interact with the Steem blockchain. According to the post, completing this vision will fulfill a major step towards the Steem ideal to "tokenize the web". The team is seeking funding through Steem's state of the art decentralized autonomous organization (DAO), the Steem Proposal System, and you can vote for the proposal here. (A 10% beneficiary setting has been attached to this post for @openseed)
In order to help bring Steem's content to a new audience, if you think this post was informative, please consider sharing it through your other social media accounts.
This post will also appear on my pre-Steem blog, chescosteve.blogspot.com.
And to help make Steem the go-to place for timely information on diverse topics, I invite you to discuss any of these links in the comments and/or your own response post.
Beneficiaries
- Burn STEEM - null - 10%
- Cited author(s) - openseed - 10%
- Posting and/or scheduling service (steempeak.com) - steempeak - 5%
- Steem/RSS services (steemrss.com) - torrey.blog - 5%
- STEM curation - steemstem - 5%
About this series
Sharing a link does not imply endorsement or agreement, and I receive no incentives for sharing from any of the content creators.
Follow on steem: remlaps-lite, remlaps
If you are not on Steem yet, you can follow through RSS: remlaps-lite, remlaps.
Thanks to SteemRSS from philipkoon, doriitamar, and torrey.blog for the Steem RSS feeds!