Phil 11.27.17

There are new versions of dom4j!!!

I had to do two things this morning that required 2FA. Instead of being another irritating step to do anything, it now feels like a comforting barrier to bad things

Today’s big thought. The Contagion metaphor of information cascade (Bikhchandani) is a bad one (These folks agree). The flocking/stampeding metaphor is more complete. For example, it provides the concept of position, alignment, and velocity in a belief space. Two equally ‘viral’ (belief space position?)  memes may be released into the wild, but the one closest to the current alignment will be the one that will succeed more. Flocking effects result in sub-stampede cascades. The ‘refractory’ period is the time it takes for the population to get into alignment again. There is also something in the notion that a flock or a stampede is a group navigating with incomplete information that is fundamentally different from disease spread, which is more like percolation. It also fits better with what’s being done in reinforcement learning.

A nice description of the contagion model, from Information Epidemics and Synchronized Viral Social Contagion

  • In network science the processes of information contagion can be studied using epidemic models. If we consider information as a disease, three different models can be used to model its proliferation through the network: SIS, SIRS and SIR (Ball 1997; Newman 2002).
In the first model an element of the population (a node in a network) can be either susceptible (S) or infected (I) and goes through these stage one after the other: susceptible – infected – susceptible. 
According to the second model, the node also has a refractory stage (R) before it becomes susceptible again, so the states are: susceptible – infected – refractory – susceptible. 
Finally, in the third model once the node is infected once it cannot be infected again, so for each node the sequence is: susceptible – infected – removed (R). Depending on the nature of information studied, a certain model can be more suitable than another (Fig 2). 

7:00 – 3:00 ASRC MKT

  • Renewed for a year. Will probably switch over to the UMBC server for the next version though
  • Starting Speaker–listener neural coupling underlies successful communication here
  • The next paper will be Mirroring and beyond: coupled dynamics as a generalized framework for modelling social interactions
    • Uri Hasson (HassonLab at Princeton)
    • Chris Frith (Scholar)
    • When people observe one another, behavioural alignment can be detected at many levels, from the physical to the mental. Likewise, when people process the same highly complex stimulus sequences, such as films and stories, alignment is detected in the elicited brain activity. In early sensory areas, shared neural patterns are coupled to the low-level properties of the stimulus (shape, motion, volume, etc.), while in high-order brain areas, shared neural patterns are coupled to high-levels aspects of the stimulus, such as meaning. Successful social interactions require such alignments (both behavioural and neural), as communication cannot occur without shared understanding. However, we need to go beyond simple, symmetric (mirror) alignment once we start interacting. Interactions are dynamic processes, which involve continuous mutual adaptation, development of complementary behaviour and division of labour such as leader– follower roles. Here, we argue that interacting individuals are dynamically coupled rather than simply aligned. This broader framework for understanding interactions can encompass both processes by which behaviour and brain activity mirror each other (neural alignment), and situations in which behaviour and brain activity in one participant are coupled (but not mirrored) to the dynamics in the other participant. To apply these more sophisticated accounts of social interactions to the study of the underlying neural processes we need to develop new experimental paradigms and novel methods of data analysis
  • Started on the use cases
  • Added sections in an HHS RFI that looks a *lot* like the research browser, with some more attention to data provenance. Which is a thing that should be added to the RB anyway.

3:00 – 4:00 Campus

  • Julie wants an ‘append’ option to the ignore.xml. Done!
  • Relaxed Fika. Learned about Stacy’s research, Julie’s CHI DC, and Andrea’s affective computing. Intimate and nice!
  • No Wayne?