A graphical, parallel biological world.....

This Wednesday (April 4) at 2pm Pacific time, I will be presenting a lecture entitled "Scenes from a grahical, parallel biological world" to the Embryo Physics groups in Second Life. The lecture (see slides here) will be an excursion into the world of GPU (graphical processing unit) computing. I have been interested in GPU computing for a few years now, and interested in computer graphics for significantly longer.

 Scene from the Embryo Physics course (between classes).

This is a different approach to using CUDA architectures, which for most scientists is all about solving their mathematically-intensive problems faster. My interest is a bit more philosophical. I wish to understand the connections between parallelism and high-powered graphics processing (e.g. image rendering) in the service of designing novel data structures for analyzing scientific data. In this case, I am interested in what we can learn from a host of biological systems (ranging from population biology to cell biology).

Perhaps this approach can be leveraged to better understand the self-organization and underlying context that characterizes many biological systems. This talk represents a first step in this synthesis, but should be interesting in any case.

UPDATE (4/4): A blogscript of the talk is available on this page.

Use, Reuse, and Use Again.....

One commonly assumed feature of the brain is that widespread plasticity is responsible for behavioral variation. While this has indeed been demonstrated for cortical regions of mammals [1,2], by and large its relationship to evolutionary processes is unclear. Related to this are the origins of cultural behaviors in humans. While cultural behaviors are not limited to humans, the scope of human dependence on culture is unique. Therefore, an important issue in the brain sciences is how cultural behaviors are represented in the brain. This involves both the origin of neuronal circuits and changes in these circuits across species.

Is this due to plasticity or some other mechanism? Some specialization of the neuronal architecture is likely required in order to enable specific cultural behaviors and subsequent cultural evolution within a species. But does this proceed in a way analogous to massive cortical plasticity, or does it build upon previously defined circuits? There are a variety of views on this in the literature, but the consensus seems to be that an evolutionary mechanism called exaptation is primarily responsible for the origins question. Exaptation is the functional redeployment of a trait that originally evolved for another purpose. In [3], the author uses the term reuse to characterize the redeployment of conserved neural circuits for cultural behaviors in humans. There are several competing theories that fall under this category, but the neuronal recycling [4] and massive redeployment [5] models suggest that these changes are related to development and evolution, respectively.

Figure 1. Comparision of anatomical modularity (TOP) with reuse (BOTTOM) as the driver of functional changes in a six-region representation of the brain. Solid and dashed lines represent two different cognitive functions. COURTESY: Figure 1 from [3].

Focusing on evolutionary origins, there are several predictions made in the massive redeployment model. One of these is that when a new function arises (such as mathematical reasoning or reading), incorporation of these functions into existing circuits should be favored over the formation of new circuits. Accordingly, there should be a correlation between the phylogenetic ancient-ness of a brain area and it's frequency of reuse. Finally, more recently derived functions should use a greater number of circuits which are more widespread in the brain. Evidence for these ideas can be seen in a meta-analysis of fMRI data [6]. Using a subtraction approach and accounting for 11 task domains, a typical cortical region is activated by up to nine of these domains [6].

Dehaene and Cohen more explicitly connect the neuronal recycling model to the origins of culture in [7]. According to their work, representations of letter and number sequences can be localized to the left occipital and bilateral intraparietal areas of the brain (see Figure 2). Importantly, these areas have two attributes that is important for redeploying existing brain areas for cultural behavior. One of these is a systematic architecture that allows for complex representations. The other involves repeatable activation patterns (as revealed by neuroimaging). The importance of repeatable activation related to other functions can be appreciated when considering cultural symbols that are invariant (e.g. a universal response) in a cross-cultural context and having their origins in non-cultural stimuli [see 8 for more information].

Figure 2. Locations for activation in the left occipital and bilateral intraparietal cortices involving both newer (letter and number processing) and older (non-word stimuli) functions for these areas. COURTESY: Figure 1 from [7].

In [7], the authors hypothesize that neuronal recycling is related to behaviors and neuronal processing observed among cultural traits (e.g. writing and numeracy) in three ways. The first is that although phylogenetically-old areas can indeed be redeployed, they are also constrained in evolution and development. Developmental constraints are particularly important in that they bias acquisition and learning during life-history. The second relation involves the existence of a neuronal niche for cultural acquisitions. It does not appear that exaptation proceeds randomly. Instead, areas that already serve a related function are more likely to be repurposed for newer functions. In the case of regions proposed by Dehaene and Cohen, previous functions related to scene and object recognition are well-suited to recognition of cultural symbols. Because of this, these previous organizational attributes are not erased upon repurposing, thus enabling future repurposing.

To understand the phenomenon of neural reuse and evolutionary relationships in another light, I will turn to a recent article in Nature Methods [9] that describes how we can determine functional equivalence across species with respect to functional evolutionary changes. If the exaptation of phylogenetically-conserved regions drives the neural representation of new behaviors, then understanding commonalities of neural representation across species is important. Here I make the distinction between "functional equivalence" and homology in part because previous research [10] has shown that neural homology (based on structure) is exceedingly hard to establish. Unlike many other morphological traits, neural traits often change position and recombine, which makes establishing their common ancestry difficult.

To assess common processing for natural scenes between monkeys (Macaca mulatta) and humans, a method called interspecies activity correlation (ISAC) was used to determine functionally-equivalent regions.Figure 3 graphically demonstrates the ISAC method in stepwise fashion.

Figure 3. Diagram of the ISAC method as applied to monkey and human brains. ROI = region of interest. COURTESY: Figure 1 from [9].

ISAC does not rely on the inference of anatomical relationships. Instead, ISAC is a statistical technique that equates correlated brain region-specific hemodynamic activity with commonalities activation patterns between species. When a common stimulus is presented (in this case natural scenes), a time-course of neural activity is collected for each species. Using both correlational and convolutional analysis techniques, species-specific hemodynamic signals can then be compared on a voxel by voxel basis. In the future, this might be one way to assess systematic architectures and repeatable activation patterns.

So are brain regions used, reused, and used again for new functions that emerge in the course of evolution? There seems to be a lot of interesting work that points in this direction. The evolutionary concept of exaptation [11] provides a mechanism for this continued reuse across phylogeny, while cutting-edge neuroimaging techniques provide a window (albeit opaquely) into this evolutionary relationship. By identifying both the conserved and derived functions of different brain regions, we might be able to fill in the boxes shown in Figure 1 with specific examples of exaptation in the brain for evolutionarily-derived cognitive functions.

References

[1] Krubitzer, L. and Kahn, D. (2003). Nature versus nurture revisited: an old idea with a new twist. Progress in Neurobiology, 70(1), 33-52.

[2] Kaas, J. and Catania, K. (2002). How do features of sensory representations develop? BioEssays, 24, 334-343.

[3] Anderson, M.L. (2010). Neural reuse: A fundamental organizational principle of the brain. Behavioral and Brain Sciences, 33, 245–313.

[4] Dehaene, S. and Cohen, L. (2007). Cultural recycling of cortical maps. Neuron, 56, 384–398.

[5] Anderson, J.R. (2007). How can the human mind occur in the physical universe? Oxford University Press, Oxford, UK.

[6] Anderson, M.L. (2007). Evolution of cognitive function via redeployment of brain areas. The Neuroscientist 13, 13–21.

[7] Dehaene, S. and Cohen, L. (2007). Cultural Recycling of Cortical Maps. Neuron, 56, 384-398.

[8] Changizi, M.A., Zhang, Q., Ye, H., and Shimojo, S. (2006). The structures of letters and symbols throughout human history are selected to match those found in objects in natural scenes. American Naturalist, 167, E117–E139.

[9] Mantini, D. et.al (2012). Interspecies activity correlations reveal functional correspondence between monkey and human brain areas. Nature Methods, 9(3), 277-282. For more on using neuroimaging to establish "homology", please see: Wager, T.D. and Yarkoni, T. (2012). Establishing homology between monkey and human brains. Nature Methods, 9(3), 237-239.

[10] Striedter, G.F. (2002). Brain homology and function: an uneasy alliance. Brain Research Bulletin, 57, 239–242. AND Sereno, M.I. and Tootell, R.B.H. (2005). From monkeys to humans: what do we now know about brain homologies? Current Opinion in Neurobiology, 15, 135–144.

[11] Okada, N., Sasaki, T., Shimogori, T., and Nishihara, H. (2010). Emergence of mammals by emergency: exaptation. Genes to Cells, 15(8), 801-812.

New Look for "Complexity Digest"

Complexity Digest (a.k.a. ComDig), the weekly update on all things complexity-oriented run by Carlos Gershenson, has a new look (besides the recursive yin-yang logo shown above). Once distributed to subscribers exclusively via e-mail, there are now two versions: one on scoop.it (a magazine-style template for organizing content, and a more conventional design.

Examples of the scoop.it design (above, with categories and profiles of recent papers).

 

  Example of the more conventional design (above, with profiles of recent papers and talks, full archives, and a forum).

Either versions is an improvement over the e-mailed digests, mainly from a community-building standpoint. Check it out if interested.

Methods of Controlling Intelligence

This blog post will focus on the recent (and not so recent) attempts to quantify, control, and augment intelligent performance-related behavior in human beings. The intersection of human intelligence and artificial intelligence by way of human performance goes by the name of Augmented Cognition. Augmented Cognition, generally regarded as a domain of Human Factors engineering, also has broad applications to human-machine systems. Relevant application domains could range from automotive and transportation performance to human interactions with information technologies and bioengineered prosthetic devices.

Augmented Cognition is distinct from traditional artificial intelligence, in which a general purpose intelligence is constructed de novo to control all aspects of intelligent behavior. Rather than machine intelligence compensating for the shortcomings of human intelligence, human intelligence compensates for the shortcomings of machine intelligence. Academic interest in this set of problems began in the 1950's [1], while contemporary approaches have included information technologies and DARPA's Augmented Cognition project. As applied to technology, this work falls into the broader category of human-assisted intelligent control.

There are two main components of augmenting human intelligence using computational means. The first is a closed-loop system which involves a feedforward and a feedback component between the individual and a technological system that enabled augmentation. This could be a heads-up display, a mobile device, or a brain-computer interface controlled by a real-time algorithm. The second is a model of human performance for a given set of cognitive and physiological functions which determines a control policy. Examples of both are provided below, along with a consideration of open problems in this field.

Closed-loop System Design

In an article from the 1950's [2], W.R. Ashby took a cybernetic approach to first-order (e.g. no intermediate variables) intelligence augmentation (Figures 1-4). While somewhat crude by modern standards (by which we use sensors to gain real-time measurements of physiological state), it does lay out a simple theoretical model for augmenting cognitive and neural function.

 

Figure 1. Highlight for the X component.

Figure 2. Highlight for the G component.

 

Figure 3. Highlight for the S component

 

Figure 4. Highlight for the U component.

In the Ashby model, the feedforward component (G) was the intelligence of the user applied to performance captured by the device. This might be driving performance, or accuracy in moving an object. While the idea that intelligence can be distilled to a single variable is controversial, modern applications have used variables such as accuracy counts or a specific electrophysiological signal to "drive forward" the system. The amplifier (S) itself gathers the feedforward elements of G and operates on them in a selective manner. This can be treated as either an optimization problem [3] or an inverse problem [4], and defines the control policy imposed on the performance data. In the Yerkes-Dodson example shown later on, a minimax-style optimization method is used. The feedback element (U) is a signal taken from the information in G and should contribute to an improvement in performance, or subsequent measurements of G.

Contemporary Models from Human Performance

More contemporary models for augmenting human performance [5,6] have involved mapping closed-loop control to a physiological response function. Figures 5 through 7 show how this works in the context of the Yerkes-Dodson curve. The Yerkes-Dodson curve is an inversely U-shaped function that characterizes arousal in the context of some physiological measurement. At both low and high values of the physiological indicator, the level of arousal is low. At moderate values of the physiological indicator, the level of arousal is high. The goal of an amplifier (also called a mitigation strategy) is to maintain performance (defined as measured arousal) among the highest range of arousal values.

 

Figure 5. Example of a physiological response function (e.g. Yerkes-Dodson curve).

Figure 6. Example of a mitigation strategy.

Figure 7. Keeping performance within an optimal range.

Two outstanding problems

There are two potential challenges to this control policy: a reliance on convexity and complete measurement of a physiological state. The example shown here has relevance to arousal and attention. It has attracted attention because of its relative ease of mitigation. The development of brain-machine interfaces has likewise focused on simple-to-characterize physiological signals (such as population vector codes for movement [7] or spectral bands of an EEG [8]). However, not all physiological response functions are so simple to characterize. In cases of significant non-convexity (or cases where the response function does not form smooth, convex gradients), it may be quite difficult to mitigate suboptimal behavior or physiological responses [9]. In such cases, there could be multiple optimal points each with very different performance characteristics.

The complete measurement of physiological state is another potential problem with this method. While fully characterizing a physiological or behavioral process is the most obvious difficulty, the adaptability of a physiological system to repeated mitigation is a more subtle but important problem. In some cases, the physiological response will habituate to the mitigation treatments and render them ineffective. In the case of presenting information on a heads-up display, users might simply tend to ignore the presented cues over long periods of time. It might also be that encouraging rapid changes in arousal level is more effective than encouraging a fixed level of performance over time. In both strength training regimens and more general physiological responses to the environment, switching between stimuli of alternating intensities can have a complex and ultimately adaptive consequences on the long-term response.

Incorporation of intelligence augmentation into the design of a technological system is an ongoing challenge. In a future post, I will focus on why certain aspects of human and animal intelligence are fundamentally different from and can potentially aid and complement current approaches to machine learning and artificial intelligence.

References:

[1] Ashby, W.R. (1952). Design for a Brain. Chapman and Hall, London.

[2] Ashby, W.R. (1958). Design for an Intelligence Amplifier. In Automata Studies. Shannon, C.E. and Ashby, W.R. Princeton University Press, Princeton, NJ.

[3] an optimization method uses some objective criterion to select a range of values thought to either minimize or maximize system properties.

[4] an inverse problem is one where the solution is known, but the route to that solution is not.

[5] Schmorrow, D. D. & Stanney, K.M. (Eds) (2008). Augmented Cognition: A Practitioner's Guide. HFES Publications.

[6] Fuchs, S., Hale, K.S., Stanney, K.M., Juhnke, J., and Schmorrow, D.D. (2007). Enhancing Mitigation in Augmented Cognition. Journal of Cognitive Engineering and Decision Making, 1(3), 309-326.

[7] Jarosiewicz, B., Chase, S.M., Fraser, J.W., Velliste, M., Kass, R.E., and Schwartz, A.B. (2008). Functional network reorganization during learning in a brain-computer interface paradigm. PNAS, 105(49), 19486-19491.

[8] Lotte, F., Congedo, M., Lecuyer, A., Lamarche, F., and Arnaldi, B. (2007). A review of classification algorithms for EEG-based Brain-Computer Interfaces. Journal of Neural Engineering, 4, 1-24.

[9] Alicea, B. The adaptability of physiological systems optimizes performance: new directions in
augmentation. arXiv Repository, arXiv:0810.4884 [cs.HC, cs.NE] (2008).

The art of a lifetime (so far)......

Even when I was a child, I had a strong interest in science and engineering. I always had a talent for drawing. As a teenager, I got interested in becoming an artist (I also got interested in becoming an economist, but that didn't have worked out purely on principle). I wanted to go to art school, but my art was not "critically acclaimed" enough to impress people. I have a very technical style of drawing that I freely interchange with "cartoonish" features and obscure references to create what I consider art. It is very much the opposite of the good aesthetic form and practice typically learned in art school.

My art avocation has been expressed in two periods. I have posted most of these works on my personal website. The first was from 1998-2003. During this period, I focused on using mathematical/ technical concepts and hand-drawn (digitized) cartoons to create abstract art. The following works are examples of this:

"Poincaré Recurrence"

 "Meltdown meets the Domino Effect"

"Sequential Spike"

"Chaz Rodders II"

"Van Gogh meets the Samurai"

Some of these works (e.g. Van Gogh meets the Samurai) resemble the narrative structure of conventional sequential art. Others (e.g. Poincare Recurrence) feature made-up characters as actors in scientific theories and real-life situations. Still others (e.g. Chaz Rodders II) feature made-up characters as actors in purely fantastical scenarios.

The is a gap comprising the years 2003 through 2008. The reason: PhD programs (e.g. coursework, finding a research voice) are time-consuming. Since I spent time in a lab that developed and experimented with virtual worlds, there is probably something in this experience to inspire some interesting artwork that I will explore over time.

The second period was from 2009-present. During this period, I started focusing more on merging pop culture and images from the internet into composite, abstract images. The following works are examples of this:

"Judgement Day"

"Obscure References, obscure references, obscure references!"

"Crazy Eyes"

"Razor-faced Spy"

"Decapod vs. Cephalopod"

These works take less time to create, and also reflect a more conventional "mash-up" style. I have also taken to interesting juxtapositions that have little semantic value but have other, more subtle relationships. For example, in "Crazy Eyes", there are three characters (Egon Spengler from "Ghostbusters", a robot avatar from Second Life, and Kramer from Seinfeld) that share a certain physical set of congruities. Other works (e.g. Razor-faced Spy, Obscure References 3x) are simply modifications to references and icons from pop culture.

Having said that, I feel that art is about more than pure subjectivity. I feel that artistic creativity has an underappreciated value in science and engineering. I had a professor at the University of Florida (Dr. Paul Fishwick) who has developed an approach to programming called Aesthetic Computing. Aesthetic Computing uses artistic representations to map out data structures, algorithms, and algebraic relationships. I have found that this approach is somewhat useful in conveying complicated scientific concepts and theoretical advances to audiences from diverse backgrounds (e.g. biologists and mathematicians). I am also a fan of using devices such as the Feynman diagram to describe advanced mathematical concepts. In the future, I would like to take my early period style of art and apply it to problems I have encountered as an academic scientist. This is fertile territory for the communication and popularization of science, mathematics, and advanced technologies, so if anyone out there is interested in developing this further, contact me.