Could a neuroscientist understand a microprocessor?

With campaigns like the BRAIN Initiative in full force, we are already producing more data than current analytical approaches can manage. So how do we go about analyzing ‘big data’? It is with this goal in mind that this week’s Neuroscience Seminar speaker, Dr. Konrad Kording, seeks to understand the brain.

Dr. Kording is a Penn Integrated Knowledge Professor at the University of Pennsylvania and a Deputy Editor for PLOS Computational Biology. He received his PhD in Physics from the Federal Institute of Technology in Zurich. His lab is a self-described group of ‘data scientists with an interest in understanding the brain’. He focuses on analyzing big data sets and maintaining a healthy skepticism towards the interpretation of results.

A brilliant example of his approach can be found in Could a neuroscientist understand a microprocessor? (Jonas and Kording 2017). This witty paper seeks to analyze the viability and usefulness of current analytical methods in neuroscience. The authors seek to glean insight into how to understand a biological system by examining a technical system with a known ‘ground truth’, a simple microprocessor (Fig 1).


Figure 1. Reconstruction of a simple microprocessor (MOS 6502).


But what does it mean to ‘understand’ a biological system? Is it the ability to fix the system? Or the ability to accurately describe its inputs, transformations, and outputs? Or maybe the ability to describe its characteristics/processes at all levels: a) computationally, b) algorithmically, and c) physically? Kording argues that a true understanding is only achieved when a system can be explained at all levels. So how do we get there?

Innovations in computational approaches are clearly required to make further progress, but it is also necessary to verify that these methods work. Jonas and Kording suggest the use of a known technical system as a test, an idea that stemmed from a critique of modeling in molecular biology, Can a Biologist Fix a Radio? (Lazebnik 2002). Kording used a reconstructed and simulated microprocessor like those used in Atari as a model system. The behavioral inputs were three games – Donkey Kong, Space Invaders, and Pitfall. The behavioral output is the boot-up of the game. The ‘recorded’ data (Fig 2) was then sent through a battery of analysis methods used on real brain data, ranging from connectomics to dimensionality reduction. Here, I outline a few of these methods.


Figure 2. a) 10 identified transistors and b) their spiking activity.


Analysis Method 1: Lesion Studies

If each transistor is removed, will the processor still boot the game? They find that there are indeed subsets of transistors that make one of the behaviors (games) impossible (Fig 3). A logical conclusion would be that these transistors are responsible for that particular game. What’s wrong with this? Transistors are not specific to a behavior, but rather implement simple functions. Moreover, these findings are unlikely to generalize to other behaviors


Figure 3. Lesioning every single transistor. a) Colored transistors were found to impact only one behavior. b) Breakdown of the impact of transistor lesion by behavioral state.


Analysis 2: Tuning Curves

How is the activity of each transistor (spike rate) tuned to behavior (luminance of last pixel upon booting)? They find that some transistors show a strong tuning curve (Fig 4). What’s wrong with this? Transistors relate in a highly nonlinear way to screen luminance; thus, apparent tuning is not insightful.


Figure 4. Examples of tuning curves for 5 transistors.


Analysis 3: Local Field Potentials

Is the processor modular? Will analyzing average activity of localized regions yield insight about functional organization? Indeed, region-specific rhythmicity is found (Fig 5). What’s wrong with this? It is hard to attribute self-organized criticality to a processor. Moreover, the temporal organization of the LFPs does not contain meaningful information about the processor’s behavior.


Figure 5. Local field potentials in 5 different regions.


This is a subset of the methods tested in this paper to check the naïve use of various approaches used in neuroscience. The authors find that the standard data analysis techniques produced results that were surprisingly similar to those found in real brains. However, in the case of the processor, its function and structure are known and the results did not lead to a satisfying understanding.

Kording suggests that using a microprocessor or artificial neural network to test all analysis methods prior to using them on the brain could be extremely useful. In my correspondence with him, Kording gave the following additional suggests for doing ‘good’ science:

  1. “When doing experiments, evaluate how far findings generalize. How much can you change the experiment until your findings go away?”
  2. “When doing data analysis, be careful about causal interpretations. There are an infinite set of models that predict the same correlations.”
  3. “When developing theories, develop theories that actually solve the problems. I.e. when your theories are linear, they are not very exciting unless they are truly fundamental.”
  4. “Be clear about the exact thing you are studying. Be clear about the question. Be clear about the implicit assumptions you are making. These are usually shared across the field and usually are wrong.”
  5. “Just because lots of people work on something does not necessarily make it more probable that it has a solid logical underpinning.”
  6. “You always believe that somewhere there are the advanced people who really understand what they are doing. Certainly I did. They don’t.”
  7. “Share. Share code. Share data. Share ideas. Share preprints. Share an appreciation for great science. If it is not worth sharing it is not worth doing.”

To hear more about Dr. Kording’s ideas on rethinking underlying assumptions and to learn about why machine learning is a useful skill that should be in every scientists’ toolbox, please attend his talk on Tuesday, October 24, 2017 at 4pm in the CNCB Marilyn G. Farquhar Seminar Room.

Jess Haley is a first-year neuroscience Ph.D. student currently rotating in the Laboratory of Dr. Sreekanth Chalasani.


An Unfair Game of Tug of War

Imagine a society where unproductive or debilitated members are captured and destroyed; no questions asked. In this society everybody must do their part efficiently and without error for this society’s main job is to cooperate with other societies for the greater good of the entire civilization. This is exactly what happens in your nervous system every single day. If proteins or other cellular materials within neurons are not perfect, they can wreak havoc on the entire cell, which could be potentially deleterious for the organism as a whole.

In a recent study laid out in the paper titled “LC3 Binding to the Scaffolding Protein JIP1 Regulates Processive Dynein-Driven Transport of Autophagosomes”, Dr. Erika Holzbaur and her lab focus on the mechanisms of directed degradation of proteins and organelles via autophagocytosis. Autophagosomes have been shown to form at the distal end of neurons and be carried toward (but not away from) the cell body whilst they mature. The problem with this is that the autophagosome is attached to both the proteins dynein (which carries it toward the cell body) and kinesin (which carries it away from the cell body), meaning there must be something promoting dynein’s activity and inhibiting kinesin’s activity and thereby regulating the retrograde movement of autophagosomes. Upon investigation of this complex system, Dr. Holzbaur and her lab demonstrate that the motor scaffolding protein, JIP1 functions broadly in axonal transport and regulates the unidirectional retrograde transport of autophagosomes in a cargo-specific manner.

In order evaluate the significance of JIP1 in retrograde transport of cargo, they knocked it down and saw significant reduction in retrograde movement of autophagosomes using the marker mCherry-LC3 (shown in figure B below). Next they asked when during the lifetime of autophagosomes JIP1 began regulating their position, and found that biogenesis of autophagosomes in the axon tip occurs independently of JIP1 but efficient exit from the distal axon does not. They then went on to show that acidification of autophagosomes in the proximal axon (which is necessary for cargo degradation) is also dependent on the interaction between autophagosomes and JIP1. There are two main mechanisms of ensuring retrograde transport of autophagosomes; the first is the inhibition of kinesins by the autophagosome-JIP1 interaction and the second is the colocalization of the phosphatase MKP1 to the autophagosome adapter LC3, because JIP1 promotes exclusive retrograde transport when it is in its non-phosphorylated state (shown in figure F below).


Dr. Holzbaur and her lab have shed a great deal of light onto the mechanisms of autophagosome transport. To learn more about her work, please attend the Neurosciences Seminar Series this Tuesday, October 10th at 4:00pm in the Center for Neural Circuits and Behavior.

Fu M, et al. LC3 Binding to the Scaffolding Protein JIP1 Regulates Processive Dynein-Driven Transport of Autophagosomes. Developmental Cell. 2014;29:577–590. DOI: 10.1016/j.devcel.2014.04.015.

Haylie Romero is a PhD student currently rotating in Dr. Takaki Komiyama’s lab in the Center for Neural Circuits and Behavior.

What can invertebrates tell us about our brains?

With its hundred billion neurons and quadrillion synapses, the human central nervous system(CNS) can seem intractably complex. Fortunately, there is a class of animals whose nervous systems and behaviors are much more easily understood.  Invertebrates, such as sea slugs and worms, have on the order of only hundreds or thousands of neurons and their connections are extremely well stereotyped. This simplicity makes them amenable to experimentation and modeling, and has allowed scientists to understand the structure and function of their neural circuits.

In his review, Allen I. Selverston, Professor Emeritus at UCSD, asks if information gained from the study of invertebrates can be translated to our understanding of the human CNS.  He focuses on a particularly well characterized type of circuit called Central Pattern Generators (CPG).  CPGs are networks of neurons which produce rhythmic outputs in the absence of sensory feedback, and often control simple motor actions such as feeding or swimming. CPGs are not only found in invertebrates but vertebrates as well, where they control certain low level functions.  An example of a CPG is the leech heartbeat network which is shown in the diagram below.


Leech heartbeat neuronal network

The study CPGs using electrical and chemical manipulation of their constituent neurons has led to three primary types of discoveries.  First, it has revealed how a complex array of ion channels contributes to the distinct activity properties of individual neurons. Second, it has shed light on the types of synapses and how they are modulated and third, how circuits produce functional outputs.

Selverston uses these three types of analysis to explain how many different CPGs from the invertebrate world work. Unfortunately, he concludes that there are very few general principles for the design of these circuits that are transferable from model to model. Each CPG has its own evolutionary history that has crafted it into a bespoke circuit for the unique function that it serves. Moreover, the experimental methods used to study CPGs are unlikely to be effective in more complicated vertebrate systems because they cannot be probed with single cell techniques. This means that while the cellular and synapse level data may broadly applicable, the further study of invertebrate CPGs is unlikely to give us much insight into the human CNS.

Selverston’s review can be found here.

Leo Breston is a first year student in the Neuroscience Graduate Program. He is currently rotating in the Navlaka lab. 


How to make a schizophrenic mouse

Dopamine is perhaps the best known neurotransmitter, almost certainly due to its association with the idea of reward. It’s often brought up to explain why we like the things we do, and how people can develop addictions to different types of rewards. However, dopamine isn’t just a reward chemical; it’s very important for a wide variety of brain processes, including voluntary movement, attention, sensory gating, evaluating the salience of a stimulus, decision making, and motivation. Given all this chemical does, it’s not too surprising that changes in dopamine signaling have been implicated in mental disorders, like schizophrenia, ADHD, and depression. But how, then, does a healthy brain regulate dopamine? And how does this system go wrong?

Larry Zweifel’s lab at the University of Washington studies these questions. Soden et al. examined the effect of a mutation in a gene called KCNN3 that was discovered in a schizophrenia patient (Bowen et al., 2001). This gene codes for an ion channel called SK3 that is activated when calcium is inside the cell and then lets potassium out of the neuron, reducing its excitability. The mutated form, however, has an early stop codon due to a frame shift, and therefore only produces a small fragment of the original protein. Interestingly, this mutation was found to be dominant in cell culture, needing only one copy to exert its full effect and suppress SK3 currents in neurons, likely because the protein fragments bind to and inactivate SK3 channels (Miller et al., 2001).

Since SK3 is expressed in dopamine neurons and was mutated in a schizophrenia patient, it seems a promising candidate for a gene regulating dopamine function. Soden et al. tested the effect of this mutation in a mouse model by adding the mutated gene into the genome of dopamine neurons using a viral vector and the Cre-lox system. Indeed, they found that dopamine neurons in the mice with the mutant gene were more excitable and fired less regularly than usual, making them more prone to firing bursts of action potentials.


These bursts are thought to be a functionally different form of dopamine signaling than the neurons’ regular spiking, causing different effects in dopamine-responsive brain regions, so this altered neuronal function should correspond to altered behavior in tasks where dopamine is important. Since dopamine is involved in sensory gating, meaning the brain’s filtering of irrelevant stimuli, the researchers tested this ability in the mutant mice. In their task, the mice were presented with two sounds, one which was was always followed by a reward (a sugary pellet), and one which was rarely followed by a reward. The mice learned to look for the pellet quickly after the more predictive sound, but not after the other. Once the mice had learned to distinguish the sounds, the researchers flashed a light at the same time the reward-predictive sound was played. The normal mice became distracted, but the mutant mice paid no attention to the novel stimulus and still proceeded quickly to the pellet, indicating that their sensory gating was altered.


The researchers also tested their mice on prepulse inhibition (PPI), a neurological process by which the startle response of an animal to a sudden, high amplitude stimulus, such as a loud sound, is reduced if the strong stimulus is preceded by a weaker one. This phenomenon occurs in both mice and humans, is affected by dopamine-modulating drugs, and is reduced in people with schizophrenia. Indeed, the control mice showed prepulse inhibition, while the mutant mice did not.


This paper is significant in that the authors were able to demonstrate a link across multiple levels of biology, from disrupted gene function to neuronal function to behavior. As the KCNN3 gene is in a chromosomal region (1q21) that is associated with schizophrenia, it’s possible that this gene, and pathological processes similar to that shown here by the author, are at play in more cases of schizophrenia. The ability understand how the brain is disrupted across different scales in psychiatric illness is crucial to developing better, targeted treatments for these conditions.

Bowen, T. et al. Mutation screening of the KCNN3 gene reveals a rare frameshift mutation. Mol. Psychiatry 6, 259–260 (2001).
Miller, M. J. Nuclear Localization and Dominant-negative Suppression by a Mutant SKCa3 N-terminal Channel Fragment Identified in a Patient with Schizorphrenia. Journal of Biological Chemistry 276, 27753–27756 (2001).
Soden, M. E. et al. Disruption of Dopamine Neuron Activity Pattern Regulation through Selective Expression of a Human KCNN3 Mutation. Neuron 80, 997–1009 (2013).
Jacob Garrett is a first-year PhD student in the neurosciences program. He has not yet narrowed his interests enough to provide any sort of useful description here.

Let’s Talk About Neural Time Travel.

Unfortunately Foster has not developed a functional time machine. Instead, he is currently working on elucidating the underlying mechanisms of neural time travel in rats and its role in goal directed navigation. Neural time travel, or chronesthesia, was first proposed by Endel Tulving in the 1980s. It refers to the ability to perceive the difference between perceived, remembered, known, and imagined time. At first glance, this appears to be a thought experiment that will leave you with a headache and not much to show for it but fear not. The basic principle of neural time travel, is the ability to distinguish now from then, whether it is a time in the past or the future. To make a grievous oversimplification, neural time travel is akin to locating memories, thoughts and imaginings in temporal space.

Now, what does this have to do with hippocampal place cells which by definition are interested in physical space?

Pyramidal place cell activity in the hippocampus encodes the spatial relationships between landmarks in the environment. The cells have spatial receptive fields which fire when the animal is in a specific location in the local environment. Each environment is independently represented by place cell activity and the relationship between the spatial fields is unique to each local environment. The firing sequences of the place cells encode the navigation of the animal as it moves through individual receptive fields, in a predictable manner. In addition to place cell activity in navigating animals, these networks of cells additionally exhibit oscillatory activity, hippocampal short wave ripple (SWR)-associated place cell sequences, in sleeping and stationary animals. This feature is the focus of David Foster’s most recent work. The SWR-associated place cell events are also referred to as “replay,” during which time the relative sequence of place cell firing generated by a navigating animal is repeated in a rapid burst.

The hippocampus of rats was implanted with forty tetrodes, to allow monitoring of network activity while the rat navigated through an open field in search of a reward in either a known location or a random location. From this, it was possible to accurately determine the location of the place cell receptive fields making it possible to determine the position and trajectory of the rat from the firing sequence. Foster and colleagues identified many brief increases in population activity in stationary animals. The firing sequence (trajectory event) of many of these events was not random, but encoded a trajectory through two dimensional space that was temporally compressed. Surprisingly, these trajectory events were not simple replay of the animal’s most recent path.


The end point of these trajectory events was more likely to be the known location of a reward than anywhere else, suggesting that the events are goal directed. Foster also demonstrated that the trajectory represented in these SWR-associated events predicted the actual navigational trajectory of the rat, the prediction was improved if the end point was the location of a known reward, suggesting the events are a reflection of future behavior. By analyzing the trajectory events when the animal was forced to learn a new location for the reward, Foster discovered that initially the trajectory events emphasize novel combinations of start and end points. This is additional evidence that this is not replay of previous firing sequences.

So now you can pretend to understand why the rats are running around in such a seemingly random manner, but how is this neural time travel? The presence of predictive neural firing suggests that the rats are able to utilize episodic memories to facilitate a goal directed future action. Establishing a trajectory from a novel start to a known reward location suggests that the rat is able to extract information from multiple other pathways and string them together in order to get somewhere in the future. There are many implications for how this may allow us to study episodic memory in the future, but I’m pretty sure that the most important finding is that your rats likely know that you were late to feed them yesterday and are probably coming up with a plan to do something about it when you get in to lab tomorrow.

Alex Smirnov is a first year student in the neuroscience graduate program currently rotating with Gentry Patrick. She is a big fan of electrophysiology and unproductive thought experiments.


  • Nyberg, L., Kim, A. S. N., Habib, R., Levine, B., & Tulving, E. (2010). Consciousness of subjective time in the brain. Proceedings of the National Academy of Sciences, 107(51), 22356–22359. doi:10.1073/pnas.1016823108
  • Pfeiffer, B. E., & Foster, D. J. (2015). Autoassociative dynamics in the generation of sequences of hippocampal place cells. Science, 349(6244), 180–183. doi:10.1126/science.aaa9633·
  • Pfeiffer, B. E., & Foster, D. J. (2013). Hippocampal place-cell sequences depict future paths to remembered goals. Nature, 497(7447), 74–79. doi:10.1038/nature12112

Exploring the Dichotomous Consciousness

“One individual studied well, and thoughtfully, might enable you to draw conclusions that apply to the entire human species.”

-David Roberts, Professor of Surgery and Neurology at Dartmouth-Hitchcock Medical Center

The fascinating story of the split-brain patient dates back to the 1940’s. You might rightfully ask: “What is a split-brain patient?”


Split-brain patients are individuals who have been plagued by intractable epilepsy — so much so that they were willing to undergo split-brain surgery, which is essentially a procedure that severs the connections between the left and right hemispheres of the brain. This surgical procedure was meant to prevent the spread of seizure activity from its site of origin, thereby controlling the occurrence of debilitating epileptic seizures. The procedure is also known as a corpus callosotomy because the anatomical structure that connects the two hemispheres of our brains is called the corpus callosum, the so-called highway system of information transfer in the brain.

“It was a total shot in the dark.”

– Michael Gazzaniga

The first group that investigated these patients in the 1940’s claimed that there were no significant cognitive or behavioral impairments as a result of split-brain surgery. Fast forward to the 1960’s and along came Michael Gazzaniga, a driven young student at Dartmouth. During his junior year, a Scientific American article on how nerves grow piqued Gazzaniga’s interest, so he wrote a letter to the author, the one and only Roger Sperry, one of the biggest names in neurobiology. In his letter, Gazzaniga inquired about research opportunities — a move he now refers to as a “shot in the dark” — and landed an NSF summer fellowship at Caltech.


Gazzaniga as a student at Caltech in 1963

Sperry’s group at Caltech had been studying split-brain rats, cats, and monkeys for some time, and were observing dramatic effects on behavior, which raised a huge question mark in their minds about why earlier assessments of split-brain humans had not revealed significant post-surgical differences. They hypothesized that surgeries done in the 1940’s had not severed the corpus callosum and anterior commissure completely. Gazzaniga was thus tasked with coming up with novel and better ways of testing split-brain patients. So he did…


And his findings introduced the notion of functional lateralization to the field of neuroscience:gazz3


When split-brain patients were presented with visual information (such as an object or a word) in their right visual field, they were able to verbally identify the stimulus. Interestingly, if visual information was presented in their left visual field, patients were unable to do so — in fact they would typically say, “I don’t know.” To understand this phenomenon we must recall the following generalizations:

  1. Information in the right visual field is known to be processed by the left hemisphere, and information in the left visual field is known to be processed by the right hemisphere (see above figure).
  2. Certain aspects of language are known to predominantly reside in the left hemisphere of the brain.

From his observations, Gazzaniga came to the conclusion that split-brain patients were unable to verbally identify stimuli presented in their left visual field because, though the information would travel to the right hemisphere, it would not be transferred to the left hemisphere where ‘language resides’ due to the severed connection between the two hemispheres.

There is a twist however. Patients who stated that they “did not know” what the stimuli presented in their right visual field was were able to draw what they saw with their left hand. These observations along with many many follow-up studies testing for part-whole relations, apparent motion detection, mental rotation, mirror image discrimination, etc. led to the idea that there is perhaps a right hemisphere dominance for visuospatial processing. These ideas are not meant to be mutually exclusive for one hemisphere or the other. In fact, one patient clearly demonstrates that certain aspects of language, such as spelling, can also reside in the right hemisphere: P.S., a teenager split-brain patient, was asked “Who is your favorite girlfriend?” with the word ‘girlfriend’ flashing only in his left visual field. He was unable to answer the question verbally because the information remained in his right hemisphere; however his left hand (controlled by the right hemisphere) was able to select Scrabble letters and align them to spell’L-I-Z.’


Split-brain patients were the key to studying the functions of the two hemispheres independently, and Gazzaniga recognized the value in capitalizing on what this unique patient population had to offer to the advancement of neuroscience. Among his many accomplishments are serving on the President’s Council on Bioethics between 2001-09, basically founding the field of cognitive neuroscience with fellow psychologist/linguist George A. Miller, and being awarded the Guggenheim Fellowship for Natural Sciences. Come hear him talk on lessons learned from split-brain research this Tuesday, January 12 at 4 pm.

Ege A. Yalcinbas is a first-year student in the neurosciences graduate program currently rotating in Dr. Chalasani’s lab. Michael Gazzaniga was one of the first neuroscientists she read about in high school so she is excited to fangirl him at his talk on Tuesday.