A Personal Aside
This is a post I have been thinking about writing for a while (since 2012). Off and on I think back to my previous life in academic science and I think there are some lessons or advice I might be able to give others who are contemplating graduate school. I will follow my course through undergraduate research in this post (Part 1). Part 2 will start with my repeated failures to get into a graduate program and then we’ll see where I go from there.
At the outset, just to clarify for those that do not know me, here is my introduction. I went to the University of Utah and studied biology from 1999-2005, went to the University of California at Davis and got a Ph.D. in neuroscience 2007-2012 (it took me two years to get into graduate school). I did my first postdoctoral stint at the UC Davis MIND Institute 2012-2013 and my second postdoctoral stint at the University of Utah in 2013-2014. After that second postdoctoral scholar position, I left academics to be an elementary school special education teacher. You can read my story about that in a previous post.
Freshman Year, 1999-2000
The start of a career in research
When I started at the University of Utah, I knew I wanted to study the brain, and I had just assumed that I would be doing so in the biology department. I registered for classes and moved into the dorms. Then I got on the internets, such as they were in 1999, and looked for a work study job that would earn me some money during college.
I looked through the options, I could work in the library, I could work taking payments in the parking lot, I could work in student services. No biology. I kept scanning and found what looked like an interesting-enough job for me. It was a job in the psychology department working with rats. The way they wrote the listing I thought I was going to be some sort of a janitor in a research laboratory. Still science. It never occurred to me I would be able to do research as an undergraduate student. I applied.
When I showed up at the interview, I met with Paul Gilbert (current lab located here) and we talked about science. He told me I would not be a janitor, but in fact, I would be a research assistant. We chatted a bit about what I wanted to be in science, which meant we talked about my autistic twin brother and the fact that I had been fascinated by scientists ever since they had come to our house to do scientific studies on my brother and my family.
Paul hired me. Told me to show up the next day when it fit my schedule and I could start.
I showed up and I was routed to the other graduate student, Inah Lee (current lab located here). Inah was still working on his English, and he was talking about something I did not yet understand (the hippocampus), so I just nodded and tried really hard to keep up. He was trying to explain that we were going to inject AP-5 into the CA3 subregion of the hippocampus to inactivate NMDA receptors and disrupt memory formation, but I heard that AP-5 was a part of the brain and the CA3 inhibits hippocampus memory… It is a good thing Inah gave me some papers to read so I could understand a bit of what I was going to be doing.
That day I was given 6 rats (Inah determined 6 rats = 20 hours a week of work), instructed me on how to take care of them, and left me to it. My job for the next week was to hang out with lab members, cuddle my rats so they were used to me, and teach rats that Froot Loops were an awesome snack.
That first year I developed a love for research
Inah was an amazing mentor. He was actively engaged in teaching me how to perform the experiments. I am very tall and Inah is not, so we cut a set of holes in the curtain at both my height and at his height. That way I could look at my rats and take data and Inah could pop his head under my arm and watch the experiment as well. He would pop by 2-3 times a day while I was there to make sure his experiment was working. Some times (actually, at least once a day), Inah would sneak up behind me and scare the bejeezus out of me. Then he would check on the experiment.
One of my great experiences that first year was when I was able to suggest there may be a problem with an experiment and was actually listened to. I mentioned to Inah that I was seeing something weird, and I thought it looked an awful lot like one of the rats was cheating. I explained that early on it looked to me like the rats were looking back and forth at the cues around the arms of the maze before the moved, and had longer latencies early on. Later on, not only were the rats running faster, it looked to me like they were not looking at the extra-maze cues. I had a hypothesis… I thought that the rats had figured out an easier way to solve the task and were just using relative brightness to solve the task.
We were using an 8 arm radial maze that was oriented with half the arms aimed at the corners of the room and half aimed at the walls. There were objects hanging down from the ceiling at rat eye height just offset from the ends of the arms to provide cues. I saw rats not using these cues. I hypothesized that the rats were using dark (corner) and not dark (walls) as the cue — and thus solving the task in a different way than we wanted. Since the rule they were learning was to go to a different arm on the test trial than the sample trial, they only had to remember dark vs not dark. This greatly simplified the task for them. Inah let me run an experiment one day to prove or disprove my hypothesis (yay for the scientific method).
Cut to the chase, the rats were cheating. I verified this by the following experiment: The rule was to go to a different arm the second time to get food. I realized I could use any arm aimed at a corner of arm 1. Then open different arms aimed at different corners and the rats avoided them–all of them, no matter how far away from the original arm. They went to any of the arms pointed at walls, no matter how close to the original arm aimed at a corner. Smart little buggers.
Inah and I put in a curtain surrounding the maze so that there were no corners. Training started again. It took the rats an extra week to learn the task, but they were clearly using the spatial cues. This was great because I felt my opinion was valid and my observations mattered. In other words, I was part of the team and we were all working to make the experiment work. (Note, the 2 experiments I worked on were later published here and here). Also, I had caught the bug. I had learned how to be a close observer of behavior and formulate testable hypotheses based on my observations.
I went to Honduras on an LDS mission from 2000-2002.
Sophomore-Senior years, 2002-2005
An unpleasant return
When I returned in 2002, Inah and Paul had left the lab for their postdoctoral work. I was referred to the graduate student in charge. I was set up to work with a rotating neuroscience graduate student. This was not so good an experience.
This graduate student was rarely around, and when they were around they were just messing around on the computers. Virtually the only time they talked to me was to tell me I was running the experiment wrong. Then they would retrain me to do the experiment exactly as I had been doing it. Ugh.
I remember being told I had to run an entire set of 12 rats (mine and another student’s rats) that night to collect data for a poster. I did it first thing in the morning. I felt exploited because I was being exploited, I was done with it.
This was a sufficiently bad experience that I talked with the head PI in charge of the lab (Ray Kesner). I told him I was going to leave at the end of the term and ask for a letter of recommendation so I could get another job. To be more precise, I was going to give up something I loved to do (research) because, as a result of being manipulated, had ceased to be fun or fulfilling.
A triumphant recovery
Ray didn’t let me quit. He would not take no for an answer. He told me he understood my experience this go round was bad, but I had done very well with Inah and Paul. He actually brought up the fact that I had helped Inah with the experiment in 1999-2000 and he did not want to lose me. He also thought it would be a bad idea for me to move over into the histology lab to work, as my skills would be wasted.
What Ray did, was he simply proposed I work with him directly instead; that way he could make sure it was a good experience. Ray gave me a list of 5 things he needed to be done in the lab to complete the aims of his grant and let me choose 1. I chose 3.
Ray let me finish up an experiment from Inah’s thesis that was not finished because Inah had left for a postdoctoral position. This was a spatial and nonspatial novelty detection task. Inah had only been able to run 3 CA3 lesioned rats on the task and they had extremely strange data. So I started over. I grabbed every rat in the lab that had a lesion to any hippocampus subregion and ran them on this task. In the end, I ran 41 rats on this behavioral task. The data were beautiful, and contradicted a few of the models prevalent in the field for hippocampus function, so we sent all the data off to Inah to write up for publication (Here).
While I was working on the paradigm from Inah’s thesis, I also worked on an experiment that Paul had tried to start when he was in the lab but was not able to perfect before he left. I jumped headlong into this project as it had not been moved past the pilot stage. This was a task to refine how the hippocampus is involved in the processing of time. This task was an object-odor paired associate task, except we decided to separate the object and the odor in time. We wanted to know if the hippocampus, particularly the CA1 subregion, was important for this temporal binding.
Ray let me refine the task, run a few control animals, and see what we had. The experiment looked good, so Ray taught me how to do ablative lesions of the rat hippocampus. I say taught, but it was more Ray did the left side, and then left the room and told me to finish it up, he would be back in 20 minutes. He didn’t come back. In fact, he had just left lab so I would finish the surgery without looking to him for help. It was a success. After that, he let me do my own surgeries so long as I ran the surgery coordinates by him before I started. I ended up doing 14 lesions to either CA3 or CA1 and we ran them on the experiment. Our hypotheses were confirmed! CA1 was involved for temporal binding of an object and an odor! I also at this point learned how to do my own histology so I could have control over the whole process. I did the surgeries, ran the experiments, sacrificed and perfused the rats, and stained the brains.
The experiment that I was working on was a success. When it was done, Ray and I actually wrote the entire paper on a plane trip from Salt Lake City to Providence, RI (Here). I was typing with the laptop precariously balanced on my knees while Ray talked a mile a minute with ideas I was frantically trying to formulate into readable sentences. How curious we two must have been talking on the plane about rats, hippocampus, memory, etc!
Around this time my future wife and I teamed up to do a behavioral task together. That is, she did the task and I provided the animals. We had to do full parietal cortex or hippocampus ablations (read, suck the tissue out) and she ran these rats on tasks evaluating the models put forth by Ken Cheng and Randy Gallistel. We extended these into hippocampus subregions as well (Here, Here, and Here).
While I was doing the surgeries for the object-trace-odor experiment, I refined and developed some surgical methods for ventral CA1 lesions in the hippocampus. When I had finished running them on the task I was working on (we included the data as a blurb in the discussion), I teamed up with Jason Rogers, a graduate student in the lab to do a quick fear conditioning experiment. We wondered if the hippocampus and amygdala were communicating information regarding fear when there was a time gap between a fear-inducing stimulus and a shock. What we found was astounding, the dorsal and ventral halves of the hippocampus could be dissociated for remembering the time gap between a tone and a shock, but not for the tone itself. Jason and I wrote the paper up (Here).
With these experiments, I was addicted to the study of time. There were not good models out for how the hippocampus processed time, and I wanted to develop one. I also had the opportunity with Ray, Inah, Paul, and Jason to write papers that fundamentally challenged the thinking of the field. The opportunity to write theory heavy papers had me set. I was going to he a hippocampus researcher studying time – at least so I thought.
When 2004 rolled around, Ray basically cut me loose on the lab. He put me unofficially in charge of the students running his side projects. He was in charge, but he left the day to day troubleshooting of the lab and experiments to me, knowing I would report back to him and we would be able to run things together. I also started to be the lab handyman, deconstructing and reconstructing lab apparatus as needed and cobbling old mazes together until we could finish experiments. Ray also put me in charge of pretty much all student training for undergraduate students and high school students entering the lab.
Also in 2004, Ray showed me an opportunity through the biology program to get $4000 from Pfizer (Summer Undergraduate Research Fellowship) to do my own research. Ray gave me a very ambitious project to propose. He gave me 2 or 3 papers that he said would help and suggested I look in the references lists to get more papers. So I spent a week in the library. And by week I mean five, 8 hour days in the library photocopying articles. In the end, I had 430 manuscripts photocopied and ready to use to write my project up.
What we were to study in my project was an output pathway from the CA3 subregion of the hippocampus to the medial and lateral septum. We had seen Mike Hasselmo’s models of cholinergic modulation of the hippocampus and we wanted to manipulate it. It also would answer a question about how information left the hippocampus after our CA1 lesions (how did CA3 information get out of the hippocampus if it only left the hippocampus via CA1?). To do this, I would have to develop a surgical method for transecting the ventrolateral half of the fornix while sparing the dorsomedial aspect under neurophysiological control. I had no idea how to do that, but it certainly did not stop me from writing the grant! Ray taught me that bold ideas are rewarded!
Ray let me write a complete draft of my proposal on my own based on a very simple template he gave me (just subheadings in the right place). Then he covered it with almost illegible notes and handed it back to me. I made changes and we had a series of 45-minute meetings to finalize the draft. Even before I was funded, Ray and I set up the logistics of the experiment.
Ray gave me some oscilloscopes and amplifiers that he had from brain stimulation experiments he performed in the 1970s. I built some electrodes. I called 3 veterinarians and 2 research colleagues of Ray’s to identify the appropriate anesthetic to do my experiments (we needed long induction of anesthetic using an injected anesthetic but would allow the rat to wake up at the end of the experiment). I developed a surgical method to precisely cut half the dorsal fimbria of the rat. I also had to email and call some colleagues of Ray’s in the Netherlands to learn how to do some histological stains and neural pathway tracing. I ended up having very difficult to understand telephone conversations and long, fruitful emails with these professors to help me get my work done. Ray and I, once I had the surgeries under control, designed the behavioral tasks and decided on experimental parameters based on our hypotheses and the computational models available in the literature. In short, we scienced like crazy that summer. I was putting in 8-10 hour days that summer and I loved every second of it.
When we were funded, Ray gave me my first official student. I say officially because I was working with all of Ray’s students, but this student was my charge. Ray gave me some advice on mentoring and left me to it. This student did an excellent job running the behavioral experiment on my rats. I considered her integral to the experiment, so I asked Ray if she could be involved in the paper, he said that she had to write the methods. She did and was a coauthor on the manuscript (Here). I consider her my first real mentoring experience for writing. I asked her to write the methods. She handed them to me and we read them and tried to run the task using her draft. A few drafts later and we had it. By now she was into it and asked if she could write more. I had a draft of the paper I had written, so I gave it to her along with the papers I was reading for theory and she ran with it. I am proud you can hear her voice in the manuscript.
I still had my time and the hippocampus ideas I was working on. I was well into attempt 3 of an experiment that was going to take 5 designs to get right. I was trying to expand on the object-trace-odor task by doing an object-trace-place task. We did eventually get it right and ended up with a neat little paper. I am proud of that one because I had 2 undergraduate students contribute to the paper and it is my “only-ever-accepted-without-revisions” manuscript ( here). It was such a clear, intuitive experiment on paper, but man was it tough to finally implement.
With these successes, Ray gave me a bunch more students and let me go with every project that we could come up with. I had 2 premed students working on one project I designed to try and determine what the opioidergic and NMDA-ergic receptors were really doing in the hippocampus to support learning. We used that first experiment Inah designed to test. We put either naloxone to block opioidergic plasticity or AP-V to block NMDAergic plasticity. Then we tested spatial and nonspatial novelty detection.
Being premed students, I asked Ray if it would be okay to teach them how to do their own surgeries and do their own drug infusions. He said that was fine, but it was my responsibility to teach them. These two premeds stepped up their game. They each did 10 cannula implantations, prepared their own reagents, and did the drug infusions unsupervised. They actually wrote an early draft of the paper for me using references I found for them (Here).
I also took the time to run some of my rats with fimbria transections to compare with data collected for Jason Rogers’ masters thesis. We combined them into a fun little paper critiquing a model of cholinergic modulation in the field, even finding a nice memory facilitation effect (Here).
I then took a quick segue back into studying time. A long term volunteer and I designed and carried out an experiment to study how rats learn sequences while running in a runway. Again, I tried to couch this study as a validation of one of the models in the field for how we use spatial information to predict temporal information, and simultaneously to refute a couple of more simplistic models (Here).
To further get into how the dorsal and ventral halves of the hippocampus work to process fear-related information, I developed a ventral CA3 lesion protocol that was reliable. It is surprisingly hard to lesion ventral CA3 because it is so close to ventral CA1. But we figured it out. We showed that ventral and dorsal CA1 and CA3 cooperate for encoding of classical fear conditioning, but they are dissociable for the retrieval of contextual fear, and ventral CA3 even seems involved for learning auditory-cued fear, something that had not been described previously (Here).
In 2005, Ray took a sabbatical in Oxford to go work with an old friend and theoretical collaborator, Edmund Rolls. Ray left me in charge of everything that was not directly under the charge of the graduate students. This included ordering supplies, upkeep in the lab, maintaining the colony, feeding, checking up on the rats, verifying students were actually running their experiments, etc.
2005-2007 was the time I got who I consider to be my best students. They were, at least, my most difficult students that showed the greatest personal and professional successes. These two students were complete opposites (I wrote about them before Here). One of them required a lot of input. She was not going to mess up. She was going to do everything right. She demanded I give her thorough protocols and descriptions of what I wanted. And if I was not precise enough, I got the 20 questions treatment until it was crystal clear. The other student was, to understate the reality, extremely independent. She wanted me to give her the broad strokes and to let her get to it. She liked to talk research but not be checked up on. So I let her know I was there whenever she wanted to give me an update or help on anything.
I also think I may view these students with fondness since they helped me on my most difficult experiments. One of them was involved with an extension of my dorsal fimbria transections work. She worked with me and even helped me to develop a method to cut the dorsal fornix output of CA1, since that also communicated to the septal nuclei, and I wanted to see what it did. I also wanted to tie the cholinergic modulation story in a neat little bow. She helped do the surgeries, independently at the end and she parameterized the experiment for me (she did and told me she had done it, so I said okay). She did both modified Hebb-Williams mazes (Here) as well as classical conditioning on her rats (Here). And one day while she was on vacation, I did a temporal ordering task, just out of curiosity (Here).
This student, as her last experiment, took on a pet project of mine. It was proposed that the different parts of the entorhinal cortex could be dissociated for the type of information. they send to the hippocampus. So this student and I designed a task. She took a high school student over her last summer and they did the experiments. She sent me the data almost 2 years later because she had forgotten. But it showed an effect, quite clearly in fact (Here).
The other student was a dynamo in her own right. She just quietly got everything done. She trained another student in how to run a temporal ordering task for me. She even helped me to develop a task by showing me that I had made a mistake/oversight in my design. She helped show that the dorsal and ventral halves of CA1 can be doubly dissociated for temporal ordering of different types of stimulus (Here). I extended one of her experiments to all hippocampus subregions out of curiosity (Here).
In another experiment, she also helped me out in my design by actually doing the design. I gave her a rough sketch and she told me what she wanted me to do. Together we showed that even the CA3 region of the hippocampus is dissociable into smaller units. She developed our rearing methods as a valid behavioral tool (Here).
Another student that had been working with us for a long time found success during this period. He had been working on our temporal ordering tasks for years, but they never panned out. We troubleshot a few experiments and he was able to show that the dorsal and ventral hippocampus are involved in different ways for sequences of odors (Here) and for olfactory working memory (Here).
When Ray returned from Oxford, we designed another experiment. Richard Morris’ lab had shown relational encoding in the hippocampus. We thought that was great, but we felt that they were incorrect because to be relational, then the association has to be symmetrical. To test this, we made a task on a board that was an object-cued location and location-cued object tasks that were interweaved. We found that CA3 was clearly involved in this relational type of processing. Supporting the earlier experiments as well as a prominent model in the field concerning how the hippocampus functioned (Here).
During my time in his lab, Ray and I spent a lot of time in front of the chalkboard in the hall. We would steal the chalk from each other and draw out diagrams and hypotheses. We would try to put the most current research paper’s results into our model. We would try to make a better model. We would try to dissect other people’s theoretical models to develop hypotheses we could test. We would propose new behavioral and surgical methods necessary to answer our questions. We would explore human neuropsychology as related to our research, and we would talk about the emotional and academic well-being of our students.
Ray also taught me some lessons I think students often miss out on. He taught me the importance of being able to simply call colleagues on the phone with questions since often times emails get buried. I was cold calling his colleagues to ask about drug targets, which drugs to use for which receptor, dosages, surgical methods, histology, neurophysiology, and everything in between. Ray taught me that scientists want to help each other. One just needs to ask and be bold enough to ask confidently. These big deal scientists soon went from “scary masterminds” to normal people that I can call good colleagues.
Ray taught me that writing papers defending our theory are fun. He cut me loose to write theoretical papers based on our conversations. The more left field the ideas the better. If they were counter to the field. Even moreover the better. If they would bother people and invite backlash, better still. Ray even got to the point he would just tell me to go have fun with it so he could see how outlandish a theory I would come up with, then we would get together and refine it until it was just crazy enough, then we would publish it! (Examples: Here, Here, Here, Here. This actually continued well after I left Ray’s lab. We still to this day, talk on the phone about new ideas and how we can make certain aspects of the field of behavioral neuroscience, particularly as relating to the brain, change their thinking to align better with our theories.
Interestingly, Ray also early on let me take care of responding to reviewers and be the corresponding author on manuscripts. He felt that if it was my idea and I did the work, then I am the one that needs to be asked about the nitty gritty details. He even supported me when I got a little terse with reviewers that were clearly taking some argument out on our paper. Got to put “Reviewer #3” in line sometimes!
When Ray was approached by researchers from other universities to learn methods or skills, Ray gave me the opportunity to teach them. He had me teach hippocampus lesion methods to a colleague from MIT and their graduate student. He had me teach cannula implantation and behavioral methods to another colleague from Quebec, Canada. When he received emails about nitty gritty technical details of an experiment, he referred the scientists to my expertise.
Ray introduced me to scientists that he collaborated with. He introduced me to scientists he disagreed with, so we could have long discussions. Ray would intentionally send colleagues that disagreed with me over to my poster to hash it out with me. He was especially quick to do this with the big deal researchers in the field that we were criticizing or whose theories we were testing. In the end, I feel I developed a positive rapport with a lot of BSD scientists. He helped me build a scientific network. He gave me a positive reputation as a behavioral neuroscientist by giving me credit when my work was essential to an experiment’s success. And he let me do the same for my students.
After I got into graduate school Ray confided in me what criticism he had of me in my applications. He told the committees when they called him that if I did not know something I needed to know for my research, they should just wait 2 weeks for me to read every paper on the subject and I would make myself an expert on the topic! What Ray was really saying was that my enthusiasm for learning and doing research might overwhelm others. I have a passion for creativity, hypothesizing, learning, exploring, etc.
After six years, I left Ray’s lab fully capable of running my own lab, mentoring students, writing papers and grants, and collaborating with other labs around the world. I got more training than most students today. There wasn’t a single aspect of research and lab management I wasn’t deeply involved with. I had 1 chapter and 9 papers published and 4 papers in press when I left. My road forward seemed clear: go to graduate school, finish my Ph.D. in 4 – 5 years, postdoc for 2 years, and finally land my first assistant professor position. By all measures of success at the time (i.e., publications), I felt I was a successful researcher and that if I continued my hard work, my future goal of becoming a professor was realistic. Boy was I wrong…
…to be continued.