From Dual Degrees to Neurosurgical Leadership: An Interview With Dr. Julie G. Pilitsis
June 19, 2024
No Comments
My name is Dr. Lorenz S. Neuwirth. I am an Associate Professor at the Department of Psychology and SUNY Neuroscience Research Institute at The State University of New York (SUNY) at Old Westbury. I have 10-years’ experience clinically as an applied behaviorist working with children and adults with autism and over 20-years’ experience working with preclinical models of neurotoxicology and neurodevelopmental disorders simulating human neuropathology. My background is in developmental-behavioral neurotoxicology, where I specialize in developing and understanding more sensitive and precise ways of analyzing and capturing animal behaviors. This is done so that we can understand and utilize preclinical models for future drug development and/or screening early drugs that are clinically promising and offer biomedical translational benefits to society.
As we look at the timely papers that come out evidencing claims of improving and/or ameliorating particular neurodevelopmental disorders or clinical symptoms of disorders (i.e., anxiety-like disorders), we noticed there were a couple of alarming trends that require serious attention and are “a bit unfortunate”. As technology has been developed very rapidly to advance preclinical studies in the behavioral neurosciences, there seems to be less of an advancement of undergraduate and graduate students as well as some early faculty and research scientists that are classically trained as behaviorists in the neurosciences. Thus, they lack regard or insights into interdisciplinary training, systems levels neuroscience, and rely too heavily on molecular biological parameters as the end all. This is concerning as the behavior of an organism is arguably the most important expression of the brain’s computations and its responsivity to a particular context and environmental stimuli with its stored lived experiences to build off the ideas put forth by Dr. Ian Q. Whishaw from the University of Lethbridge, Canada.
When we randomly look at some peer-reviewed journals, most of their Editorial Boards fail to include a behaviorist in the Associate or Section Reviewer’s area for quality controls and oversight of publication. This situation creates real risks of publishing manuscripts that may not even have somebody with the expertise overseeing and reviewing the articles that come through the process. Additionally, some top peer-review journals that pressure authors to pay their article processing charge fees can also care more for incentivization than having these quality and integrity checks in place regarding the science. These issues are very important for both the scientific community and the public to understand the root phenomenon of what is being actually examined by the researchers to advance public interests, advocacy, and care of people facing real world issues. What these select “peer-review journals” translate into society in terms of next steps or advancements, improvements, or issues if no such quality controls and integrity checks are in place is obscure at best? There is a greater need for research and more clarity of the behavioral science methodologies to inform a wide-range of researchers to build upon the work and go into different directions that will help society expand its understanding of a given topic and to collectively and responsibly work towards a reduction and/or cure for a given neuropathology.
To shed some further light on this topic, when my students Michael T. Verrengia, Zachary, I. Harikinish-Murrary, Jessica E. Orens, and Oscar E. Lopez and I were looking at some published papers regarding our own research initiatives, we soon realized that a series of problems were occurring every time we looked at an article that we were using for our own research purposes where they failed to indicate lighting effects and the tests that they use when studying preclinical animal models of anxiety-like disorders.
Lighting is an important variable that behaviorists look for in terms of controlling baseline measures and manipulations of anxiety and/or stress, as well as using it to set establishing operations for the motivational principles to cause a rodent to respond as naturally as possible to a light stimulus to evoke these behaviors to assess their function in a very specific way in a particular behavioral apparatus.
One wonders how much anxiety-like, stress-like, or other motivational responses are being elicited, by these particular experimental setups that deviate from the classical behavioral methodologies and how aligned these experiments are with the principles of behavioral testing? What is more concerning is the question that arises as to how many artifacts of false positives or false negatives are being produced in the field of behavioral neuroscience through these journals that fail to have these necessary behavioral controls at the peer-review levels.
When we started reviewing these suspect publications, we went through this massive literature review covering 10 years of research articles looking at Elsevier’s database and used the top 20 research articles that came up based on downloadable PDF availability to read and assess the full methods used in 420 publication. We further had an inclusion criterion for reporting lighting measures, sex, animal strain etc. and found that after screening for criteria from the 420 articles, we ended up with roughly about 11% or 46 of the articles actually meeting criteria that actually using lighting in the proper way. That’s an alarming statistic because that tells us that from this estimate roughly 89% to 90% of all preclinical animal research being done in behavioral neurosciences that require lighting for controls may not be well controlled at the preclinical level but are trying to advance drug development to address neuropathological phenomenon without proper utilization of these behavioral tests. We found that research evaluating social behaviors should be done in lighting less than or equal to 30 Lux and research evaluating anxiety-like behaviors should be done in lighting equal to 300 Lux but no greater than 1,000 Lux.
The question that is raised is how can preclinical and clinical researchers overcome these problems to advance better drug treatments in the field to help society. Behavior is the most important aspect of the translation of the brain’s expression and by not properly controlling for that, researchers may be limiting the organism’s functionality in both normal and contrived test conditions. Researchers can contrive and play with specific stimuli they want in terms of in-vitro experimentation and then come up with, at best, speculative conclusions that what is found to be manipulated at the cell biological level may or may not translate in the behaving organism. In the end, it is all in the behaving organism under specific learning, adaptive, and contextual stimuli that permit researchers proper evaluation of the complexity of the brain in a given environment. If that environmentally situated behavior doesn’t match or is artificially inflated, researchers may be offering conclusions that are unfortunately false positives.
When you think of animal behavior and not just in the classically trained definition of observation techniques, but in the ethological sense (i.e., the natural observations in the field), the goal is to try to simulate as much of that as closely as possible within an artificial laboratory environmental setting.
There’s a tradeoff of how much precision can you get within the artifacts that you’re working with for a given behavior test. Essentially we create the conditions for the background noise and try to dial in on the behaviors while increasing the gain for detecting those behaviors and reducing the noise simultaneously. That’s what really underscores the testament of comparing paper-to-paper with reliability and external validity checks from one research lab to another, replicating the phenomenon, and believing beyond assumptions that these results are real. It also begets a situation where if done correctly others will repeat it correctly, whereas if done incorrectly problems in the field can quickly and pervasively be replicated. When researchers look at any behavioral apparatus that comes out and its associated software, they find themselves getting into a lot of these issues with trying to understand what the behavioral phenomenon is and what parameters the software are measuring to really indicate what’s happening in an organism when under their experimental test conditions.
One clear example of the work we are trying to do to help researchers become more conscious of classical behaviorism in a modern neuroscience approach to testing of anxiety-like behavior is through the Light Dark Box where there’s a light chamber and a dark chamber that assesses the animal’s ability to avoid or tolerate “anxiety-inducing stimuli” (i.e., anxiogenic stimuli). The “anxiety-inducing stimuli” is a bright white light at 300 lux to 1000 lux. The Light Dark Box is used to understand whether the animal has a tendency to go out of the dark chamber and its natural motivation to go into the light chamber when you treat it for anxiety.
The Light Dark Box has been a standard tool for preclinical anxiety-like behavioral assessment for many years, but it’s very contrived in a light-on, light-off condition. In the early 90s, Dr. Jackie Crawley’s lab at the National Institutes of Health (NIH) created this wonderful “three-chamber social interaction test” that tried to induce these low light conditions while offering opportunities to either compare the mouse’s ability to be near or in proximity to another captive mouse underneath a wire mesh cage in the right chamber location versus a novel object in the left chamber location. They had a center chamber for the animal to freely traverse. The problem is that all the behavioral video tracking through those analyses were done from a bird’s eye view and none of the videos ever showed what the captive animal or mouse was doing in the wire cage relative to the behaviors of the actual mouse being tested to fully understand the context of these behaviors before making any claims of translations.
At that time, they didn’t know what socialization behaviors were besides this very poorly defined theory of being in the same chamber in local proximity to the captive mouse. Now that mouse could be grooming itself, its back turned, could be sleeping, you wouldn’t know, but time near meant it was socializing. This would be the equivalent of you sitting next to a stranger on a public park bench and both of you are on your cellphones and never make eye contact or talk, but you would consider this socialization between the two of you since you sat next to one another. Despite these limitations, this test became a golden standard.
When I was pursuing my doctorate at The City University of New York (CUNY) Graduate Center with my advisor, Dr. Abdeslem El Idrissi from The College of Staten Island CUNY , we started looking at this problem in the field in a similar preclinical model, Fragile X Mice. We did a mashup of the light-dark social interaction chamber with the three-chamber social interaction test. This combination of testing controls served to detect whether the animal was genuinely not interested and it left the center chamber to go into a completely dark box to confine itself or alleviate the social stress and anxiety-induced by the presence of the other mouse. The apparatus also had the middle chamber to socially explore the stranger mouse through a restricted social hole. The test was conducted using a side view animal behavior tracking system to see if they would go snout-to-snout, if they would attempt to go into the social hole while the other mouse went away, if they were both socializing at the same time or if the stranger was trying to socialize in the social hole, but the test mouse avoided it.
We were able to get actual data that mimicked what we saw on Fragile X patients from data produced by Dr. Vicki Sudhalter from the New York State Institute for Basic Research at the George Gervis Clinic that indicated that when you work with human Fragile X patients, they don’t make eye contact when you make eye contact with them. Still, when you look away, patients with Fragile X turn their heads towards you to try to make eye contact because they want to socialize, but differences in their brain circuitry suggest that they don’t feel socially comfortable enough with the joint eye contact to lock in for that moment of time. You can see in the behavioral recordings of these interview sessions with patients that have Fragile X that they’re looking to engage in attempts to socialize, but when you make eye contact with them, they socially avoid it.
We tried to mimic that clinical phenomenon by using this side view behavioral tracking approach in our version of the Light/Dark Social Interaction Test and we were quite effective at capturing the preclinical analogous model of this human social behavioral phenomena. We are in the process of finally finishing up a paper on it that actually illustrates this because we compared the three-chamber Jackie Crawley version of the Light Dark Chamber with our version of the test for reliability and showed that we got the same time in the chamber, but we got more sensitive data with that corroborated with the actual clinical diagnostic criteria for Fragile X as a snippet of the larger spectrum of autism spectrum disorders.
We suggested from this data that our Light/Dark Social Interaction Test is a more sensitive tool that may advance what has been attempted in the current genomic studies of mice looking for autism candidate genes or clusters of genes, but may have been skipping over some critical factors prematurely due to ineffective testing methods.
Yes. This particular apparatus is called the Light/Dark: Social Interaction Test and was invented by myself and my doctoral advisor Dr. Abdeslem El Idrissi. We see this test being multifaceted where you can use it for both mice and rats, so we’ve made a larger scale model for the rats because in the field it’s mostly mouse genetic models that are cheaper and more easily engineered, but for cost-prohibitive reasons in rats, it’s just too expensive. However, it doesn’t preclude people from studying this phenomenon on larger organisms. Second, the Light/Dark Social Interaction Test really controls for ecological motivation in the rodent due to whatever treatment or genetic manipulation they have, it is either interested or not interested in socialization, or it is found to be stressed or depressed, etc.
Many clinical applications besides autism could be investigated through this Light/Dark Social Interaction Test because of the natural tendency for rodents to be in the dark and not have the motivation to leave. Additionally, you can control for other factors in the Light/Dark Social Interaction Test such as the mouse or rat of choice versus a stranger of the same or opposing sex, differences in socialization as a function of age, versus familiar versus unfamiliar rodents. There’s a range of possibilities that still have not been assessed correctly in our field because we focused on static social parameters or protocols: the alone condition, the stranger condition, enrichment conditions, and then you wrap the experiment up as an autism index of behavior. There is much more to investigate with different protocols using the Light/Dark Social Interaction Test.
This is to be argued no differently than B. F. Skinner who created the Operant Conditioning Box and people using it with different schedules of reinforcement. The protocol originally published was not the protocol used. You can use the original apparatus to invite the creativity of the researchers and their teams to investigate real-world problems through different utilization of the same apparatus. That’s what we are trying to do here with the Light/Dark Social Interaction Test, making people become a little bit more open-minded and not functionally fixed to this social behavioral apparatus because it is not meant to work that way. Researchers should never think of a behavioral apparatus being fixed as it then limits their scientific creativity and advancement of science in the field if it is done in that capacity.
We get to a point where we tried to revisit some of the 1950s and 1960s efforts that researchers had to go into by designing their own equipment, and today it’s a little bit more intriguing because we have a lot of open neuroscience tools, including software and hardware that competes with these commercial vendors. One of the things that really expanded the market in the last 20 years was 3D printing patents which became open access to the public. This changed the game significantly and researchers can now go with a couple of open-source tools, hardware, and software, and 3D print their own apparatus for whatever purposes they need and build them with some training in basic computers and coding.
Researchers can have their own equipment without vendors, but the problem is that it requires a lot of troubleshooting. It requires a lot of training in some capacity to link everything to the actual concept in the neuroscience field that researchers may be trying to study. When it comes down to behaviorism, you can make all these toggling components all you want, which is no different from many of the current vendors in the field, but it’s how you make them to precisely explore the functions of the particular behaviors that you really need to study. I think it goes back down to those classical behavioral definitions and how we approached organisms early in the field from the early days of the roots of biopsychology that actually led to modern psychology and translational clinical psychology. If you look at the parallel structures that emerged from early biopsychology, chemistry and physics, that was the beginnings of medical physiology that came about in that capacity.
When we look at these early integrations, they had to use a lot of hardware for real-time capture of observation data points \ that they were using to make their claims of every investigation that they were doing. But now we’re in a time-period where we have a lot of high-tech equipment, and the problem that comes up is that people tend to lean towards more advanced equipment than very simple ways of studying behavior. The more complex you make a problem; you move further and further away from the rule of Parsimony that the simplest explanation is the best.
We see a lot of these issues in bioinformatics right now; there are so many color graphs and color charts and data on very small hypotheses, and it takes six or seven pages of data to explain what the question was and how they went through the data analysis to get to the data endpoint. In the end, it still doesn’t lead to a functional outcome through a living behavioral animal and the more data extractions and visualizations required pose opportunities for more human error.
We have all that information underscoring the big data, but what does it translate into besides this kind of intervention and response, or a stimulus response, or cause and effect outcome on a gene, or a protein or some RNA, or what have you? All important things…but until it goes back to the functional behavior of the organism, it’s an isolated event and it’s as good as for example, a yeast growing in a culture flask or literally a static set of brain cells in culture in a Petri dish. Until you put them in a living organism, you won’t know how they work correctly. If these models don’t change with those interventions and those manipulations in any aspect of the organism’s behavior, then the reality is something in the neurobiological circuitry compensated and those effects don’t hold.
We get lost in these translations. It’s probably best to start with behavior and work backwards, which is really where we see the clinical phenomenon of symptomatology and what the annals of neurology suggest. Oftentimes, we get caught up with bioengineering mutant animals and then going from that direction forward, and sometimes we miss the cross-intervention point. Additionally, if the mutant models create inadvertent side effects, we may be spending many years assuming the direct gene manipulation is causing these outcome measures when it may be some other indirect gene influence. Thus, we don’t have any parallel structures to work off.
Well, I have two little boys at home, a three and a five year old, which rightfully take up much of my time. But that’s mostly my daytime routine and in between getting them to school and addressing their needs I am in a variety of work meetings to help the University function. We are a social justice-empowered University with a specific mission. Every one of our researchers is dedicated and committed to changing the world through their field of expertise, even if it seems so esoteric or far-fetched. It is directly linked to a social justice initiative. As faculty, we have a lot of creative think tanks trying to marry concepts together from within a discipline to cross-disciplinary outcomes; this is also seen at larger scale and collaborative institutions as well. Meetings like these take up quite a bit of time for such a complex attempt at dealing with society’s problems and bridging it to neuroscience in society.
We can study everything in a lab, but if it doesn’t match what people are concerned about or care about, then we have a mismatch that doesn’t help, not only to fund projects but also doesn’t deal with real concerns the public is having at any given moment. We need to have that dialogue to stay relevant in the ever changing moments of where the needs are and have a push-pull system with society, research institutions, and higher educational institutions so we can get more people interested in pursuing careers in STEM.
There’s a teaching responsibility where we are delivering that social justice curriculum, enticing young minds, even returning minds to the field, to really think differently in a more traditional liberal arts way where it’s across the curriculum. However, it is also done in a focused way of what could you do with these skills, even though you might not be in practical technical training in every single one of your courses, how could you use these particular skills to either upskill or reskill yourself to go back into your areas of society and fix these social justice problems either now or later?
It’s a tremendous challenge to go from a curriculum of that kind to translational thinking in the moment every single time we have a lecture. Outside of teaching, I do basic preclinical research to try to advance our initiatives. I’m particularly interested in environmental lead poisoning as my social justice issue and to try and help kids who are still facing these problems stemming from a man-made disease of antiquity that can easily be removed from society, yet continues to persist. I have been working on this for the past 20 years, developing effective preclinical stages of drug therapies to mitigate against symptoms of environmental lead poisoning.
We know kids are still being exposed to lead by living in houses that contain lead whether those sources are outdated water and plumbing irrigations or living next to industrialized plants or Super fund sites (which are basically electronic waste dump mines), that they will be exposed to these neurtoxins for years to come and in their earliest years this could rob them from their future quality of life.
Our goal is to see if we can give them some type of drug treatment compound that’s very affordable and that they could take every single day to essentially mitigate the negative influences that lead exposure has on the developing brain. Our early research in the preclinical stages have shown that through such a therapeutic intervention, we can actually recover a substantial amount of intellectual disability, emotional problems, and social behavioral problems in our lab rats exposed to lead through taurine drug therapy.
We hope that within the next year or two, we’re going to get this work into a clinical/pilot study on lead poisoned children to see if we can move it to the FDA to see its value and actually get the evidence to solidify it as a standard pediatric therapy for kids with lead poisoning nationally and internationally.
I’ve worked with a number of different tools and components, in particular behavioral neuroscience tools. Right now I use BehaviorCloud as a more versatile behavioral software system that allows you to capture behavior in the moment anywhere you are using your phone or another type of video capture device or webcam, which is a little bit different from prior tools I have used in the past. BehaviorCloud allows you to use it in different rooms or with different cameras, which is unique and offers limitless capabilities. There are more functionalities and accessibility advantages in BehaviorCloud than its competitors. In terms of electrophysiology, I have about 80 instruments I work through, so I have a range of different electrophysiological applications for in-vivo recordings, and in-vitro slice recordings.
When it comes down to imaging systems, it depends on what you’re able to get your hands on. The imaging world is really costly in terms of million-dollar confocal microscopes etc. For me, working at a primarily undergraduate institution (PUI) and a minority-serving institution (MSI), we don’t have budgets for those items. So, I have to collaborate with local universities to obtain such access to these resources, where right now I’m collaborating with Dr. Randy Stout who is the Director of The New York Institute of Technology in their Center for Biomedical Innovation and Imaging Center. At this center I’m able to use a super-resolution confocal microscope for my brain imaging project, this is a multimillion-dollar confocal microscope to capture the fine details in the fibers of neurons.
We have to leverage things like this to get tools and access as it doesn’t allow my students who are in the pipeline to see these kinds of cutting-edge technology. It goes back down to where we started in the conversation of “How do you leverage these massive gaps?” Current students don’t have access to “tier-one” universities in the upper 10% of the nation and, 90% of them are going to other universities with similar challenges like my own. They won’t ever see this high-tech microscopes until they get to a doctorate or post-doctoral educational training opportunity, which is just too late in today’s research pipeline to be successful and to remain in the STEM fields just to have this experience. We try our best to leverage as much as we can as faculty that take substantial time and effort to either train them indirectly or somehow have them chaperone us to be able to access such equipment.
It’s really hard to say outside of the everyday products, what products we use because there’s also much product turnover occurring at a rapid pace. There’s also product cancellation and in the sense of businesses getting bought out, taken over, divisions closed down, or products don’t get extended anymore, and then they become obsolete, new scientists are often clueless and just seek out money to buy equipment that they were simply trained on. I still even use programs like WinLTP for long-term potentiation that was built in the UK probably 20 years ago. I still use Grass amplifiers from the 1970s to do my electrophysiology experiments because some of the technology back then is far better than the stuff you commercially buy today. Although today, the technology is easier to setup and use with plug and play options, but very difficult if not impossible to trouble shoot when a problem arises and has expensive tech support services. Further, the new technology may not have the nuanced control dials that you need for equipment sensitivity so there are some trade-offs, researchers need to think about. We can also make our own equipment in which we don’t have to succumb to the limitations of the actual equipment in the field.
Years ago, in the early 1990’s you would subscribe to an old school, three-and-a-half-inch hard disc stored/pre-loaded program called Current Contents and have it mailed to you every single week, and you would get an update of all the articles that were there. Nowadays it is setting yourself up on public databases like ResearchGate, and LinkedIn to get some news updates through social media, or using typical journal database searches like Elsevier and Science Direct. Searching what you’re interested in every couple of days, weeks, or monthly keeps you aware of what is happening in your field. It becomes a timely and overwhelming process.
We essentially set these systems up like speed dating where you match your interest needs and get blasted with emails of different things that might be on your radar. Those tools with these matching algorithms have been very helpful to promote multitasking. They’ve also been able to use the same matching algorithms with grant-related opportunities such as programs like SPIN and PIVOT where you could set up the same kind of algorithms rather than you searching databases, you put in your preferences, and the databases return hits, and it makes it easier for you to go through the process and find grants to seek out funding.
I don’t, but I’ve participated in some podcasts in the past. Sometimes I listen to NPR on the radio. It’s a way to keep abreast with certain things, but news moves very quickly unless you’re at a specialized segment for things that may not match at the time you are listening. Podcasts don’t give you the details that you need so I’ve pulled away from them over the years, although sometimes it can become very interesting. I will use some for my teaching demonstrations or to make an example in a classroom, to become a bit livelier with the expert in the room. In that same regard, I use TED Talks to supplement content, which I think delivers the curriculum and gives an expert in the room where students would never see this person in the flesh otherwise.
I think the real challenge is how you pick and choose what to steer towards and what to steer away from, and balancing your approach in this kind of tech-heavy world where everything is in your face. This situation is very different than the private domain of behind the paywall of academic knowledge and then breaking through the paywall and talking to people about the stuff that now has such a language or field-specific disconnect and making that not just interesting for the student, but tangible enough that they can walk away and say, I got something out of that. I’m enticed and I want to know more. For example, when you watch a popular TV show such as “Stranger Things” and you find yourself watching the next scene and then the next episode…academics want to achieve that for our students with knowledge. That’s the way I try to get people to understand science. If we can’t get them to come with us for the next episode, the next class, the next lecture, or the next experiment, we’re failing because we’re not captivating their attention enough regarding the topic. It’s not that we must entertain you, we have to captivate you to bring you with us. That’s the hardest part about this job.
Start early and be intentional. One thing to be aware of in this field is the boom of a very diverse next-generation workforce that comes through the pipeline. Many of them have a lot of unique and different talents that we may not see as traditional talents, but if you take an intentional step back and you looked at what they have to offer, you can help them develop those traditional talents on top of their unique offerings.
You have the advantage of these other talents they bring to the table, and I think students get distilled or filtered for not being that template of the traditional student, which is the longstanding problem in academia that is slowing becoming undone. We’re starting to see shifts away from that and I think students have to be very intentional about what they want. They must commit, they must be willing to work, and just let themselves shine through the process.
If students find really good mentors who are willing to work with them; they will get every advantage coming to them through that process. The one thing that I try to even tell everyone, even my own colleagues is to get away from this idea of “grit”. It’s all over the field. I think it’s basically an insult to people, you shouldn’t be acknowledged for how hard you struggle but rather be acknowledged that you dealt with the struggle.
You should be acknowledged for putting in the time the effort and that intentionality to move to the next level, but you also need to be shown how to succeed. I think the efforts and the commitments by individuals in the next generation of STEM need to be reconsidered in a different model or a better definition, avoid the grit definition and reimagine people as really hard workers that are trying to overcome a substantial traditional to non-traditional academic gap, a shift in the way academia is functioning, and how to train the next generation of STEM workers, educators, and inspirers; the classroom is very different from research training and these environments are more critical than ever.
Most of us have never been trained fully for research. We’ve been given a bottle in the mouth, then a pipette in the hand and expected to make up the gap. Today’s educators have to go to great depths to build relationships with future STEM students and in must be a valued and appreciated for of return on investment. What faculty put into students today is what the world will gain as next generation STEM workers tomorrow.
Monday – Friday
9 AM – 5 PM EST
DISCLAIMER: ConductScience and affiliate products are NOT designed for human consumption, testing, or clinical utilization. They are designed for pre-clinical utilization only. Customers purchasing apparatus for the purposes of scientific research or veterinary care affirm adherence to applicable regulatory bodies for the country in which their research or care is conducted.
