Page 1


Volume 7 Issue 1

| Fall 2017

The Journal of Undergraduate Neuroscience

Photo By: Eunhyuk Chang, Francis Szele Laboratory, University of Oxford

Featured Articles

Identification of Target Stimuli in Speech Quilts Utilizing Neurobiology to Diagnose Psychosis

Fall 2017 | Volume 7 | Issue 1 | 1

2 | Issue 1 | Volume 7 | Fall 2017

The Undergraduate Journal of Neuroscience

Volume 7 Issue 1 Fall 2017

Copyright © 2018

Fall 2017 | Volume 7 | Issue 1 | 3

Editorial Board Editor-In-Chief

Connor Hile Class of 2018

Publishing Editors Tina Zhao Class of 2018 Kathy Dai Class of 2018 William Chen Class of 2019 Nidhila Masha Class of 2019

Managing Editors

Jackson Xu Class of 2018 Megbana Vagwala Class of 2018

Kanav Chhabra Class of 2019

Design Team Gehua Tong Class of 2018

Riya Dange Class of 2019 Esther Liu Class of 2019 Michelle Dalson Class of 2018

Faculty Advisor Leonard White, Ph.D. Duke University School of Medicine Director of Education Duke Institute for Brain Sciences

Rohini Paul Class of 2018 Chris Lai Class of 2019 4 | Issue 1 | Volume 7 | Fall 2017

*We would like to thank the John Spencer Bassett Memorial Fund for their generous support of this publication.

The Undergraduate Journal of Neuroscience

Letter from the Editor The human brain, while only making up two-percent of one’s body weight, is the most dynamic organ in the body, constantly generating action potentials and responding both with and without our awareness to external stimuli. The production of these neurons, also known as neurogenesis, was thought to have ceased in adults. However, recent studies have overthrown this antiquated dogma with the advent of new technological innovations and methods. This process continues throughout our lives with constant fine-tuning and refinement in response to everyday experiences. Our cover – as well as the inspiration for our publication title – illustrates this intricate process with an example of postnatal neurogenesis. These recent technological innovations had led to novel methods toward studying various disorders and cognitive processes. This issue of Neurogenesis introduces a variety of these techniques from transcranial magnetic stimulation to deep brain stimulation to neuroimaging. These tools have been used as therapeutic tools for treating disorders ranging from Parkinson’s disease to obsessive compulsive disorder. Furthermore, these studies have even led to answering more philosophical questions, such as the neural correlates of atheism and religion. Next, we would like to highlight our two featured articles in this edition. In Aurelio Falconi’s paper, he answers the question of speech perception by analyzing various segments of sound using EEG. He deciphers the differences between segment length and language familiarity and their relationship to target identification. Our other featured article by Geoffrey Bocobo discusses how one can synthesize various diagnostic tools for treatment and understanding of psychiatric illnesses. Bocobo utilizes eye tracking, electrophysiology, and diffusion tensor imaging to tease these tools apart in order to generate a more objective diagnostic approach. Neurogenesis functions as a platform for students to showcase their research in neuroscience to a global audience. Publications come from all over the United States and world, and this participation helps make the journal continue to be a success and pleasure for all of the readers and editors involved. Because of this, we would like to thank and share the work of students from Massachusetts Institute of Technology, Emory University, Trinity College, DePaul University, and Washington University in St. Louis. While from different universities, all of us share the common goal of exploring and understanding the inner workings of the brain. Sincerely, Connor Hile Editor-in-Chief

Fall 2017 | Volume 7 | Issue 1 | 5

6 | Issue 1 | Volume 7 | Fall 2017

The Undergraduate Journal of Neuroscience


TABLE OF CONTENTS ARTICLES 8 Identification of Target Stimuli in Speech Quilts: Analysis of Language Familiarity and Segment Length Aurelio Falconi

14 The Effects of Repetitive Transcranial Magnetic Stimulation on Insula-Based Functional Connectivity Folasade Abiodun

20 Utilizing Neurobiology to Diagnose Psychosis: Comparing DTI Derived White Matter Connectivity in DSM Diagnoses vs. in Biological Psychoses Subtypes Geoffrey Bocobo

27 Shedding Light on the Role of ipRGCs in the Human Eye During Light Exposure Lissa Neira

REVIEWS 32 The Role of Epigenetics in Alzheimer's Disease Aysswarya Manoharan

36 What Drives Us? Chasing Reward in a Dopaminergic Society Solana Liu

OPINION 41 Heritability and Neurophysiology of Psychopathy Implicate Deep Brain Stimulation as a Possible Treatment for Psychopathy Schuyler Gaillard

HOT TOPICS 45 The Brain on Atheism: An Analysis of Religion and Its Neural Roots David Graykowski

Fall 2017 | Volume 7 | Issue 1 | 7



Identification ofTarget Stimuli in Speech Quilts: Analysis of Language Familiarity and Segment Length Aurelio Falconi Duke University, Durham, North Carolina 27708 Correspondence should be addressed to Aurelio Falconi ( 1

Speech possesses temporal structure characteristics that must be processed at the cortical level to fully comprehend language. To better understand the neural process behind speech perception, we used sound quilts of speech—segments of sound reordered into a novel stimulus. Target stimuli were embedded into these speech quilts into a 2x2 factorial design of language (English vs. Korean) and segment length (30 ms vs. 960 ms). Our goal was to test whether familiarity with a certain language or exposure to speech quilts created with longer segments increases P300 amplitude and decreases reaction time. Using EEG and a behavioral analysis, P300 peak amplitudes and reaction times were compared across all conditions. We found that the human brain is more adept at identifying targets in speech quilts of longer segment length, F(1, 7) = 5.55, p = 0.05. This finding suggests that segment length influences target identification more than language and that language familiarity does not significantly influence speech perception.


In a world filled with auditory information, we must, from a very young age, learn to identify which sounds are meaningful. From an early stage in development, humans learn to classify a significant portion of these sounds into what is known as language, and response to language gives rise to communication among individuals. What is perplexing to researchers is the symbiotic nature of the brain and language, and how each may have played a role in influencing the other. The high specialization of language areas in human brains originated from a preexisting language system in the primate brain (Aboitiz & Garcia, 1997). We can even express ourselves in a multitude of languages including those that do not require an aural component such as ASL or Braille. The temporal structure of sound can be defined by three characteristics: envelope, periodicity, and fine-structure. These three attributes are found to 8 | Issue 1 | Volume 7 | Fall 2017

have a relation to speech and linguistic perception (Rosen et al., 1992). Envelope specifically refers to the changes in amplitude or intensity of a certain sound that can represent linguistic correlates in identification of language. Periodicity of a sound describes the tendency of sound to recur at intervals; the fluctuation of periodicity is directly related to the frequency of sound which contributes to our perception of pitch. The fine structure of speech refers to the small variations in wave shape in a period of sound, a more time-specific characteristic than periodicity. Fine structure influences formant pattern, the preferred resonating frequencies that give rise to the perception of different vowels or different voices being spoken (Delattre, Liberman, Cooper, & Gerstman, 2015). These characteristics work in tandem and are responsible for much of our perception and understanding of speech. The study which inspired this one found that the superior temporal sulcus is sensitive to the temporal The Undergraduate Journal of Neuroscience

Falconi | ARTICLE

structure of speech from an unfamiliar language (Overath, 2015). Thus, sensitivity to speech seems to be independent of the semantic content of the language from which it derives. Temporal structure is therefore suggested to be the more important factor in describing the acoustic structure of speech. In order to have meaningful communication, we also must be able to discriminate one language from another—more specifically, one word from another. Languages are constructed through phonemes— the smallest unit of speech in a language that can be distinguished from another. They are used in combination with other phonemes to create purposeful dialogue. A human is capable of producing a wide variety of phonemes, yet many similarities in phonemic inventories are found across different languages (Kohler, 1996). However, auditory experiments such as this one can still make use of the subtle differences between languages. In a previous study, Japanese and English listeners were asked to listen to Japanese words and identify target stimuli within the audio. Japanese listeners produced quicker and more accurate responses than the English listeners suggesting that speech processing seems to be language-specific (Cutler & Otake, 1994). Therefore, a great question is whether subtle phonemic and temporal structure differences between languages have a significant effect on an individual’s response to target stimuli. Target detection is becoming a widely used approach in neuroscience research. Different types of neuroimaging can reflect target processing in cortical areas. Electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) are often used together to create a more integrated understanding, both temporally and spatially, of cortical activity with target detection (Mulert et al., 2003). Experiments using auditory target detection can therefore tell us more about how auditory stimuli such as speech is processed in the brain. Research shows that through event related potentials (ERP), the audio target detection elicits the positive waveform known as the P300 (Murray, 2012). EEG measures the electrical activity in the brain and graphs the brain’s responses to stimuli. The neural component, the P300, seems to be independent of the modality in which it is detected in, implying that it is not modality-specific (Kiehl et al., 2001). The P300 is a specific waveform that can be observed from event related potentials (ERP). The presence and amplitude of the P300 can be used to assess cognitive function among individuals. This

waveform is elicited during decision-making and target detection making it an endogenous component. P300s are often found in oddball paradigms where the participant is instructed to detect a target stimulus embedded in a train of standard stimuli (Di Rollo et al., 2016). As can be seen through ERPs, the P300 is a notable positive increase in the electric potential specifically from the parietal lobes. The P300 has a latency that ranges from 250 to 500 milliseconds (ms) but is often found at 300 ms (Picton, 1992). The amplitude of this waveform is found to vary depending on the certainty of target stimulus detection (Polich, et al., 2003). For this reason, the P300 has been found to be an important indicator of specific cognitive processes such as attention and working memory—though this study focuses more on the cognitive task of attention (Linden, 2005). P300s consistently seem to reflect the idea that higher-level operations are involved in target processing (Gamble & Woldorff, 2015). A more explicit way to analyze the nature of a response to a stimulus is through analysis of the reaction time of target stimuli. Reaction times (RT) are used to indicate cognitive abilities and cognitive deficiencies. RTs tend to decrease as a person’s attention towards certain stimuli increases, because their attentional load becomes more dedicated to the task at hand. Thus, comparing RTs can give us information as to what conditions individuals can more easily focus on. Here, we investigated audio target stimuli through a 2x2 factorial design with factors language (English vs. Korean) and segment length (30 ms vs. 960 ms). We were curious as to whether these different conditions—language and segment length— have a significant effect on the peak amplitude of the P300. We embedded target stimuli in speech quilts to elicit the P300. We ask participants to listen attentively and respond to the target stimuli. The P300 amplitudes and RTs were then analyzed across languages to further investigate language familiarity and segment length, and their role in attention tasks. It was hypothesized that quilts comprised of the longer segment lengths in English would elicit a heightened response because they contained more familiar and naturalistic temporal structure.


Participants Eight individuals (four females) with a mean age of 23.5 and an age range of 18 to 40 were recruited Fall 2017 | Volume 7 | Issue 1 | 9

ARTICLE | Target Stimuli in Speech Quilts

for the paradigm. These individuals were affiliated through Duke University as students or employees. All volunteers were found to have normal auditory capabilities and to be native English speakers with no prior knowledge of how to speak or comprehend Korean. Participants consented and were given the opportunity to leave the experiment at any point in time. The EEG procedure was approved by the Duke University Institutional Review Board. No participant was found to have extremely poor task behavior or excessive eye artifacts in the EEG. Thus, all eight individuals were included in the analyses.

Stimuli and Procedure Bilingual speakers fluent in both English and Korean recorded the original audio sources of speech to be used in the study. Acoustic structure was manipulated and reordered into novel sound quilts. A special algorithm was used to synthesize these sound quilts into making smooth transitions between segments of speech. Segments of different lengths were stitched together in a manner where their segment-to-segment transitions were of similar temporal structure so that they resembled the original sound source (Overath, 2015). These resulting clips were thus manipulated into speech quilts each six seconds long. Segment length was adjusted so that half of the speech quilts were reordered with 30 ms long segments and the other half with 960 ms long segments. Thus, four conditions emerged from this 2x2 factorial design: English 30 ms, English 960 ms, Korean 30 ms, and Korean 960 ms. Embedded in a quarter of these speech quilts was a target stimulus. For the stimulus, we used sinusoidally amplitude modulated (10 Hz) white noise inserted at ~-18 dB lasting one second long. This stimulus was placed at exactly one, two, three, four, or five second(s) into the speech quilt. Participants were instructed to press the “t” button on a keyboard placed in front of them immediately after they heard the target stimulus. Their reaction times were recorded. The participants then listened to five runs—each possessing a total of 64 speech quilts containing equal amounts of all four conditions. Twenty-five percent of the speech quilts were target trials. Each speech quilt (or trial) was presented bilaterally via headphones while the volunteer was being monitored through EEG. Participants were asked to keep their eyes closed during the paradigm to reduce the likelihood of eye artifacts in the data. At the midpoint of each run, participants received a break and 10 | Issue 1 | Volume 7 | Fall 2017

were instructed to continue through the rest of the trial at their own leisure. They received an additional break at the end of each run. The entire paradigm (including set-up) lasted two hours on average; the experimental portion (from the beginning of the first run to the end of the final run) lasted roughly 50 minutes.

Recording and Analysis The EEG was recorded using a 64-electrode elastic cap (ActiChamp, Brain Vision). The EEG data was sampled at 1000 Hz. Electrode 31 was used to monitor vertical eye movements (VEOG) and was placed beneath the left eye. Electrode 32 was used as a reference electrode and was placed on the right mastoid. Only the speech quilts with target audios embedded in them were extracted from the data (25% of the total amount of speech quilts presented to the participant). The potential of each target condition was plotted across time with epoch limits of 100 ms before the onset of the target (-100 ms) to 1000 ms after the onset of the target (+1000 ms). A data scroll was manually performed through the extracted epochs of each subject where epochs with excessive noise or a large number of electrodes drifting were removed. There was a baseline correction applied to all epochs from -100 ms to the onset of the target stimulus at 0 ms. The data was bandpass filtered from 0.3 Hz to 15 Hz. Independent component analysis was run on all the epochs and specific components were removed if there seemed to be excessive eye artifacts present, though there were not many since this paradigm was run with eyes closed. The data from all eight subjects were combined to make grand average ERPs for each condition. The P300s were measured and compared across the four conditions to determine whether or not the differences between the P300 amplitude were statistically significant. Statistical analysis was used through the EEGlab interface in MATLAB. A behavioral analysis was also conducted using the RTs of each condition of all the participants combined. SPSS was used to conduct a two-way repeated measures analysis of variance (ANOVA) to analyze the effect of language, segment length, and overall interaction effect between both factors.


Behavioral Analysis Participants were able to accurately detect the embedded target stimulus in the speech quilts in 96.25% of the trials. The average RT for all trials The Undergraduate Journal of Neuroscience

Falconi | ARTICLE

was 637.11 ms. It is to be noted that because the target stimulus was placed in various parts of the speech quilt, the RT was measured from the onset of the target stimulus to the moment the participant pressed the “t” key. RTs were recorded and categorized into each of the four conditions with English 30 ms having a mean RT of 651.83 ms, English 960 ms of 632.72 ms, Korean 30 ms of 653.45 ms, and Korean 960 ms of 610.42 ms (Figure 1). The behavioral data was then analyzed through a two-way repeated measures ANOVA with language and segment length as the independent variables and the mean reaction time of each condition as the dependent variable. Using SPSS, a significant main effect was found for segment length condition, F(1, 7) = 5.55, p = 0.05, η2p = 0.44, where participants were quicker in discriminating target stimuli embedded in a speech quilt created from 960 ms segments than they were from 30 ms segments. There was no significant main effect for language condition, F(1, 7) = 0.57, p = 0.47, η2p = 0.08. There was not a significant interaction effect, F(1, 7) = .87, p = .38, η2p = 0.11 (Table 1). An additional analysis of segment length was performed to verify its significance. A paired samples t-test further specified that there was only a significant effect of segment length in Korean (30 ms: M = 653.45, SD = 145.09) (960 ms: M = 610.42, SD = 134.58), t(7) = 3.864, p = .006. However, no such significance was found for varying segment lengths in English (30 ms: M = 653.45, SD = 145.09) (960 ms: M = 610.42, SD = 134.58), t(7) = 0.812, p = 0.44.

Table 1: Two-way ANOVA of reaction times.





Language Segment Length Language * Segment Length P300 Amplitude Analysis We examined the amplitude of the P300 across all conditions expecting a statistically significant difference for P300s with varying language and segment length condition. Statistical analyses were run on the MATLAB EEGlab interface specifically looking at the differences in peak amplitudes across conditions. The two-way repeated measures ANOVA yielded no significant results. Though there was not a significant effect of language or segment length on P300 amplitude, the group average ERPs presented a P300 of around +12-13 microvolts for all conditions (Figure 2). The change in potential from baseline near 300 ms suggests the waveform present was the P300. This was further verified through the topographical ERP mapping of all four conditions (Figure 3). There was a positive effect at around the peak time (~600 ms) where the parietal electrodes were located.

Figure 2: Group Average ERPs for all 4 conditions in electrode 37 (placed on the parietal lobe). There is a notable P3 in each condition signifying a response to the target stimuli.

Figure 1: Plot of mean reaction times in each condition.

Figure 3: Topographical ERP plots of four conditions. The regions in red indicate there is a positive effect in the parietal area of the brain.

Fall 2017 | Volume 7 | Issue 1 | 11

ARTICLE | Target Stimuli in Speech Quilts


In this study, we investigated auditory target identification across a 2x2 factorial design of language and segment length to research how these factors can influence neuronal responses to speech quilts. Specifically, we hoped to elucidate a clear difference in peak amplitude of the P300 across the varying languages and segment length that comprised each speech quilt. Though no significant difference of P300 amplitude was found in any of the conditions, the observation of the P300 itself was promising. There was a significant decrease in the RT of individuals when listening to speech quilts of 960 ms segment length. However, there was no compelling difference of RT between English and Korean speech quilts. Through Figures 2 and 3, it can be inferred that the paradigm was successful in eliciting the P300. This finding is hopeful as it provides opportunities for future studies to investigate other languages or more extreme segment lengths. Different variables involved in speech processing could also be considered. Additionally, more can be done to improve auditory paradigms such as this one to continue to investigate the effects of target detection on P300s. Segment length had a significant effect where participants had shorter RTs for speech quilts of longer segment lengths, and more investigation should be done to verify these effects. Though the data analysis cannot prove that the parietal area is more sensitive to segment length in speech clips than it is to different languages, the behavioral analysis implies there is a difference in human response. It can be argued that though no neural differences were discovered in this study, other neuroimaging techniques can be utilized to fully investigate all components of the response. Additionally, because of the length of the experiment itself, the attention of the participant may have decreased as more runs progressed. This possible confound was not further explored as most of the participants claimed they had no issue running through the entire paradigm. Regardless, variations of the experiment can be conducted in order to counteract the possible decline in attention. The more naturalistic speech becomes, the easier it is for an individual to identify target stimuli. Attentional resources may explain this claim, as less naturalistic speech becomes more of a complex stimulus for the brain to process. Using Ornstein’s storage size metaphor to estimate reaction time, the duration of a reaction time is a direct function 12 | Issue 1 | Volume 7 | Fall 2017

of the amount of attention allocated to the task (Zakay, 1989). Therefore, processing time of the target stimuli shortens when a person is presented with more familiar speech sounds. A similar theory is that the brain cannot detect such small and fast changing fluctuations in speech such as those present in the 30 ms speech quilts, and thus, it cannot easily discern the target stimuli from the speech quilt itself. In conditions of speech quilts composed of longer segment lengths, RT decreased—suggesting that there are certain characteristics of speech that are easier to focus on. More of the auditory system must be explored in order to fully understand how speech processing works. Future studies using similar target detection methods can hopefully highlight other important findings on how speech processing functions in the auditory cortex. An interesting approach may be to analyze non-human responses to these quilts, particularly those of primates, to explore the origins of language specialization in the brain. If more evidence could be provided that the human language system originated from the primate brain, researchers can focus more on how these corresponding regions developed over time. Alternatively, sound quilts other than speech could illuminate whether segment length response is only speech specific. This can determine whether humans are highly sensitive to the temporal structure of sound as well as speech. Knowing how sound is processed can help us further understand the full scope of the auditory cortex and help improve treatment for speech and auditory deficiencies. Whatever direction researchers take to further understand the auditory system and target detection will ideally benefit the scientific community at large. REFERENCES

Aboitiz, F., & Garcı́a V., R. (1997). The evolutionary origin of the language areas in the human brain. A neuroanatomical perspective. Brain Research Reviews, 25(3), 381–396. Cutler, A., & Otake, T. (1994). Mora or Phoneme? Further Evidence for Language-Specific Listening. Journal of Memory and Language, 33(6), 824–844. Delattre, P., Liberman, A. M., Cooper, F. S., & Gerstman, L. J. (1952). An Experimental Study of the Acoustic Determinants of Vowel Color; Observations on One- and Two-Formant Vowels Synthesized from Spectrographic Patterns. WORD, 8(3), 195–210. Gamble, M. L., & Woldorff, M. G. (2015). Rapid context-based identification of target sounds in an auditory scene. Journal of Cognitive Neuroscience, 27(9), 1675–1684. Kiehl, K. A., Laurens, K. R., Duty, T. L., Forster, B. B., & Liddle, P. F. (2001). Neural sources involved in auditory target detection and novelty processing: An event-related fMRI study. Psychophysiology, 38(1), 133–142.

The Undergraduate Journal of Neuroscience

Falconi | ARTICLE Linden, D. E. J. (2005). The P300: Where in the Brain Is It Produced and What Does It Tell Us? The Neuroscientist, 11(6), 563–576. Mulert, C., Jäger, L., Schmitt, R., Bussfeld, P., Pogarell, O., Möller, H.-J., … Hegerl, U. (2004). Integration of fMRI and simultaneous EEG: towards a comprehensive understanding of localization and time-course of brain activity in target detection. NeuroImage, 22(1), 83–94. Murray, M. M., & Wallace, M. T. (Eds.). (2012). The Neural Bases of Multisensory Processes. Boca Raton (FL): CRC Press/Taylor & Francis. Overath, T., McDermott, J. H., Zarate, J. M., & Poeppel, D. (2015). The cortical analysis of speech-specific temporal structure revealed by responses to sound quilts. Nature Neuroscience, 18(6), 903–911. Polich, J., Alexander, J. E., Bauer, L. O., Kuperman, S., Morzorati, S., O’Connor, S. J., … Begleiter, H. (1997). P300 topography of amplitude/latency correlations. Brain Topography, 9(4), 275–282. Rosen, S. Temporal information in speech: acoustic, auditory and linguistic aspects. Phil. Trans. R. Soc. Lond. B 336, 367–373 (1992). Zakay, D. (1989). Chapter 10 Subjective Time and Attentional Resource Allocation: An Integrated Model of Time Estimation*. In I. L. and D. Zakay (Ed.), Advances in Psychology (Vol. 59, pp. 365–397). North-Holland.

Fall 2017 | Volume 7 | Issue 1 | 13



The Effects of RepetitiveTranscranial Magnetic Stimulation on Insula-Based Functional Connectivity Folasade Abiodun1 Duke University, Durham, North Carolina 27708 Correspondence should be addressed to Folasade Abiodun ( 1

Low-frequency Repetitive Transcranial Magnetic Stimulation (rTMS) is thought to have inhiitory effects on various regions of the brain. With this knowledge in mind, we sought to understand the cognitive and emotional effects of low-frequency rTMS on the posterior insula and its funcitonal connectivity, through observing the changes in distress tolerance, mood, and functional connectviity. Although 1 Hz vs. 10 Hz TMS stimulation had no significantly different effects separately, statistical analysis revealed that the presence of TMS stimulation in general did result in a decrease in functional connectivity.


TMS Transcranial Magnetic Stimulation (TMS) has been at the forefront of discussion in neuroscience as one of the leading methods of non-invasive neuromodulation. It is a method that utilizes the electromagnetic nature of the circuits in the brain, manipulating their tendency to fire and not fire by the application of an external magnetic field with a magnetic coil connected to TMS probe (McClintock, 2011). Various studies have compared the nuances of TMS, from the differences between repetitive and single pulse stimulation (Fitzgerald et al., 2006; Thickbroom et al., 2006) to the qualifiers of inhibitory versus excitatory stimulation (Watanabe et al., 2014; Caparelli et al., 2012). One major factor in determining the effects of TMS is target selection and the network connectivity of that region. It is believed that the downstream effects of stimulation – reliant on the ‘where’ and ‘what’ of regional connectivity – determines the efficacy of TMS (Fox et al., 2010; Fox et al., 2012; Watanabe et al., 2014). TMS targets are usually selected with this in mind 14 | Issue 1 | Volume 7 | Fall 2017

with regions such as the sensorimotor cortex being popular targets of choice due to their high functional connectivity (Cordes et al., 2000). Functional connectivity is also of great benefit to researchers as it allows for assessment that superficial research tools cannot achieve on their own. This is one of the limitations of TMS, as it can only directly impact neural stimulation as far as the initial strength of the magnetic field, which is roughly 2 Tesla for a brief pulsed magnetic field (Mishra, 2011). Rather than switching out this instrument for more invasive deep brain stimulation probes, however, we can use what we know of the maps and networks of functional connectivity (FC) to measure the effects of stimulation in the deeper regions. The Insula One area of interest to researchers is the insular cortex, more commonly known as the insula. The insula is located deep in the cerebral cortex, folded within the lateral sulcus. Prior stroke and lesion studies have shed some light on the specifics of insula function, especially with its role in addiction The Undergraduate Journal of Neuroscience

Abiodun | ARTICLE

and dependence (Paulus et al., 2003; Craig, 2009). It has also been shown that the posterior insula is connected to regions such as the postcentral gyrus and thalamus and is an integral component of pain perception and emotional response. This lends itself to our theories about the link between distress, stimulation, and insula-based FC. (Sabarato et al., 2006). Previous research techniques had not yet developed enough to manipulate insula-based FC without adversely affecting the natural processes (whether through lesions or other methods). With the emergence of TMS, however, we now have a technique to temporarily and non-intrusively influence and observe the insula’s network according to our will. This allows us to see its direct impact on processes related to distress tolerance, emotion, and awareness. This Study There are two goals in mind. First, we will aim to understand the effect of high frequency versus low frequency stimulation on the functional connectivity of the insula. We stimulated the postcentral gyrus – a target area known to be functionally connected to the insula (Addicott et al., 2015) – and observed the changes between these two areas on a measurable scale. Second, we will study the role of the insula in distress tolerance (DT) and whether manipulation of the insula results in either the strengthening or weakening of the subject’s threshold of endurance. We hypothesize that low-frequency (1Hz) stimulation will reduce resting state FC between the TMS target site and the insula in comparison to high-frequency (10Hz) stimulation. Additionally, we predict that low-frequency rTMS will increase distress tolerance in our subjects in comparison to high-frequency stimulation. We hope that such an experiment will not only help us understand the exact effects of TMS on the insula neural network but also shed further light on the downstream effects of TMS.


Participant Selection Participants were recruited from the Research Triangle Area (Raleigh/Durham/Chapel Hill). Subjects were right-handed, healthy individuals between 1855 years old and with no history of chronic physical or mental conditions. All participants provided written consent forms, as per compliance with the Institutional Review Board (IRB) at Duke University. Each subject underwent 2 magnetic resonance

imaging (MRI) scans, 5 repetitive TMS sessions, and behavioral DT tests.

MRI During the MRI, subjects underwent a structural anatomical and resting-state FC scan. Participants were in the MRI scanner for 1 hour. Functional data processing was modeled from the parameters of Addicott et al. (2015) and preprocessed using SPM12. DT As a measure of distress tolerance, subjects underwent a Paced Auditory Serial Addition Test (PASAT). Subjects had the option to quit the task at any point in time with the simple press of a button, but greater lengths of endurance/perseverance through the task represent greater affective distress tolerance. Participants completed the mood-related survey (PANAS) before and after each scan session and the DT tasks. TMS During the first rTMS session, resting motor threshold (MT) was determined by placing a figure-eightshaped TMS coil on the subject’s scalp and applying single pulses to the right motor cortex. MT was defined as the minimum magnetic flux needed to reliably elicit an electromyographic response in a target muscle. Individual MT was used to determine the intensity of stimulation for each individual.

Excitatory rTMS Parameters 10 Hz rTMS was applied for 5 seconds followed by a 20-second inter-train interval over the course of 16 minutes. Inhibitory rTMS Parameters 1 Hz rTMS was applied for 960 continuous pulses. During the rTMS application, the subject performed the hard phase of the PASAT task. The 5 TMS sessions lasted an hour.

Data Analysis The resting-state FC imaging data was preprocessed according to standard protocols in SPM12. The CONN toolbox (Whitfield-Gabrieli & Nieto-Castaton, 2012) was used to assess resting-state FC. The final sample size consisted of 17 participants (mean years of education 16 ± 2; mean age 29 ± 12). All 17 participants completed the necessary PANAS, PASAT and TMS treatments, with 10 receivFall 2017 | Volume 7 | Issue 1 | 15

ARTICLE | Repetitive Transcranial Magnetic Stimulation

ing 1Hz stimulation and 7 receiving 10Hz stimulation. Before any specific analyses were carried out, a between-group differences comparison using a two-sample t-test was conducted to investigate any between-group differences. We compared age and years of education with the subsequent subject scores for the PANAS, PASAT and Functional Connectivity Measures after Scan 1.

than subjects under 25 (N=7), especially in the 1Hz TMS population.

Mood Effects Analysis was conducted using a Repeated Measure ANOVA (p ≤ 0.05) comparing difference scores for positive symptoms and difference scores for negative symptoms before and after scans. ROI Analysis was conducted using a Repeated Measure ANOVA (p ≤ 0.05) between pre- and post-stimulation ROI scores for the 1Hz and 10Hz groups.


Results showed a significant correlation between age and PASAT DT t (15) = 0.50, p < 0.05, indicating that age difference between subjects correlated with differences in DT scores. With this in mind, age was included as a covariate of interest in the subsequent PASAT analysis.

Distress Tolerance Results for the distress tolerance variable were calculated using the ratings from the PASAT protocol during the first and second functional MRI (fMRI) scans. Three categories were compared: baseline DT scores of the 1Hz versus 10Hz group (prior to stimulation), DT scores of the 1Hz versus 10Hz group after stimulation, and the mean difference in DT scores for each group before and after TMS. Since it was previously noted that age was a covariate with DT Ratings, both a repeated measures ANOVA and two-sample t-test were conducted. No significance was observed between time (preto post-stimulation) and DT levels or between stimulation frequency. There was, however, a significant interaction effect between time and stimulation group (Figure 1a), with a notable decrease in DT for the 1Hz group after TMS (t (15) = 2.35, p < 0.05). An exploratory analysis of the covariate effects of age was also conducted. With age factored in, there was an observable difference in the DT scores not only based on TMS frequency but also on age. This is demonstrated in Figure 1b where subjects older than 25 (N = 9) had a lower mean change in DT 16 | Issue 1 | Volume 7 | Fall 2017

Mood Effects Results for the mood ratings were calculated using pre- and post-PANAS ratings for the first and second fMRI scans. No significant difference was seen between positive mood ratings before and after stimulation in the either stimulation group. There was, however, a slight decrease in positive affect scores for the 1Hz stimulation (Figure 2). Most notably, a significant change was seen in negative The Undergraduate Journal of Neuroscience

Abiodun | ARTICLE

affect levels (t (15) =2.18, p<0.05), with a stark increase in negative mood scores for the 10Hz group in MRI Session 2.

ROI: Functional Connectivity Results for insula-based FC were calculated using the pre and post-TMS Insula Region of Interest (ROI) scores. While there were no notable variances in the post-stimulation ROI scores between the 1Hz and 10Hz groups, analysis revealed that there was a substantial effect of time on the observed decline in functional connectivity (Figure 3b). Between pre and post stimulations, the combined ROI score decreased significantly, a finding which was supported by the results of our ANOVA test (F(1,15)=25.63, p<0.001). This indicates that TMS stimulation did have some effect on FC, although frequency of stimulation may not have played as large of a role as was previously hypothesized.


Our aim was to investigate the effects of rTMS on resting state functional connectivity, both in relation to the subsequent influence on insula-mediated distress tolerance and mood, and the anatomical implications of either increasing or decreasing connectivity of the insula to its related regions. Firstly, we looked at the effects of TMS on DT, observing that while the frequency of TMS did initially seem to have some impact, participant age demographics dampened the statistical significance.

It was shown that older subjects tended to a smaller decrease in tolerance levels between the pre and post scans, as well as the fact that subjects under 25 who received 1Hz TMS had a larger decrease in DT than any other group in either category. This finding contradicts our initial hypothesis, as 1Hz stimulation was shown to have negative effects on DT. It is clear that several other factors may have contributed to this result, including subject demographics, sensitization, and the relationship between age and TMS efficacy. In our subject pool there were considerably fewer subjects in the over 25 that received 1Hz TMS in than those who received 10Hz (the ratio was more even for the under 25 group). Additionally, we cannot discount the probability that Fall 2017 | Volume 7 | Issue 1 | 17

ARTICLE | Repetitive Transcranial Magnetic Stimulation

some of the decreases in DT could have been due to subject’s sensitization to the task, rather than the direct effects of TMS stimulation. This is one way in which this study is limited: without the incorporation of a sham-TMS control variable, it is much more difficult to decisively conclude what effects can be attributed to the presence of TMS itself. It would be useful, in future studies, to include a placebo condition in order to test this. Secondly, we looked at mood ratings directly from participants before and after task completion and prior to and following TMS. Although the majority of changes were nonsignificant, there was a significant increase in negative affect in subjects who received 10Hz TMS stimulation. Although the post-stimulation mood ratings were similar between the 10Hz and 1Hz groups, there is an observable difference when comparing the numbers to their baseline scores – the 10Hz scores decreased while the 1Hz group negative ratings were distinctly elevated. With no other noticeable covariance with our noted variables, it would appear that this result provides some evidence that frequency difference had some influence on mood, although it will not be clear how superficial this impact is without further research. This theory is corroborated by research done by Caparelli et. al, where 1Hz rTMS was shown to induce a change in activation pattern, although the inhibitory vs. excitatory dichotomy was also insufficient to describe the effect. (Caparelli et. al 2012) Lastly, we analyzed the main variable of this study – functional connectivity. Once again, results showed no notable differences between 1Hz and 10Hz stimulation across any of the categories. Surprisingly, it was demonstrated that although frequency did not have any particular impact, TMS stimulation in general did cause a significant change. It was originally hypothesized that 1Hz stimulation specifically would result in a more decreased FC than the 10Hz dose, but this was not the exact case: when the 1Hz and 10Hz stimulation results were grouped together in comparing baseline to post-stimulation results, there was a clear overall decrease in FC. This result is intriguing for two reasons: firstly, it confirms that rTMS does have an observable effect on FC. Interestingly enough, however, this effect did not take place in the direction that we originally expected, as 1Hz TMS had no substantially unique effect on FC in comparison to the 10Hz, in the positive or negative direction. The findings of our study are exciting, as they shed more light on the initial question at hand. In 18 | Issue 1 | Volume 7 | Fall 2017

some way shape and form, Transcranial Magnetic Stimulation can influence insula-based functional connectivity. While it is still unclear the exact extent to which this influence also branches into distress tolerance and mood, it does provide important insight into the variables to manipulate in order to investigate their relationship. It is clear, for instance, that age is a factor to be controlled, as it may affect a subject’s ratings of distress, mood, or overall susceptibility to TMS. Some studies have shown that due to age-related decreases in neuroplasticity – the ability of the brain to develop and utilize inter-neuronal connections – TMS is more effective in various treatment and stimulation conditions in younger subjects than in older subjects (Levkovitz 2001; Pascual-Leone 2011), which would agree with what we found in our own study. Further investigation is needed into the specifics of all of the variables measured, however this investigation serves as a promising foundation for such future studies. REFERENCES

Addicott, Sweitzer, Froegiler, Rose, McClernon. (2015). Increased Functional Connectivity in an Insula-Based Network is Associated with Improved Smoking Cessation Outcomes. Behzadi et al. (2007). A component based noise correction method (CompCor) for BOLD and perfusion based fMRI. Neuroimage, 37(1): p. 90-101. Neuropsychopharmacology, 40, p. 2648-2656 Caparelli, Backus, Telange, Wang, Maloney, Goldstein & Henn. (2012) Is TMS Always Inhibitory in Healthy Individuals? The Open Neuroimaging Journal(6) p. 69-74 Craig (2009). How Do You Feel - Now? The Anterior Insula and Human Awareness. Nature Reviews, Neuroscience, 10, p. 99 Dietmar Cordes, Victor M. Haughton, Konstantinos Arfanakis, Gary J. Wendt, Patrick A. Turski, Chad H.Moritz, Michelle A. Quigley, M. Elizabeth Meyerand. Mapping Functionally Related Regions of Brain with Functional Connectivity MR Imaging. American Journal of Neuroradiology Oct 2000, 21 (9) 1636-1644 Dinur-Klein et al (2014). Smoking cessation induced by deep repetitive transcranial magnetic stimulation of the prefrontal and insular cortices: a prospective, randomized controlled trial. Biol Psychiatry, 76(9): p. 742-9. Fox, Halko, Eldaif, Pascual-Leone (2012). Measuring and Manipulating Brain Connectivity with Resting State Functional Connectivity Magnetic Resonance Imaging (fcMRI) and Transcranial Magnetic Stimulation (TMS). NeuroImage, 62, p. 2232-2243 K. Bornhövd, M. Quante, V. Glauche, B. Bromm, C. Weiller, C. Büchel; Painful stimuli evoke different stimulus - response functions in the amygdala, prefrontal, insula and somatosensory cortex: a single-trial fMRI study. Brain 2002; 125 (6): 1326-1336. Levkovitz, Segal (2001). Aging Affects Transcranial Magnetic Modulation of Hippocampal Evoked Potentials. Neurobiology of Aging 22, p. 255263 Maldjian et al., (2003). An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets. Neuroimage, 19(3): p. 1233-9. McClintock, S. M., Freitas, C., Oberman, L., Lisanby, S. H., & Pascual-Leone, A. (2011). Transcranial Magnetic Stimulation: A Neuroscientific Probe of Cortical Function in Schizophrenia. Biological Psychiatry, 70(1), 19-27.

The Undergraduate Journal of Neuroscience

Abiodun | ARTICLE Mishra, B. R., Sarkar, S., Praharaj, S. K., Mehta, V. S., Diwedi, S., & Nizamie, S. H. (2011). Repetitive transcranial magnetic stimulation in psychiatry. Annals of Indian Academy of Neurology, 14(4), 245-251. Pascual-Leone, A., Freitas, C., Oberman, L. et al. Characterizing Brain Cortical Plasticity and Network Dynamics Across the Age-Span in Health and Disease with TMS-EEG and TMS-fMRI Brain Topography (2011) 24: p. 302 Paulus, Rogalsky, Simmons, Feinstein, Stein (2003). Increased Activation in the Right Insula During Risk-taking Decision Making is Related to Harm Avoidance and Neuroticism. Neuroimage, 19, p. 1439-1448 Sambataro, F., Dimalta, S., Di Giorgio, A., Taurisano, P., Blasi, G., Scarabino, T., Giannatempo, G., Nardini, M. and Bertolino, A. (2006), Preferential responses in amygdala and insula during presentation of facial contempt and disgust. European Journal of Neuroscience, 24: p. 23552362. Watanabe, Hanajima, Shirota, Ohminami, Tsutsumi, Terao, Ugawa, Hirose, Miyashita, Konishi, Kunumatsu, Ohtomo (2014). Bidirectional Effects on Interhemispheric Resting-State Functional Connectivity Induced by Excitatory and Inhibitory Repetitive Transcranial Magnetic Stimulation. Human Brain Mapping, 22, p. 1896-1905. http://doi. org/10.1016/j.biopsych.2011.02.031

Fall 2017 | Volume 7 | Issue 1 | 19



Utilizing Neurobiology to Diagnose Psychosis: Comparing DTI Derived White Matter Connectivity in DSM Diagnoses vs. in Biological Psychoses Subtypes Geoffrey A. Bocobo1, Shashwath Meda, M.S.2, and Godrey Pearlson, M.D.2,3 Trinity College, Hartford, Connecticut 06106 Correspondence should be addressed to Geoffrey Bocobo ( 2 The Olin Neuropsychiatry Research Center of Harford Hospital, CT 3 Yale University School of Medicine, CT 1

Supposedly distinct mental disorders within the psychosis spectrum possess large overlaps with one other not only in symptomology, but also in terms of treatment response. As these diseases are classified and diagnosed solely based upon the presence or absence of pre-determined clinical manifestations, the exploration of psychoses categorizations independent of the Diagnostic and Statistical Manual of Mental Disorders (DSM) is warranted. Advances in brain imaging have allowed scientists and clinicians to characterize the biological architecture and function of the brain more precisely. Owing to this fact, biological objectivity may serve as a tool in future diagnosis, and potentially even classification of said psychiatric illnesses. The current study introduces and explores the potential for neuro-biologically based disease classes. Patients with various forms of psychoses and healthy controls underwent a panel of functional brain exams involving electrophysiology, eye tracking, and tests of other cognitive domains. By using the meaningful differences seen in the yielded results of these tests, psychoses subtypes were re-categorized into â&#x20AC;&#x153;biotypes.â&#x20AC;? Here, we validate the use of biotype categorizations by utilizing Diffusion Tensor Imaging, a measure of fractional anisotropy within white matter tracts of the brain. More between group heterogeneity and within group homogeneity was seen in the biotype groupings when compared to the DSM-groups. Thus, subscribing to a more objective based approach for the diagnosis of cognitive disease may prove to be most effective for clinicians in the years to come.


Due to its proven reliability, the Diagnostic and Statistical Manual of Mental Disorders (DSM) has been the gold standard for clinical diagnosis (Regier et al., 2013). Although the manual avoids subscribing to a single ideology and endeavors to fully include a wide array of disciplines such as cognition, biology, and psychodynamics, it is still a phenomenologically grounded manual. Thus, the DSM method of diagnosis may overlook important physiological underpinnings of various psychiatric conditions 20 | Issue 1 | Volume 7 | Fall 2017

(Pearlson, 2016). Examining neurobiology may lead to a better understanding of the pathogenesis of certain disorders, especially those in the psychosis spectrum: illnesses that possess large amounts of homogeneity. Psychoses are debilitating brain disorders. While behavior based diagnosis remains the primary vehicle for identifying the condition, there is considerable evidence that this method may not completely capture biologically meaningful differences between each psychosis category (Clementz et al., The Undergraduate Journal of Neuroscience

Bocobo | ARTICLE

2014). As a result, antipsychotics, the hallmark treatment for these conditions, are only effective 75% of the time (Johnsen & Kroken, 2012). The pitfall of relying on the DSM manual for these conditions is the overlap in these diseases. Both bipolar disorders with psychosis and schizophrenia possess vast similarities in pathogenesis, pathophysiology, and most markedly, symptomologies, such as hallucinations, delusions, formal thought disorder, and many others. The two disorders also show similar treatment responses. Additionally, more than one-third of schizophrenic patients meet the criteria for Major Depressive Disorder (Pearlson, 2016). Further, bipolar disorder and schizophrenia are genetically anomalous in terms of predictive heritability, despite their paralleling risk genes (Rijsdijk et al., 2010). This creates yet another barrier towards ascertaining a differential diagnosis, because disease classes that possess similar genetic origins should be highly heritable within families. Put more simply, diagnostic boundaries between supposedly distinct psychotic disorders can no longer be supported, because there exists too much criteria overlap. An exemplification of this is the creation of schizo-affective disorder, a condition identical to schizophrenia, except that patients also possess some symptoms of either depression or bipolar disorder. If phenomenology cannot distinguish psychoses, what can? This is an important question, as healthcare professionals need to be able to treat their patients effectively.

Figure 1: The Overlap Between Psychosis Disorders: Characteristics of bipolar disorder, schizo-affective disorder, and schizophrenia. Taken from:

With the growing knowledge of neurobiology, neurological disease classifications are constantly changing. This is especially helpful when clinical manifestations are indistinguishable, as biological research has resulted in the classification of diseases with similar behavioral presentations

and pathologies, into distinct disorders (Pearlson, 2016). Therefore, knowledge of neurobiological architecture may enhance diagnostic outcomes, and possibly even help individualize the treatment of different psychoses. Unfortunately, efforts to validate bio-signatures or bio-markers for psychotic illnesses have been largely overshadowed to this day. This is possibly due to clinicians being trained to unilaterally rely on symptomology as a standard practice when diagnosing certain neurological illnesses (Clementz et al., 2014). Despite this, some studies show that the field of research and medicine is slowly moving in the right direction, away from diagnosing based on only phenomenology. Due to advancements in brain imaging techniques, there has been a recent push for objectivity in the diagnosis of mental illness. Studies have explored whether brain scans can more readily classify psychoses (Skudlarski et al., 2013). Diffusion Tensor Imaging (DTI) is a promising method of brain imaging that elucidates upon microstructural changes within the brain. It is a tool that some clinicians use to characterize the current state of patientsâ&#x20AC;&#x2122; brains (Alexander et al., 2007). Fractional anisotropy (FA) is the metric employed by DTI to measure the directionality of water flow in tracts of the brain. It is a scalar value between zero and one, with numbers closer to one indicating directional nature (anisotropy), and numbers closer to zero depicting a diffuse nature (isotropy). Healthy white matter tracts are anisotropic with respect to their water flow, and therefore have FA values close to one. These healthy axons enable grey matter to communicate with one another in a normal fashion. DTI is a powerful tool to measure the integrity of white matter connections, the structure of a brain, and since diffusion traits are controlled by cellular structures, it is a powerful tool for quantifying the effects of disease and detecting pathophysiological changes at the cellular and molecular level (Alexander et al., 2007). Clinicians are beginning to utilize this method to study brains more and more. One study (Barysheva et al., 2013) examined bipolar brains with respect to their myelination. As bipolar disorder is characterized by a down-regulation in oligodendrocytic genes, the investigators hypothesized that some white matter tracts would have lower than normal FA values. Results demonstrated significantly isotropic water diffusion and demyelination within the corpus callosum, frontal cortex and limbic system (Barysheva et al., 2013); their hypothesis was fully Fall 2017 | Volume 7 | Issue 1 | 21

ARTICLE | Utilizing Neurobiology to Diagnose Psychosis

supported. The white matter tract similarity seen within different patient’s brains of those with bipolar disorder set a strong precedent for the potential of neurobiology in diagnosing disorders.

Figure 2: Teipal M.D., Stephan: Fiber tract integrity as evidenced by diffusion tensor imaging; coronal views of the temporal lobe in a healthy brain (top) and Alzheimer’s brain (bottom). Taken from:

Acknowledging that schizophrenia and bipolar disorder have large overlaps in terms of cognition and brain function, another group of investigators in 2013 (Skudlarski et al., 2013), looked at the brain images of 513 individuals with either schizophrenia or bipolar disorder. The investigators found that although the diseases manifest themselves similarly, they had distinct white matter tract abnormalities. All subjects had generally lower FA values in their white matter tracts, but the bipolar data set displayed a significantly fewer number of affected brain regions (Skudlarski et al., 2013). Here, the researchers were able to distinguish bipolar individuals from schizophrenic individuals using diffusion tensor imaging, despite the behavioral similarities that the two conditions possess. The current study attempts to build upon this. This exploratory research uses DTI as an external validator for the use of neurobiology to supplement phenomenology: objectivity to complement subjectivity. In order to explore the strength of neurobiology in the diagnosis of psychosis, new DSM-independent categories must be created. These groupings are made by placing all psychoses patients together 22 | Issue 1 | Volume 7 | Fall 2017

in one group, and administering to them a battery of neurobiological measures. These tests measure variables of known relevance to psychosis and the characterization of functional brain activity including eye tracking, other functional brain scans such as magnetoencephalography, and electrophysiology (Clementz et al., 2014). These re-categorized “Biotype” classifications (B1, B2, B3) are objective based brain markers created to transcend traditional DSM boundaries. Thus, when a group of patients has clear neurobiological distinctiveness, they can be classified as a certain biotype. With this method, two schizophrenic patients can have very different results from biotype analysis and therefore fall into two different biotype categories. Consequently, the two patients would have vastly different treatment plans since one would have more severe or different neurobiological brain abnormalities. (It should be noted that B1 is the most severe and B3 the least).

Figure 3: Taminga M.D., Carol & Clementz Ph.D., Brett: Novel Biotype Categorizations Taken from:

The biotype approach seems promising and may have the ability to more ably classify patients. If neurobiological distinctness is seen, this method may mitigate the problem of overlap in psychotic behavioral criteria. The purpose of this study to determine whether the biotype categories possess diagnostic power. If true, the present research will further explore whether biotype categorization may be an aid to or superior to DSM categorization. We are hypothesizing the following: all psychosis individuals will generally show decreased FA values when compared to healthy individuals, though The Undergraduate Journal of Neuroscience

Bocobo | ARTICLE

the biotype categories will possess more individual distinctiveness than the DSM pro-bands; there will also exist more widespread FA disturbances in the biotypes when compared to DSM. Imaging the biotype categories will show even more distinction than imaging the DSM categories. This is an exploratory study, and a novel hypothesis in regard to the validation of neurobiology for diagnosing psychiatric illness (Pearlson, 2016). The imaging analysis will be done at the voxel and region of interest level, as derived from the International Consortium of Brain Mapping.


Participants The brain scans of 454 subjects were taken from the NIH’s Bipolar-Schizophrenia Network on Intermediate Phenotypes Initiative (Tamminga et al., 2014). Data was collected from Healthy Subjects (n = 211) and from pro-bands of psychosis (n = 343), namely bipolar disorder with psychotic episodes, schizo-affective disorder, and schizophrenia. Relatives used in a previous study were excluded, though some met the criteria of re-categorization into pro-bands groupings. The diagnoses were established by trained clinical diagnosticians using DSM-IV criteria. All of the subjects did not have any previous neurological illness, and patients with a history of substance abuse were excluded. The participants were initially recruited by means of advertisements and community groups. They completed informed consent as approved by the Institutional Review Boards of Hartford Hospital, Yale University, The University of Maryland, and Johns Hopkins University. Measures A pool of neurobiological measures was used to characterize brain function of all individuals. This included auditory paradigms, eye tracking, electrophysiology, psychophysiological measures, and many other measures of cognitive domains. Afterwards, the subject data was re-categorized into three biotype groups along with their original DSM groupings. The DTI imaging data was obtained by 3-Tesla MRI scanners using single shot spin echo planar imaging, with a twice refocused balance echo sequence at The Olin Neuropsychiatry Research Center and at Johns Hopkins University. Hartford used TR/TE=6,300/85 ms, field of view=220 m, b=1,000 s/mm2 along 32 directions, 45 contigu

ous slices, three imaging series, and a voxel size of 1.731.733 mm. Baltimore used TR/TE=6,700/92 ms, field of view=230 m, b=1,000 s/mm2 along 30 directions, 48 contiguous slices, two imaging series, and a voxel size of 1.831.833 mm. The data were corrected for motion and noise artifacts with eddy current correction and gradient direction correction, respectively. All of the data analysis was done at Hartford Hospital and confounding covariates were eliminated by adjusting for age, sex, ethnicity, and site of scanning by inclusion in the general linear model.

Data Analysis Voxel-Based Whole Skeleton Analysis was performed on FMRIB Software Library (FSL) and Statistical Parametric Mapping (SPM) for the visual analysis of the voxel-wise data. By using this analytical method, all brain scans were projected onto a common, 4D white matter skeleton space, which enabled us to quantify the white matter integrity (FA) of significant brain regions, averaged from the whole group. This was done by the means of Tract-Based-Spatial-Statistics (TBSS), which skeletonized the DTI images and ran an analysis of covariance (ANCOVA) on the data. TBSS allowed us to analyze the brains in a common space, at the voxel level, and to see the level of difference between each DSM and biotype group (and compare those groups to controls). All psychosis pro-bands were compared to the control groups and all pro-bands, biotype or DSM, were compared to each other. Multiple comparisons were controlled for via Unix enabledThreshold-Free Cluster Enhancement, a nuanced and complex statistical method employed for functional brain imaging originally described by Smith and Nichols in their 2009 NeuroImage paper. The Statistical Package for the Social Sciences (SPSS) was used for the statistical analysis of region of interest (ROI) data. This data included both the right and left hemisphere of the brain, as averaging the two would yield non-specific results. General Linear Model (GLM) multivariate analysis was run to determine which ROI’s were significant by examining levels of variance. Pillai’s Trace, Wilk’s Lambda, Hotelling’s Trace, and Roy’s Largest Root significance levels were all calculated to determine regions that displayed differences between biotype groupings and DSM groupings. Healthy controls were excluded for this reason. Fall 2017 | Volume 7 | Issue 1 | 23

ARTICLE | Utilizing Neurobiology to Diagnose Psychosis


Voxel-wise Whole Brain Visual Analysis When comparing schizophrenic pro-bands to healthy controls, many regions of the brain showed significantly decreased fractional anisotropy values (p=0.05), which is consistent with our hypothesis that a typical pro-band of psychosis would display unhealthy white matter tracts.

Figure 5: Biotype Grouping 1 vs. Healthy Controls: Mean FA skeleton is seen in red; yellow delineates areas of significantly lower FA values when comparing the biotype category to the healthy control group.

Figure 4: Schizophrenia Pro-bands vs. Healthy Controls: Mean FA skeleton is seen in red, whereas yellow delineates areas of significantly lower FA values when comparing pro-bands to controls.

Biotype categorizations showed more significant results overall. B1 and B3 possessed significantly lowered FA values when compared to healthy controls (p = 0.05 & p = 0.04). Biotype category 2 even showed trendy significance with decreased FA values: p = 0.076. These are consistent with our hypotheses of generally decreased FA values, and even more significantly decreased FA values than DSM categorizations. Lastly, unlike any of the DSM groupings, biotype categorizations showed distinctive individual homogeneity. One comparison between two biotype groups showed regions of significance that easily distinguished the two categories (p = 0.05). This is consistent with our hypothesis of more distinctiveness between the biotype groupings. The two other comparisons (B1 vs. B2 and B1 vs. B3) did not show significant differences (p > 0.15). There was not more widespread tract damage in the biotype groups when compared to DSM or control. Survival analysis showed significant differences to P = 0.05 in all groups. 24 | Issue 1 | Volume 7 | Fall 2017

Figure 6: Biotype Grouping 3 vs. Healthy Controls: Mean FA skeleton is seen in red; yellow delineates areas of significantly lower FA values when comparing the biotype category to the healthy control group.

Figure 7: Biotype Grouping 3 vs. Biotype Grouping 2: Mean FA skeleton is seen in red; yellow delineates areas of significantly lower FA values when comparing the B2 to B3 patients.

The Undergraduate Journal of Neuroscience

Bocobo | ARTICLE

Region of Interest Pro-band Analysis After conducting a General Linear Model MANOVA, Pillai’s Trace (PT), Wilk’s Lambda (WL), Hotelling’s Trace (HT), and Roy’s Largest Root (RLR) significance levels were produced. No ROI’s were significant in distinguishing between any of the DSM groups (PT = 0.882, WL = 0.88, HT = 0.879, RLR = 0.385). Many ROI’s were significant in distinguishing between the biotype categories, which supports our hypothesis of greater pro-band group differences and distinctness (PT = 0.075, WL = 0.07, HT = 0.066, RLR = 0.01). Because of this significance, post-hoc pairwise T-Tests were performed on the areas of decreased FA values. This elucidated what specific areas of the brain make biotypes effective at being distinctive. As seen in the table, the anterior corona radiata, the anterior limb of the internal capsule, the left corona radiata, the external left capsule, the inferior cerebellar peduncle, the posterior corona radiata, the posterior limb of the internal capsule, the posterior thalamic radiation, and the sagittal striatum were all of the white matter tracts that significantly accounted for greater biotype pro-band intra-group differences. B3 contains very homogenously affected brain areas. As B3 is supposed to represent the least affected psychosis group, it is also consistent with the unstated hypothesis of B3 being the least “affected” group and therefore having the highest FA values and thus the least damage to its white matter tracts. Cohen’s D effect sizes were all about 0.5: the statistical magnitude of this analysis is moderately impactful.


Our hypotheses were supported. All of the probands, regardless of biotype or DSM categorization, displayed lowered fractional anisotropy levels (Figure 4). Only one DSM group (schizophrenia) showed significantly lower FA values when compared to control. This is possibly intuitive, as schizophrenia does classically have the most neurobiological abnormalities of the three classifications. On the other hand, two biotype groups (B1 & B3) showed significantly lower FA values when compared to control (see Figures 5 & 6). The biotype categories were more distinguishable than the DSM categories as seen by voxel-wise whole brain analysis. Moreover, there were no significantly distinguishing brain regions between DSM groups, whereas the biotype groupings had regions that showed significant differences between (Figure 7). Therefore, the DSM groups were not as individually homogenous as the biotype groups were, again consistent with our hypothesis. The implications of this are large, as these groupings may have less variance and could be more useful than traditional, gold-standard DSM groupings. Lastly, biotype groupings were more sensitive to FA analysis. Region of Interest analysis focuses on major white matter tracts pre-determined to be significant. ROI data excluded controls and was used to measure only intra-group differences between biotypes and DSM. When looking at the GLM multivariate region of interest data, no brain regions were significantly different in the DSM group, also supportive of our hypothesis that these groups would possess a large

Table 1: Post Hoc, Pairwise T-Tests extracted derived from general PT, WL, HT, RLR: The table shows the region of “distinguishing” significance, the two biotypes compared, the significance level, the mean difference or directionality and the magnitude of difference as yielded from standard deviation and demonstrated by Cohen’s D effect size.

Fall 2017 | Volume 7 | Issue 1 | 25

ARTICLE | Utilizing Neurobiology to Diagnose Psychosis

amount of overlap. The biotype categories had ten significantly distinct brain regions that accounted for the large differences between B3 and the other biotype categories at the ROI level (Table 1). Further research should be performed on the function of these brain regions, to explain why the tract may have been more anisotropic in these psychotic illnesses (e.g. the striatum). The statistical magnitude of these differences was validated by the Cohen test. This supports our hypothesis of biotype categorizations being more individually homogenous. Overall, there were more significant differences and findings in the biotype groups, and this supports our claim of neurobiology’s potential. Biotypes could be a useful tool for clinicians to use as a DSM manual aid. It should be noted that this study’s sample size was relatively small, and could be replicated with a larger population pool to increase the statistical power. Biotypes may prove to be an efficient way of validating clinical phenomenology, though it warrants further research. In the long run, objectivity could become even more useful in diagnosing psychosis than subjectivity. We have validated the use of neurobiology in the diagnosis of psychosis. It should be noted that this research is exploratory in nature, and very novel. Neurobiology is a powerful tool, and biotypes may solve the clinical problem of overlap in psychotic illnesses.


Alexander, A. L., Lee, J. E., & Field, A. S. (2007). Diffusion Tensor Imaging of the Brain. Neurotherapeutics, 4(3), 316-329. Barysheva, M., Jahanshad, N., Foland-Ross, L., Altshuler, L., & Thompson, P. (2013). White matter microstructural abnormalities in bipolar disorder: A whole brain diffusion tensor imaging study. NeuroImage, 2, 558-568. Bhati, M. T. (2013). Defining Psychosis: The Evolution of DSM-5 Schizophrenia Spectrum Disorders. Current Psychiatry Reports Curr Psychiatry Rep, 15(11). Clementz, B., Sweeney, J., Hamm, J., Ivleva, E., Ethridge, L., Pearlson, G., Tamminga, C. (2014). Identification of Distinct Psychosis Biotypes Using Brain-based Biomarkers. The American Journal of Psychiatry. Johnsen, E., & Kroken, R. A. (2012). Drug treatment developments in schizophrenia and bipolar mania: Latest evidence and clinical usefulness. Therapeutic Advances in Chronic Disease, 3(7), 287-300. Pearlson, G. (2016). Does Biology Transcend the Symptom-based Boundaries of Psychosis. Psychiatric Clinics of North America. Regier, D. A., Kuhl, E. A., & Kupfer, D. J. (2013). The DSM-5: Classification and criteria changes. World Psychiatry, 12(2), 92-98. Rijsdijk, F. V., Gottesman, I. I., Mcguffin, P., & Cardno, A. G. (2010). Heritability estimates for psychotic symptom dimensions in twins with psychotic disorders. Am. J. Med. Genet. American Journal of Medical Genetics Part B: Neuropsychiatric Genetics, 156(1), 89-98. Skudlarski, P., Schretlen, D.J., Thaker, GK., Stevens, MC., Keshevan, MS., Sweeney, JA., Tamminga, CA., Clementz, BA., O’Neil, K., Pearlson, GD. (2013). Diffusion tensor imaging white matter endophenotypes in patients with schizophrenia or psychotic bipolar disorder and their relatives. The American Journal of Psychiatry, 8(170), 886-898. Tamminga, C. A., Pearlson, G., Keshavan, M., Sweeney, J., Clementz, B., & Thaker, G. (2014). Bipolar and Schizophrenia Network for Intermediate Phenotypes: Outcomes Across the Psychosis Continuum. Schizophrenia Bulletin, 40 (Suppl 2).


I would like to thank Prof. Raskin and Kathy for again coordinating this year’s Health Fellows Program at Trinity; it was a wonderful semester of scientific inquiry and healthcare related dialogue. Moreover, I would like to acknowledge my mentor Shashwath. Without his guidance, I couldn’t have completed the scripting, data mining, and the running of neuro-imaging programs on the command line and in Matlab. Finally, I must thank Dr. Pearlson for introducing me to this research and leading The Olin Neuropsychiatry Research Center.

26 | Issue 1 | Volume 7 | Fall 2017

The Undergraduate Journal of Neuroscience



Shedding Light on the Role of ipRGCs in the Human Eye During Light Exposure Lissa Neira Trinity College of Arts and Sciences, Duke University, Durham, North Carolina 27708 Correspondence should be addressed to Lissa Neira ( 1

People with no light perception (NLP) suffer from a circadian rhythm disorder called Non-24 as a result of not being able to sense daylight. Currently there does not exist an effective treatment for regulating the circadian rhythms of people with blindness, much less those with NLP. This paper explores the possibility of having high-energy blue light therapy during daylight hours in order to see if there is an activation of intrinsically photosensitive retinal ganglion cells (ipRGCs), which may be the functioning photoreceptors at work in this population. By measuring the amount 6-sulphatoxymelatonin (aMT6s) through urine and saliva samples, we can find the concentration of melatonin which can help to determine whether high-energy blue light exposure can help to regulate the circadian rhythm and pose as a potential treatment for Non-24.


For many people across the world, sleep has become an impossible task. Our busy lives keep us awake at night and we cannot seem to relax enough to fall into a restful sleep. Many of us do not even have a consistent sleep schedule that allows us to fall asleep at the same time every night. For most people, their problem is that they are much too preoccupied to develop a consistent sleep cycle; however, there are others who cannot develop a sleep cycle for biological reasons such as blindness. When someone becomes blind, their ability to distinguish between night and day weakens and thus their brains do not know when to go to sleep. There are two types of blindness: light perception (LP) and no light perception (NLP). Someone with LP is not able to distinguish between objects and sees the world as a massive blur. However, majority of blind people with LP are still able to process the presence of light because there are still functional photoreceptors in their eyes that allow them to register the different intensities of light. Those with LP are still able to establish consistent sleep cycles since their brains can sense the presence of daylight, which is not the case for people with NLP. People who are

blind with NLP are not able to directly â&#x20AC;&#x153;seeâ&#x20AC;? light; however, very recent research has shown that blind people with NLP can still biologically register the presence of light regardless of their inability to be consciously aware of it. The proposed experiment explores the possibility of NLP sleep cycle regulations through the exposure of light. Many living organisms are able to sleep at night and remain awake during the day because of the presence of light. Cells in our eyes are able to detect light and send messages to our brains to indicate that we are currently awake in the day (Berson, 2002). The lack of light thus causes our brains to release a sleep-inducing hormone called melatonin (Taillard et al., 2012). We then become tired and are able to sleep during the night. After a while, a person can train himself or herself to go to sleep and wake up at the same time every day because of the regulated release of melatonin right before bedtime. This consistent secretion of melatonin thus creates a sleep cycle referred to as the circadian rhythm. Naturally, humans have a circadian rhythm that is slightly longer than 24 hours and thus without the presence of light, the circadian rhythm has the tendency to get misaligned with Fall 2017 | Volume 7 | Issue 1 | 27


the regular time of day, which is the case for most patients with NLP (Berson, 2002). Over time their circadian rhythms can get disarranged with regular daytime hours causing tremendous fatigue and exhaustion throughout the day. Therefore, many people who have NLP suffer from a circadian rhythm disorder called Non-24-Hour Sleep Wake Disorder (Non-24). Of all blind patients with NLP, 70% suffer from Non-24 (Vanda Pharmaceuticals Inc., 2015). Many blind people with Non-24 are not able to keep their jobs as a result of this disorder because their exhaustion is so extreme that they can fall asleep at unexpected times without control. According to the National Sleep Foundation (2015), there are three treatments for Non-24: phototherapy, the intake of a drug called Tasimelteon, and the treatment of melatonin. Specifically for blind people with NLP, phototherapy is the least useful because the patient does not detect the light used in the treatment at all since the light used is purposely weak in order to not damage the eyes of people who can still see. Tasimelteon is effective in helping to regulate the circadian rhythm; however, there are many side effects that can be harmful for older blind patients. The use of melatonin is thus the most reliable source of treatment for patients with NLP, but the patient would have to continuously take melatonin for the rest of their lives, a feat that may become financially difficult for some people. Thus the introduction of a new treatment must be implemented, one that is both economically friendly and effective. We are then left with a treatment that must trigger an internal response without the use of drugs. The result? Phototherapy again. Only this time, the phototherapy will not depend on weak white light, but a powerful blue light. From the visible light spectrum, blue light has the shortest wavelength and most energy. The problem with regular phototherapy for people with NLP is that the treatment uses relatively low-energy light to prevent any damage to a sensitive eye); however, most people with NLP do not have sensitive eyes and thus cannot register this weak light (Vandewalle et al., 2013). For this reason, a powerful blue light will be used in our experiment. How can the exposure of blue light lead to the regulation of the circadian rhythm? The answer lies within our eyes. Initially it was believed that humans were able to perceive light through two photoreceptors in the eye called rods and cones. Rods are cells that respond to low-light vision and transmit black and white shapes to the brain. Cones, on 28 | Issue 1 | Volume 7 | Fall 2017

the other hand, are cells in the eyes that are able to register color. It was thought that light was only perceived through these two photoreceptors until Clyde E. Keeler accidentally bred mice that were born with no rods and cones (Keeler, 1924). These mice were completely blind, yet when light was shone into their eyes, their bodies responded to the stimuli and their pupils dilated. The study showed implications that there is a nonvisual way in which we perceive light. Yet how was this possible if none of the mice were born with rods and cones? Keeler wondered over this enigma for years but could not figure out the reasoning behind this phenomenon. Unfortunately, his work was left untouched for decades and the mystery of the perception of light was left unsolved for many decades to come until Russell G. Foster dug up Keeler’s work in 1991. After extensive experiments, Foster’s work showed a correlation between the circadian rhythm of blind mice and the exposure of light, indicating that the circadian rhythm is still able to regulate itself without the presence of rods and cones (Foster et al., 1991). In his experiment, Foster realized the mice that were exposed to light had slightly different sleep cycles than those who were not exposed to light. Therefore, there was an unknown cell somewhere within the eye that was sensing the presence of light (Vandewalle et al., 2013; Owens et al., 2012); however, it would take yet another couple of years before these mysterious cells were found. After far-reaching research, the discovery of these mysterious cells was finally made. These special cells are called intrinsically photosensitive retinal ganglion cells (ipRGCs) and transmit light signals to the brain (Weng et al., 2009; Owens et al., 2012). To this day, there is still limited research on how ipRGCs function in human eyes and even less on how the stimulation of these ipRGCs can be used to help regulate the circadian rhythms of blind patients. The predicted general outlook of the role of ipRGCs is that these cells send an electrical message to the suprachiasmatic nuclei (SCN, located in the hypothalamus within the brain) when light comes in contact with the eye. Once in the hypothalamus, the message is transferred to the pineal gland where melatonin is either released or suppressed depending on the initial message (Schmoll et al., 2010); thus, ipRGCs may very well have a direct correlation with the regulation of sleep (Owens et al. 2012, Weng et al., 2009; Flynn-Evans et al., 2014). However, Owens’ study showed that there is The Undergraduate Journal of Neuroscience

ARTICLE | Shedding Light on the Role of ipRGCs

little relevance between the presence of light and ipRGCs. Rather than using long-term light training during the day, the blind mice in their experiment were flashed with intense light for only seconds. The results showed that there was no correlation between the stimulation of ipRGCs and the circadian rhythm in accordance to the flashes of light. These results could have been due to the fact that light-induced ipRGCs may require a longer period of time to stimulate. Thus, the quick flashes may be insufficient stimuli and were unable to transduce a signal in the brain. Long-term exposure of light may potentially lead to a response to the stimulus and warrants the need for a study in which participants have complete light exposure during the time that they are awake in the day. In this research project, light training will take place in order to help mimic the hours of daylight through the exposure of blue light since ipRGCs are most sensitive to the blue light of about 480 nm (Vandewalle et al., 2013; Taillard et al., 2012). By obtaining urine and saliva samples to measure the amount of melatonin and functional magnetic resonance imaging (fMRI) to detect the activation of the pineal gland, results can be used to correlate the function of ipRGCs with the circadian rhythm. The sample group will consist of participants who are blind with NLP and currently suffering from Non-24. We hypothesize that by shining light into the participants’ eyes consistently throughout the day and preventing light during the night, the participant will slowly build a circadian rhythm that matches 24 hours. This could be due to the activation of ipRGCs that send signals to the pineal gland, which in turn releases melatonin in the absence of light. Such melatonin makes the participant sleepy causing a better night’s rest. By shining blue light into the participant’s eyes, we can observe if there is a direct response from the pineal gland. Thus if our hypothesis is correct, light can still be sensed by the brain of a blind person with NLP even though he or she may not perceive the presence of light. Such a correlation could prevent the development of Non24 in patients who are in the process of losing all of their eyesight. By exposing patients to this blue light and activating their ipRGCs, we can help decrease insomnia in people with NLP.


Participants Participants in the experiment include blind people with NLP. In order to control which part of the eye is

damaged, these participants will all have gone blind due to age-related macular degeneration (AMD), a disease that attacks the macular (a part of the retina). All participants will also be female and between the ages of 65 and 85 years in order to prevent the presence of a confounding variable. There will be two kinds of control groups: a negative control which undergoes no light exposure and a control group that receives light continuously even during the night to see if the presence of light is the defining factor that regulates the circadian rhythm. The total population size for this experiment will be 60: 20 for the negative control, 20 for the other control and lastly 20 for the experiment. This experiment will last for about one year. Procedure We can measure whether or not ipRGCs are being stimulated by light through the presence of melatonin. If ipRGCs are being stimulated, there will be no production of melatonin at that time since the brain will think the person is currently in daylight. Thus the production of melatonin indicates that the brain believes the person is experiencing nighttime and releases melatonin in order to make the person sleepy and ready for bed. In order to test for the stimulation of ipRGCs two types of measurements will thus be made. The first test involves measuring the concentration of melatonin in the body. Urine samples and saliva samples will be tested. In urine, the concentration of melatonin is determined based on the amount of urinary metabolite (a substance present in urine) called 6-sulphatoxymelatonin (aMT6s). There is a direct correlation between aMT6s and melatonin since aMT6s is released after the secretion of melatonin (Perez-Rico, De La Villa, Blanco, Germain, Paz-Moreno, & Arribas-Gomez, 2009). The participants will provide saliva samples because melatonin is also present in human saliva. Each day for 3 months, the participant will provide 5 of these saliva samples: one in the morning, one in the middle of the day and three at nighttime. The first two samples during the day are used to ensure that there is no melatonin secretion in the day. The last three samples at night are to see which times the participant has the highest concentration of melatonin. In a normal human being with regular eyesight, melatonin is typically released two hours before bedtime (Perez-Rico et al., 2009). Therefore, these three samples will occur 3 hours, 2 hours and 1 hour before bedtime to identify if melatonin secretion occurs at about the same time as people with Fall 2017 | Volume 7 | Issue 1 | 29


regular eyesight. Once a consistent time has been identified for the peak secretion of melatonin, the experimentation will now begin and only one urine sample will be necessary daily. One urine sample will be used daily instead of one saliva sample because it is more accurate to measure the amount of melatonin in urine than in saliva (Perez-Rico et al., 2009). The experiment itself involves 20 participants with specialized glasses to have blue light shone directly into their eyes for 16 hours beginning at 7 AM with a light intensity of about 20,000 lux. Light intensity is measured in lux, a measurement of illumination. Since a regular sunny day has about 10,752 lux on average, the illumination on the blue light will be almost twice that amount (20,000 lux) to ensure that the light is potent enough to travel the length of the eye and hit the ipRGCs in the retina (The Engineering ToolBox). The light in the glasses will continue to shine throughout the day at the same intensity until 7 PM when the light intensity will gradually decrease to tell the body that the sun is going down. By 11 PM all light has completely seized to be emitted from the glasses and the participant is no longer exposed to any light. The reasoning behind the gradual decrease in light intensity at 7 PM is because when the pineal gland is stimulated and melatonin is released, it takes the hormone a couple hours to travel to the rest of the body and make the person sleepy. Using the data collected from the saliva samples, one urine sample will be taken at the optimal time of melatonin secretion. Therefore, in order to connect the presence of melatonin and the pineal gland, fMRI will be used to identify the stimulation of the pineal gland. One fMRI scan will take place each week after the blue light has stopped shining for the day to see if the lack of light has triggered the pineal gland. The bedtime for each participant will be 11:30 PM. In order to see if the light training has an effect on the participants, the concentration of aMT6s will be compared as time passes. In conjunction with aMT6s levels, fMRI scans will also be compared to see if the pineal gland is stimulated to a greater extent as a regular circadian rhythm establishes.


By measuring the concentration of melatonin in the body using urine and saliva samples, we can determine whether or not high-energy blue light has the potential to stimulate ipRGCs. If the experiment proves successful, people with NLP will be able to 30 | Issue 1 | Volume 7 | Fall 2017

regulate their circadian rhythms and significantly improve the quality of their sleep. However, despite the potential implications of this experiment, there are some limitations that must be considered. In order to find the time when the optimal melatonin secretion takes place, saliva samples are used instead of urine samples because of the unfeasibility of taking five urine samples a day within a short period of time. However, by choosing to use saliva samples, the results may vary in accuracy since the test conducted for measuring the amount of melatonin in saliva has a much less analytical sensitivity (Perez-Rico et al., 2009). Also, we make the assumption that all of the participants need about the same amount of sleep, but this statement is not entirely true. People who need less sleep may have a higher concentration of melatonin that allows them to reach quality sleep faster. This discrepancy may cause for a slight variation in data; however, we have taken these limitations into account and have expanded our population size in order to decrease the significance of these variations. The general workings of this experiment thus still remain practical and sensible. If the expected results from this experiment prove to be successful, there are many advances that can be made to help those with NLP. Not only does this experiment broaden the number of available resources for blind people to improve their quality of life, but also the science behind this experiment will help to further understand what is happening inside the human eye. Much of the current research that exists is on rats or other animals with the assumption that humans will behave the same way. For this experiment we will be able to make a direct correlation between the function of human ipRGCs and light which has never been done before. With this relationship established, we can investigate other, even more efficient ways to stimulate ipRGCs. We can even branch out to people with no problems in their vision but use ipRGC stimulation as a form of deterrence for Non-24 or other circadian rhythm disorders. This experiment would mark the first of many future attempts at minimizing the effects of being blind. REFERENCES

Berson, D. (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science, 454, 1070-1073. Flynn-Evans, E., Tabandeh, H., Skene, D., & Lockley, S. (2014). Circadian rhythm disorders and melatonin production in 127 blind women with and without light perception. Journal of Biological Rhythms, 29, 215-224.

The Undergraduate Journal of Neuroscience

ARTICLE | Shedding Light on the Role of ipRGCs Foster, R., Provencio, I., Hudson, D., Fiske, S., Grip, W., & Menaker, M. (1991). Circadian photoreception in the retinally degenerate mouse (rd/rd). J Comp Physiol A Journal of Comparative Physiology A, 169, 39-50. Keeler, C. (1921). Blind mice. J. Exp. Zool. Journal of Experimental Zoology, 495-508. Owens, L., Buhr, E., Tu, D., Lamprecht, T., Lee, J., & Gelder, R. (2012). Effect of circadian clock gene mutations on nonvisual photoreception in the mouse. Investigative Opthalmology & Visual Science Invest. Ophthalmol. Vis. Sci., 53, 454-454. Perez-Rico, C., De La Villa, P., Blanco, R., Germain, F., Paz-Moreno, J., & Arribas-Gomez, I. (2009). Alterations in nocturnal melatonin levels in patients with optic neuropathies. ARCH SOC ESP OFTALMOL, 84, 251-258. Schmoll, C., Lascaratos, G., Dhillon, B., Skene, D., & Riha, R. (2011). The role of retinal regulation of sleep in health and disease. Sleep Medicine Reviews, 15, 107-113. Taillard, J., Capelli, A., Sagaspe, P., Anund, A., Akerstedt, T., & Philip, P. (2012). In-car nocturnal blue light exposure improves motorway driving: A randomized controlled trial. PLoS ONE, 7(10). Retrieved from Vandewalle, G., Collignon, O., Hull, J., Daneault, V., Albouy, G., Lepore, F., Phillips, C., Doyon, J., Czeisler, C., Dumont, M., Lockley, S., & Carrier, J. (2013). Blue light stimulates cognitive brain activity in visually blind individuals. Journal of Cognitive Neuroscience, 25, 2072-2085. Weng, S., Wong, K., & Berson, D. (2009). Circadian modulation of melanopsin-driven light response in rat ganglion-cell photoreceptors. Journal of Biological Rhythms, 24, 391-402.

Fall 2017 | Volume 7 | Issue 1 | 31



The Role of Epigenetics in Alzheimer’s Disease Aysswarya Manoharan1, Alfredo Harb-De la Rosa2, and Amirthaa Suntharalingam3 Washington University at St. Louis, St. Louis, Missouri 63130 University of Miami, Miami, Florida 33146 2 Byrd Alzheimer’s Institute, University of South Florida, Tampa, Florida 33620 Correspondence should be addressed to Aysswarya Manoharan ( 1


AD is a progressive disorder characterized by increasing behavioral changes, resulting in loss of cognitive function and ultimately, death. A variety of genome wide and locus specific epigenetic changes have been reported. Epigenetic changes are involved in the pathogenesis of AD and are potential targets for therapeutic manipulation in patients with AS. Epigenetics is defined as inherited changes in gene expression that are not encoded within the DNA structure. Epigenetic alterations include DNA methylation, Histone modifications and miRNA regulation. At present, evidence directly linking epigenetics changes and AD is lacking. However, non-causal and indirect correlative evidences have been established. Epigenetics based tests could prove to be a reliable biomarker for early diagnosis of AD. Several therapeutic molecules to target these modifications have been developed and the possibility of reversing epigenetic modifications is encouraging and continues to motivate research in the development of novel targeted therapies.


Geriatric population is currently the most rapidly growing segment of the population throughout the developed world due to increasing life expectancy and declining fertility (Bennett et al., 2015). Alzheimer’s disease is one of the most common causes of Dementia in old age. Alzheimer’s disease (AD) is a progressive neurodegenerative disorder, with over 35 million cases worldwide (Selkoe, 2012). AD is a progressive disorder characterized by increasing behavioral changes, resulting in loss of cognitive function and ultimately, death. The hallmark of AD pathogenesis appears to be the production, accumulation and oligomerization of amyloid-beta protein (Aβ), forming extracellular amyloid plaques that lead to formation of tangles of intracellular hyper-phosphorylated tau protein, gliosis, synaptic dysfunction and eventually cell death (Bennett et al., 2015; Hardy & Selkoe, 2002). Though few genetic alterations have been reported in the development of AD, in most cases genetic basis is not established including the identical twins. This led to the search for non-genetic factors affecting the AD. A variety of genome wide 32 | Issue 1 | Volume 7 | Fall 2017

and locus specific epigenetic changes have reported (Sanchez-Mut & Graff, 2015). Epigenetic changes including DNA methylation and histone modifications are involved in the pathogenesis of AD and are potential targets for therapeutic manipulation in patients with AS. Drugs such as methyl donors and histone deacetylase inhibitors are being investigated for possible therapeutic benefits (Adwan & Zawia, 2013). In this article we review contemporary literature evaluating the role of epigenetics in AD and therapeutic options utilizing these epigenetic targets.


In this article we review contemporary literature evaluating the role of epigenetics in AD and therapeutic options utilizing these epigenetic targets.


Epigenetics is defined as inherited changes in gene expression that are not encoded within the DNA structure (La Rosa, 2015). Epigenetic alterations include DNA methylation, Histone modifications and miRNA regulation. DNA methylation is the most The Undergraduate Journal of Neuroscience

Manoharan et al. | REVIEW

extensively studied epigenetic change. It refers to the formation of a covalent bond by the addition of a methyl group (CH3) to the carbon-5 (C-5) in cytosine residues of CpG dinucleotides. Regions of the genome containing a high density of CpG dinucleotides are commonly known as “CpG islands” (Goldberg et al., 2007). This process is carried out by a group of enzymes known as DNA methyltransferases (DNMT) which use S-adenosyl-methionine (SAM) as the methyl donor (Jeronimo & Henrique, 2014). Methylation of these regions may result in reduced expression or even complete silencing of the genes involved. Another important epigenetic mechanism is the Histone modification. This includes acetylation, methylation, phosphorylation, ubiquitination, sumoylation and ADP ribosylation of histone proteins. Of these, acetylation and deacetylation of lysine residues play a major role (Liep et al., 2012). There is evidence indicating that histone modifications play a role in neurodegenerative disorders by altering expression patterns of regulated genes. The third main epigenetic mechanism involved in carcinogenesis is miRNA regulation. These are small, non-coding RNAs that regulate the degradation and translation of messenger RNAs (mRNA). These miRNAs can carry out opposing functions within the same gene by behaving variably depending on the tissue type and specific targets (Williams et al., 2016).


ylation of CpG islands may result in the inactivation of transcription, impaired gene expression and gene silencing. (Figure 1). Loss of gene expression may affect several mechanisms of cell homeostasis and memory resulting in AD (Qazi et al., 2017; Andres et al., 2013). Various DNA methylation related genes involved in AD is shown in Table 1. The process of DNA methylation is carried out by a group of enzymes known as DNMT’s. These three active DNMT’s (DNMT1, DNMT3a and DNMT3b) and one related protein lacking catalytic activity (DNMT3L), are present in mammals. DNMT1 shows a preference for hemimethylated DNA. It preserves the methylation patterns during cell division by maintaining methylation of hemimethylated CpG dinucleotides produced by DNA replication. DNMT3a and DNMT3b have been traditionally designated as de novo DNA methyl-transferases. They are responsible for the establishment of DNA methylation patterns in germ cells and during mammalian development. As previously mentioned, DNMT3L is catalytically inactive and functions as a regulatory factor in cells (Jurkowska et al., 2011). DNMT are crucial for memory formation in the adult brain (Levenson et al., 2006). Oliveira et al. (2012) have demonstrated there is decline in DNMT3a2 gene expression with age and this is linked to memory decline. More importantly this memory decline can be reversed by restoring the DNMT3a2 levels.

The following is a discussion on the association between the three main epigenetics mechanisms, (DNA methylation, Histone modification, miRNA regulation) and AD.

DNA Methylation DNA methylation was first reported by Hotchkiss in 1948 when he discovered what he thought were artifacts designated as “epicytosine” and “epiguanine” from calf thymus DNA (Hotchkiss, 1948). DNA methylation consists of the addition of a methyl group to nucleotide bases. Three types of base methylation have been described: adenine at the N-6 position, cytosine at the N-4 position and cytosine at the C-5 position which is the most common among all organisms (Wiedlecki & Zielenkiewicz, 2006). Methylation of these cytosine residues usually takes place in short CpG-rich regions with a CpG density of > 60%, known as CpG islands. These are found in the promoter regions of 50% of all genes (Nettersheim et al., 2011). Hyper meth

Figure 1: DNA methylation of CpG Island carried out by DNA methyl-transferase: A) Attachment of transcription factor to DNA initiates RNA polymerase transcription and gene expression. B) Methyl group prevents transcription factor attachment to DNA leading to transcription inactivation and gene silencing.

Fall 2017 | Volume 7 | Issue 1 | 33

REVIEW | Epigenetics and Alzheimer’s Disease

Despite various studies, strong evidence for causal relationship between DNA methylation and AD is lacking. Sanchez-Mut and Graff (2015) caution that though the results from DNA methylation is promising, further studies are needed to validate these findings using cell-type specific studies. Table 1: DNA methylation genes in AD (Sanchez-Mut & Graff, 2015)


Change Increase Increase Decrease Increase Decrease Increase Increase Increase Increase Increase

Genomic Region Gene Body Downstream Promoter Gene Body Promoter Promoter Promoter Gene Body Promoter Gene Body

MicroRNA Regulation MicroRNAs (miRNAs) are small non-coding RNAs that are typically 19-35 nucleotides in length and regulate post-transcriptional gene expression (La Rosa et al., 2015; Rounge et al., 2015). Alteration of gene expression by regulation of miRNAs is thought to be distinct from DNA methylation. Therefore, the mutual influence between these two epigenetic mechanisms is largely overlooked (Liep et al., 2012). Characterization of miRNAs is performed by using two approaches: studying expression of known miRNAs by hybridization-based techniques or discovery of novel miRNAs molecules by cloning and sequencing (La Rosa et al., 2015; Ahmed, 2007). miRNA have been studied as biomarkers in AD and it has been demonstrated that several miRNAs, such as miR-34a and miR-181b, were upregulated in peripheral blood mononuclear cells in AD patients (Grasso et al., 2015; Schipper et al., 2007). Expression of miRNA in human brain of AD patients have been studied. the expression of miR-9 and miR-128 was found to be elevated in the hippocampus of AD patients. Another study showed that miR-9, -26a, -132, and -146b expression is downregulated and miR-27, -29, -30, -34, and -125b were upregulated in the frontal gyrus of AD patients (Basavaraju & de Lencastre, 2016). Brain tissue are rich in miRNA. 34 | Issue 1 | Volume 7 | Fall 2017

They are highly stable in CSF and regulate several pathways that are involved in the neurodegenerative diseases. miRNAs serve as promising new biomarkers or as a base for therapeutic strategies to treat AD (Williams et al., 2016).

Histone Modification Histones are proteins intertwined with coiled, double stranded genomic DNA that form a structure known as nucleosome, the basic unit of eukaryotic chromatin. Each nucleosome is composed by a core of four pairs of histone molecules (octamer) H2A, H2B, H3 and H4, and the histone linker H1 (Harbde la Rosa et al., 2015). All of the histone molecules shared a common structure which consists of a central “fold domain” and “tails” that contain NH2 and COOH terminals. Histone modifications typically target the “tail” domain at different amino acid positions. These modifications include acetylation, methylation, phosphorylation, ubiquitination, sumoylation and ADP ribosylation. The modification of DNA and histones can result in inhibition or activation of genes. This occurs by a process that regulates chromatin remodeling between the transcriptionally active euchromatin and transcriptionally silent heterochromatin (La Rosa et al., 2015; Dere et al., 2013). Common histone modification related genes involved in AD is shown in Table 2. In an AD mouse model, increased HDAC2 levels have been associated with cognitive impairment. HDAC2 levels are increased within hippocampus of AD patients in postmortem studies (Graff et al., 2012).


At present, evidence directly linking epigenetics changes and AD is lacking. However, non-causal and indirect correlative evidences have been established. DNA methylation based tests could prove to be a reliable biomarker for early diagnosis of AD. Epigenetics markers such as glutathione S-transferase pi 1(GSTP1) have been shown as both diagnostic and prognostic biomarker in prostate cancer. Epigenetic changes are more upstream in AD pathology than other targets such as tau proteins and thus could be beneficial in early stages of the disease to prevent further transcription and accumulation of pathological changes (Adwan & Zawia, 2013). It is reasonable to view epigenetic modifications as a potential therapeutic target given their feasible reversibility. Several molecules to target these modifications have been developed, most notable being DNA methyltransferases inhibitors (DNMTi) The Undergraduate Journal of Neuroscience

Manoharan et al. | REVIEW

and histone deacetylase inhibitors (HDACi). The possibility of reversing epigenetic modifications is encouraging and continues to motivate research in the development of novel targeted therapies.

Table 2: Histone acetylated genes in AD (Sanchez-Mut & Graff, 2015)


Jurkowska, R.Z., T.P. Jurkowki, and A. Jeltsch. (2011). Structure and function of mammalian DNA methyltransferases. Chembiochem, 12(2): p. 206-22. La Rosa, A.H., et al. (2015). The role of epigenetics in kidney malignancies. Cent European J Urol, 68(2): p. 157-64. Levenson, J.M., et al. (2006). Evidence that DNA (cytosine-5) methyltransferase regulates synaptic plasticity in the hippocampus. J Biol Chem, 281(23): p. 15763-73. Liep, J., A. Rabien, and K. Jung. (2012). Feedback networks between microRNAs and epigenetic modifications in urological tumors. Epigenetics, 7(4): p. 315-25. Nettersheim, D., et al. (2011). NANOG promoter methylation and expression correlation during normal and malignant human germ cell development. Epigenetics, 6(1): p. 114-22. Oliveira, A.M., T.J. Hemstedt, and H. Bading (2012). Rescue of aging-associated decline in Dnmt3a2 expression restores cognitive abilities. Nat Neurosci, 15(8): p. 1111-3. Qazi, T.J., et al., Epigenetics in Alzheimer’s Disease: Perspective of DNA Methylation. Molecular Neurobiology, 55(2): p. 1026-1044. Rounge, T.B., et al. (2015). Profiling of the small RNA populations in human testicular germ cell tumors shows global loss of piRNAs. Mol Cancer, 14: p. 153. Sanchez-Mut, J.V. and J. Graff. (2015). Epigenetic Alterations in Alzheimer’s Disease. Front Behav Neurosci, 9: p. 347. Schipper, H.M., et al. (2007). MicroRNA expression in Alzheimer blood mononuclear cells. Gene Regul Syst Bio, 1: p. 263-74. Selkoe, D.J. (2012). Preventing Alzheimer’s disease. Science, 337(6101): p. 1488-92. Siedlecki, P. and P. Zielenkiewicz. (2006). Mammalian DNA methyltransferases. Acta Biochim Pol, 53(2): p. 245-56. Williams, J., et al. (2017). Are microRNAs true sensors of ageing and cellular senescence? Ageing Research Reviews, 35: p. 350-363.


Adwan, L. and N.H. Zawia. (2013). Epigenetics: a novel therapeutic approach for the treatment of Alzheimer’s disease. Pharmacol Ther, 139(1): p. 41-50. Ahmed, F.E. (2007). Role of miRNA in carcinogenesis and biomarker selection: a methodological view. Expert Rev Mol Diagn, 7(5): p. 569603. Andres, G., et al. (2013). The study of DNA methylation in urological cancer: present and future. Actas Urol Esp, 37(6): p. 368-75. Basavaraju, M. and A. de Lencastre. (2016). Alzheimer’s disease: presence and role of microRNAs. Biomol Concepts, 7(4): p. 241-52. Bennett, D.A., et al. (2015). Epigenomics of Alzheimer’s disease. Transl Res, 165(1): p. 200-20. Dere, E., et al. (2013). Biomarkers of chemotherapy-induced testicular damage. Fertil Steril, 100(5): p. 1192-202. Goldberg, A.D., C.D. Allis, and E. Bernstein. (2007). Epigenetics: a landscape takes shape. Cell, 128(4): p. 635-8. Graff, J., et al. (2012). An epigenetic blockade of cognitive functions in the neurodegenerating brain. Nature, 483(7388): p. 222-6. Grasso, M., et al. (2015). Circulating microRNAs in Neurodegenerative Diseases. EXS, 106: p. 151-69. Harb-de la Rosa, A., et al. (2015). Epigenetics application in the diagnosis and treatment of bladder cancer. Can J Urol, 22(5): p. 7947-51. Hardy, J. and D.J. Selkoe. (2002). The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science, 297(5580): p. 353-6. Hotchkiss, R.D. (1948). The quantitative separation of purines, pyrimidines, and nucleosides by paper chromatography. J Biol Chem, 175(1): p. 315-32. Jeronimo, C. and R. Henrique. (2014). Epigenetic biomarkers in urological tumors: A systematic review. Cancer Lett, 342(2): p. 264-74.

Fall 2017 | Volume 7 | Issue 1 | 35



What Drives Us? Chasing Reward in a Dopaminergic Society Solana Liu1 Emory University, Atlanta, Georgia 30322 Correspondence should be addressed to Solana Liu ( 1

An individual’s level of motivation is often heavily attributed as a personality trait-it’s a quality we group with self-control to embody the ideal achiever capable of success in our society. However, seemingly discrete identification of personality traits understates a major component of the motivational driving force: the dopamine mesolimbic reward system, commonly known as the reward circuit. Activation of this circuit triggers the projection of the neurotransmitter dopamine to the nucleus accumbens and feelings of pleasure (Niehaus et al., 2009). Dopamine also critically mediates feelings of motivational desire, an important determinant of incentive drive, in goal-directed behaviors (Robinson et al., 2000; Palmiter et al., 2008). This mechanism is also common to many of our everyday activities. For instance, exercising and listening to music have been associated with increased levels of dopamine (Petzinger et al., 2015). An evolutionary predisposition to chase reward further suggest dopamine’s strength in cultural behavioral manifestations. From the reinforcing effects of cocaine, alcohol, nicotine, food, and music through mediation of the mesolimbic circuit, we’re able to connect a variety of seemingly separate intrinsic and extrinsic factors to provide explanatory and predictive power for dopamine’s increasing influence on the cultural development of the dopaminergic society we live in today. FUNCTIONAL IMPORTANCE OF DOPAMINE Countless studies emphasize the importance of dopamine in many life sustaining behaviors. Nigrostriatal dopamine regulates motor control and goal-directed movement and mesolimbic dopamine plays an invaluable role in executive functions like attention which are important in learning (Korchounov et al., 2010). Along with eustress-“good”, manageable levels of life stress, mesolimbic dopamine also invigorates action towards desired goals when we face mild challenges. Dopamine activates the behavioral explorative approach response, and thus through feelings such as a sudden creative insight or the feeling of anticipation that precedes highly focused engagement in a task, the challenge is perceived as exciting (Ikemoto 2007). Executive functions used to maintain attention to the task at hand is coordinated by the prefrontal cortex and is mediated by proper dopamine func36 | Issue 1 | Volume 7 | Fall 2017

tioning as well (Hosenbocus et al., 2012). Furthermore, reaching the goal-directed outcome results in a sense of mastery and accomplishment–feelings highly dependent on the release of dopamine (Yau & Potenza, 2013). It is nevertheless evident how dopamine is highly involved in all aspects of behavior from the creation of energizing incentive drive and maintenance of mental focus and concentration to the subsequent production of rewarding pleasant feelings as behavioral reinforcement. The reward seeking behavior maintains a seductive dynamic as the release of dopamine triggers the feeling of “wanting”, an anticipatory craving for additional dopamine-releasing behaviors. This subconsciously manifests in a behavioral bias towards dopamine releasing behaviors, which can comprise everyday activities such as indulging in your favorite comfort food. These small stress relieving activities serve as natural rewards that satisfy physioThe Undergraduate Journal of Neuroscience


logical drives such as hunger and reproduction and activate the mesolimbic circuit (Blum et al., 2013). Pleasure can also be derived from unnatural rewards, such as alcohol, drugs, or other thrill-seeking behaviors-all which fuel the rush of dopamine (Michaelides et al., 2013).

EVOLUTION OF A DOPAMINERGIC SOCIETY Dopamine not only plays a key role in our everyday functions, but is also attributed as the key to the development of unique human cognitive skills such as abstract reasoning, temporal analysis, and working memory. These physiological adaptations to social, cultural, and environmental changes were enabled 80,000 years ago through dietary changes such as the inclusion of fish oils and increased meat consumption (Adams et al., 2011). Dietary changes overcame previous developmental constraints and resulted in increased development of dopamine receptors allowing increased dopamine processes or improved functioning at the receptor itself. Hyperdopaminergic effects can be associated with a variety of cultural behavioral manifestations. For instance, hyperdopaminergic effects functionally result in masculine and less emotional behavior (Wersinger & Rissman 2000). Boldness and social risk taking, traits useful in climbing hierarchical ladders, are characterized by the “fearless dominance” trait used to characterize psychopaths. The rising number of psychopaths each year intriguingly parallels the general population’s perception of psychopaths through increasingly normalized psychopathic media portrayal (Smith et al., 2013). In addition, the number of disorders correlated to excessive dopamine such as bipolar disorder, OCD, schizophrenia, and Tourette’s syndrome rapidly continues to increase yearly, which offers further evidence to support an increasingly hyperdopaminergic society (Healy 2006, Nestadt 2010). A “hyped up” motivation system exaggerates dopaminergic effects and therefore encourages fast paced, driven, competitive, and novelty seeking behaviors-characteristic of our hyperdopaminergic society. This evolutionary predisposition to seek rewards explains increasing obsession with achievements in a highly competitive environment in which we continuously set goals to chase the feeling of satisfaction and fulfillment. High dopamine individuals thrive in these societies and through their influence, shape and further encourage the society’s hyperdopaminergic qualities. The rush to seek rewards as well as the intense

culture constructed by conquest and competition can fuel chronic restlessness within our high stress competitive hyperdopaminergic society. Countless neurochemical studies have demonstrated that in addition to relaxing activities and thrill seeking behaviors, stressful stimuli activate the dopamine system (Belujon & Grace, 2015, Piazza & Le Moal, 1998). Our homeostatic abilities to maintain a constant internal environment in response to environmental changes establishes a subconscious search for external dopamine releasing behaviors. The drive to fulfill immediate cravings may account for the presence of high risk, thrill-seeking activities which maintain a significant cultural presence today. Thrill seeking activities can behaviorally manifest as binge drinking, gambling, or substance use and all are able to fuel to the satisfying rush of dopamine (Michaelides et al., 2013, Avena et al., 2008, McHugh et al., 2014, Michaelides et al., 2013, Adinoff et al., 2004, Gosnell, 2005). Although societal standards intuitively categorize these activities as separate, this understates the common circuit activation between seemingly unrelated behaviors. Even sugar can be classified as addictive substance (Avena et al., 2008, Michaelides et al., 2013) and the cross-sensitization in movement from sugar to cocaine indicates some degree of potentiation between sucrose and psychostimulants (Gosnell, 2005). As dopamine additionally plays a role in stress relief, subconscious reward seeking behavior also suggests a neurobiological incentive for trending stress-relieving cultural practices such as yoga, meditation, and spiritual practices (Krishnakumar et al, 2015, Newberg 2014). Common mesolimbic dopamine activation in both everyday and thrill seeking activities clarifies how culturally encouraged stress behaviorally nurtures the manifestation of a range of dopamine-releasing “self-medicating” behaviors. And thus, in response to acute cravings, we naturally gravitate towards rewarding activities to maintain a homeostatic balance.

DOPAMINE DEFICIENCY The proper functioning and processing of dopamine communicates a message of calm and wellbeing to the rest of the brain. However, it’s a delicate balance that the brain battles to maintain. Disruption of dopamine-induced engagement underpins a variety of behavioral disorders which further supports the necessity of properly functioning dopamine. Excessive dopamine can lead to a false sense of euphoria and schizophrenic symptoms whereas insufficients Fall 2017 | Volume 7 | Issue 1 | 37

REVIEW | Chasing Reward in a Dopaminergic Society

dopamine levels can result in Parkinsonâ&#x20AC;&#x2122;s disease, depression, and propensities to addiction (Ducci & Goldman, 2012; Edenberg et al., 2006; Bevilacqua & Goldman, 2009, Korchounov et al., 2010). As dopamine plays crucial roles as the pleasure and anti-stress molecule, inadequate dopaminergic activity stresses out other brain functions resulting in feelings of stress, pain, discomfort, and agitation (Zai et al., 2012). Physical changes in brain areas critical in judgement, decision making, learning, memory and behavior control also occur which behaviorally manifest as motivational deficits, impulsivity, novelty seeking, and a short attention span (Blum et al., 2015). Altered dopaminergic function affects the brainâ&#x20AC;&#x2122;s resting state network connectivity, and is also suggested to impact the efficiency of functional network updates that occurs between tasks (Cole et al., 2013, Schultz & Cole 2016). This is consistent for why those with attention deficit/ hyperactivity disorder (ADHD), for example, would have trouble sustaining focus on a task. Problems in dopamine processing can stem from innate genetic differences, chronic drug use, or long term stress (Aschacher 2013, Weiss et al., 2001, Reist et al., 2007). A specific variant of the DRD2 receptor gene has been posited as the anomaly for resultant neurochemical issue and exists in 30% of the population today (Blum et al., 2008). Regardless of the source of dysfunction, a heightened reward threshold results in a chronically underexcited reward system. Loss in ability to feel various dopaminergic effects in comparison to individuals with properly functioning dopamine processes manifests behaviorally in an inability to derive pleasure from everyday activities. The brain attempts to adjust to the deficiency by searching for external dopamine releasing behaviors. Thus, these individuals are more likely to self-medicate with thrill-seeking activities which would be able to trigger the release of supa-threshold dopamine. Cocaine, alcohol, nicotine, food, and music are all able to fulfill the reinforcing reward through the common mesolimbic mechanism (Blum et al., 2013). In addition to aberrant reward seeking behavior, the loss of dopamine-induced feelings of calm and well-being compounded by downstream effects from the disrupted cascade impacts the ability to respond to stressors. The consequential outcome automatically predisposes individuals to a higher risk for multiple impulsive, compulsive, and addictive behavioral tendencies ranging from mild anxiety and irritability to attention deficit hyper38 | Issue 1 | Volume 7 | Fall 2017

activity disorder, alcoholism, eating disorders, and smoking (Blum et al., 2008). Evidence associating a certain variant of the DRD2 gene that results in deficient dopamine functioning with various substance use disorders, pathological gambling, sex addiction, eating disorders, alcoholism, attention deficit hyperactivity disorder, and smoking supports this hypothesis (Noble et al., 1993; Stice et al., 2012; Noble et al., 1991; Nisoli et al., 2007; Need et al., 2006. In fact, the umbrella term Reward Deficiency Syndrome (RDS) conceptually posits the A1 variant as the anomaly that results in the neurochemical inability to drive pleasure from normal, everyday activities (Blum et al., 2008). Additional studies propose other genetic polymorphisms that may predispose individuals to reward seeking behavior, but itâ&#x20AC;&#x2122;s imperative to note that any reduction in dopamine function can lead to reward deficiency and results in aberrant substance-seeking behavior and a lack of wellness, which can manifest in a variety of behaviors (Reist et al., 2007). CONCLUSION Recognizing the driving force of dopamine notes the biological basis underlying our habits and conceptually redefines perceptions of reward seeking behavior within this dopaminergic society. Predisposition for mesolimbic activation clarifies cultural novelty seeking as well as the lure of thrill seeking extrinsic factors. Connecting the biological basis within all reward seeking behaviors moves away from societal misconceptions of various compulsions attributed to problems in the dopaminergic cascade which predispose to unnatural rewards and explains comorbidity between seemingly separate disorders. Furthermore, it anticipates progression away from the social stigma attached to substance use disorders, ADHD, alcoholism, and the spectrum of reward-seeking behaviors that comprise RDS. Understanding how strongly dopamine is interwoven into our actions can embody a raw perspective of the dopaminergic society we live in. It questions the degree to which our civilization serves as an artificial construct nurtured by our natural human predisposition to chase reward. Western moral values of efficiency, encouraged by dopaminergic projections, shapes many social pressures and expectations. For instance, the ideal of maximum efficiency contributes to shaping the social construction of insomnia and sleep disorders, which continuously increase in number each year (Nunn et al., 2016). The Undergraduate Journal of Neuroscience


The fascinating implications of understanding our cognitive capabilities and motivational sources provides invaluable perceptual perspective on the complex interactions that influence our thoughts, feelings, and actions. Mediation of these feelings significantly contribute to our comprehension of the psychological notion of pleasure. Studies attribute happiness to finding the right hedonic balance-a state of pleasure without too much “wanting” (Kringelbach & Berridge, 2010). And so despite individual differences, the right ideological views powerfully allow us to drive our own motivational mindset within this dopaminergic society. REFERENCES

Adams DK, Sewell MA, Angerer RC, Angerer LM. Rapid adaptation to food availability by a dopamine-mediated morphogenetic response. Nat Commun. 2011;2:592. Adinoff, B. (2004). Neurobiologic Processes in Drug Reward and Addiction.Harvard Review of Psychiatry, 12(6), 305–320. André Nieoullon, Dopamine and the regulation of cognition and attention, Progress in Neurobiology, Volume 67, Issue 1, May 2002, Pages 53-83, ISSN 0301-0082 Aschbacher, K., O’Donovan, A., Wolkowitz, O. M., Dhabhar, F. S., Su, Y., & Epel, E. (2013). Good Stress, Bad Stress and Oxidative Stress: Insights from Anticipatory Cortisol Reactivity. Psychoneuroendocrinology, 38(9), 1698–1708. Avena NM, Rada P, Hoebel BG. Evidence for sugar addiction: behavioral and neurochemical effects of intermittent, excessive sugar intake. Neurosci Biobehav Rev. 2008;32(1):20–39. Belujon, P. & Grace, A. A. Regulation of dopamine system responsivity and its adaptive and pathological response to stress. Proc. Biol. Sci. 282, 20142516 (2015). Bevilacqua, L., & Goldman, D. (2009). Genes and Addictions. Clinical Pharmacology and Therapeutics, 85(4), 359–361. Blum, K., Chen, A. L.-C., Braverman, E. R., Comings, D. E., Chen, T. J., Arcuri, V., … Oscar-Berman, M. (2008). Attention-deficit-hyperactivity disorder and reward deficiency syndrome. Neuropsychiatric Disease and Treatment, 4(5), 893–918. Blum, K., Febo, M., Thanos, P. K., Baron, D., Fratantonio, J., & Gold, M. (2015). Clinically Combating Reward Deficiency Syndrome (RDS) with Dopamine Agonist Therapy as a Paradigm Shift: Dopamine for Dinner? Molecular Neurobiology, 52(3), 1862–1869. Blum, K., Thanos, P. K., & Gold, M. S. (2014). Dopamine and glucose, obesity, and reward deficiency syndrome. Frontiers in Psychology, 5, 919. Blum, K., Thompson, B., Oscar-Berman, M., Giordano, J., Braverman, E., Femino, J., … Schoenthaler, S. (2013). Genospirituality: Our Beliefs, Our Genomes, and Addictions. Journal of Addiction Research & Therapy, 5(4), 162. Blum, K., Thompson, B., Oscar-Berman, M., Giordano, J., Braverman, E., Femino, J., … Schoenthaler, S. (2013). Genospirituality: Our Beliefs, Our Genomes, and Addictions. Journal of Addiction Research & Therapy, 5(4), 162. Bowirrat, A. and Oscar-Berman, M. (2005), Relationship between dopaminergic neurotransmission, alcoholism, and reward deficiency syndrome. Am. J. Med. Genet., 132B: 29–37. David M. Cole, Nicole Y. L. Oei, Roelof P. Soeter, Stephanie Both, Joop M. A. van Gerven, Serge A. R. B. Rombouts, Christian F. Beckmann; Dopamine-Dependent Architecture of Cortico-Subcortical Network Connectivity. Cereb Cortex 2013; 23 (7): 1509-1516. Douglas H. Schultz, Michael W. Cole; Higher Intelligence is Associated with Less Task-Related Brain Network Reconfiguration. Journal of Neuroscience; 36 (33): 8551-8561.

Ducci, F., & Goldman, D. (2012). The Genetic Basis of Addictive Disorders. The Psychiatric Clinics of North America, 35(2), 495–519. Edenberg, H. J. and Foroud, T. (2006), REVIEW: The genetics of alcoholism: identifying specific genes through family studies. Addiction Biology, 11: 386–396. Gosnell B. A. (2005). Sucrose intake enhances behavioral sensitization produced by cocaine. Brain Res. 1031 194–201 Healy D (2006) The Latest Mania: Selling Bipolar Disorder. PLoS Med 3(4): e185. Hosenbocus, S., & Chahal, R. (2012). A Review of Executive Function Deficits and Pharmacological Management in Children and Adolescents. Journal of the Canadian Academy of Child and Adolescent Psychiatry, 21(3), 223–229. Ikemoto S. Dopamine reward circuitry: Two projection systems from the ventral midbrain to the nucleus accumbens-olfactory tubercle complex. Brain Res. Rev. 2007;56:27–78. Koob GF, Volkow ND. Neurocircuitry of addiction.Neuropsychopharmacology. 2010;35(1):217–238. Korchounov, A., Meyer, M. F., & Krasnianski, M. (2010). Postsynaptic nigrostriatal dopamine receptors and their role in movement regulation. Journal of Neural Transmission, 117(12), 1359–1369. Kringelbach, M. L., & Berridge, K. C. (2010). The Neuroscience of Happiness and Pleasure. Social Research, 77(2), 659–678. Krishnakumar, D., Hamblin, M. R., & Lakshmanan, S. (2015). Meditation and Yoga can Modulate Brain Mechanisms that affect Behavior and Anxiety-A Modern Scientific Perspective. Ancient Science, 2(1), 13– 19. McHugh MJ, Demers CH, Salmeron BJ, Devous MD, Sr, Stein EA, Adinoff B. Cortico-amygdala coupling as a marker of early relapse risk in cocaine-addicted individuals. Front Psychiatry. 2014;5:16. Michaelides M., Miller M. L., Subrize M., Kim R., Robison L., Hurd Y. L., et al. (2013). Limbic activation to novel versus familiar food cues predicts food preference and alcohol intake. Brain Res. 1512 37–44 Modestino, E. J., Blum, K., Oscar-Berman, M., Gold, M. S., Duane, D. D., Sultan, S. G. S., & Auerbach, S. H. (2015). Reward Deficiency Syndrome: Attentional/Arousal Subtypes, Limitations of Current Diagnostic Nosology, and Future Research. Journal of Reward Deficiency Syndrome, 1(1), 6–9. Need A. C., Ahmadi K. R., Spector T. D., Goldstein D. B. (2006).Obesity is associated with genetic variants that alter dopamine availability. Ann. Hum. Genet. 70 293–303 Nestadt, G., Grados, M., & Samuels, J. F. (2010). Genetics of OCD. The Psychiatric Clinics of North America, 33(1), 141–158. Newberg, A. B. (2014). The neuroscientific study of spiritual practices. Frontiers in Psychology, 5, 215. Niehaus, J. L., Cruz-Bermúdez, N. D., & Kauer, J. A. (2009). Plasticity of Addiction: a Mesolimbic Dopamine Short-Circuit? The American Journal on Addictions / American Academy of Psychiatrists in Alcoholism and Addictions,18(4), 259–271. Nisoli E., Brunani A., Borgomainerio E., Tonello C., Dioni L., Briscini L., et al. (2007). D2 dopamine receptor (DRD2) gene Taq1A polymorphism and the eating-related psychological traits in eating disorders (anorexia nervosa and bulimia) and obesity. Eat. Weight Disord. Noble E. P., Blum K., Khalsa M. E., Ritchie T., Montgomery A., Wood R. C., et al. (1993). Allelic association of the D2 dopamine receptor gene with cocaine dependence. Drug Alcohol Depend. 33 271–285 Noble E. P., Blum K., Ritchie T., Montgomery A., Sheridan P. J. (1991).Allelic association of the D2 dopamine receptor gene with receptor-binding characteristics in alcoholism. Arch. Gen. Psychiatry 48 648–65 Nunn, C. L., Samson, D. R., & Krystal, A. D. (2016). Shining evolutionary light on human sleep and sleep disorders. Evolution, Medicine, and Public Health, 2016(1), 227–243. Palmiter, R. D. (2008). Dopamine Signaling in the Dorsal Striatum Is Essential for Motivated Behaviors: Lessons from Dopamine-deficient Mice. Annals of the New York Academy of Sciences, 1129, 35–46. Petzinger, G. M., Holschneider, D. P., Fisher, B. E., McEwen, S., Kintz, N., Halliday, M., Jakowec, M. W. (2015). The Effects of Exercise on Dopamine Neurotransmission in Parkinson’s Disease: Targeting Neuroplasticity to Modulate Basal Ganglia Circuitry. Brain Plasticity, 1(1), 29–39.

Fall 2017 | Volume 7 | Issue 1 | 39

REVIEW | Chasing Reward in a Dopaminergic Society Piazza PV, Le Moal M (1998). The role of stress in drug self-administration. Trends Pharmacol. Sci. 19, 67–74. Reist C, Ozdemir V, Wang E, Hashemzadeh M, Mee S, Moyzis R. 2007. Novelty Seeking and the Dopamine D4 Receptor Gene (DRD4) Revisited in Asians: Haplotype Characterization and Relevance of the 2-Repeat Allele. Am J Med Genet Part B 144B:453–457. Robinson T. E., Berridge K. C. (2000). The psychology and neurobiology of addiction: an incentive-sensitization view. Addiction95(Suppl. 2), S91–S117 Smith, S. F., Lilienfeld, S. O., Coffey, K., and Dabbs, J. M. (2013). Are psychopaths and heroes twigs off the same branch? Evidence from college, community, and presidential samples. J. Res. Pers. doi: 10.1016/j. jrp.2013.05.006 Stice E., Dagher A. (2010). Genetic variation in dopaminergic reward in humans. Forum Nutr. 63 176–185 10.1159/000264405 Stice E., Yokum S., Zald D., Dagher A. (2011). Dopamine-based reward circuitry responsivity, genetics, and overeating. Curr. Top. Behav. Neurosci. 6 81–93 Sullivan D., Pinsonneault J. K., Papp A. C., Zhu H., Lemeshow S., Mash D. C., et al. (2013). Dopamine transporter DAT and receptor DRD2 variants affect risk of lethal cocaine abuse: a gene-gene-environment interaction. van Strien T., Snoek H. M., van der Zwaluw C. S., Engels R. C. (2010).Parental control and the dopamine D2 receptor gene (DRD2) interaction on emotional eating in adolescence. Appetite 54 255–261 Weiss F, Koob GF. Drug addiction: Functional neurotoxicity of the brain reward systems. Neurotox Res. 2001;3:145–56. Wersinger, S., Rissman, E. (2000). Dopamine Activates Masculine Sexual Behavior Independent of the Estrogen Receptor α. Journal of neuroscience, 20(11), 4248-4254. Yau, Y. H. C., & Potenza, M. N. (2013). Stress and Eating Behaviors. Minerva Endocrinologica, 38(3), 255–267. Young R. M., Lawford B. R., Nutting A., Noble E. P. (2004). Advances in molecular genetics and the prevention and treatment of substance misuse: implications of association studies of the A1 allele of the D2 dopamine receptor gene. Addict. Behav. 29 1275–1294 Zai C. C., Ehtesham S., Choi E., Nowrouzi B., de Luca V., Stankovich L., et al. (2012). Dopaminergic system genes in childhood aggression: possible role for DRD2. World J. Biol Psychiatry 13 65–74

40 | Issue 1 | Volume 7 | Fall 2017

The Undergraduate Journal of Neuroscience



Heritability and Neurophysiology of Psychopathy Implicate Deep Brain Stimulation (DBS) as a PossibleTreatment for Psychopathy Schuyler Gaillard1 Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 Correspondence should be addressed to Schuyler Gaillard ( 1

Psychopathy is a serious disorder, which affects approximately one percent of the general population and is characterized by antisocial traits and behaviors. Among criminal offenders, psychopathy is associated with high rates of recidivism, and psychopaths comprise up to twenty percent on US prison inmates. Cognitive behavioral and psychopharmaceutical treatments have been unsuccessful in cases of psychopathy, with no currently known effective treatment for the disorder. To date, research has revealed that psychopathy is heritable, begins manifesting during childhood with antisocial behaviors becoming worse over time, and is associated with reduced amygdala volume, decreased grey matter density in the ventromedial prefrontal cortex (vmPFC), and impaired vmPFC-amygdala functional connectivity. Taken together, these findings implicate deep brain stimulation (DBS) as a possible treatment for or mitigation of psychopathy, with the amygdala or vmPFC-amygdala connections as a potential DBS target. DBS target validity, potential candidates, and future research directions are discussed. Psychopathy is a disorder characterized by superficial charm and glibness, calloused or flat affect, lying, manipulation of others, excessive boredom, impulsiveness, thrill seeking, poor planning, and lack of empathy, remorse, and guilt (Brook et al., 2010; Kiehl & Hoffman, 2011; Tuvblad, Bezdjian, Raine, & Baker, 2014). Psychopaths constitute 1% of the general population and approximately 20% of prison inmates (Kiehl & Hoffman, 2011). They are notoriously difficult to treat, but their overrepresentation in the criminal justice system and their extremely high rates of recidivism remain strong motivators for finding a treatment (N. E. Anderson & Kiehl, 2014; Kiehl & Hoffman, 2011). Using recent functional, structural, and lesion studies, the current paper argues that deep-brain stimulation (DBS) of the vmPFC or amygdala is a potential treatment for at-risk youths with significant psychopathic traits and for adult psychopaths. Traditional treatment attempts for psychop

athy, such as correctional or behavioral therapy, have been largely unsuccessful (N. E. Anderson & Kiehl, 2014; Kiehl & Hoffman, 2011). There are several noteworthy theories as to why we have not been able to treat psychopaths, the first of which is that they are â&#x20AC;&#x2DC;untreatableâ&#x20AC;&#x2122; (Kiehl & Hoffman, 2011). There are several treatment methods that have yet to be used for psychopathy. Until all potential, scientifically valid treatment options have been exhausted, it is unreasonable to assume psychopathy is untreatable. A second explanation for why no treatment has been very successful is that psychopaths may not stay in therapy long enough for treatment to be effective, possibly due to their manipulation of others into believing that they have improved (N. E. Anderson & Kiehl, 2014; Kiehl & Hoffman, 2011). This is a possible problem with treatment implementation, and suggests that previously attempted treatment regimens, such as cognitive behavioral therapy, be updated and furFall 2017 | Volume 7 | Issue 1 | 41

Gaillard | OPINION

ther studied; however, it does not reveal whether those treatments will be effective given ample time. Therefore, it is necessary to continue investigating other potential treatments, such as DBS. DBS treatment of psychopathy should include counter-measures against manipulation, such as long-term use and close monitoring of device use. To the author’s knowledge, DBS has not yet been studied as a treatment for psychopathy. Twin and adoption studies suggest that psychopathy is heritable, a notion supported by different assessment methods. Heritability estimates for overall psychopathic personality are as high as 70%, with some traits (e.g. interpersonal dominance, fearlessness) being more heritable than others (e.g. impulsivity, poor planning, aggression, social alienation) ( BLONIGEN et al., 2005; Brook et al., 2010; Burt, 2009; Tuvblad et al., 2014). There are several psychopathy measures, including the Hare Psychopathy Check List – Revised (PCL-R) and its Youth Version (PCL:YV), the Multidimensional Personality Questionnaire (MPQ), the Child Psychopathy Scale (CPS), the Antisocial Process Screening Device (APSD), and others. Discrepancies in individual trait heritability measures may be affected by psychopathy measure. However, a study that analyzed five different psychopathy assessments, including self-reported, parent-reported, and interview-based methods, concluded that all assessments suggested similar heritability of overall psychopathy (Tuvblad et al., 2014). The characteristic brain abnormalities of psychopaths suggest many potential DBS targets. Psychopaths have significantly different brain morphology and connectivity than non-psychopaths (N. E. Anderson & Kiehl, 2014). Psychopaths have decreased grey matter in the vmPFC, including the orbitofrontal cortex, and in limbic and paralimbic structures (N. E. Anderson & Kiehl, 2014; Ermer, Cope, Calhoun, Nyalakanti, & Kiehl, 2012). Additionally, researchers have found surface deformations and total volume reduction in the amygdala, and reduced functional connectivity between the vmPFC and the amygdala in psychopaths (Motzkin, Newman, Kiehl, & Koenigs, 2011; Yang, Raine, Narr, Colletti, & Toga, 2009). While some studies have found minor structural differences between female and male psychopaths and between different psychopathic subtypes, they still reveal reduced pre-frontal cortical thickness and vmPFC-amygdala functional connectivity in all psychopaths when compared to normal controls (Motzkin et al., 2011; 42 | Issue 1 | Volume 7 | Fall 2017

Yang et al., 2009). DBS is most commonly used to treat Parkinson’s Disease, a disease characterized by frontal cortical thinning and degeneration of dopaminergic neurons in the substantia nigra (Kalia & Lang, 2015; Mak et al., 2015). Because DBS is usually used to stimulate atrophied neurons, underactive neurons, or their targets (Liu et al., 2014), DBS for psychopathy should target regions of reduced volume or density. The regions of reduced grey matter in psychopathic brains cover a large portion of the brain, but converging evidence most heavily implicates the vmPFC and the amygdala as potential DBS targets. Lesion and functional studies suggest that reduced vmPFC and amygdala activation are linked to major psychopathic traits. Lesion studies reveal that damage to the vmPFC results in psychopathic traits, including impulsivity, proneness to boredom, and, if incurred at a developmental age, moral inhibition and egocentricity (S. W. Anderson, Bechara, Damasio, Tranel, & Damasio, 1999; Moretto, Làdavas, Mattioli, & Pellegruno, 2010; Taber-Tomas et al., 2014). These studies implicate the vmPFC in altruistic behaviors and suggest that it may be part of a “moral pathway.” Functional studies suggest that the vmPFC and the amygdala are involved in empathy for others, a feature that psychopaths lack – psychopathic inmates had less activity in the vmPFC and in the amygdala than non-psychopathic inmates when viewing others experience pain . Furthermore, higher psychopathy scores in the general public correlate with lower mPFC and amygdala activation when interpreting emotion via facial expressions (Gordon, Baird, & End, 2004). Finally, a combined functional-structural MRI study revealed reduced functional vmPFC-amygdala connectivity in psychopaths (Motzkin et al., 2011). Reduced vmPFC and amygdala activation in psychopaths suggests that the vmPFC and the amygdala are critical contributors to psychopathic traits. Therefore, the vmPFC-amygdala pathway is a likely target for DBS for treatment of psychopathy. The exact mechanism of DBS unclear, but it appears to disrupt abnormal neural signals, and has excitatory and inhibitory effects dependent on signal frequency and placement of the lead. (Murrow, 2014; Okun et al., 2013; Chiken et al., 2016) Additionally, recent rodent studies suggest that DBS may increase synaptic density and local blood vessel size. Therefore, DBS may partially reverse reduced activity observed in psychopaths by altering neuronal circuits. (Chakravarty et al., 2016; The Undergraduate Journal of Neuroscience

OPINION | DBS as a Possible Treatment for Psychopathy

Veerakumar et al., 2014). While there are few current examples of vmPFC and amygdala DBS in humans, case studies and review of other DBS applications suggest the amygdala as a potential target (Langevin, 2012; Sturm et al., 2013). Of particular interest is a case in which DBS to the BL amygdala successfully reduced self-harm behaviors and improved emotional, social, and cognitive symptoms in a child with Kanner’s Autism. While autism and psychopathy are very different diseases, this shows that the amygdala is a valid target for DBS, including for reducing anti-social behaviors. High heritability and early-onset of behavioral manifestations suggest DBS treatment should be used from a young age. Behavioral indicators of future antisocial behavior appear as early as 27 months (Hyde et al., 2016). In a study with over 561 adopted infants whose biological mothers had antisocial personality disorder (ASPD), Incidence and severity of unemotional-like behaviors at the age of 27 months were positively correlated with severity of anti-social behaviors at the age of seven (Hyde et al., 2016). Behaviors measured included unresponsiveness to punishment, lack of affection, fear, and guilt, and diminished response to affection, were assessed using the Child Behavior Checklist, and are related to traits characteristic of ASPD and psychopathy. More intense psychopathic traits, and even criminal behavior, are often present by adolescence (Brook et al., 2010; Burt, 2009; Hyde et al., 2016; Kiehl & Hoffman, 2011; Tuvblad et al., 2014). Therefore, early intervention may be necessary to mitigate the progression of psychopathic traits. While more commonly used in adults, DBS is also used in pediatric patients with movement disorders (DiFrancesco, Halpern, Hurtig, Baltuch, & Heuer, 2012; Kaminska et al., 2016; Keen, Przekop, Olaya, Zouros, & Hsu, 2014; Lipsman, Ellis, & Lozano, 2010). Pediatric patients with DBS have slightly higher rates of infection and exhibit faster battery drainage than adult patients, but DBS remains an effective treatment for this pediatric cohort (DiFrancesco et al., 2012; Kaminska et al., 2016; Keen et al., 2014; Lipsman et al., 2010). Therefore, early intervention with DBS for high-risk individuals is a valid treatment option. In summary, psychopathy has thus far remained resistant to many forms of treatment, but its high impact on society and the criminal justice system necessitates finding an effective treatment. Imaging studies have revealed reduced grey matter and activity in the vmPFC, as well as reduced functional

connectivity between the vmPFC and amygdala in psychopaths. Lesion studies further implicate vmPFC involvement in psychopathy by revealing that developmental vmPFC damage results in increased impulsive behaviors, egocentricity, and antisocial behaviors; diminished empathy; and a significantly impaired “moral compass”. Therefore, increasing function or activity of the vmPFC-amygdala pathway may mitigate psychopathic traits. DBS has been shown to increase synaptic density and normalize disrupted signals. Furthermore, it may increase activity of surrounding neural pathways and is used to target specific, small areas of the brain. Therefore, DBS in the vmPFC or amygdala is a promising treatment for psychopathy, as it may partially-reverse reduced or impaired activity of the vmPFC and amygdala observed in psychopaths. Moreover, DBS has been successfully used in the amygdala for treatment other diseases, suggesting that, with proper placement, the amygdala is a safe target for DBS. Because psychopathy is highly heritable and psychopathic behaviors manifest at a young age, early treatment may be key in treatment of the disease. DBS use in pediatric cohorts with movement disorders has no apparent negative impacts on normal development. Thus, DBS can be used in children or adolescents who exhibit psychopathic traits. However, DBS implant surgery has several complications, including risk of hemorrhage and infection. Additionally, though rare, DBS leads can migrate over time, resulting in undesired physical or behavioral side effects. Therefore, further understanding of DBS’s specific mechanisms of action and pinpointing a specific DBS target for psychopathy is necessary before subjecting any psychopathic individuals to neurosurgery. Finally, as for any other patients with DBS implants, close monitoring is necessary so that lead displacement can be corrected immediately. ACKNOWLEDGEMENTS Thank you to Professor Rebecca Saxe, Jessie Stickgold-Sarah, and Dr. William Gaillard for your encouragement of my interests and guidance through my writing process.

Fall 2017 | Volume 7 | Issue 1 | 43

Gaillard | OPINION


Anderson, N. E., & Kiehl, K. A. (2014). Psychopathy: Developmental Perspectives and their Implications for Treatment. Restorative Neurology and Neuroscience, 32(1), 103–117. RNN-139001 Anderson, S. W., Bechara, A., Damasio, H., Tranel, D., & Damasio, A. R. (1999). Impairment of social and moral behavior related to early damage in human prefrontal cortex. Nature Neuroscience, 2(11). BLONIGEN, D. M., HICKS, B. M., KRUEGER, R. F., PATRICK, C. J., & IACONO, W. G. (2005). Psychopathic personality traits: heritability and genetic overlap with internalizing and externalizing psychopathology. Psychological Medicine, 35(5), 637–648. S0033291704004180 Brook, M., Panizzon, M. S., Kosson, D. S., Sullivan, E. A., Lyons, M. J., Franz, C. E., … Kremen, W. S. (2010). PSYCHOPATHIC PERSONALITY TRAITS IN MIDDLE-AGED MALE TWINS:A BEHAVIOR GENETIC INVESTIGATION. Journal of Personality Disorders, 24(4), 473–486. https://doi. org/10.1521/pedi.2010.24.4.473 Burt, S. A. (2009). Are there meaningful etiological differences within antisocial behavior? Results of a meta-analysis. Clinical Psychology Review, 29(2), 163–178. Chakravarty, M. M., Hamani, C., Martinez-Canabal, A., Ellegood, J., Laliberté, C., Nobrega, J. N., … Lerch, J. P. (2016). Deep brain stimulation of the ventromedial prefrontal cortex causes reorganization of neuronal processes and vasculature. NeuroImage, 125, 422–427. https:// Chiken, S., & Nambu, A. (2016). Mechanism of Deep Brain Stimulation: Inhibition, Excitation, or Disruption? The Neuroscientist: A Review Journal Bringing Neurobiology, Neurology and Psychiatry, 22(3), 313–322. DiFrancesco, M. F., Halpern, C. H., Hurtig, H. H., Baltuch, G. H., & Heuer, G. G. (2012). Pediatric indications for deep brain stimulation. Child’s Nervous System: ChNS: Official Journal of the International Society for Pediatric Neurosurgery, 28(10), 1701–1714. https://doi. org/10.1007/s00381-012-1861-2 Ermer, E., Cope, L. M., Calhoun, V. D., Nyalakanti, P. K., & Kiehl, K. A. (2012). Aberrant Paralimbic Gray Matter in Criminal Psychopathy. Journal of Abnormal Psychology, 121(3), 649–658. a0026371 Gordon, H. L., Baird, A. A., & End, A. (2004). Functional differences among those high and low on a trait measure of psychopathy. Biological Psychiatry, 56(7), 516–521. Hyde, L. W., Waller, R., Trentacosta, C. J., Shaw, D. S., Neiderhiser, J. M., Ganiban, J. M., … Leve, L. D. (2016). Heritable and Nonheritable Pathways to Early Callous-Unemotional Behaviors. The American Journal of Psychiatry, 173(9), 903–910. ajp.2016.15111381 Kalia, L. V., & Lang, A. E. (2015). Parkinson’s disease. Lancet (London, England), 386(9996), 896–912. Kaminska, M., Perides, S., Lumsden, D. E., Nakou, V., Selway, R., Ashkan, K., & Lin, J.-P. (2016). Complications of Deep Brain Stimulation (DBS) for dystonia in children - The challenges and 10 year experience in a large paediatric cohort. European Journal of Paediatric Neurology: EJPN: Official Journal of the European Paediatric Neurology Society. Keen, J. R., Przekop, A., Olaya, J. E., Zouros, A., & Hsu, F. P. K. (2014). Deep

44 | Issue 1 | Volume 7 | Fall 2017

brain stimulation for the treatment of childhood dystonic cerebral palsy. Journal of Neurosurgery. Pediatrics, 14(6), 585–593. https:// Kiehl, K. A., & Hoffman, M. B. (2011). THE CRIMINAL PSYCHOPATH: HISTORY, NEUROSCIENCE, TREATMENT, AND ECONOMICS. Jurimetrics, 51, 355–397. Langevin, J.-P. (2012). The amygdala as a target for behavior surgery. Surgical Neurology International, 3(Suppl 1), S40–S46. https://doi. org/10.4103/2152-7806.91609 Lipsman, N., Ellis, M., & Lozano, A. M. (2010). Current and future indications for deep brain stimulation in pediatric populations. Neurosurgical Focus, 29(2), E2. Liu, Y., Li, W., Tan, C., Liu, X., Wang, X., Gui, Y., … Chen, L. (2014). Meta-analysis comparing deep brain stimulation of the globus pallidus and subthalamic nucleus to treat advanced Parkinson disease. Journal of Neurosurgery, 121(3), 709–718. JNS131711 Mak, E., Su, L., Williams, G. B., Firbank, M. J., Lawson, R. A., Yarnall, A. J., … O’Brien, J. T. (2015). Baseline and longitudinal grey matter changes in newly diagnosed Parkinson’s disease: ICICLE-PD study. Brain: A Journal of Neurology, 138(Pt 10), 2974–2986. brain/awv211 Moretto, G., Làdavas, E., Mattioli, F., & Pellegruno, G. di. (2010). A Psychophysiological Investigation of Moral Judgment after Ventromedial Prefrontal Damage. Journal of Cognitive Neuroscience, 22(8), 1888– 1899. Motzkin, J. C., Newman, J. P., Kiehl, K. A., & Koenigs, M. (2011). Reduced prefrontal connectivity in psychopathy. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 31(48), 17348– 17357. Murrow, R. W. (2014). Penfield’s Prediction: A Mechanism for Deep Brain Stimulation. Frontiers in Neurology, 5, 213. fneur.2014.00213 Okun, M. S., & Oyama, G. (2013). [Mechanism of action for deep brain stimulation and electrical neuro-network modulation (ENM)]. Rinsho Shinkeigaku = Clinical Neurology, 53(9), 691–694. Sturm, V., Fricke, O., Bührle, C. P., Lenartz, D., Maarouf, M., Treuer, H., … Lehmkuhl, G. (2013). DBS in the basolateral amygdala improves symptoms of autism and related self-injurious behavior: a case report and hypothesis on the pathogenesis of the disorder. Frontiers in Human Neuroscience, 6. Taber-Tomas, B. C., Asp, E. W., Koenigs, M., Sutterer, M., Anderson, S. W., & Tranel, D. (2014). Arrested development: early prefrontal lesions impair the maturation of moral judgment. Brain, 137, 1254–1261. Tuvblad, C., Bezdjian, S., Raine, A., & Baker, L. A. (2014). The Heritability of Psychopathic Personality in 14 to 15 year Old Twins: A Multi-Rater, Multi-Measure Approach. Psychological Assessment, 26(3), 704–716. Veerakumar, A., Challis, C., Gupta, P., Da, J., Upadhyay, A., Beck, S. G., & Berton, O. (2014). Antidepressant-like effects of cortical deep brain stimulation coincide with pro-neuroplastic adaptations of serotonin systems. Biological Psychiatry, 76(3), 203–212. https://doi. org/10.1016/j.biopsych.2013.12.009 Yang, Y., Raine, A., Narr, K. L., Colletti, P., & Toga, A. W. (2009). Localization of Deformations Within the Amygdala in Individuals With Psychopathy. Archives of General Psychiatry, 66(9), 986–994. https://doi. org/10.1001/archgenpsychiatry.2009.110distance between males and females. The Journal of Neuroscience, 32(46), 16074-9.

The Undergraduate Journal of Neuroscience


H OT TO P I C S The Brain on Atheism: An Analysis of Religion and Its Neural Roots David Graykowski1 DePaul University, Chicago, Illinois 60604 Correspondence should be addressed to David Graykowski ( 1

Questioning the roots of religion has been a prevalent topic of interest in neuroscience since the turn of the century. In “The Brain on Atheism,” I analyze studies which support the “God Spot” theory and delve into more recent studies which oppose this concept and give rise to new thought. In addition, I propose a new theory concerning the existence of an “Anti-God Spot” and question if it is possible that humans are “hardwired” to be atheistic. Allow me to begin with a hypothetical situation: Imagine a group of Martian voyagers zipping past moons and stars on their way to planet Earth. They arrive somewhere in the Middle East and witness a completely foreign sight: a group of strange looking creatures—each bowing its head to the ground and voicing outwards to someone or something. What these aliens have just witnessed for the first time is our human concept of religion. Throughout history, religion has played an important if not vital role in human existence and in the creation of human cultures. From Rama defeating Ravana in the famous Hindu text to the Ramayana (Valmiki, 500 B.C.) to the Buddha achieving enlightenment under the Bodhi tree, we humans have created all sorts of stories which theorize about where life originates from, how it is possible, and where the practice of it will ultimately lead us to in the end. We have always felt the need, if not the responsibility, to serve a higher power to give us a sense of comfort and security in our lives. Religion has been, in multitudes of occurrences, the impetus of great art, literature, music, architecture, voyages, inventions, and numerous other human achievements. There is no doubt that religion has had a significant effect on seemingly every aspect of what we call life. It can be rather baffling and sometimes even intimidating to contemplate religion and the

consistent, constant, and awe-inspiring role it has played throughout history. But what about the opposition? Anti-religion. Atheism. “Is man merely a mistake of God’s? Or God merely a mistake of man?” (Nietzsche, 1998, p. 2) Friedrich Nietzsche, a self-proclaimed “Antichrist” and one of history’s most renowned proponents of atheism, has been the face of this ideology in countercultures for many years. However, atheism dates further back than Nietzsche’s works. In fact, atheism has been around as long as its counterpart: God-worshipping, life-fulfilling religious vocations. The first accounts of anti-religious teaching date back to the 5th century B.C. (ironically before Christ) and are attributed to Diagoras of Melos. While religious ideology has been the spark of much inspiration and creation, atheistic dogma has been an impetus as well. However, inspiration seems to be the wrong word choice here for atheism is a general disbelief in the existence of God and even religion as a whole. The inspiration that atheism provides stems from the overall dogma of disbelief that coincides with this ideology—translating into works inspired by this ideology. Are advocates of the concept that there is no higher power in existence purely existentialists looking to be apart from the crowd? Were they raised in this manner? Or is there some other yet to be deciphered defining Fall 2017 | Volume 7 | Issue 1 | 45

Graykowski | HOT TOPICS

factor? Is religion purely of a spiritual nature (our human attempt to seek acceptance and comfort), or has evolution led us to become “hardwired” for religion on a biological, internal level? Have our incessant years of practicing and preaching the idea that there is something bigger than ourselves— governing our lives, looking over us, and guiding us to success—actually worked its way into the neural mapping of our brains? Or is all of this just scientific babble? Is religion purely a mystical and mysterious entity that not even science can explain? Recently, researchers from many different fields such as biology, neuroscience, psychology, and even the newly emerged field of neurotheology have been asking such questions, and the topic has been receiving more and more attention. With the turn of the century, many scientists, more specifically those of the ever-growing field of contemporary neuroscience, have theorized that religion can be explained biologically by a neurological phenomenon that affects our internal world. An area in our right parietal lobe dubbed the “God Spot” has been pinpointed as a possible source of our concept of religion. The way in which neuroscientists have explored the “God Spot” theory is similar across the board. The researchers asked a group of carefully selected patients (with religious or non-religious background) to pray or presented them with a series of questions concerning their viewpoints on religious and spiritual concepts. Meanwhile, the patients’ neurophysiological data was recorded by either magnetic resonance imaging (MRI), an electroencephalogram (EEG), or a machine of a similar nature. Researchers found that when patients were asked to do these tasks, their right parietal lobes showed an increased amount of activity in the area of their cerebral cortex corresponding to the “God Spot”. (Beauregard, Paquette & et al., 2009). For some time, this theory was thought to be the explanation for our notion of religion that everyone was waiting for; unfortunately, this proved not to be the case. A group of researchers from the University of Missouri have recently put an end to this religious debacle. Their research on the topic in 2012 disproved the “God Spot” theory—the idea that an isolated area of the human brain (specifically in the right parietal lobe) was responsible for spirituality . These researchers studied 20 patients with severe damage to the right parietal lobe and asked the selected patients’ questions about their level of 46 | Issue 1 | Volume 7 | Fall 2017

spirituality after the damage. Many actually reported an increased amount of belief in a higher power post-damage (Johnstone et al., 2012). This intriguing but disconsolate research sent progress in the wrong direction and left us yearning for an explanation to such an outrageously important part of human life and existence. However, as our faith in finding the root of faith diminished, Dr. Keise Izuma from the University of York performed a study which seemed to support the theory that there was a biological component to religion. Dr. Izuma used transcranial magnetic stimulation to disrupt the posterior medial frontal cortex temporarily, and this led him to a rather baffling finding. In the experiment, Izuma split a group of participants into two halves. Half of the participants received a low level “sham” procedure that did not affect the brain, and the other half received a shock that lowered the activity of the posterior medial frontal cortex—which is essential in solving problems. Following this, the patients were then asked to think about death and their feelings on religious beliefs. Izuma found that 32.8% of the participants who received the non-placebo shock reported less belief in God, angels, or heaven (Holbrook et al., 2015). Time and time again it has been shown that people tend to seek the security blanket of religion when presented with a problem, especially with death. But Dr. Izuma’s findings also gives rise to the following observations. The “God Spot” theory has already been proven implausible. There is not one specific spot in our neural programming that gives rise to our concept of spirituality and religion. For this reason, I suggest that religion can be generalized into a system of pure decision-making and problem-solving. Either religion is a fitting solution to the problem, or it is not. However, I also believe that we cannot completely shut down the “God Spot” theory. No single area of the brain may be the sole origin of our ability to believe in the concept of religion. However, the right parietal lobe and posterior medial frontal cortex may still play roles in conjuring up our human idea of religion. Certain theories state that many different areas of our brain work together when one is feeling spiritual (Cristofori et al., 2016). If an atheist were to stumble upon the findings of Dr. Keise Izuma, they would probably become a little worried about themselves and their fellow atheists. Lowered activity in the brain results to a disbelief in God. However, I would like to present my own theory on the matter of atheism—a theoThe Undergraduate Journal of Neuroscience

HOT TOPICS | The Brain on Atheism

ry that may reassure any atheists familiar with the works of Dr. Izuma. Scientists of nearly every field have already pondered the question of where religion originated from in our internal and possibly purely external world. But from where does atheism originate? Possibly the thought is that if we find out the origin of religion in our physiology, we will also have an explanation for the disbelief in religion. It is assumed that diminished activity in part(s) of the brain vital for our ability to conceptualize religion will result in disbelief in God and spirituality as a whole. But what if this is not the case? Some scientists and even the average citizen may find the following proposal absurd, but it is a question that deserves attention. Does atheism have its fixed part in the brain, the “Anti-God Spot”? Why should a belief in God be examined and disbelief ignored when atheism is just as popular as its counterpart and has been practiced by human beings for nearly as long. Some may say it is because religion is abstract and requires higher levels of thinking. However, I find that precisely because religion is so abstract as to prevent a single, localized area of the brain from being responsible for it, atheism is of a more elementary, pure nature and makes the possibility of an “Anti-God Spot” plausible. While evolutionists will argue that we have become “hardwired” for spiritual belief, I do not see why a hard-wiring for atheistic belief cannot also be possible. Atheism has been around for a very long time, and maybe certain cultures that do not necessarily believe in a higher power or have a communal spiritual identity have become programmed for this thought. Possibly the concept of “Nature versus Nurture” can provide an explanation for disbelief in God and religion. Maybe the disbelief in religion stems purely from how you were raised: Did you go to church every Sunday, or did you stay at home and watch cartoons? Possibly synaptic pruning has resulted from the act of religious apathy—from religious connections in the brain becoming irrelevant to the “unpracticed” individual. It would be interesting to analyze patients who were once spiritual but who formed a disbelief in God after experiencing damage to a part of the brain other than the right parietal lobe or the posterior medial frontal cortex. In Clark Elliott’s book, “The Ghost in My Brain,” he speaks of how after his concussion, he lost his ability to have an inner dialogue with God—an ability he had since childhood. His prayers became empty words without meaning, going outwards to nobody but the empty void of

the room (Elliott, 2015, pgs. 107-109). Clark Elliott had severe visual and spatial reasoning issues as a result of his concussion. A lot of spatial reasoning comes from the parietal lobe, but what if his loss of dialogue with God came from the visual center? While this seems highly unlikely, it is intriguing to consider. It would also be intriguing to take an atheist who has experienced damage to a part of the brain and see if the patient then “found God.” It would possibly further the studies to take a patient from either a spiritual or atheistic background who has experienced a lesion to the right parietal lobe or posterior medial frontal cortex and see the impact that this lesion has had on the patient and his/her spiritual identity. Many patients who have suffered temporal lobe seizures have been reported to state that they have “seen God” during these episodes of unusual brain activity. Maybe there is a spot in this area of the brain that is directly related to our ability to conjure up ideology. My intentions are not to discredit any of the research that has already been done. However, finding the root of one problem or question often gives rise to a whole slew of answers to other intriguing questions. While this has worked many times in the past, I believe that in order to push the boundaries of science, abstract thought—which can be pursued by conducting studies which explore hypotheses similar to my proposed “Anti-God Spot” theory—is necessary. While the answer to why, how, and where our concept of religion comes from is still in the works, I present my own theory above. Though much attention has been paid to spirituality, I believe it would be worthwhile to look into its opposition: atheism. On many occasions, science has “blown our minds” and proved us wrong for making assumptions. I believe this is one of those occasions. I do believe we are hardwired to either believe or to not believe in religion—exactly what religion consists of, however, is still unknown. To better understand ourselves, it is necessary to better understand this vital contributor to the formation of our cultural identities—religion: that which makes us, us. REFERENCES

Barber, N. (2012, August 09). The ‘God spot’ revisited. Retrieved from Beauregard, M., Courtemanche, J., & Paquette, V. (2009). Brain activity in near-death experiencers during a meditative state. Resuscitation, 80(9), 1006-1010.

Fall 2017 | Volume 7 | Issue 1 | 47

Graykowski | HOT TOPICS Belief and the brain’s ‘God spot’. (2009). The Independent. Retrieved from Cristofori, I., Bulbulia, J., Shaver, J. H., Wilson, M., Krueger, F., & Grafman, J. (2016). Neural correlates of mystical experience. Neuropsychologia, 80, 212-220. Elliott, C. (2015). The ghost in my brain. Penguin Press. Holbrook, C., Izuma, K., Deblieck, C., Fessler, D. M., & Iacoboni, M. (2015). Neuromodulation of group prejudice and religious belief. Social cognitive and affective neuroscience, 11(3), 387-394. Hsu, J. (2009). Scientists locate ‘God spot’ in human brain. Retrieved from Johnstone, B., Bodling, A., Cohen, D., Christ, S. E., & Wegrzyn, A. (2012). Right parietal lobe-related “selflessness” as the neuropsychological basis of spiritual transcendence. International Journal for the Psychology of Religion, 22(4), 267-284. MailOnline, R. O. (2016, February 11). Feeling mystical? Blame your brain’s ‘God spot’: Study discovers why some people find spiritual connections with events and others don’t. Retrieved from http:// Nietzsche, F. W. (1998). Twilight of the idols. New York, New York: Oxford University Press Raushenbush, P. B. (2012, April 20). No ‘God spot’ in brain, spirituality linked to right parietal lobe. Retrieved from Sample, I. (2006, August 29). ‘God spot’ researchers see the light in MRI study. Retrieved from aug/30/medicalresearch.neuroscience Scientists seek religious experience – in subjects’ brains. (n.d.). Retrieved from V., & Sen, M. L. (1927). The Ramayana. Calcutta: Oriental Publ. Wellington, V. U. (2015, December 03). Exploring the brain’s role in mystical experiences. Retrieved from news/2015/12/exploring-the-brains-role-in-mystical-experiencess

48 | Issue 1 | Volume 7 | Fall 2017

The Undergraduate Journal of Neuroscience

Neurogenesis Editors Editor-in-Chief

Connor Hile

Publishing Editors

Tina Zhao

Kathy Dai

William Chen

Nidhila Masha

Managing Editors

Jackson Xu

Rohini Paul

Meghana Vagwala

Chris Lai

Kanav Chhabra

Design Editors

Gehua Tong

Riya Dange

Esther Liu

Michelle Dalson Fall 2017 | Volume 7 | Issue 1 | 49

Undergraduate Publications Board Duke University

50 | Issue 1 | Volume 7 | Fall 2017

The Undergraduate Journal of Neuroscience

Fall 2017 | Volume 7 | Issue 1 | 51

Neurogenesis is a member of the Duke Undergraduate Publications Board Š Copyright 2018

52 | Issue 1 | Volume 7 | Fall 2017

The Undergraduate Journal of Neuroscience

Neurogenesis Fall 2017  
Neurogenesis Fall 2017