Skip to main content

Frontiers in Neuroscience, January 27th, 2011: Emory’s Dr. Elaine Walker on “Neurodevelopmental Mechanisms in the Emergence of Psychosis”

Psychotic disorders like schizophrenia affect about 5% of people and often result in life-long disability. Identifying at-risk individuals and predicting disease onset are crucial, and present a challenge to the development of preventative treatments. Understanding the biological mechanisms underlying psychosis are also extremely important in identifying risk factors and designing treatments.

Because psychotic disorders are so disabling and usually irreversible, research interests in this field have shifted toward prevention and early intervention. Subtle pre-clinical deficits in psychosocial and neurocognitive functioning have been reported for many years and are now being extensively studied. Elucidating this pre-illness state, known as the “prodromal” period, is one area of research for Dr. Elaine Walker, the Samuel Candler Dobbs Professor of Psychology and Neuroscience at Emory University. Dr. Walker spoke at the Frontiers in Neuroscience Seminar Series on Friday, January 27th, about her research on prodromal psychosis as a part of the NAPLS (North American Prodrome Longitudinal Study).

Dr. Walker began her talk with a discussion of the prodromal period and risk factors for developing a psychotic disorder. Simply defined, the prodrome is a “period of functional decline that precedes the clinical onset of psychosis.” It typically occurs from adolescence to young adulthood, and is characterized by “attenuated positive symptoms,” such as perceptual distortions, paranoia, and disorganized communication. While prodromal individuals do not always develop full-blown psychosis (although many actually do), they are at substantial risk. But as Dr. Walker pointed out, the etiology of psychosis is always a complex interaction between an individual’s genetics and her environment.

Risk factors for psychosis typically begin before one even takes her first breath. Prenatal stressors, such as maternal stress or viral infection, play a significant role in fetal brain development, likely through epigenetic mechanisms (epigenetics are heritable, functionally-relevant modifications to the genome that do not involve changes in DNA sequence). Along with gene mutations, these factors can coalesce in vulnerability of certain neural circuits. One likely candidate, known as the frontostriatal circuit, connects areas of the prefrontal cortex (involved in executive function) to subcortical structures involved in emotional processing. Aberrant firing or connectivity of neurons within this circuit is believed to underlie some neuropsychiatric disorders, such as depression, obsessive-compulsive disorder, and substance abuse (Bonelli and Cummings, 2007).

During puberty and into young adulthood, the developing brain is guided by changes in sex hormones like testosterone and estrogen, as well as by stress hormones like cortisol. Sex hormones play a pivotal role as regulators of gene expression, meaning they can activate or repress the production of certain proteins. Changes in brain protein levels can affect processes like neurotransmitter function and neuronal organization, which eventually manifest through behavior. That is essentially the rationale behind your parents calling you “hormonal” as a kid. Recent studies have also shown that adolescents are particularly vulnerable to stress. Throughout one’s life, the peak stress response is found during adolescence, which is thought to be partly a result of peaking stress hormone receptor densities (like the glucocorticoid receptor). What all this means is basically that adolescence is a time of heightened sensitivity to neurological processes moderated by hormones. So if every little social stressor (like asking someone to prom) seemed like a huge deal to you back then, that’s probably because, to your developing brain, it was.

The links between hormones, stress, and adolescent vulnerability to psychosis might leave you feeling a bit guilty about the things you did or said in high school and how they might have affected your classmate’s brain development. But for Dr. Walker and the NAPLS research team, they’re a promising starting point from which to identify mechanisms and risk-factors within the prodrome. The first part of NAPLS, completed in 2007, tracked about 300 prodromal individuals over 30 months and found that 35% became clinically psychotic. Using those conversion rates, coupled with patient data on their genetics, positive symptoms, social impairments, and substance abuse history, researchers were able to develop a predictive algorithm that was far more accurate than previous models (Cannon et al., 2008). With the new model, a baseline assessment can predict whether a patient will or will not convert to psychosis over a 2.5 yr period with about 80% accuracy. This is a big improvement over previous models, which could only predict with 35% accuracy. As a result, physicians and researchers can now be more selective about which patients to admit or not admit into prevention programs or other research studies on the prodrome.

The second part of NAPLS is at its halfway point, but Dr. Walker was generous enough to share some exciting preliminary findings with the audience. Phase II is more focused on the biological changes occurring during the prodrome. The hope is that by collecting neuroimaging (like fMRI, EEG, and DTI), genetic, and hormonal data every six months, researchers will be able to paint a much clearer and more comprehensive picture of the biological changes that lead to psychosis. These will then hopefully serve as biomarkers and targets for designing future therapies that could prevent or limit conversion.

The story from NAPLS II begins with stress. Compared to healthy, age-matched controls, prodromal subjects report higher daily stress scores and an increased number of stressful life events. In controls, self-reported stress declines with age; but for the prodromal group, life is stressful and stays that way. Their heightened stress is also reflected in higher levels of cortisol, a stress hormone that acts on the same glucorticoid receptor mentioned earlier. These data illustrate that non-prodromal adolescents become better at managing their stress with age, whereas prodromal individuals continue struggling to cope with the complex, challenging world around them, and that cortisol may be a key player in this story.
Increases in psychotic symptoms correlate with reductions in temporal lobe volume.

From Takahashi et al., 2009.
The story continues with differences in brain structure. Your brain, like everyone else’s, increases in total volume up until about age 12 (grey matter accounts for this and is mainly composed of neurons), after which its volume slowly decreases, a normal process called synaptic pruning. This form of neural plasticity is believed to be a good thing because it lets the brain “trim away the fat,” so to speak, in order to refine connections needed for ongoing learning and memory. But over the years, multiple lines of evidence illustrate that the pruning process is defective in schizophrenic brains, and that in fact “hyperpruning” might be decreasing their grey matter volumes (Faludi and Mirnics, 2011). Finding decreased grey matter volumes within the prodrome could therefore be an early indicator of this aberrant process at work. Indeed, NAPLS found that prodromal subjects exhibited a greater decrease in grey matter density in the prefrontal cortex and temporal lobe over the course of the study as compared to controls. Interestingly, those subjects who eventually converted to psychosis showed an even greater rate of grey matter volume decline. Tying the whole story together and again showing how stress is important in this process, higher levels of cortisol were strongly correlated with decreased brain volume. Previous work found a negative correlation between cortisol levels and neuronal growth factor levels in the prefrontal cortex, lending credence to the NAPLS data which suggest that higher cortisol may be mechanistically related to decreased brain volume (Issa et al., 2010). But as Dr. Walker suggested, more research is needed to tease apart the cortisol-brain structure relationship, such as looking for variations in genes involved in cortisol sensitivity.

The preliminary findings of NAPLS show promise for identifying biomarkers and mechanisms of conversion from prodrome to psychosis. However, inherent in this conversation are ethical questions about diagnosis, intervention, and treatment of prodromal individuals, some of whom will never experience a psychotic episode. Previous posts by our scholarly contributors on The Neuroethics Blog have addressed many of these questions already (see 10/21/11; 12/5/11; 12/6/11; 12/8/11; 12/9/11), but I’ll briefly summarize them here.

As Dr. Walker described in her talk, NAPLS’ new algorithm can predict conversion from prodrome to psychosis with much better accuracy than before (80% versus the previous 35%). This sounds promising for clinicians wanting to intervene, but keep in mind that their algorithm has yet to be empirically tested. Additionally, “prodrome risk syndrome” doesn’t even exist on the books (the books in this case being The Diagnostic and Statistical Manual of Mental Disorders or DSM), meaning that clinicians have no officially recognized criteria for making such a diagnosis. Even if they did, they would likely be concerned about the inevitable false positives. If antipsychotic drugs begin to be used in prodromal patients, it could negatively impact the misdiagnosed in a big way: not only in terms of side effects, but also how that individual is treated by her friends, family, and society. Misdiagnosis could also increase anxieties or paranoia that may themselves fuel a psychotic episode. While early detection is important in this and any other disease, we must remain extremely cautious about formally diagnosing an individual as prodromal until our tools to reliably discern a prodromal period are validated and effective treatments become available.

–Jordan Kohn
Emory Neuroscience Graduate Student

Want to cite this post?

Kohn, J. (2012). Frontiers in Neuroscience, January 27th, 2011: Emory’s Dr. Elaine Walker on “Neurodevelopmental Mechanisms in the Emergence of Psychosis”. The Neuroethics Blog. Retrieved on
, from


 1. Bonelli, R. and Cummings, J. (2007) Frontal-subcortical circuitry and behavior.  Dialogues Clin. Neurosci. 9(2): 141-151.

2. Cannon, T., Cadenhead, K., Cornblatt, B., Woods, S., Addington, J., Walker, E., Seidman, L., Perkins, D., Tsuang, M., McGlashan, T., Heinssen, R. (2008) Prediction of psychosis in youth at high clinical risk. Arch Gen Psychiatry. 65(1): 28-37.

3. Faludi, G. and Mirnics, K. (2011) Synaptic changes in the brain of subjects with schizophrenia. Int J Dev Neurosci. 29(3): 305-9.

4. Issa, G., Wilson, C., Terry, A., Pillai, A. (2010) An inverse relationship between cortisol and BDNF levels in schizophrenia: data from human postmortem and animal studies. Neurobiol Dis. 39(3): 327-33.

5. Takahasi, T., Wood, S., Yung, A., Soulsby, B., McGorry, P., Suzuki, M., Kawasaki, Y., Phillips, L., Velakoulis, D., Pantelis, C. (2009) Progressive gray matter reduction of the superior temporal gyrus during transition to psychosis. Arch Gen Psychiatry. 66(4): 366-376.


Emory Neuroethics on Facebook