Schizophrenia Research Forum - A Catalyst for Creative Thinking

Monkey Model of Schizophrenia Debuts

September 26, 2013. Monkeys under the influence of ketamine show brain signals reminiscent of schizophrenia, reports a study published online August 19 in the Proceedings of the National Academy of Sciences. Led by Thomas Albright at the Salk Institute for Biological Studies in La Jolla, California, the study developed non-invasive electroencephalography (EEG) that identified comparable brain signals from humans and monkeys. Two signals linked to impairments in cognitive and sensory processing found in schizophrenia—mismatch negativity (MMN) and P3a—were identified in macaque monkeys, and were reduced by ketamine, a blocker of the NMDA type of glutamate receptor that is thought to mimic the underactive glutamate systems in schizophrenia. A similar reduction is seen in people with schizophrenia, which suggests that monkeys can model aspects of the disorder.

The research comes at a time when drug development for psychiatric disorders has stalled, which some have blamed on a shortage of animal models (Hyman, 2012). With the rise of genomic research, most animal models these days include rodents carrying disease-related genetic glitches, but linking their various behaviors to psychiatric symptoms can be tenuous. The approach in the new study combines the complexity of the nonhuman primate brain with a pharmacological manipulation to simulate schizophrenia by using EEG signals as outcomes. Based on evidence of underactive NMDA receptor signaling in schizophrenia (see hypothesis by Bita Moghaddam) as well as the schizophrenia-like state the drug causes, ketamine has been used to mimic aspects of the disorder in healthy humans (see SRF related news story).

But tracking ketamine’s effects in the brains of nonhuman primates is complicated by the fact that their brains are shaped differently from human brains. This means that signals arising from, say, auditory cortex, could be located in a different place than in humans or combined with signals from another region. An invasive type of EEG using electrodes resting on the surface of the brain has found an MMN-equivalent in monkeys (Javitt et al., 1992), but the new work tests non-invasive EEG electrodes placed on the scalp instead.

First author Ricardo Gil-da-Costa and colleagues studied the EEG responses of five humans and two macaque monkeys during an auditory processing task that taps into attention. The humans or monkeys listened to a series of identical tones that were occasionally interrupted by an “oddball” tone louder or quieter than the rest. In human brains, the oddball stimulus evokes a larger than usual voltage signal compared to the standard stimuli. The difference between the deviant and the standard stimuli is the MMN, which indicates the brain’s detection of the deviant stimulus. The P3a signal appears only during an oddball stimulus, after the MMN signals, and is thought to reflect reorienting one’s attention to the oddball tone.

The researchers reported similar MMNs between humans and monkeys. In the humans, a 64-electrode array measured a standard-looking MMN at 56-188 milliseconds after stimulus onset, peaking at -1.83 μV. In the monkeys, a 22-electrode array picked up an MMN-like signal 48-120 milliseconds after stimulus onset, peaking at -1.62 μV. The P3a signal that followed was also similar across species. The high density of scalp electrodes allowed the researchers to infer where the MMN and P3a signals were coming from, and these analyses pointed to the superior temporal gyrus and homologous frontal regions in both species.

A sub-anesthetic dose of ketamine given to the monkeys reduced their MMN and P3a signals compared to saline infusions. The signals returned to normal after five hours, when ketamine had dropped to low levels.

The researchers propose that this model could be used to screen drugs for schizophrenia and get a better picture of their action in the brain. And as MMN and P3a are widely studied, monkey EEG may well prove useful in testing treatments for other brain disorders.—Michele Solis.

Gil-da-Costa R, Stoner GR, Fung R, Albright TD. Nonhuman primate model of schizophrenia using a noninvasive EEG method. Proc Natl Acad Sci U S A. 2013 Sep 17;110(38):15425-3. Abstract

Comments on News and Primary Papers
Comment by:  Dan Javitt, SRF Advisor
Submitted 30 September 2013
Posted 30 September 2013

This is an important paper that confirms the role of NMDA receptors in the generation of mismatch negativity (MMN) and, by extension, the potential role of NMDA receptors in the pathophysiology of schizophrenia. As in prior studies with MMN in monkeys, the latency of MMN in monkeys appears to obey the 2/3 rule, which allows cross-species scaling of sensory ERP.

Since our initial report of PCP effects on MMN in monkeys in the late 1990s (Javitt et al., 1996) and subsequent reports on ketamine effects on MMN in humans shortly thereafter (Umbricht et al., 2000), the findings have been extensively replicated in humans. This, however, is the first replication in monkeys and the first to use primarily surface electrodes, and so opens the door to more widespread investigation. In particular, studies in humans are limited to acute administration. However, acute administration of NMDA receptor antagonists only captures a portion of the syndrome. In monkeys, chronic administration of NMDA receptor antagonists is possible and is associated with progressive development of negative-like symptoms (Linn et al., 2007).

As in our earlier report, ketamine treatment reduced MMN-related activity but did not affect responses to rapidly presented, repetitive, standard stimuli, reproducing the pattern of deficit observed in schizophrenia. In this initial study, no other classes of compounds were tested. However, establishment of this model permits testing of a wide range of compounds, including pharmacological probes for other classes of glutamate receptors, or from other transmitter systems (e.g., dopaminergic, cholinergic, GABAergic) that have also been implicated in schizophrenia. Especially during subchronic treatment, the ability to reverse MMN deficits may be an important screening model for potentially psychotherapeutic compounds in schizophrenia and other NMDA receptor-related disorders.

Another important issue that can be addressed using monkey models is the nature and identity of the "frontal generator." It is clear from both MEG and intracranial recording studies that primary generators for MMN are in auditory regions of the superior temporal cortex. Additional, frontal generators are also sometimes reported based upon source analysis. However, unless constrained through physiological means, source localizations can easily produce spurious results. This study uses LORETTA, a common source-localization approach, and identifies sources in frontal and anterior cingulate cortices (ACC), as well as in auditory cortex.

Generators in ACC are unlikely in humans, because they should be detectable by MEG. Nevertheless, an obvious follow-up of this study is to implant intracranial electrodes in those regions detected using LORETTA. If local generators are found, it will give renewed understanding about the relationship between auditory and frontal interaction during MMN generation. If local generators are not found, it will permit refinement of the LORETTA approach and reduction in "false positive" localizations that may, of themselves, complicate understanding of disorders such as schizophrenia.

Finally, a goal of biomarker research is the development of measures that can be implemented in relatively simple species, such as rodents. For complex disorders such as schizophrenia, however, it could be that more complex, primate models are required. As opposed to most primate paradigms, MMN can be obtained even in untrained animals, permitting a development path from rodents through primates and into humans.


Javitt DC, Steinschneider M, Schroeder CE, Arezzo JC. Role of cortical N-methyl-D-aspartate receptors in auditory sensory memory and mismatch negativity generation: implications for schizophrenia. Proc Natl Acad Sci U S A. 1996;93(21):11962-7. Abstract

Umbricht D, Schmid L, Koller R, Vollenweider FX, Hell D, Javitt DC. Ketamine-induced deficits in auditory and visual context-dependent processing in healthy volunteers: implications for models of cognitive deficits in schizophrenia. Arch Gen Psychiatry. 2000;57(12):1139-47. Abstract

Linn GS, O'Keeffe RT, Lifshitz K, Schroeder C, Javitt DC. Behavioral effects of orally administered glycine in socially housed monkeys chronically treated with phencyclidine. Psychopharmacology (Berl). 2007;192(1):27-38. Abstract

Javitt DC, Spencer KM, Thaker GK, Winterer G, Hajos M. Neurophysiological biomarkers for drug development in schizophrenia. Nature reviews. 2008;7(1):68-83. Abstract

View all comments by Dan Javitt