Find A PhysicianHome  |  Library  |  myDownstate  |  Newsroom  |  A-Z Guide  |  E-mail  |  Contact Us  |  Directions
curve gif


« Previous | Next »


Janina Ferbinteanu, PhD

Research Assistant Professor 

Dept of Physiology and Pharmacology

Tel: (718) 270-1796


Research Interests:

Learning and memory, memory systems (and their interactions), neurophysiology, and memory-guided behavior.

Background And Overview

Through my training, I have developed expertise in behavioral methods, stereotaxic surgery, intracerebral drug infusion, high-density electrophysiological recording, and digital signal processing. My doctoral dissertation demonstrated that anatomically distinct hippocampal circuits play specific roles in spatial navigation and context encoding, a theme that has re-surfaced in my more recent work. Subsequently, I used high-density unit recording methods to investigate the activity of hippocampal neural populations in rats performing a hippocampal-dependent spatial navigation task which is considered a model of declarative memory in humans (Ferbinteanu and Shapiro, 2003). This work showed that as the animals utilize memory to perform the task, activity of CA1 hippocampal cells encodes the current location of the animal in conjunction with the origin or destination of journeys. Further investigation revealed that this type of activity, referred to as hippocampal journey-dependent activity (HJDA), is modulated when the animal changes from spatial navigation, to cue response, a different type of behavioral strategy that involves procedural memory (Ferbinteanu et al., 2011). Because of its temporal component (encoding of past and future), HJDA could be part of the neurophysiological substrate of the temporally organized episodic memory. In order to broaden my knowledge of large-scale analysis methods suitable to apply to simultaneous recording of the firing of a large number of neurons across multiple brain regions, I joined Dr. Joseph Francis’ lab at SUNY Downstate in 2011. Soon after my move to Downstate, I received independent NIH funding and in 2012 moved on to establish my own lab.   

One research line I pursue aims to clarify the neural mechanism that produces HJDA. The central hypothesis of this work is that fully functional HJDA in hippocampal CA1 is generated by a combination of input from the entorhinal cortex, the source of cortical inflow into the hippocampus, with input from upstream (CA3) hippocampal neurons. The first study in this line showed that relative to entorhinal cortex lesions, damage to the CA3 hippocampal field causes more severe and permanent impairment in the spatial navigation task I previously used to record HJDA (O’Reilly et al., 2014). The same experiment also yielded two unexpected results. The behavioral paradigm of this experiment was initially developed to study the memory modulation of HJDA (Ferbinteanu et al., 2011) and utilized concurrent training in a spatial navigation task, which models declarative memory, with training in a cue-response task, which models procedural memory. The paradigm very clearly isolates the memory strategy the animal uses at a particular point in order to behave successfully. Intriguingly, in contrast to hippocampal lesion effects in rats trained on cue-response task alone, the CA3 and medial entorhinal cortex lesions affected performance in the cue-response task when it was learned concurrently with the spatial task. Because cue-response is thought to be guided by procedural memory and involve only the lateral dorsal striatum, these data suggested that the hippocampus and dorsal striatum memory systems interact in more complex ways than any previous research suggested. Second, slice electrophysiological recording in lesioned, sham trained, and sham untrained animals revealed a previously unknown heterosynaptic interaction between the two cortical inputs to CA1, the CA3-CA1 and the medial entorhinal cortex-CA1 pathways. CA3-CA1 input in animals with entorhinal cortex lesions was greatly enhanced, while the converse was not the case. Further, we found a similar although smaller enhancement in the CA3-CA1 pathway of normal animals that had undergone extensive behavioral training. Thus, heterosynaptic interaction between the two cortical inputs to CA1 may support normal learning processes, may participate in post-lesion compensatory processes, and may be critical to functional HJDA.


Theoretical framework

The general aim of my work is to understand the brain processes that allow encoding, storage, and utilization of information in order to generate organized, goal-directed behavior. Theoretically, my work is framed by the memory systems model, which postulates that distinct types of mnemonic representations, with different properties and relevance for behavior, are formed through activity in separate brain networks which share some, but not all, of their constituent areas. Currently, it is well established that the hippocampus is involved in supporting declarative memory, dorsal striatum in habits, and amygdala in emotional memory (see diagram below, from Squire, 2004).

I am interested in how memory traces guide behavior. Specifically, I am investigating in the rat animal model 1) the neurophysiological substrate of declarative memory, which in humans is the type of memory that can be consciously recollected and expressed through language, and encompasses memory for facts (semantic) and personally experienced events (episodic); and 2) the neural correlates of interactions between different types of memories. Any given experience results in the formation of distinct memory traces, with different property and dynamics, in separate neural circuits. Memory-guided behavior is influenced by all these representations, albeit to different extents, and the normal balance is affected in pathological situations. Despite intense focus on hippocampal activity, we currently do not know how the hippocampal representation guides behavior and how it interacts with other mnemonic representations. My research rigorously and systematically combines neurophysiological recording and behavioral design. Neuroimaging studies have conclusively shown that activity in functionally distinct brain areas changes with cognitive demand. Thus, to detect the neural activity underlying a brain function such as memory, neurophysiological recording in animal models ideally should be coupled with experimental manipulations that isolate and identify cognitive processes.


Research in my laboratory currently follows three main themes:

1. Hippocampal-striatal interactions. Following the results described above, which revealed an unexpected contribution of the hippocampus to what was thought to be a striatum-mediated behavior, I am currently investigating the contributions of the hippocampus, medial dorsal striatum, and lateral dorsal striatum to the performance of spatial navigation and cue-response tasks when introduced separately vs. concurrently to the rats. The design of the behavioral task isolates and identifies the cognitive strategies the animal uses to behave successfully (i.e., spatial navigation cannot be solved through cue response, and vice-versa); and neither task can be solved through other approaches such as path integration (dead reckoning). The results of this study strongly suggest that the hippocampal and striatal contributions to behavior are not set, but depend on past experience. As suggested by ours and others’ previous work, when animals learn only one task at a time, there is a double dissociation in the neurobiological basis of cue response and spatial navigation:  cue response involves the lateral dorsal striatum and is independent of hippocampal activity, and conversely, spatial navigation involves hippocampal activity and is independent of lateral dorsal striatum. However, when the two tasks are learned concurrently, cue-response additionally involves hippocampal activity and spatial navigation additionally involves lateral dorsal striatum activity. Medial dorsal striatum has an important contribution to both spatial navigation and cue response regardless of the method of training; possibly reflecting its proposed role in behavioral flexibility, which both tasks require. This finding is totally novel and has important implications for our understanding of how memory is stored in the brain and how it is utilized in order to guide behavior. Additionally, the current results clarify the findings of our previous recording study (Ferbinteanu et al., 2011), which show that numbers and proportions of HJDA fields do not change when animals shift between spatial navigation and cue-response (as it turns out, the hippocampal activity is necessary in both cases) and entail a reconsideration of the previous statement that HJDA is present even if the hippocampus is not necessary to support behavior. This paradigm is now being used to further investigate the neurophysiological basis of the interaction between hippocampal and dorsal striatum memory systems.

2. Cognitive and physiological mechanisms of HJDA. I continue pursuing the HJDA mechanisms. Currently, it is difficult to selectively manipulate activity in distinct hippocampal pathways. Thus, I plan to combine in vivo experiments with in silico modeling. The central hypothesis of this research is that spatial navigation involves HJDA in the hippocampal CA1, which results from combining the context-sensitive input from CA3 region, with the journey-selective input from the entorhinal cortex. A project that will address this question empirically is recording in the CA1 of animals with CA3 or entorhinal cortex temporary or permanent lesions. A second, complementary approach that can provide insight into the link between neural activity, brain function and behavior is synthetic neural modeling: using large-scale neural simulations that implement anatomical and physiological findings in neural models to query how structure and activity can generate brain functions and behavior. Synthetic neural models can be embodied in virtual or robotic phenotypes and engaged in behavioral tasks homologous to those used in animal experiments. By virtue of the homology between the anatomy, dynamics, and behavior of synthetic neural models and animals, this approach can make predictions that in turn can then be tested in biological experiments. Remarkably, a synthetic neural model of the large-scale connectivity in hippocampus and its surrounding regions has demonstrated the emergence of HJDA in an autonomous robot engaged in a plus maze task. I am very pleased to be collaborating with Drs. Jason Fleischer (NSI) and Jeff Krichmar (UCI), the authors of this work. Together, we envision pursuing a series of experiments to address how HJDA is generated in the brain and how it may guide behavior. Together we have submitted a grant application currently pending review with NSF.

3. Neurophysiological substrate of vascular cognitive impairment. I am actively interested in applying my experimental methods to models of human disease. Through interaction with Dr. Frank Barone’s group at Downstate, I have become interested in studying the cognitive and neurophysiological aspects of hypertensive-hypoperfusion vascular cognitive impairment. Temporary occlusion of middle cerebral artery, a significant cause of ischemic stroke in humans and rodent models, produced deficits in rats performing passive avoidance and active place avoidance learning tasks, both of which involve hippocampal functions. Intriguingly, Dr. Barone’s team found that these learning deficits were not associated with hippocampal neural loss. Administration of thrombopoietin, a hematopoietic growth factor responsible for platelet production, reduced the memory deficits. The neurophysiological substrate of these modifications is currently completely unknown. I am working with Dr. Barone’s group to develop a battery of cognitive tests that would characterize the cognitive deficits of rats with experimental focal stroke. Devising relevant behavioral tasks adequate for electrophysiological recording is the first step necessary for future projects that would investigate neurophysiological modifications following stroke and changes introduced by treatments that improve ischemic brain injury and improve functional outcome. 

Education and training:
  • 1994 B.Sc. Specialist in Neuroscience. University of Toronto, Scarborough College, Canada
  • 1995 M.A. Psychology. University of Toronto, Scarborough College, Canada
  • 2000 PhD  Psychology, with specialization in Neuroscience. University of Toronto, Canada
  • 2000-2005 Postdoctoral Fellow, Mt. Sinai School of Medicine, New York, NY

John M, Ikuta T, Ferbinteanu J. (2016) Graph analysis of structural brain networks in Alzheimer's disease: beyond small world properties  Brain Struct Funct. 2016 Jun 29. [Epub ahead of print]


Ferbinteanu J (2016) Contributions of hippocampus and striatum to memory-guided behavior depend on past experience. J Neurosci. 2016 Jun 15;36(24):6459-70. doi: 10.1523/JNEUROSCI.0840-16.2016.


M. John, T. Lencz, J. Ferbinteanu, J.A. Gallego, and D. Robinson (2015) – Applications of temporal kernel canonical correlation analysis in adherence studies. Stat. Methods in Med. Res., in press [PDF]


K. C. O’Reilly, J.M. Alarcon, and J. Ferbinteanu (2014) - Relative Contributions of CA3 and Medial Entorhinal Cortex to Memory in Rats. Front Behav Neurosci, 8:292 [PDF]


J. Ferbinteanu, P. Shirvalkar, and M.L. Shapiro (2011)– Memory modulates journey-dependent coding in the rat hippocampus. J. Neurosci.  31(25): 9135-46. [PDF]


M.L. Shapiro, P.J. Kennedy, and J. Ferbinteanu (2006) -Representing episodes in the mammalian brain. Curr. Opin. Neurobiol. 16(6): 701-709. Review.


J. Ferbinteanu, P.J. Kennedy, and M.L. Shapiro (2006) – Episodic memory – from brain to mind. Hippocampus, 16(9): 691-703. Review. [PDF]


M. L. Shapiro and J. Ferbinteanu (2006) – Relative spike timing in pairs of hippocampal neurons distinguishes the beginning and end of journeys. Proc Natl Acad Sci USA, 103(11): 4287-4292 [PDF


J. Ferbinteanu and M. L. Shapiro (2003) – Prospective and retrospective memory coding in the hippocampus. Neuron, Dec 18; 40(6): 1227-1239 [PDF]


J. Ferbinteanu, C. Ray, and R. J. McDonald (2003) – Both dorsal and ventral hippocampus contribute to spatial learning in Long-Evans rats. Neurosci.Lett. Jul 17; 345(2): 131-135 [PDF]


J. Ferbinteanu and R. J. McDonald (2001) – Dorsal/ventral hippocampus, fornix, and conditioned place preference. Hippocampus, 11(2): 187-200. [PDF]


J. Ferbinteanu and R. J. McDonald (2000) Dorsal and ventral hippocampus: Same or different? Psychobiol 28(3): 314-324 [PDF]


J. Ferbinteanu, R. M. D. Holsinger, and R. J. McDonald (1999) - Lesions of the medial or lateral perforant path have different effects on hippocampal contributions to place learning and fear conditioning to context. Behav. Brain Res., 101(1): 65-84 [PDF]


Michalakis M., Holsinger D., Ikeda-Douglas C., Cammissuli S., Ferbinteanu J., DeSouza C., DeSouza S., Fecteau J., Racine R. J., and Milgram N. W. (1998) – Development of spontaneous seizures over extended electrical kindling. I. Electrographic, behavioral, and transfer kindling correlates. Brain Res. 793(1-2): 197-211


J. Ferbinteanu and N. W. Milgram (1997) - Origin of the lateral perforant path; letter to the editor. Hippocampus , 350


Milgram N.W., Michael M., Cammisuli S., Head E., Ferbinteanu J., Reid C., Murphy P.M., and Racine R.(1995)- Development of spontaneous seizures over extended electrical kindling. II. Persistence of dentate inhibitory suppression. Brain Res. 670, 112-120