Research Interests

The goal of my research is to understand how cognitive functions are mediated by interactions of cortical and subcortical circuits. The primary focus is to elucidate the roles of basal forebrain neuronal populations in top-down attention. To gain a broad perspective on this issue, I investigate how cortical activity is dynamically shaped by basal forebrain inputs, how basal forebrain neuronal activity may be controlled by cortical and other inputs, and how this interaction may serve important behavioral functions such as attention. 

Non-cholinergic basal forebrain neurons transiently enhance cortical activity

The mammalian basal forebrain (BF) is one of the largest neuromodulatory systems and plays critical roles in controlling cortical activity and plasticity, arousal, as well as top-down attention. While BF corticopetal projections are traditionally thought to consist exclusively of cholinergic (ACh) neurons, recent anatomical studies have established that the majority of the BF corticopetal projections are in fact non-ACh neurons, consisting mostly of GABAergic neurons and a smaller subset of glutamatergic neurons. During my dissertation research, I discovered a novel mechanism by which these non-ACh BF neurons may transiently enhance prefrontal cortex (PFC) activity, highlighting for the first time the important physiological functions of this poorly understood system [Pubmed]. By simultaneously recording PFC LFPs and the activity of up to 50 BF neurons throughout the natural wake-sleep cycle of rats, I identified a homogeneous population of non-ACh BF neurons which, unlike ACh neurons, do not change their average firing rate across wake-sleep states. I demonstrated that these non-ACh neurons engage in spontaneous ensemble bursting events lasting on average 150 msec. Crucially, I discovered that BF ensemble bursting events are phase-locked with PFC LFP oscillations and tightly coupled with transient (~200 msec) increases in PFC gamma oscillation power. This BF-mediated fast cortical modulation is remarkably consistent with the likely actions of BF corticopetal GABAergic neurons, which are known to preferentially innervate intra-cortical interneurons and may transiently enhance cortical activity via disinhibition.

A novel candidate mechanism for attention


This research pointed to the intriguing hypothesis that ensemble bursting of non-ACh BF neurons may serve as a novel neural mechanism for top-down attention. This hypothesis was supported by the observation that BF-mediated fast modulation of cortical activity closely resembles how top-down attention enhances cortical activity, as indexed by enhanced event-related potential (ERP) and gamma oscillation. To directly test this hypothesis, I continued my postdoctoral training with Dr. Nicolelis to investigate whether BF ensemble bursting can be activated while rats attend to motivationally salient sensory stimuli. I demonstrated, for the first time, that ensemble bursting of non-ACh BF neurons encodes the motivational salience of attended stimuli and signals whether
and when rats pay attention to sensory cues [Pubmed]. Using a Go/Nogo task, I showed that both the innate salience conveyed by primary reward (sucrose) and punishment (quinine), as well as the learned salience of environmental cues that predict rewards or punishments, are encoded by the same non-ACh BF neurons with similar bursting responses. This encoding of motivational salience is present irrespective of the cue’s sensory modality, the associated motor response or the hedonic valence of the expected outcome (reward or punishment). In a separate experiment, I further demonstrated that the presence of BF ensemble bursting predicts successful detection of near-threshold tones on a trial-by-trial basis and likely plays a causal role in improving behavioral performance. Together, these results indicate that ensemble bursting of non-ACh BF neurons may translate the encoding of motivational salience of attended stimuli into transient enhancement of cortical activity, and thus is capable of mediating the influences of attention on enhancing cortical processing and on improving behavioral performance.


Neuromodulatory influences on forebrain dynamics
During my PhD, my research focused on neuromodulatory systems, which play important roles in normal cognitive functions and various neuropsychiatric disorders, to elucidate how neuromodulatory inputs dynamically shape the activity of forebrain networks in different behavioral states. 

I investigated how coordinated changes of neuromodulatory activities in different wake-sleep states modulate forebrain local field potential (LFP) oscillatory patterns in rats, and developed a novel state-space method that reveals stable regimes of global forebrain dynamics [Pubmed] (see this book chapter for details). This novel method has become an important tool to investigate state-dependent neuronal processes (see examples [1][2][3][4][5]).



In collaboration with Dr. Rui Costa, I also investigated how dopamine (DA) modulates corticostriatal neuronal ensemble dynamics using a pharmaco­genetic approach to acutely deplete DA in DAT-KO mice. We showed that acute DA depletion results in profound alterations in the coordination of cortico-striatal ensembles, manifested as dramatic changes in LFP oscillations that mimic those observed in Parkinson’s disease [Pubmed].



Visual attention in humans

I also studied visual attention and oculomotor control in human subjects, focusing on an attentional phenomenon: inhibition of return (IOR). I demonstrated that IOR originates from a delay in perceptual processing but not motor programming [Pubmed]. I further discovered that the perceptual delay in IOR can be demonstrated in a perceptual temporal order judgment task, but only when temporal order judgment is reported by a speeded saccadic eye movement [Pubmed] .