Habib Cognition Lab

Research Program

We use fMRI, PET, and behavioral paradigms to understand memory, impulsivity, and the neuroscience of addiction — with a particular focus on pathological gambling.

Core Research Area

Pathological Gambling & the Near-Miss Effect

Our most widely-cited research concerns a deceptively simple question: why does almost winning at a slot machine make you want to play more? Slot machines are designed to produce near-miss outcomes — events that look like a win but pay out nothing. We have shown, using fMRI, that these near-miss outcomes activate the brain's reward circuitry in a manner that resembles a genuine win, even though no money changes hands.

This work has significant implications for understanding why gambling disorder is so difficult to treat. If the brain treats near-misses as partial rewards, then every "almost" on a slot machine is neurologically rewarding the behavior we want to extinguish.

Near-Miss Effects in the Brain

Using fMRI and behavioral paradigms, we have demonstrated that near-miss slot machine outcomes engage the ventral striatum — a region central to reward processing — in a way that genuine losses do not. Problem gamblers show an exaggerated response compared to recreational gamblers, suggesting a neurobiological mechanism underlying the disorder.

Gambling Disorder Neuroscience

We compare brain activation patterns between problem gamblers and matched controls during gambling tasks, monetary reward tasks, and resting state. These studies help identify which neural differences precede gambling disorder onset and which are consequences of extended gambling.

Behavioral Indicators of Gambling Risk

We have developed and validated behavioral tasks that distinguish problem gamblers from controls based on their responses to win and loss sequences. These tasks are increasingly used by other labs as screening instruments in clinical populations.

Research Area

Memory & Episodic Recall

A core strand of the lab's work concerns how the human brain encodes and retrieves episodic memories — memories of specific events tied to time and place. We trained under Endel Tulving, whose distinction between episodic and semantic memory remains one of the most influential frameworks in cognitive neuroscience.

Current work examines how episodic memory interacts with future planning, and how memory systems are disrupted in aging and addiction. We use fMRI to map the default mode network structures — particularly the hippocampus, medial prefrontal cortex, and retrosplenial cortex — that support autobiographical memory.

Episodic Memory Networks

We study how hippocampal and neocortical systems cooperate during memory encoding and retrieval, and how the strength of this coordination predicts subsequent recall. Functional connectivity analyses reveal which regions form a coherent "memory network" and how that network differs across individuals and age groups.

Memory & Future Thinking

The same brain systems that support episodic memory also support imagination and future planning — a phenomenon called "mental time travel." We explore how these systems are recruited during prospective cognition and whether impairments in episodic memory predict poor future planning in clinical populations.

Research Area

Impulsivity & Response Inhibition

Impulsivity — the tendency to act without adequate forethought — is a transdiagnostic risk factor for addiction, ADHD, and several personality disorders. Our lab studies both trait impulsivity (stable individual differences) and state impulsivity (situational fluctuations in control), and the neural systems that regulate each.

Neural Correlates of Inhibitory Control

Using stop-signal and go/no-go paradigms in the fMRI scanner, we identify the prefrontal and subcortical circuits that support the ability to cancel a planned action. Problem gamblers consistently show reduced activation in inferior frontal gyrus regions during successful inhibition trials.

Impulsivity as a Vulnerability Marker

We study whether elevated impulsivity predates gambling disorder onset (vulnerability marker) or emerges as a consequence of extended gambling (scar hypothesis). Longitudinal designs and at-risk population studies are used to address this question.

Research Area

Decision-Making Under Uncertainty

How do people evaluate probabilistic outcomes? And how do those evaluations go wrong in individuals who are prone to gambling disorder? We use classic decision-making paradigms — the Iowa Gambling Task, probability discounting tasks, and novel slot-machine simulations — alongside neuroimaging to map the brain systems involved.

Probability and Reward Discounting

We examine how people discount the value of uncertain vs. certain rewards, and how this discounting rate differs across clinical populations. Problem gamblers typically show steeper discounting of probabilistic rewards — meaning they demand more expected value to justify choosing an uncertain option.

Framing Effects and Loss Aversion

Gambling environments are designed to highlight potential gains while obscuring losses. We study how framing manipulations influence choice behavior and neural activity, and whether problem gamblers are differentially susceptible to these effects.

Methods

How We Work

fMRI

Functional magnetic resonance imaging for whole-brain mapping of task-related activation and functional connectivity.

PET

Positron emission tomography for examining dopamine receptor availability and metabolic activity in reward circuits.

Behavioral Paradigms

Computerized gambling tasks, stop-signal paradigms, probability discounting, and ecological momentary assessment.

EMA / Mobile Methods

Experience sampling and ecological momentary assessment via smartphones to capture behavior in the real world — central to the OpenEMA project.

Clinical Populations

Studies with diagnosed problem gamblers, recreational gamblers, older adults, and healthy controls recruited from the local community.

Statistical Modeling

GLM-based fMRI analysis, connectivity modeling, mixed-effects models, and machine learning classification of imaging data using R, Python, and SPM.