Professor Redmond O'Connell

How do we make reliable decisions about noisy sensory inputs? This fundamental question has drawn intense interest from mathematicians, experimental psychologists, cognitive neurophysiologists and engineers alike, culminating in the emergence of a rich theoretical framework for understanding perceptual decision making. At the center of this framework lies the notion of a "decision variable," which integrates sensory evidence over time and triggers the appropriate action upon accruing a criterion amount. Monkey electrophysiology studies have propelled the field forward by identifying neural signals in the brain that fit this role. However, progress in establishing the neural underpinnings of human decision making has been severely hampered by technical challenges. This has presented an enormous obstacle to the field, not only because more elaborate decision making situations can be examined more feasibly in humans, but because decision making, even when based on simple perceptions, may be fundamentally different in humans than in animals. The present proposal builds on a recent paradigm breakthrough that enables isolation of a genuine "decision variable" in a freely-evolving electrical scalp signal that precisely determines the timing and accuracy of perceptual reports. This provides an unprecedented, direct neurophysiological window onto the distinct parameters of the decision process (e.g. "drift rate", baseline and criterion levels) such that the underlying mechanisms of several major behavioral phenomena can be finally resolved. This proposal seeks to develop a systems-level understanding of perceptual decision making in the human brain by tackling four major questions: 1) How are decisions reached more quickly when based on stronger evidence? 2) How is the decision process strategically adapted to deal with prior information and speed pressure? 3) How does top-down spatial attention facilitate faster and more accurate decision making? 4) How does practice strengthen perceptuo-motor interactions and promote intentional decision making? Each of the experiments described in this proposal will definitively test key predictions from prominent theoretical models using robust neurophysiological measurements that have exquisite temporal resolution, are unencumbered by model assumptions, and require minimal signal transformation.

Background The cognitive impairments exhibited by individuals suffering from attention-deficit/hyperacitivity disorder (ADHD) vary considerably from person to person and consequently a major priority for research has been to identify core impairments that can be linked to disease-specific abnormalities. One of the most reliable hallmarks of ADHD is a difficulty maintaining consistent performance levels over time: known as 'intra-subject variability' (ISV). ISV has been revealed to be a stable and heritable trait but its neural origins remain poorly understood. A key challenge in the effort to explain ISV is that all cognitive tasks requires a complex series of brain mechanisms whose specific contribution to performance can be difficult to conclusively establish. Recently I developed a novel technique that enables the simultaneous monitoring of the critical processing stages of sensory encoding, decision formation and motor preparation, as well as crucial supporting factors of attention and arousal while participants perform a simple perceptual task. The signals can be measured on a trial-by-trial basis allowing them to be directly linked to the timing and accuracy of the participant's responses. In the present project I will utilize this technique to pinpoint the specific neural abnormalities that give rise to elevated ISV in ADHD. Method A group of 35 children and adolescents diagnosed with ADHD and a non-ADHD comparison group will undergo electroencephalogram (EEG) recording while they perform a simple perceptual task that requires detection of gradual contrast changes in a patterned stimulus. Group-level and single-trial neural signal analyses will be conducted to establish precisely which processing stages contribute to inconsistent performance in ADHD. Benefits of this research project Several competing explanatory accounts of elevated ISV in ADHD have been proposed but there has been a lack of conclusive evidence to adjudicate between them. Identifying the neural origins of ISV in ADHD is essential to our understanding of how this disorder impacts on brain function. This research has the potential to identify disease-specific biomarkers for ADHD that can contribute to improved strategies for earlier and more accurate diagnosis as well as providing novel targets for pharmacological and non-pharmacological intervention.