Biomedical subjects
Publications and source records attributed to Jonathan D Cohen.
Dopamine and the prefrontal cortex are critical for thought and behaviour. Recently, computational models have tried to elucidate the specific and intricate roles of dopamine in the prefrontal cortex, at the neurophysiological, system and behavioral levels, with varying degrees of success.
We present a computational model of the intradimensional/ extradimensional (ID/ED) task (a variant of the Wisconsin card sorting task) that simulates the performance of intact and frontally lesioned monkeys on three different kinds of rule changes (Dias et al., 1997, J Neurosci 17:9285-9297). Although Dias et al. interpret the lesion data as supporting a model in which prefrontal cortex is organized into different processing functions, our model suggests an alternative account based on representational content. A key aspect of the model is that prefrontal cortex representations are organized according to different levels of abstraction, with orbital areas encoding more specific featural information and dorsolateral areas encoding more abstract dimensional information. This representational scheme of the model is integrated with two additional key elements: (i) activation-based working memory representations controlled by a dynamic gating mechanism that simulates the hypothesized phasic actions of dopaminergic neuromodulation in prefrontal cortex, which acts to stabilize or destabilize frontal representations based on success in the task; and (ii) a weight-based associative learning system simulating posterior cortex and other subcortical areas, where the stimulus-response mappings are encoded. Frontal cortex contributes to the task via top-down activation-based biasing of task-appropriate features and dimensions in this posterior cortex system - this top-down biasing is specifically important for overcoming prepotent associations after a sorting rule reverses. The ability of the model to capture the double-dissociation observed by Dias et al. with orbital versus dorsolateral lesions supports the validity of these principles, many of which have also been useful in accounting for other frontal phenomena.
OBJECTIVE: To assess whether the respiratory rate to tidal volume ratio (RVR) measured while receiving automatic tube compensation (ATC) [RVRATC] would have a better predictive value as a weaning measure than unassisted RVR. DESIGN: Prospective cohort study. SETTING: General ICU of a tertiary-care university hospital. PATIENTS: Forty-three patients who received mechanical ventilation for > 24 h and were considered ready for weaning. INTERVENTIONS: All patients underwent a 60-min spontaneous breathing trial (SBT) [positive end-expiratory pressure of 5 cm H(2)O; ATC, 100%]. Patients tolerating the trial (n = 35) were extubated immediately. The following parameters were measured at the onset and end of the SBT: RVR, RVRATC, peak airway pressure (Paw), airway occlusion pressure, and minute ventilation. The outcome measure was successful extubation (ability to maintain spontaneous breathing for > 48 h). MEASUREMENTS AND RESULTS: Median age was 55 years (range, 25 to 88 years), median APACHE (acute physiology and chronic health evaluation) II score was 15.5 (range, 3 to 29), and median duration of mechanical ventilation prior to the SBT was 7 days (range, 1 to 40 days). Extubation was successful in 25 patients (72%). There were no significant differences in baseline characteristics between patients successfully extubated (group 1) and those requiring reintubation. On multivariate analysis, RVRATC measured at 60 min (RVR(60)ATC) was most predictive of successful extubation (p = 0.03). The area under the receiver operator characteristic curve was also highest for RVR(60)ATC (0.81 +/- 0.03) as compared to RVR (0.77 +/- 0.03), RVRATC (0.75 +/- 0.04), and RVR measured at 60 min (0.69 +/- 0.05). The ratio of RVR(60)ATC to Paw was the best predictor (0.84 +/- 0.02). CONCLUSIONS: RVRATC measured at the end of the SBT was the best predictor of successful extubation. A new ratio (ratio of RVRATC to Paw) was most predictive and deserves further study.
In choice reaction time tasks, response times and error rates demonstrate differential dependencies on the identities of up to four stimuli preceding the current one. Although the general profile of reaction times and error rates, when plotted against the stimulus histories, may seem idiosyncratic, we show that it can result from simple underlying mechanisms that take account of the occurrence of stimulus repetitions and alternations. Employing a simple connectionist model of a two-alternative forced-choice task, we explored various combinations of repetition and alternation detection schemes in an attempt to account for empirical results from the literature and from our own studies. We found that certain combinations of the repetition and the alternation schemes provided good fits to the data, suggesting that simple mechanisms may serve to explain the complicated but highly reproducible higher order dependencies of task performance on stimulus history.
A growing body of evidence from functional neuroimaging and computational modeling studies indicates that the anterior cingulate cortex (ACC) detects the presence of response conflict and conveys this information to other brain regions, enabling subsequent adjustments in cognitive control. The present study examined previous empirical findings of increased ACC for low-frequency stimuli across three distinct speeded response tasks (two-alternative forced choice, go/no-go, and oddball). Simulations conducted in a neural network model incorporating sequential priming mechanisms (developed in Cho et al., 2002) confirmed that a computational measure of response conflict was higher on low-frequency trials across all three tasks. In addition, the model captured detailed aspects of behavioral reaction time and accuracy data, predicted the dynamics of ACC activity related to trial sequence effects, and provided evidence for the functional role of conflict information in performance monitoring and optimization. The results indicate that the conflict-monitoring hypothesis, augmented by mechanisms for encoding stimulus history, can explain key phenomena associated with performance in sequential speeded response tasks.
Neurophysiological work in monkeys has shown that changes in tonic and stimulus-induced activity in the noradrenergic brainstem nucleus locus coeruleus (LC) are tightly correlated with fluctuations in behavioral performance in a visual discrimination task. Simulation work suggests transitions between observed modes of LC activity may be mediated by changes in the level of coherent firing among individual LC neurons. We have simplified this simulation by abstracting the LC to a simple, excitable system in two variables modeled after the FitzHugh-Nagumo relaxation oscillator. This abstracted LC simulates population-level dynamics of the nucleus relevant to its influence on behavior. Coherence within the nucleus was simulated as a single parameter which amplified external input to the abstracted LC while attenuating tonic, uncorrelated activity. Simulated results captured LC dynamics and their relationship with behavior observed in monkeys. Behavior was further simulated over a range of potential LC states not, as yet, directly observed. These results suggest a reverse sigmoidal relationship between false alarm rate and coherence, and a more complex, non-monotonic relationship between response time and coherence.
This paper presents a novel method for volume rendering of unstructured grids. Previously, we introduced an algorithm for perspective-correct interpolation of barycentric coordinates and computing polynomial attenuation integrals for a projected tetrahedron using graphics hardware. Here, we enhance the algorithm by providing a simple and efficient method to compute the projected shape (silhouette) and tessellation of a tetrahedron, in perspective and orthographic projection models. Our tessellation algorithm is published here for the first time. Compared with works of other groups on rendering unstructured grids, the main contributions of this work are: 1) A new algorithm for finding the silhouette of a projected tetrahedron. 2) A method for interpolating barycentric coordinates and thickness on the faces of the tetrahedron. 3) Visualizing higher-order attenuation functions using GPU without preintegration. 4) Capability of applying shape deformations to a rendered tetrahedral mesh without significant performance loss. Our visualization model is independent of depth-sorting of the cells. We present imaging and timing results of our implementation, and an application in time-critical "2D-3D" deformable registration of anatomical models. We discuss the impact of using higher-order functions on quality and performance.