J.D. Cohen’s research while affiliated with Princeton University and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (46)


Figure 1. Neural basis of moral decision-making. In Greene et al. (2001), subjects were asked to make moral judgments to a series of personal (involving direct physical contact) and impersonal (requiring causally distant interaction) moral dilemmas. Two sets of brain areas are activated when subjects make these moral judgments. One set is composed of brain areas implicated in cognitive processing and is activated preferentially by impersonal problems. This set includes the DLPFC and angular gyrus. The other set of includes brain areas commonly implicated in social and emotional processes and includes the superior temporal sulcus, posterior cingulate cortex, and medial prefrontal cortex. Adapted from Greene et al., 2001.
Figure 3. Cognitive and emotional systems in intertemporal choice. When people choose between different monetary payments available at different time delays, they tend accept a larger reduction in value in order to obtain payment immediately. (A) In the brain, deciding in any intertemporal choice leads to increased activity in the lateral prefrontal cortex (lateral orbitofrontal and DLPFC) as well as in the posterior parietal cortex. (B) In addition, choices that involve immediate reward are associated with enhanced activity in several brain areas tied to reward and emotion including the ventral striatum (VStr), the orbitofrontal cortex (OFC) posterior cingulate cortex (PCC), and the medial prefrontal cortex (MPFC). For choices involving an immediate and a delayed reward, the relative activity in the emotional (A) and cognitive (B) systems correlates with subjects' choices. Adapted from McClure et al., 2004a.
Figure 6. Response conflict in intertemporal choice. (A) Response time in intertemporal choice correlates with activity in the ACC. (B) When choices are separated into three equally sized sub-samples on the basis of response time, activity in the ACC (centered on the time of response) is seen to scale with increased deliberation time.
Conflict monitoring in cognition-emotion competition
  • Article
  • Full-text available

January 2006

·

1,300 Reads

·

47 Citations

·

·

·

[...]

·

J.D. Cohen
Download


Fig. 1. Model and example stimuli. ( a ) The model with the complete PFC system. Stimuli are presented in two possible locations (left, right). Rows represent different stimulus dimensions (e.g., color, size, shape, etc., labeled A–E for simplicity), and columns represent different features (red, orange green, and blue; small, medium, etc., numbered 1– 4). Other inputs include a task input indicating current task to perform (NF, name feature; MF, match feature; SF, smaller feature; LF, larger feature), and, for the ‘‘instructed’’ condition (used to control for lack of maintenance in non-PFC networks), a cue to the currently relevant dimension. Output responses are generated over the response layer, which has units for the different stimulus features, plus a ‘‘No’’ unit to signal nonmatch in the matching task. The hidden layers represent posterior cortical pathways associated with different types of inputs (e.g., visual and verbal). The AG unit is the adaptive gating unit, providing a temporal differences (TD) based dynamic gating signal to the PFC context layer. The weights into the AG unit learn via the TD mechanism, whereas all other weights learn using the Leabra algorithm that combines standard Hebbian and error-driven learning mechanisms, together with k-winners-take-all inhibitory competition within layers and point-neuron activation dynamics (26) (also see supporting information). ( b ) Example stimuli and correct responses for one of the tasks (NF) across three trials where the current rule is to focus on the Shape dimension (the same rule was blocked over 200 trials to allow networks plenty of time to adapt to each rule). The corresponding input and target patterns for the network are shown below each trial, with the unit meanings given by the legend in the lower left. The network must maintain the current dimension rule to perform correctly. 
Fig. 2. Representations (synaptic weights) that developed in four different network configurations. ( a ) Posterior cortex only (no PFC) trained on all tasks. ( b ) PFC without the adaptive gating mechanism (all tasks). ( c ) Full PFC trained only on task pairs (name feature and match feature in this case). ( d ) Full PFC (all tasks). Each image shows the weights from the hidden units ( a ) or PFC ( b – d ) to the response layer. Larger squares correspond to units (all 30 in the PFC and a random and representative subset of 30 from the 145 hidden units in the posterior model), and the smaller squares within designate the strength of the connection (lighter ϭ stronger) from that unit to each of the units in the response layer. Note that each row designates connections to response units representing features in the same stimulus dimension (as illustrated in e and Fig. 1). It is evident, therefore, that each of the PFC units in the full model ( d ) represents a single dimension and, conversely, that each dimension is represented by a distinct subset of PFC units. This pattern is less evident to almost entirely absent in the other network configurations (see text for additional analyses). 
Fig. 3. Generalization and learning results. ( a ) Crosstask generalization results (% correct on task-novel stimuli) for the full PFC network and a variety of control networks, with either only two tasks (Task Pairs) or all four tasks (All Tasks) used during training ( n ϭ 10 for each network, error bars are standard errors). Overall, the full PFC model generalizes substantially better than the other models, and this interacts with the level of training such that performance on the All Tasks condition is substantially better than the Task Pairs condition (with no differences in numbers of training trials or training stimuli). With one feature left out of training for each of four dimensions, training represented only 31.6% (324) of the total possible stimulus inputs (1,024); the Ϸ 85% generalization performance on the remaining test items therefore represents good productive abilities. The other networks are: Posterior, a single large hidden unit layer between inputs and response, a simple model of posterior cortex without any special active maintenance abilities; P ϩ Rec, posterior ϩ full recurrent connectivity among hidden units, allows hidden layer to maintain information over time via attractor dynamics; P ϩ Self, posterior ϩ self-recurrent connections from hidden units to themselves, allows individual units to maintain activations over time; SRN, simple recurrent network, with a context layer that is a copy of the hidden layer on the prior step, a widely used form of temporal maintenance; SRN-PFC, an SRN context layer applied to the PFC layer in the full model (identical to the full PFC model except for this difference), tests for role of separated hidden layers; NoGate, the full PFC model without the AG adaptive gating unit. ( b ) The correlation of generalization performance with the extent to which the units distinctly and orthogonally encode stimulus dimensions for the networks shown in Fig. 2. This was computed by comparing each unit’s pattern of weights to the set of five orthogonal, complete dimensional target patterns (i.e., the A dimension target pattern has a 1 for each A feature, and 0s for the features in all other dimensions, etc.). A numeric value between 0 and 1, where 1 represents a completely orthogonal and complete dimensional representation was computed for unit i as: d i ϭ max k ͉ w i ⅐ t k ͉͞ ͚ k ͉ w i ⅐ t k ͉ ; where t k is the dimensional target pattern k , and w i is the weight vector for unit i , and ͉ w i ⅐ t k ͉ represents the normalized dot product of the two vectors (i.e., the cosine). This value was then averaged across all units in the layer and then correlated with that network’s generalization performance. ( c ) Relative stability of PFC and hidden layer (posterior cortex) in the model, as indexed by Euclidean distance between weight states at the end of subsequent epochs (epoch ϭ 2,000 trials). The PFC takes longer to stabilize (i.e., exhibits greater levels of weight change across epochs) than the posterior cortex. For PFC, within-PFC recurrent weights were used. For Hidden, weights from stimulus input to Hidden were used. Both sets of weights are an equivalent distance from error signals at the output layer. The learning rate is reduced at 10 epochs, producing a blip at that point. 
Prefrontal Cortex and the Flexibility of Cognitive Control: Rules Without Symbols.

January 2005

·

139 Reads

·

138 Citations

Human cognitive control is uniquely flexible, and has been shown to depend on prefrontal cortex (PFC). But exactly how the biological mechanisms of the PFC support flexible cognitive control remains a profound mystery. Existing theoretical models have posited powerful task-specific PFC representations, but not how these develop. We show how this can occur when a set of PFC-specific neural mechanisms interact with breadth of experience to self-organize abstract, rule-like PFC representations that support flexible generalization in novel tasks. The same model is shown to apply to benchmark PFC tasks (Stroop and Wisconsin card sorting), accurately simulating the behavior of neurologically intact and frontally-damaged people.






Dissociating the contributions of DLPFC and anterior cingulate to executive control: An event-related fMRI study

October 2001

·

54 Reads

·

9 Citations

Brain and Cognition

Both DLPFC and the anterior cingulate (ACC) show increased activation during executive control; however, the specific contributions of each area remains controversial. Two classes of processes underlie control. Strategic processes provide top-down support for task operations; evaluative processes monitor ongoing performance. Using event-related fMRI and a task-switching Stroop paradigm we examined whether the strategic/evaluative distinction could be used to dissociate DLPFC and ACC. LDLPFC showed cue-related activity which was greater for color naming than word reading, with greater activation correlating with smaller Stroop effects (r = -.63). ACC showed only response-related activity which was greater for incongruent color-naming trials and correlated positively with the RT Stroop effect (r = .41). These data suggest DLPFC contributes a strategic function and ACC an evaluative one to executive control.


Fig. 1. Functional magnetic resonance imaging (fMRI) activity across the course of a trial in the left DLPFC (L. DLPFC) (BA 9; Talairach coordinates x – 41, y 18, z 28; maximum F 7.14; 36 pixels) and ACC (BA 24 and BA 32; Talairach coordinates x 4, y 1, z 43; maximum F 7.98; 10 voxels). At the beginning of each 25-s trial, subjects were given an instruction to either read the word or name the color of the following stimulus. A colored word was presented 12.5 s after the beginning of each trial. Significant differences between conditions were detected in the left DLPFC across scans 1 to 5 (lower left quadrant) and in the ACC across scans 7 to 10 (upper right quadrant).  
Dissociating the Role of the Dorsolateral Prefrontal and Anterior Cingulate Cortex in Cognitive Control

July 2000

·

3,310 Reads

·

2,861 Citations

Science

Theories of the regulation of cognition suggest a system with two necessary components: one to implement control and another to monitor performance and signal when adjustments in control are needed. Event-related functional magnetic resonance imaging and a task-switching version of the Stroop task were used to examine whether these components of cognitive control have distinct neural bases in the human brain. A double dissociation was found. During task preparation, the left dorsolateral prefrontal cortex (Brodmann's area 9) was more active for color naming than for word reading, consistent with a role in the implementation of control. In contrast, the anterior cingulate cortex (Brodmann's areas 24 and 32) was more active when responding to incongruent stimuli, consistent with a role in performance monitoring.



Citations (31)


... It is difficult to overstate the grip on current research of the control account. The fad of conflict monitoring and control is unprecedented within the Stroop milieu; following Schmidt's (2019) observation, the first few articles published between 1998 and 2004 now combine for over 30,000 citations in the literature (e.g., Carter et al., 1998;Botvinick et al., 1999Botvinick et al., , 2001Botvinick et al., , 2004MacDonald et al., 2000;Miller and Cohen, 2001;Kerns et al., 2004;see Schmidt, 2019, for an extensive bibliography). The upshot is, the Stroop effect has been appropriated from being an index of inputdriven selective attention to a tool for generating conflict and measuring control. ...

Reference:

Reclaiming the Stroop Effect Back From Control to Input-Driven Attention and Perception
ANTERIOR CINGULATE CORTEX, ERROR DETECTION AND PERFORMANCE MONITORING: AN EVENT RELATED fMRI STUDY
  • Citing Article
  • May 1998

NeuroImage

... The working memory task (N-back task) has been described elsewhere in full detail (Cohen et al., 1994). In this study, we used the 0-back and 2-back tasks to activate brain regions. ...

Activation of prefrontal cortex in a non-spatial working memory task with fMRI
  • Citing Article
  • January 1994

Human Brain Mapping

... Particularly, recent studies have focused on identifying biomarkers that may be useful for identifying at-risk individuals and avenues for future treatment research (Gottesman and Gould, 2003). Research suggests that the anterior cingulate cortex (ACC) -a frontostriatal brain region-mediated performance monitoring processes, particularly those indexed by event-related potentials (ERPs), may represent one such set of important biomarkers (Olvet and Hajcak, 2008;Botvinick et al., 2004;Kerns et al., 2004). Importantly, recent research has demonstrated that exaggerated ACC activation is characteristic of many anxious populations. ...

Cognitive control and the anterior cingulate cortex: an update
  • Citing Article

... Eine Grundlage zum Verständnis der verwendeten neuropsychologischen Verfahren ist der Ansatz der kognitiven Kontrolle der Arbeitsgruppe um Cohen, Servan-Schreiber und Braver [1,[16][17][18][19]. Kognitive Kontrolle ist die "Fähigkeit der angemessenen Aufrechterhaltung und Aktualisierung der internen Repräsentation aufgabenrelevanter Kontextinformation" [16]. ...

An integrated computational model of dopamine function in reinforcement learning and working memory.
  • Citing Article
  • March 1998

Journal of Cognitive Neuroscience

... The results replicated an earlier adult fMRI study showing inferior and middle frontal gyrus activity using the same paradigm (COHEN et al., 1994). Several groups (CASEY et al., 1997a;ORENDI et al., 1996;TRUWIT et al., 1996) have also examined brain activity during a spatial working memory task. One goal of this research was to equate task difficulty across age groups. ...

A developmental functional MRI study of cortical activation during a spatial working memory task
  • Citing Article
  • January 1997

... In the current study, emotional experience and expression were affected by brain activity within the attention system for sadness and anxiety. The relationship between negative emotions (anxiety and anger) and the attention system is associated with increased emotional lability [30]. Emotional lability is thought to be related to hyperconnectivity of the amygdala network, which includes the anterior cingulate cortex [31]. ...

Selective pharmacological activation of limbic structures in human volunteers: A positron emission tomography study
  • Citing Article
  • March 1998

The Journal of Neuropsychiatry and Clinical Neurosciences

... Of interest, these studies evaluated both the graymatter and white-matter regions of the ACC. Functional abnormalities in the ACC brain region have been associated with schizophrenia; these changes commonly involve altered cognition, including executive control and attention [99][100][101][102]. The involvement of the ACC white and gray matter in schizophrenia is becoming more apparent with the findings of structural, genetic and morphological changes in schizophrenia [103][104][105]. ...

Anterior cingulate cortex and cognitive disability in schizophrenia
  • Citing Article
  • April 1999

Biological Psychiatry

... Prior studies have shown functional dissociations between the dACC and DLPFC. While activation in the ACC have been associated with conflict evaluation and monitoring, the DLPFC is associated with attentional allocation and top-down cognitive control [15] and plays a critical role in conflict resolution [30]. With regard to language control, Abutalebi et al. [1] proposed that the dACC is involved in the detection of language conflict and, through projection to the DLPFC, the dACC reinforces (i.e., raises the activation levels) of neural representations of the target language in order to resolve competition between the target and non-target language. ...

Dissociating the contributions of DLPFC and anterior cingulate to executive control: An event-related fMRI study
  • Citing Article
  • October 2001

Brain and Cognition

... Broca's area is called by the name of "syntax" in one study (Caplan et al., 1999;Heim et al., 2003), "semantics" in another (Homae et al., 2002), "phonology" in yet another (Fiez et al., 1993). Then the plot thickens and we hear the identical persona called "imitation" (Heiser et al., 2003), "motor preparation" (Krams et al., 1998), "planning" (Fincham et al., 2002) and "imagery" (Binkofski et al., 2000), "action understanding" (Buccino et al., 2004;Hamzei et al., 2003), "visuomotor coordination" (Müller et al., 2003), "sequence learning" (Haslinger et al., 2002), "tonal discrimination" (Müller et al., 2001), "artificial grammar learning" (Bahlmann et al., 2008), "working memory" (Nystrom et al., 1998), "rule shifting" (Konishi et al., 1998), "response selection" (Thompson-Schill et al., 1997), "response inhibition" (Collette et al., 2001) and so on. As there is no technique allowing neuroscientists to probe for functional preferences of individual neurons in the living human being, it remains theoretically possible that each of these specializations is entirely separate from linguistic specializations, and that LIFG consists of a large array of functionally discrete modules. ...

Dynamics of fMRI: Broca's Area Activation Reflects Independent Effects of Duration and Intensity of Working Memory Processes
  • Citing Article
  • May 1998

NeuroImage

... . CC-BY-NC-ND 4.0 International license available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The ability of animals to form and use contextdependent, hierarchical associative structures (or context-informed cognitive maps) is arguably the foundation of high-level cognition 48,49 . This contextual knowledge contributes to the regulation of elicited behavior and influences learning. ...

Context, cortex, and dopamine: A Connectionist approach to behavior and biology in schizophrenia
  • Citing Article
  • January 1992

Psychological Review