Functions Of The Cingulate Cortex A Associative Attention

Sobriety Success

At Home Drug Withdrawal

Get Instant Access

As students of cognition are aware, the term attention has multiple interpretations. Research in the area of behavioral neuroscience with rabbits and rats indicates that the attention supported by cingulate cortex is selective attention or attention focused on particular stimuli. The stimuli that are selectively processed by cingulate cortical neurons are those that predict important outcomes, such as reward or aversion. Also, in most if not all cases they are "task-relevant" stimuli (i.e., stimuli that require a particular action).

The selective processing of stimuli by cingulate cortical neurons emerges as the subjects learn that a given stimulus predicts an important outcome and requires action. Thus, the attentional process of cingulate cortex that is invoked by such predictive stimuli is a learned or "associative" process of the brain. Stimuli that signal important outcomes and call for action produce large neuronal activations in cingulate cortex, whereas stimuli that predict that no important event will occur and that no action is needed do not elicit ample cingulate cortical neuronal discharges. An apt characterization is that cingulate cortex mediates associative attention to significant stimuli. Findings in support of these propositions are reviewed next.

1. Cingulate Cortical Neuronal Activity a. Discriminative Neuronal Activity Data supporting the hypothesis that cingulate cortex mediates associative attention to significant stimuli are afforded by studies of neuronal activity in multiple brain sites during discriminative instrumental learning of rabbits. In these studies the rabbits occupied a large rotating wheel apparatus. They learned to step in response to a tone cue, a positive conditional stimulus (CS+), in order to prevent a foot-shock scheduled for delivery 5 sec after the CS + . They also learned to ignore a different tone, the negative conditional stimulus (CS_), that was not followed by the foot-shock.

Recordings of neuronal activity in multiple areas of cingulate cortex and in related areas of the thalamus exhibited the development, during behavioral learning, of massive discriminative multi-, and single-unit neuronal responses. Discriminative neuronal responses are significantly greater firing frequencies in response to the CS+ than to the CS. They occur at brief latencies (15-100 msec) and persist for varied durations, with some neurons exhibiting discrimination during the full interval preceding response or shock (Fig. 2).

The discriminative neuronal activity represents an associative or learned activity of cingulate cortical neurons. It is not simply a frequency-specific response of the monitored neurons because the discriminative response is observed in many subjects given different (counterbalanced) sets of acoustic frequencies as CS + and CS. Discriminative activity develops as a result of the associative pairing of the CS + with the foot-shock reinforcer. A different neuronal response (either no response or one of significantly reduced magnitude) occurs to the CS, which is not predictive of the foot-shock. Thus, the activity of cingulate cortical neurons in these studies may be viewed as a neuronal code for the associative significance of the discriminative cues. It can be said that cingulate cortical neurons "learn" to respond selectively to stimuli that predict the occurrence of significant events and require action, which in these studies is the locomotor response needed to avoid the foot-shock.

Discriminative activity also occurs during learning of a reward-based response, wherein rabbits approach and make oral contact with a drinking spout inserted into the experimental chamber following CS+ presentation. The rabbits also learn to inhibit responding when the spout is inserted following CS~ presentation and oral contact with the spout does not yield water reward. Once again, neurons in cingulate cortex encode the associative significance of the conditional stimuli, i.e., they produce greater discharges in response to the CS+ than to the CS. These data rule out the possibility that cingulate cortical neurons encode the emotional value (pleasant or unpleasant) of the reinforcer predicted by the CS + , i.e., differential firing to stimuli that predict pleasant versus unpleasant consequences. Rather, cingulate cortical neurons encode stimuli that predict important events and call for action, whether the events are good or bad.

Discriminative neuronal activity in cingulate cortex has also been reported by Shirley Buchanan, Donald Powell, and Charles Gibbs during classical Pavlovian conditioning of heart rate and eyeblink responses in rabbits. Also, early studies of Menachem Segal and James Olds and recent studies of Takenouchi and colleagues demonstrated the occurrence of neuronal responses in anterior and posterior cingulate cortex that are specific to stimuli that predict reinforcement during appetitive conditioning.

Figure 2 Average anterior (area 24) and posterior (area 29, cellular layers IV and V) cingulate cortical integrated unit activity elicited by CS + and CS" in rabbits during pretraining, first exposure session, first significant behavioral discrimination, and criterial behavioral discrimination in a discriminative avoidance task. The neuronal activity for area 24 is plotted in the form of standard scores normalized with respect to the pre-CS baseline for 40 consecutive intervals following CS onset. Area 29 data are plotted starting 100 msec after tone onset.

Figure 2 Average anterior (area 24) and posterior (area 29, cellular layers IV and V) cingulate cortical integrated unit activity elicited by CS + and CS" in rabbits during pretraining, first exposure session, first significant behavioral discrimination, and criterial behavioral discrimination in a discriminative avoidance task. The neuronal activity for area 24 is plotted in the form of standard scores normalized with respect to the pre-CS baseline for 40 consecutive intervals following CS onset. Area 29 data are plotted starting 100 msec after tone onset.

The development of discriminative neuronal activity during conditioning supports the idea that cingulate cortex mediates associative attention to significant cues. The fact that the activity is time-locked to the onset of the CS+ and CS" and occurs as early as 15 msec after tone onset, at least 2 sec earlier than the learned avoidance response, suggests that it promotes learning-based stimulus selection. The early, stimulus-locked, CS + -specific brain activation could be involved in capturing attention and thereby enhancing stimulus evaluation and response mobilization processes.

b. Duration Coding and Salience Compensation The brief-latency discriminative neuronal responses are positioned temporally to maximize resources for subsequent processing of and responding to the eliciting cue. Accordingly, Stephen Sparenborg and Michael Gabriel showed that testing with non-salient cues, such as a brief (200-msec) CS + and CS", is associated with greater brief-latency anterior and posterior cingulate cortical discriminative neuronal responses than are observed when testing is carried out with more enduring and therefore more salient cues (e.g., a 500- or 5000-msec CS+ and CS"; Fig. 3). The enhanced neuronal coding of brief stimuli is referred to as duration coding. Duration coding may be thought of as an instance of salience compensation, i.e., a process of cingulate cortex that amplifies the neural representation of nonsalient but associatively significant stimuli in order to increase the likelihood that such stimuli receive processing that is commensurate with their associative significance.

c. Premotor Activity As discussed later, neuroa-natomical and neurophysiological data indicate a close association between cingulate cortex and the brain's motor system. One illustration of this involvement in motoric processing is the finding that approximately half of all single neurons studied in anterior and posterior cingulate cortex during discriminative avoidance learning exhibited premotor firing ramps, i.e., progressive increments of firing frequency during the 5-sec interval from CS + to the scheduled foot-shock,

Figure 3 Average multi-unit spike frequency recorded in well-trained rabbits in anterior cingulate (area 24) and posterior cingulate (area 29) during separate counterbalanced sets of three training sessions in which brief (200-msec), intermediate (500-msec), or long (5000-msec) CSs were presented. The neuronal activity following CS onset is in the form of standard scores normalized with respect to the pre-CS baseline in six consecutive 100-msec intervals after CS onset. Data shown for area 29 were obtained from records that exhibited discriminative TIA in the later stages of learning.

Figure 3 Average multi-unit spike frequency recorded in well-trained rabbits in anterior cingulate (area 24) and posterior cingulate (area 29) during separate counterbalanced sets of three training sessions in which brief (200-msec), intermediate (500-msec), or long (5000-msec) CSs were presented. The neuronal activity following CS onset is in the form of standard scores normalized with respect to the pre-CS baseline in six consecutive 100-msec intervals after CS onset. Data shown for area 29 were obtained from records that exhibited discriminative TIA in the later stages of learning.

in anticipation of the behavioral avoidance response (Fig. 4). Also, in recent studies by Takenouchi and colleagues, neuronal firing in ventral portions of the anterior cingulate cortex was correlated with the onset of consummatory behavior (licking a drinking tube) during appetitive conditioning of rats. This premotor firing of cingulate cortical neurons may represent a neural ''command volley'' projected from cingulate cortex to cortical and striatal motor areas to trigger the learned response. Thus, in addition to the encoding of significant stimuli, cingulate cortical neurons appear to be involved in the initiation or triggering of learned motor responses.

2. Brain Damage

An equivalent impairment of learning was found following bilateral combined lesions of the anterior and MD thalamic nuclei (i.e., areas of the thalamus that reciprocate connections with cingulate cortex). The interdependence of the thalamic and cingulate cortical processes was indicated by studies that showed that the training-induced, attention-related activity of cingulate cortical neurons did not develop in subjects that had been given thalamic lesions.

The fact that the lesions prevented the rabbits from learning to respond behaviorally to the CS + could be interpreted as a motor problem such as an inability to initiate locomotion on cue. That this was not the case was indicated in a study that showed that rabbits with lesions were substantially impaired in learning to inhibit the previously described reward-based approach response to a drinking spout that was inserted into the experimental chamber following CS" presentation (when approach and oral contact with the spout did not yield water reward). The rabbits with lesions responded equally often to the spout, whether its insertions were preceded by the CS + or the CS", whereas sham lesion controls discriminated significantly between the CS + and CS". These results show that cingulothalamic circuitry is critical for rabbits' ability to base behavioral responding on particular discriminative cues (as in the case of discriminative avoidance learning) or learning to inhibit a well-established response on cue (as in the approach learning task). These findings are in keeping with the notion that the cingulothalamic circuitry is not specialized for particular kinds of behavioral outputs, such as active movement or ''inhibitory'' omission of movement. Rather, this circuitry enables subjects to predicate their performance of context-appropriate, goal-directed instrumental behavior on the occurrence of discrete cues that predict significant outcomes and call for action. A wide variety of distinct forms of behavior can come under stimulus control as a result of cingulate cortical processing. This stimulus-oriented contribution of cingulate cortex is consistent with the hypothesis of an involvement of cingulothalamic circuitry in associative attention to significant stimuli.

a. Experimental Lesions If cingulate cortical neurons mediate associative attention to significant stimuli, then damage in cingulate cortex should interfere with learning about such stimuli. This expectation was confirmed by studies that showed that combined lesions of both the anterior and the posterior cingulate cortex severely impaired the ability of rabbits to exhibit discriminative avoidance learning.

b. Exposure to Cocaine in Utero Exposure of human fetuses to cocaine during gestation is associated with a variety of developmental, neurocognitive deficits, including impaired attention, habituation, arousal, recognition memory, and language development. Adult rabbits exposed to cocaine in utero (4 mg/ kg of cocaine given intravenously twice daily to gestating dams) exhibited morphological and

Figure 4 Histograms indicating anterior cingulate cortical single-unit activity related to CS + /CS onset and avoidance responses, where each bar indicates the average firing rate in hertz for the cell during a 40-msec interval. (Top) CS onset-related activity; (middle) premotor discharges preceding the conditioned response (CR); (bottom) neuronal firing on CS+ and CS" trials in which no response occurred and the trial terminated (EOT). All histograms represent data obtained from the same neuron.

Figure 4 Histograms indicating anterior cingulate cortical single-unit activity related to CS + /CS onset and avoidance responses, where each bar indicates the average firing rate in hertz for the cell during a 40-msec interval. (Top) CS onset-related activity; (middle) premotor discharges preceding the conditioned response (CR); (bottom) neuronal firing on CS+ and CS" trials in which no response occurred and the trial terminated (EOT). All histograms represent data obtained from the same neuron.

biochemical abnormalities in the anterior cingulate cortex relative to controls exposed to saline injections. No changes were found in the visual cortices of the subjects exposed to cocaine.

In addition, exposure to cocaine was associated with attenuated anterior cingulate cortical training-induced discriminative neuronal activity and deficient avoidance learning. Specifically, when brief (200-msec) and therefore nonsalient discriminative stimuli (CS+ and CS") were employed for training, cocaine-exposed rabbits performed significantly fewer learned responses than saline-exposed controls in the first session of conditioning. The first-session learning deficit found with brief stimuli could have resulted from compromised salience compensation mechanisms in anterior cingulate cortices of the rabbits exposed to cocaine. Note, however, that with continued training beyond the first session the cocaine-exposed rabbits did attain normal asymptotic performance levels as rapidly as did saline controls. Moreover, learning was entirely normative, even in the first session, in cocaine-exposed subjects when the CS + was more enduring (500 msec) and thus more salient.

Very similar results were obtained in studies of Pavlovian conditioning of rabbits' eyeblink response. Rabbits exposed to cocaine in utero were able to acquire the conditioned eyeblink response as rapidly as controls when a salient CS+ and a nonsalient CS" were used. However, acquisition was significantly retarded in cocaine-exposed rabbits when a nonsalient CS+ and a salient CS" were used. These results were obtained when the stimuli were of different modalities (a salient tone and a less salient flashing light) and when they were of the same modality (tones of varying intensity).

The absence of behavioral learning during the first discriminative avoidance conditioning session with the brief (200-msec) stimuli was accompanied by an absence of discriminative neuronal activity in the anterior cingulate cortex. Moreover, when behavioral discrimination did occur in later stages of training, so also did discriminative neuronal activity in the anterior cingulate cortex in rabbits exposed to cocaine. Thus, the loss of discriminative neuronal activity in anterior cingulate cortex was, arguably, the neural basis of the impaired discriminative behavior. This possibility received further support from previous demonstrations that lesions confined to the anterior cingulate cortex or to the MD thalamic nucleus impaired learning in the early stages of behavioral acquisition, but these lesions also allowed learning to occur to normative asymptotic levels.

A particularly intriguing finding during training of cocaine-exposed rabbits with the brief conditional stimuli was a dramatic alteration of the poststimulus firing profiles of anterior cingulate cortical neurons. Specifically, the neuronal response profiles of the cocaine-exposed rabbits lacked the firing pause, which occurred robustly in controls from 40 to 80 msec after CS onset and which is a consistent feature of the CS + -and CS- elicited poststimulus histograms of neurons in cingulate cortex (Fig. 5). Loss of the brief-latency inhibitory "pause" component of neuronal response was a most robust cocaine-related neuronal phenom-

Figure 5 Average anterior cingulate cortical multi-unit spike frequency in 40 consecutive 10-msec intervals after onset of a brief (200-msec) CS in rabbits exposed to cocaine in utero and in salineexposed controls. Asterisks indicate the occurrence of a significantly greater discharge for the indicated interval compared to the discharge in the corresponding interval for the other experimental group (cocaine or saline). Reproduced by permission from Harvey and Kosofsky, 1998, Cocaine: Effects on the Developing Brain, Ann. N.Y. Acad. Sci. 846, 208.

Figure 5 Average anterior cingulate cortical multi-unit spike frequency in 40 consecutive 10-msec intervals after onset of a brief (200-msec) CS in rabbits exposed to cocaine in utero and in salineexposed controls. Asterisks indicate the occurrence of a significantly greater discharge for the indicated interval compared to the discharge in the corresponding interval for the other experimental group (cocaine or saline). Reproduced by permission from Harvey and Kosofsky, 1998, Cocaine: Effects on the Developing Brain, Ann. N.Y. Acad. Sci. 846, 208.

enon. Indeed, the magnitude of the inhibitory pause was significantly increased in the saline-exposed rabbits trained with 200-msec stimuli compared to saline-exposed rabbits trained with 500-msec stimuli. These findings suggested that the inhibitory pause is a dynamic feature that reflects the associative atten-tional processing "demand" that is operating in a given situation. The absence of the inhibitory pause in rabbits exposed to cocaine may thus be a direct neurological indicant of impaired associative attention, which is in turn the basis for the observed retardation of the discriminative neuronal activity and behavioral learning in rabbits exposed to cocaine in utero.

The inhibitory pause in anterior cingulate cortex may function to reset active neurons by halting ongoing firing, thereby maximizing the number of neurons available for processing the incoming stimulus. Inhibitory feedback produced by activation of GABAergic neurons in response to cue-driven inputs is likely to be involved in resetting. The failure of the resetting mechanism in rabbits exposed to cocaine means that neurons already engaged in rapid firing could not contribute to stimulus processing. The resulting reduction in the number of participating neurons could impair processes such as the recruitment of existing modified synapses involved in classifying the incoming stimulus or retardation of synaptic plasticity development needed for the production of discriminative neuronal activity.

Research of Eitan Freidman and colleagues has shown that D1 dopamine receptors in the anterior cingulate cortex are decoupled from their G proteins in rabbits exposed to cocaine in utero. Thus, the failure of the resetting mechanism in exposed rabbits could be a result of impaired activation of GABA neurons normally mediated by stimulation of D1 dopamine receptors in the anterior cingulate cortex.

B. Executive Attention

New neuroimaging techniques have yielded an explosion of data in cognitive neuroscience concerning brain activation accompanying task engagement of human subjects. Studies employing position emission tomography (PET), functional magnetic resonance imaging (fMRI), high-density electroencephalography (EEG) for these analyses have repeatedly indicated an involvement of anterior cingulate cortex in cognition-relevant processing.

Results and interpretations converge intriguingly with the aforementioned findings in rabbits, rats, and primates in indicating a critical involvement of anterior cingulate cortex in processes subserving attention. Michael Posner and Gregory Di Girolamo provided an integrative account of anterior cingulate cortical function derived from the results of imaging studies and other findings in cognitive neuroscience. Building on the prior theoretical work of Donald Norman and Timothy Shallice, Posner and DiGirolamo proposed that anterior cingulate cortex mediates executive attention, which is part of a more general executive control function. Executive attention comes into play when routine or automatic processing is insufficient for the task at hand, such as when novel or conflict-laden situations are encountered. Guided by the individual's overall plans and goals, executive attention selectively activates and inhibits particular inputs, schemas, and behaviors to deal with the problematic circumstances. Paraphrasing Norman and Shallice, Posner and DiGirolamo argue that executive attention will likely be recruited in situations that involve planning and decision making, error detection, novelty and early stages of learning, difficult and threatening situations, and overcoming habitual behavior.

1. Review of Evidence

Activation of the anterior cingulate cortex as assessed by PET and fMRI is significantly enhanced in response to stimuli that require a particular response from multiple alternative responses, such as when subjects verbally generate uses of visually or acoustically presented words denoting familiar objects (e.g., the response "drive" to the stimulus "car"). In most of these studies a subtraction method is employed whereby the brain activation found during a control condition (merely reading and pronouncing the stimulus words) is subtracted from the scores obtained in the "generate uses" condition. Thus, anterior cingulate cortical activation above and beyond that involved in pronouncing visualized words is associated with the need to generate a response from a set of possible alternative responses. It is argued that executive attention reflected by anterior cingulate cortical activation is brought into play as a result of the conflict created by the multiple response alternatives in the generate uses condition. The activation produced in the generate uses condition declined as the subjects were repeatedly exposed to the same words and thus generated the same uses. According to the theory, the generation of uses became routinized with repetition and thus no longer required executive attention.

Additional support for a role of the anterior cingulate cortex in executive attention is the observed activation of cingulate cortex during performance in Stroop tasks, wherein subjects are required to name the color of visually presented words in a congruent condition (e.g., the word "red" printed in red ink) or in an incongruent condition (e.g., the word "green" printed in red ink). In some of these studies, a neutral condition is also employed in which, for example, subjects must give the ink color of noncolor words. The results of multiple Stroop experiments suggest the following generalization: The anterior cingulate cortex is activated significantly in all three of the aforementioned conditions. However, the extent of activation appears to depend on the degree to which the irrelevant dimension (word meaning) corresponds to the relevant dimension (word color). High correspondence, such as when both dimensions refer to color (as in the incongruent condition), creates maximal conflict and invokes substantial anterior cingulate cortical processing.

In keeping with the indication that the degree of conflict determines the magnitude of anterior cingulate cortical activation, several studies have shown substantial activation in association with dual-task performance, such as when subjects concurrently perform a generate-use task and a motor sequencing task. The results showed that performance in the dual-task situation elicited greater activation of anterior cingu-late cortex than did performance in either task singly. Because the two tasks did not conflict at the sensory or motor levels, the dual-task-specific activation was assumed to reflect competition for central processing resources rather than sensory or motor processes. These studies also showed declining anterior cingulate cortical activation as the tasks became well learned, but ample activation was reestablished simply by instructing the subjects to attend to the well-learned tasks. These results are consistent with the hypothesis that anterior cingulate cortical processing is recruited in response to conflict and interference among central aspects of task-relevant processing.

Recent work using event-related potential recordings derived from multiple scalp locations in human subjects has yielded an intriguing result. A marked electrical negativity occurs 100 msec after an erroneous key press in a discriminated reaction time task (Fig. 6). Brain electrical source analysis of data yielded by large arrays of simultaneously recorded scalp EEG records was employed in attempts to localize the brain areas in

Figure 6 A three-dimensional and sagittal view of the human brain illustrating the source of error-related negativity (ERN) found after brain electric source analysis of event-related brain potentials. (Bottom) The results of several ERN studies that have demonstrated that the source of the ERN is not affected by response modality (subjects responding with their feet or hands) or error feedback modality (visual, auditory, and somatosensory). Also shown is the ERN source for two reaction time experiments, one involving a decision of whether a number was "smaller than/larger than'' (RT Exp.1) and another involving a classification of words into semantic categories (RT Exp. 2). [Reprinted from Holroyd, C. B.,Dien, J., and Coles, M. G. (1998). Error-related scalp potentials elicited by hand and foot movements: Evidence for an output-independent error-processing system in humans. Neurosci. Lett. 242, 65-68, with permission from Elsevier Science].

Figure 6 A three-dimensional and sagittal view of the human brain illustrating the source of error-related negativity (ERN) found after brain electric source analysis of event-related brain potentials. (Bottom) The results of several ERN studies that have demonstrated that the source of the ERN is not affected by response modality (subjects responding with their feet or hands) or error feedback modality (visual, auditory, and somatosensory). Also shown is the ERN source for two reaction time experiments, one involving a decision of whether a number was "smaller than/larger than'' (RT Exp.1) and another involving a classification of words into semantic categories (RT Exp. 2). [Reprinted from Holroyd, C. B.,Dien, J., and Coles, M. G. (1998). Error-related scalp potentials elicited by hand and foot movements: Evidence for an output-independent error-processing system in humans. Neurosci. Lett. 242, 65-68, with permission from Elsevier Science].

which the error-related negativity (ERN) is generated. The results ofseparate studies by William Gehring and colleagues and by Stanislas Dehaene and colleagues indicated that anterior cingulate cortex is a likely intracranial source of the ERN (Fig. 6). These results are consistent with the hypothesis that anterior cingulate cortex is involved in mediating processes of executive attention that are recruited by the occurrence of errors.

2. Executive Attention and Response Selection

One problem with the concept of executive attention is that it includes multiple processes, including selection of input channels, schemas, and responses in accordance with the individual's overall plans and goals. It seems unlikely that all these processes are strictly localized within the cingulate cortex. Even the most cingulate-centered view of the universe must acknowledge the likely importance of continuous exchange of information between cingulate cortex and other brain modules that comprise a larger circuitry and supply information needed for the computation of executive functions. A clear understanding of how the components of executive attention are allocated among cingulate cortex and related areas of brain circuitry remains to be worked out. Nevertheless, a series of recent findings, reviewed in this section, indicate that response selection is an important aspect of cingulate cortical function.

First, there appears to be considerable regional specialization in cingulate cortex with respect to particular response modalities. Studies by Nathalie Picard and Peter Strick have indicated the existence of multiple areas of cingulate cortex, each containing neurons that project to the primary motor cortex in primates. Studies employing PET in human subjects indicate that some of the projecting areas may be associated with distinct response modalities. Similarly, Thomas Paus and colleagues showed that different areas of the anterior cingulate cortex were activated in the same subjects when they performed tasks involving oculomotor, manual, and spoken responses. Also, neurophysiological investigations by Keisetsu Shima and colleagues using primate subjects indicated the existence of two distinct cingulate cortical areas involved in mediating self-paced and stimulus-triggered forelimb movements.

In addition, U. Turken and Diane Swick reported studies of a patient (DL) who had undergone surgery for removal of a brain tumor. The surgery resulted in a circumscribed right hemispheric lesion in a region of the anterior cingulate cortex that had been implicated by neurophysiological studies in hand and arm movements of primate subjects. DL exhibited entirely normal performance in Stroop-like and divided attention tasks when responses were orally reported. However, DL showed a dramatic deficit in the same tasks when manual responses were required. Because

DL showed only a manual impairment, the authors argued that executive functions were intact in DL. Thus, they interpreted the results as favoring the idea that command signals originating in lateral prefrontal areas are sent to motor output areas through the anterior cingulate cortex, where correct response output is facilitated and incorrect response output is suppressed. They concluded that executive functions reside in the prefrontal areas, whereas cingulate cortex performs a ''final check''on the already-selected output.

A somewhat different view is afforded by recent findings of Michael Millham, Marie Banich, and colleagues, who employed fMRI imaging and a variant of the color-word Stroop task wherein participants were required to respond selectively to information about the color of a word while disregarding the word's meaning. They sought to address the issue of whether activity in the anterior cingulate cortex is engaged in attentional selection generally or is more specifically related to response conflict. To disentangle these two possibilities, these authors asked whether cingulate cortical processing on incongruent trials is engaged by both response and nonresponse conflict or by only one of these varieties of conflict. Participants indicated the ink color of the word (yellow, green, or blue) via a keypress. Half of the incongruent trials were response eligible in that the word named one of the eligible responses (e.g., the word ''blue'' printed in green ink). These trials engender conflict at both the response and nonresponse levels. The other half of the incongruent trials were response ineligible in that the word did not name an eligible response (e.g., the word ''purple'' printed in green ink). These trials engendered conflict at the nonresponse level but not at the response level (Fig. 7).

The pattern of results indicated that the anterior cingulate cortex is specifically involved in detecting the potential for error at the response level. As found in previous studies, both dorsolateral prefrontal regions and the anterior cingulate exhibited greater activation in response to incongruent trials than in response to neutral trials. However, further analysis indicated that the enhanced activation that occurred in anterior cingulate cortex on incongruent trials occurred only on response-eligible trials, not on response-ineligible trials. (In contrast, dorsolateral prefrontal cortex showed the difference on both response-eligible and response-ineligible trials.) These results thus suggest that anterior cingulate activation is driven exclusively by response conflict, and not by conflict related to stimulus-selection processes.

Time (sec)

Figure 7 A saggital fMRI image and corresponding line graph depicting the increased activation of the anterior cingulate during response-eligible vs response-ineligible trials in a modified Stroop task.

Time (sec)

Figure 7 A saggital fMRI image and corresponding line graph depicting the increased activation of the anterior cingulate during response-eligible vs response-ineligible trials in a modified Stroop task.

Was this article helpful?

0 0
Drug Free Life

Drug Free Life

How To Beat Drugs And Be On Your Way To Full Recovery. In this book, you will learn all about: Background Info On Drugs, Psychological Treatments Statistics, Rehab, Hypnosis and Much MORE.

Get My Free Ebook


Post a comment