chapter 11: executive function and higher-order thinking
Higher-Order Thinking: Abstract and Conceptual Thinking
A meta-analysis of neuroimaging data indicated that a variety of regions are activated during analogical reasoning, regardless of whether verbal or nonverbal problems are solved. • These include frontopolar and dorsolateral prefrontal regions, as well as the anterior insula and parietal cortex • In addition, distinct areas in frontopolar cortex activate for visuospatial analogies as compared to semantic analogies, which tend to be in a verbal format. • This pattern suggests both common mechanisms for reasoning as well as more specific regions that vary with problem type. • Anterior temporal regions involved in semantic processing may also be important for analogical reasoning as anatomical variation in these regions is associated with an individual's level of analogical reasoning ability
Judgment and Decision Making
Almost any clinician who treats or works with people who have executive dysfunction will invariably report that the judgment and decision-making abilities of such people are compromised. • The knowledge base of these people usually remains relatively unaffected and in some cases this knowledge can be used effectively. • However, if the task becomes a bit more abstract, they begin to exhibit difficulties. • The neural underpinnings of decision making have been an especially vibrant topic of research lately. • This work tends to focus on a particular psychological phenomenon, such as how people evaluate risk versus reward or how much time people are willing to wait for a reward. • The frontopolar cortex has been implicated as being particularly important for abandoning the current strategy and trying a new one, consistent with its role in processing novelty. • Relatedly, frontopolar cortex appears to be able to hold information about "the road not taken" in reference to the current course of action.
sequencing and planning
Another important aspect of sequencing behavior is the ability to choose which sequence or strategy best allows a goal to be attained. • Patients with frontal lobe damage are less likely to report that they use strategies • When they do use a strategy, it tends to be ill-defined or invoked inconsistently • One task used to measure strategy and sequencing is the Tower of London task • Participant is shown an initial position and a goal position to which the balls must be moved one at a time in as few moves as possible. • Dorsolateral prefrontal activity increases with increased difficulty of the task. • Performance on this task can be altered by tDCS over dorsolateral prefrontal regions
Organization of the Brain for Executive Function
Another model suggests that selection or control processes involve prefrontal regions that exert control at different times • from task preparation to stimulus processing to response output and then response evaluation. • This model views control as implemented in a cascade involving lateral and medial prefrontal regions • If control is not well implemented at a prior "waystation," more control will be required at the next point in the circuit. • For example, if posterior dorsolateral prefrontal cortex does a poor job of imposing a top-down attention set, then it will be harder to select the correct response, which will require the anterior cingulate to work harder to do so
Organization of the Brain for Executive Function
Another posits a hierarchy based on the nature of representations that are competing for control over action selection. • More posterior regions may select actions on the basis of more concrete dimensions, and the representations used for selection become more abstract as one moves in the anterior direction For example, if trying to decide what flowers to buy, Most posterior regions may be involved in selecting the flowers closest to your hand More anterior regions may select based on color the person prefers Most anterior regions may exert control with regard to the context (
creation and maintenance of a goal or task set
If more than one task set needs to be maintained, frontopolar cortex is likely to play a role. This area likely integrates processing for two or more separate operations in service of a higher goal. • In individuals with damage to frontopolar cortex, greater extent of damage predicts greater impairment in the management of multiple goals. • Frontopolar regions become more active when individuals choose which task to perform rather than being instructed which task to do, consistent with the idea that this region is involved in selecting which of the potential task sets is to be employed.
Theoretical Perspectives: Controlled Versus Automatic Processes
Like Shallice's theory, the theory of Stuss and Benson (1986) suggests that the frontal lobes are especially important in regulating behavior in nonroutine situations or in situations in which behavior must be carefully constrained. • Their model links the degree of control to particular neural substrates in a hierarchical manner. • At the lowest level, sensory information and simple tasks are processed by posterior regions of the brain in a relatively automatic manner that varies little from day to day. • The next level of control is associated with the executive, or supervisory, functions of the frontal lobe. • The highest level of control involves self-reflection (i.e., self-awareness and understanding the relationship of the self to the environment) and metacognition (i.e., the ability to reflect upon a cognitive process).
Organization of the Brain for Executive Function
One model argues that there is a nested hierarchy of control from posterior to anterior portions of frontal cortex • The most posterior regions are influenced by the most immediate aspects of a situation and more anterior regions are affected by the larger context. • This nested-hierarchy model focuses on the type of information that is used to guide control. Sensory control: Your phone rings, you read your text message Contextual control: Your phone rings in class, you do not read your text message Episodic control: On this particular occasion, since you are waiting to hear about the outcome of your parent's surgery, even though you are in class, when your phone rings, you read your text message Branching control: If you really want to show the person with whom you are speaking that you care about her/him, you do not read your text message when your phone rings. But if both of you are lost and trying to get to a restaurant where you are meeting friends, when your phone rings, you read your text message as it may contain information on how to get to the restaurant
creation and maintenance of goal or task set
One of the most basic prerequisites for meeting a goal is the ability to stay on task. Patients with frontal lobe damage are notorious for "wandering off task." • Neuroimaging studies have shown that dorsolateral prefrontal regions aid in creating and maintaining an attentional set, the process that designates which information is task- relevant • For example, activity in dorsolateral prefrontal regions depends on how large the change in attention set must be from one trial to another
Judgment and Decision Making
One paradigm often used to examine decision making is the delay discounting paradigm, also sometimes referred to as an intertemporal choice task • In this paradigm, an individual must decide whether they would like a smaller reward now ($5) or a larger reward in the future ($10 in a week). • Some researchers argue that the value of both present and future rewards are calculated by medial orbitofrontal regions • Other research suggests that dorsolateral prefrontal regions are engaged when an individual must overcome the temptation to take the immediate monetary reward or to choose a tasty piece of cake rather than the healthy but less mouthwatering carrot. • This idea is supported by findings that TMS over left dorsolateral prefrontal cortex biases people towards taking the immediate reward
Theoretical Perspectives: Goal-Centered Processing
Other approaches to understanding executive control emphasize that it allows one to guide behavior toward a goal. • Some models draw from work in artificial intelligence, which has investigated what type of computational structures are required to reach a goal. • These approaches suggest that reaching a goal requires the creation of a hierarchy of simpler and more solvable subgoals. • It has been argued that frontal lobe damage disrupts the ability to create such a hierarchical goal list. • Other models emphasize the role of the prefrontal cortex in providing a bias signal to other brain systems depending on the current goal and context. • Much like setting the switches on a train track, prefrontal cortex is thought to bias other brain regions involved in sensory processing, memory, emotion, response selection and so forth. • These models also emphasizes the role of prefrontal cortex in maintaining the task goal
Theoretical Perspectives: Multifactor Models
Some models argue that executive function may consist of separable factors. • One model, the unity and diversity model, has three components. • The ability to maintain task goals (common factor) • The ability to switch between tasks or subgoals (specialized factor) • The ability to update information being held in working memory (specialized factor) • Another model suggests three types of executive abilities based on the pattern of deficits in patients with frontal lobe damage • Initiating and sustaining a response; thought to rely on medial prefrontal regions. • Task-setting, which enables a task to be chosen and provides the steps required to perform the task; thought to rely on left lateral regions of prefrontal cortex. • Monitoring, which is the checking of behavior over time to ensure that it is being produced correctly, and to make any adjustment of behavior that might be required; thought to rely on right lateral regions of prefrontal cortex.
inhibition
Studies of patients and neuroimaging studies indicate that inhibition is associated with increased demands on cognitive control and working memory, activities supported by lateral prefrontal cortex. • From this perspective, response inhibition may be a specific example of a more general function, that of interference resolution • This is the ability to resolve conflict between competing or distracting information that might interfere with performing a task. • Consistent with this idea, when participants have to engage across three types of inhibitory tasks—one which requires stopping of a motor response, one which requires stopping of memory retrieval, and one which requires stopping of an emotional reaction—overlapping regions of the dorsolateral prefrontal cortex become active • This suggests a general mechanism for cognitive control over inhibitory processes and not one that is specific to motoric inhibition
self-monitoring and evaluation
The ability to evaluate one's own behavior is affected by frontal lobe lesions. • Patients with frontal lesions show disrupted metacognitive awareness, an overall evaluation of one's performance. • Frontal damage, especially right frontal damage, impairs the ability to detect errors and to modify ongoing behaviors to take a corrective action. • The error-related negativity (ERN) is an ERP component that occurs after an error is made; it has been linked to error monitoring because its amplitude is larger when accuracy is emphasized and when larger errors are made. • A variety of converging evidence suggests that the ERN component arises from rostral regions of the anterior cingulate cortex
inhibition
The ability to stop, interrupt, or abort inappropriate responses is referred to as response inhibition. • These abilities are compromised in patients with frontal lobe damage. • Two common response inhibition tasks • In the Go/No-Go task, the person responds by pushing a button when certain visual stimuli appear (Go trials) and withholding response to other stimuli (No-Go trials). • In the Stop-signal task, the person must respond as quickly as possible to a stimulus that appears on the screen, but abort their response when the stimulus is followed by a stopsignal. • Both of these tasks activate a large network of brain regions, spanning lateral prefrontal cortex, the anterior cingulate, SMA, pre-SMA, insula, and parietal regions
executive functions
include: The ability to plan actions to reach a goal • The ability to use information flexibly • The ability to think abstractly • The ability to make inferences • Difficulties in executive function can arise despite normal functioning in other domains of intellectual processing such as those measured by IQ tests • Classic work on patients with brain damage and research with monkeys has linked the frontal lobe with executive function • Executive deficits are often observed after damage to white-matter tracts, indicating that executive function likely relies on a network of brain structures
shifting set and modifying strategies
• A classic neuropsychological test used to examine set- shifting is the Wisconsin Card Sorting Test (WCST). • Participants are not told on what dimension (color, number, shape) cards should be sorted, but only whether their response is correct or not. • After the participant correctly sorts 10 cards, the rule for sorting the cards is changed. • Neurologically intact people quickly realize that their previous behavior no longer leads to a correct response. Hence, they adjust their responses accordingly. • In contrast, people with executive dysfunction perseverate, which is the action of continuing to engage in the same behavior. • Damage to dorsolateral prefrontal cortex is associated with poor performance on the WCST. • However, evidence from neuroimaging studies using this task and others suggests that set-shifting is likely subserved by a large network of brain regions
Higher-Order Thinking: Response to Novelty
• A related issue is that people with executive dysfunction have trouble being cognitively flexible - that is, looking at situations from a variety of vantage points and/or producing a variety of behaviors. • Research across many species, including rats, monkeys, and humans implicates the orbitofrontal cortex with aiding in flexible behaviour. • Orbitofrontal damage impedes the ability to exhibit normal reversal learning, in which an individual reverses a previous response. • For example, in a reversal trial, after learning to press the right-hand key when a blue light appears and the left-hand key when a yellow light appears, one would have to press the right-hand key when the yellow light appears and the left-hand key when the blue light appears. • Frontopolar cortex is also implicated in dealing with novel situations. • Monkeys with damage to frontopolar cortex appear to be "hyperfocused" on current tasks and impaired at learning quickly about novel rules or alternative rules • This pattern of behavior suggests that their ability to reorient potential task goals toward novel situations or opportunities is compromised
Higher-Order Thinking: Rules and Inference
• Another form of abstract thinking is the ability to deduce or invoke a rule. • Some frontal lobe patients have difficulty with abstraction and rule-like understanding • For example, whereas some people with frontal lobe damage exhibit only perseverative tendencies on the WCST, others cannot even figure out the criterion by which cards should be sorted. • In neuroimaging studies, a variety of tasks have been used to provide insights into the neural substrates of abstract rule-governed behavior in neurologically intact individuals. • Some studies compare activation in conditions in which a person is told what rule should govern his or her behavior, compared to when the person must deduce that rule. • Another way to examine rule-governed behavior is for people to make decisions about whether inferences are valid.
sequencing and planning
• Another process involved in reaching a goal is determining what steps must be taken to attain the goal, and the order in which those steps must be taken. • One task that reveals deficits in sequencing in patients with frontal lobe damage is the self-ordered pointing task, in which individuals are shown an array of items and must select a unique item on each trial. • Dorsolateral prefrontal regions may be important in this task because they support executive processes that act on information being maintained in working memory
higher-order thinking
• Executive function is often conceptualized to include a set of abilities known as higher-order thinking. • This term describes more complicated aspects of thought, such as • being able to think in an abstract and conceptual rather than concrete manner, • the ability to deduce rules or regularity, and • the ability to be flexible and respond to novelty. • The frontal lobes make many contributions to higher-order thinking
goal-directed behaviors
• It is not clear which of these theoretical models is most accurate, but the following observations appear to be true: • Goal-directed behavior is a central aspect of executive function • A variety of subprocesses are invoked when working toward a goal • Therefore, one way to think about executive function is as a group of "goaldirected behaviors," which include: • Initiation of behaviour • Creation and maintenance of a goal or task set • Sequencing and planning • Set-shifting • Self-monitoring • Evaluation and inhibition • Many of these behaviors rely on overlapping portions of the frontal lobe
Higher-Order Thinking: Response to Novelty
• Novelty is when an event, a situation, or an action has a low probability of occurring given a particular context. • The ventral attentional system is proposed to allow novel stimuli to capture attention. • Electrophysiological studies implicate frontal regions as playing an important role when a novel stimulus captures attention. • Novel stimuli elicit a P3a, while rare stimuli elicit a P3b • Neuroimaging studies show greater activity in prefrontal regions in response to novel versus familiar stimuli
Higher-Order Thinking: Abstract and Conceptual Thinking
• One deficit exhibited by patients with executive dysfunction is an inability to process material in an abstract rather than a concrete manner. • When reading metaphorical sentences, activity is greater in many areas of prefrontal cortex compared to when reading literal sentences. • Another way to examine the issue of abstract thinking is to examine analogical reasoning
Higher-Order Thinking: Rules and Inference
• One model argues that a number of prefrontal regions, along with their interactions with posterior cortex, are required when one uses rules to guide actions. • Ventrolateral prefrontal cortex, by virtue of its connections to regions of the middle temporal gyrus, plays a role in retrieving stored knowledge that allows the retrieval of rules. • Dorsolateral prefrontal cortex is more involved in selecting or influencing how rules should be used to guide responding than in actually selecting the rules. • Other viewpoints argue that dorsolateral prefrontal cortex is important because it plays a direct role in abstracting the rules, while still others argue that dorsolateral prefrontal cortex just holds rules in working memory. • Some theorists suggest that under certain situations problem solving is better off without our frontal cortex. • Their argument is that reduced frontal cortex control frees us from the restrictions provided by the rule-based behaviour we have learned, allowing and enhancing creativity
initiation of behavior
• Patients with executive dysfunction are often poor at starting an action or a behavior, but once engaged in it, they have great difficulty stopping it, a symptom called psychological inertia. • Patients with damage to medial prefrontal regions, including the supplementary motor area and anterior cingulate, show particular difficulty in responding quickly compared to patients with damage in other frontal regions. • Growing evidence also implicates regions of the medial prefrontal cortex as being involved in determining how much "effort" will be exerted to reach a goal. • This evidence suggests patients with medial prefrontal damage may have difficulty initiating behavior because the goal does not seem worth the effort
A Central Role for Working Memory in Executive Function
• Researchers have argued that DLPFC may play a central role in executive function because it supports working memory. • If one cannot maintain information in working memory, one may not be able to keep a goal in mind • Such a deficit would interfere with a person's ability to direct behavior toward a goal or to formulate a strategy for attaining the goal. • In addition, difficulty in keeping information on-line may disrupt a person's understanding of temporal relations between items and events. • If what has just happened cannot be kept on-line, its relation to subsequent happenings will be lost. In such cases, sequencing would be quite difficult. • Moreover, if one is not able to retain multiple pieces of information simultaneously in working memory, there will be difficulty in creating or following rules, in making inferences, and/or in understanding the relations between items in the world. • If one is unable to clear out what is held in working memory, perseveration will result • Accordingly, difficulties in working memory may account for a number of those observed in executive dysfunction
inhibition
• Some scientists have suggested that the right inferior frontal cortex may be specifically involved in response selection and override. • Damage to this region impairs performance on the stop-signal task • The degree of right inferior frontal cortex damage predicts the degree of impairment • A pathway from the right inferior frontal cortex to the subthalamic nucleus may support ceasing motor action • However, other research has questioned the idea that the right inferior frontal cortex is specifically involved in inhibiting responses, suggesting rather that it plays a more general role in altering responses. • This region is also activated when people have to push a button twice rather than not respond • According to these researchers, right inferior frontal regions are involved in monitoring the current environmental context so as to provide information about how goals can be met.
self-monitoring and evaluation
• The exact computation that the medial prefrontal regions perform to detect and correct errors is currently a source of debate. • One viewpoint is that medial prefrontal cortex does not so much detect errors as it monitors for conflict. • It is linked to error detection because errors usually occur when there is conflicting information. • When the cingulate detects conflict, it signals the dorsolateral prefrontal cortex to ramp up the top- down control, so as to reduce conflict, and hence errors, in subsequent behavior. • Others argue that the medial prefrontal cortex determines whether exerting control is worth the effort and cost in a given situation. • Others argue that the cingulate is focused on determining whether the action you just took leads to a useful or good outcome. • Still others argue it calculates the difference between the expected outcome of an action and the actual outcome regardless of whether that outcome is positive or negative. • From this viewpoint, the ERN may, in part, reflect the subjective evaluation of performance ("Oh damn, I think I just messed up and didn't get what I wanted"), rather than whether one did indeed make an error.
Organization of the Brain for Executive Function
• The organization of the frontal lobe for executive function is quite complicated. • There is unlikely to be a strict subregion-to-function mapping. • Rather, overlapping regions are involved in most of the functions we have discussed. • Nonetheless, researchers have attempted to derive some principles or general trends concerning how prefrontal cortex might be organized for executive function
self-monitoring and evaluation
• There is unlikely to be a one-to-one mapping between a particular piece of frontal cortex and a specific function such as self-evaluation. • At least some abilities relating to evaluation and monitoring are intact in patients with damage to the anterior cingulate cortex • Patients with cingulate damage are able to report errors, but are slow in correcting their errors • Patients with damage to other regions have trouble with self-monitoring and evaluation • Patients with damage to lateral prefrontal cortex may not show self-corrective actions after making an error and may not show evidence of using less force on error trials as people usually do when they make an error • And still other evidence from brain-damaged patients suggests that error evaluation can be altered by damage to the basal ganglia. • Evidence from EEG recordings suggests a cascade of activity occurs across a variety of brain regions that unfolds over time as we make choices and act upon the world. • These brain regions detect and weight the accumulating evidence that provides evaluation of our actions and use that to determine the subsequent actions or reactions that are likely to be required.
Theoretical Perspectives: Controlled Versus Automatic Processes
• Two classic theories view the frontal lobes as playing an important role in executive function because they are critical for controlled as compared to automatic processing. • Shallice suggested that a two-component system influences the choice of behavior. • One part, contention scheduling, is a cognitive system that enables relatively automatic processing, which has been learned over time. • The second part, the supervisory attentional system, is the cognitive system required to effortfully direct attention and guide action through decision processes. • According to this theory, frontal lobe damage disables the supervisory attentional system and thereby leaves actions to be governed totally by contention scheduling
