Executive functions

Executive functions (also known as cognitive control and supervisory attentional system) is an umbrella term for the management (regulation, control) of cognitive processes, including working memory, reasoning, flexibility, and problem solving as well as planning and execution. The prefrontal areas of the frontal lobe are necessary but not solely sufficient for executive functions.

Neuroanatomy

Historically, the executive functions have been seen as regulated by the prefrontal regions of the frontal lobes, but it is still a matter of ongoing debate if that really is the case. Even though articles on prefrontal lobe lesions commonly refer to disturbances of executive functions and vice versa, a review found indications for the sensitivity but not for the specificity of executive function measures to frontal lobe functioning. This means that both frontal and non-frontal brain regions are necessary for intact executive functions. Probably the frontal lobes need to participate in basically all of the executive functions, but it is not the only brain structure involved.

Neuroimaging and lesion studies have identified the functions which are most often associated with the particular regions of the prefrontal cortex.

• The dorsolateral prefrontal cortex (DLPFC) is involved with "on-line" processing of information such as integrating different dimensions of cognition and behaviour. As such, this area has been found to be associated with verbal and design fluency, ability to maintain and shift set, planning, response inhibition, working memory, organisational skills, reasoning, problem solving and abstract thinking.

Side view of the brain, illustrating dorsolateral prefrontal and orbitofrontal cortex

• The anterior cingulate cortex (ACC) is involved in emotional drives, experience and integration. Associated cognitive functions include inhibition of inappropriate responses, decision making and motivated behaviours. Lesions in this area can lead to low drive states such as apathy, abulia or akinetic mutism and may also result in low drive states for such basic needs as food or drink and possibly decreased interest in social or vocational activities and sex.
• The orbitofrontal cortex (OFC) plays a key role in impulse control, maintenance of set, monitoring ongoing behaviour and socially appropriate behaviours. The orbitofrontal cortex also has roles in representing the value of rewards based on sensory stimuli and evaluating subjective emotional experiences. Lesions can cause disinhibition, impulsivity, aggressive outbursts, sexual promiscuity and antisocial behaviour.

Furthermore, in their review, Alvarez and Emory state that: "The frontal lobes have multiple connections to cortical, subcortical and brain stem sites. The basis of "higher-level" cognitive functions such as inhibition, flexibility of thinking, problem solving, planning, impulse control, concept formation, abstract thinking, and creativity often arise from much simpler, "lower-level" forms of cognition and behavior. Thus, the concept of executive function must be broad enough to include anatomical structures that represent a diverse and diffuse portion of the central nervous system."

Hypothesized role

The executive system is thought to be heavily involved in handling novel situations outside the domain of some of our 'automatic' psychological processes that could be explained by the reproduction of learned schemas or set behaviors. Psychologists Don Norman and Tim Shallice have outlined five types of situations in which routine activation of behavior would not be sufficient for optimal performance:

1. Those that involve planning or decision making
2. Those that involve error correction or troubleshooting
3. Situations where responses are not well-rehearsed or contain novel sequences of actions
4. Dangerous or technically difficult situations
5. Situations that require the overcoming of a strong habitual response or resisting temptation.

A prepotent response is a response for which immediate reinforcement (positive or negative) is available or has been previously associated with that response. The executive functions are often invoked when it is necessary to override these prepotent responses that might otherwise be automatically elicited by stimuli in the external environment. For example, on being presented with a potentially rewarding stimulus, such as a tasty piece of chocolate cake, a person might have the automatic response to take a bite. However, where such behavior conflicts with internal plans (such as having decided not to eat chocolate cake while on a diet), the executive functions might be engaged to inhibit that response.

Although suppression of these prepotent responses is ordinarily considered adaptive, problems for the development of the individual and the culture arise when feelings of right and wrong are overridden by cultural expectations or when creative impulses are overridden by executive inhibitions.

Historical perspective

Although research into the executive functions and their neural basis has increased markedly over recent years, the theoretical framework in which it is situated is not new. In the 1940s, the British psychologist Donald Broadbent drew a distinction between "automatic" and "controlled" processes (a distinction characterized more fully by Shiffrin and Schneider in 1977), and introduced the notion of selective attention, to which executive functions are closely allied. In 1975, the US psychologist Michael Posner used the term "cognitive control" in his book chapter entitled "Attention and cognitive control".

The work of influential researchers such as Michael Posner, Joaquin Fuster, Tim Shallice, and their colleagues in the 1980s (and later Trevor Robbins, Bob Knight, Don Stuss, and others) laid much of the groundwork for recent research into executive functions. For example, Posner proposed that there is a separate "executive" branch of the attentional system, which is responsible for focusing attention on selected aspects of the environment. The British neuropsychologist Tim Shallice similarly suggested that attention is regulated by a "supervisory system", which can override automatic responses in favour of scheduling behaviour on the basis of plans or intentions. Throughout this period, a consensus emerged that this control system is housed in the most anterior portion of the brain, the prefrontal cortex (PFC).

Psychologist Alan Baddeley had proposed a similar system as part of his model of working memory and argued that there must be a component (which he named the "central executive") that allows information to be manipulated in short-term memory (for example, when doing mental arithmetic).

Development

Further information: Neurobiological effects of physical exercise § Cognitive control and memory

When studying executive functions, a developmental framework is helpful because these abilities mature at different rates over time. Some abilities peak maturation rate in late childhood or adolescence while others' progress into early adulthood. The brain continues to mature and develop connections well into adulthood. A person's executive function abilities are shaped by both physical changes in the brain and by life experiences, in the classroom and in the world at large. Furthermore, executive functioning development corresponds to the neurophysiological developments of the growing brain; as the processing capacity of the frontal lobes and other interconnected regions increases, the core executive functions emerge. As these functions are established, they continue to mature, sometimes in spurts, while other, more complex functions also develop, underscoring the different directions along which each component might develop.

Early childhood

Inhibitory control (see Inhibitory control test) and working memory act as basic executive functions that makes it possible for more complex executive functions like problem-solving to develop. Inhibitory control and working memory are among the earliest executive functions to appear, with initial signs observed in infants, 7 to 12-months old. Then in the preschool years, children display a spurt in performance on tasks of inhibition and working memory, usually between the ages of 3 to 5 years. Also during this time, cognitive flexibility, goal-directed behavior, and planning begin to develop. Nevertheless, preschool children do not have fully mature executive functions and continue to make errors related to these emerging abilities - often not due to the absence of the abilities, but rather because they lack the awareness to know when and how to use particular strategies in particular contexts.

Preadolescence

Preadolescent children continue to exhibit certain growth spurts in executive functions, suggesting that this development does not necessarily occur in a linear manner, along with the preliminary maturing of particular functions as well. During preadolescence, children display major increases in verbal working memory; goal-directed behavior (with a potential spurt around 12 years of age); response inhibition and selective attention; and strategic planning and organizational skills. Additionally, between the ages of 8 to 10, cognitive flexibility in particular begins to match adult levels. However, similar to patterns in childhood development, executive functioning in preadolescents is limited because they do not reliably apply these executive functions across multiple contexts as a result of ongoing development of inhibitory control.

Adolescence

Many executive functions may begin in childhood and preadolescence, such as inhibitory control. Yet, it is during adolescence when the different brain systems become better integrated. At this time, youth implement executive functions, such as inhibitory control, more efficiently and effectively and improve throughout this time period. Just as inhibitory control emerges in childhood and improves over time, planning and goal-directed behavior also demonstrate an extended time course with ongoing growth over adolescence. Likewise, functions such as attentional control, with a potential spurt at age 15, along with working memory, continue developing at this stage.

Adulthood

The major change that occurs in the brain in adulthood is the constant myelination of neurons in the prefrontal cortex. At age 20-29, executive functioning skills are at their peak, which allows people of this age to participate in some of the most challenging mental tasks. These skills begin to decline in later adulthood. Working memory and spatial span are areas where decline is most readily noted. Cognitive flexibility, however has a late onset of impairment and does not usually start declining until around age 70 in normally functioning adults. Impaired executive functioning has been found to be the best predictor of functional decline in the elderly.

Models

Top-down inhibitory control

Aside from facilitatory or amplificatory mechanisms of control, many authors have argued for inhibitory mechanisms in the domain of response control, memory, selective attention, theory of mind, emotion regulation, as well as social emotions such as empathy. A recent review on this topic argues that active inhibition is a valid concept in some domains of psychology/cognitive control.

Working memory model

One influential model is Baddeley’s multicomponent model of working memory, which is composed of a central executive system that regulates three other subsystems: the phonological loop, which maintains verbal information; the visuospatial sketchpad, which maintains visual and spatial information; and the more recently developed episodic buffer that integrates short-term and long-term memory, holding and manipulating a limited amount of information from multiple domains in temporal and spatially sequenced episodes.

Supervisory attentional system (SAS)

Another conceptual model is the supervisory attentional system (SAS). In this model, contention scheduling is the process where an individual’s well-established schemas automatically respond to routine situations while executive functions are used when faced with novel situations. In these new situations, attentional control will be a crucial element to help generate new schema, implement these schema, and then assess their accuracy.

Self-regulatory model

Primarily derived from work examining behavioral inhibition, Barkley’s self-regulatory model views executive functions as composed of four main abilities. One element is working memory that allows individuals to resist interfering information. A second component is the management of emotional responses in order to achieve goal-directed behaviors. Thirdly, internalization of self-directed speech is used to control and sustain rule-governed behavior and to generate plans for problem-solving. Lastly, information is analyzed and synthesized into new behavioral responses to meet one’s goals. Changing one’s behavioral response to meet a new goal or modify an objective is a higher level skill that requires a fusion of executive functions including self-regulation, and accessing prior knowledge and experiences.

Problem-solving model

Yet another model of executive functions is a problem-solving framework where executive functions is considered a macroconstruct composed of subfunctions working in different phases to (a) represent a problem, (b) plan for a solution by selecting and ordering strategies, (c) maintain the strategies in short-term memory in order to perform them by certain rules, and then (d) evaluate the results with error detection and error correction.

Lezak’s conceptual model

One of the most widespread conceptual models on executive functions is Lezak’s model. This framework proposes four broad domains of volition, planning, purposive action, and effective performance as working together to accomplish global executive functioning needs. While this model may broadly appeal to clinicians and researchers to help identify and assess certain executive functioning components, it lacks a distinct theoretical basis and relatively few attempts at validation.

Miller & Cohen's (2001) model

In 2001, Earl Miller and Jonathan Cohen published their article 'An integrative theory of prefrontal cortex function' in which they argue that cognitive control is the primary function of the prefrontal cortex (PFC), and that control is implemented by increasing the gain of sensory or motor neurons that are engaged by task- or goal-relevant elements of the external environment. In a key paragraph, they argue:

We assume that the PFC serves a specific function in cognitive control: the active maintenance of patterns of activity that represent goals and the means to achieve them. They provide bias signals throughout much of the rest of the brain, affecting not only visual processes but also other sensory modalities, as well as systems responsible for response execution, memory retrieval, emotional evaluation, etc. The aggregate effect of these bias signals is to guide the flow of neural activity along pathways that establish the proper mappings between inputs, internal states, and outputs needed to perform a given task.

Miller and Cohen draw explicitly upon an earlier theory of visual attention that conceptualises perception of visual scenes in terms of competition among multiple representations - such as colors, individuals, or objects. Selective visual attention acts to 'bias' this competition in favour of certain selected features or representations. For example, imagine that you are waiting at a busy train station for a friend who is wearing a red coat. You are able to selectively narrow the focus of your attention to search for red objects, in the hope of identifying your friend. Desimone and Duncan argue that the brain achieves this by selectively increasing the gain of neurons responsive to the color red, such that output from these neurons is more likely to reach a downstream processing stage, and, as a consequence, to guide behaviour. According to Miller and Cohen, this selective attention mechanism is in fact just a special case of cognitive control - one in which the biasing occurs in the sensory domain. According to Miller and Cohen's model, the PFC can exert control over input (sensory) or output (response) neurons, as well as over assemblies involved in memory, or emotion. Cognitive control is mediated by reciprocal PFC connectivity with the sensory and motor cortices, and with the limbic system. Within their approach, thus, the term 'cognitive control' is applied to any situation where a biasing signal is used to promote task-appropriate responding, and control thus becomes a crucial component of a wide range of psychological constructs such as selective attention, error monitoring, decision-making, memory inhibition, and response inhibition.

Miyake and Friedman’s model of executive functions

Miyake and Friedman’s theory of executive functions proposes that there are three aspects of executive functions: updating, inhibition, and shifting. A cornerstone of this theoretical framework is the understanding that individual differences in executive functions reflect both unity (i.e., common EF skills) and diversity of each component (e.g., shifting-specific). In other words, aspects of updating, inhibition, and shifting are related, yet each remains a distinct entity. First, updating is defined as the continuous monitoring and quick addition or deletion of contents within one’s working memory. Second, inhibition is one’s capacity to supersede responses that are prepotent in a given situation. Third, shifting is one’s cognitive flexibility to switch between different tasks or mental states.

Miyake and Friedman also suggest that the current body of research in executive functions suggest four general conclusions about these skills. The first conclusion is the unity and diversity aspects of executive functions. Second, recent studies suggest that much of one’s EF skills are inherited genetically, as demonstrated in twin studies. Third, clean measures of executive functions can differentiate between normal and clinical or regulatory behaviors, such as ADHD. Last, longitudinal studies demonstrate that EF skills are relatively stable throughout development.

Banich's (2009) "Cascade of control" model

This model integrates theories from other models, and involves a sequential cascade of brain regions involved in maintaining attentional sets in order to arrive at a goal. In sequence, the model assumes the involvement of the posterior dorsolateral prefrontal cortex (DLPFC), the mid-DLPFC, and the posterior and anterior dorsal ACC.

The cognitive task used in the article is selecting a response in the Stroop task, among conflicting color and word responses, specifically a stimulus where the word "green" is printed in red ink. The posterior DLPFC creates an appropriate attentional set, or rules for the brain to accomplish the current goal. For the Stroop task, this involves activating the areas of the brain involved in color perception, and not those involved in word comprehension. It counteracts biases and irrelevant information, like the fact that the semantic perception of the word is more salient to most people than the color in which it is printed.

Next, the mid-DLPFC selects the representation that will fulfill the goal. The task-relevant information must be separated from other sources of information in the task. In the example, this means focusing on the ink color and not the word.

The posterior dorsal anterior cingulate cortex (ACC) is next in the cascade, and it is responsible for response selection. This is where the decision is made whether you will say green (the written word and the incorrect answer) or red (the font color and correct answer).

Following the response, the anterior dorsal ACC is involved in response evaluation, deciding whether you were correct or incorrect. Activity in this region increases when the probability of an error is higher.

The activity of any of the areas involved in this model depends on the efficiency of the areas that came before it. If the DLPFC imposes a lot of control on the response, the ACC will require less activity.

Recent work using individual differences in cognitive style has shown exciting support for this model. Researchers had participants complete an auditory version of the Stroop task, in which either the location or semantic meaning of a directional word had to be attended to. Participants that either had a strong bias toward spatial or semantic information (different cognitive styles) were then recruited to participate in the task. As predicted, participants that has a strong bias toward spatial information had more difficulty paying attention to the semantic information and elicited increased electrophysiological activity from the ACC. A similar activity pattern was also found for participants that had a strong bias toward verbal information when they tried to attend to spatial information.

Assessment

Assessment of executive functions involves gathering data from several sources and synthesizing the information to look for trends and patterns across time and setting. Apart from formal tests, other measures can be used, such as standardized checklists, observations, interviews, and work samples. From these, conclusions may be drawn on the use of executive functions.

There are several different kinds of tests (e.g., performance based, self-report) that measure executive functions across development. These assessments can serve a diagnostic purpose for a number of clinical populations.

3

Experimental evidence

The executive system has been traditionally quite hard to define, mainly due to what psychologist Paul W. Burgess calls a lack of "process-behaviour correspondence". That is, there is no single behavior that can in itself be tied to executive function, or indeed executive dysfunction. For example, it is quite obvious what reading-impaired patients cannot do, but it is not so obvious what exactly executive-impaired patients might be incapable of.

This is largely due to the nature of the executive system itself. It is mainly concerned with the dynamic, "online" co-ordination of cognitive resources, and, hence, its effect can be observed only by measuring other cognitive processes. In similar manner, it does not always fully engage outside of real-world situations. As neurologist Antonio Damasio has reported, a patient with severe day-to-day executive problems may still pass paper-and-pencil or lab-based tests of executive function.

Theories of the executive system were largely driven by observations of patients having suffered frontal lobe damage. They exhibited disorganized actions and strategies for everyday tasks (a group of behaviors now known as dysexecutive syndrome) although they seemed to perform normally when clinical or lab-based tests were used to assess more fundamental cognitive functions such as memory, learning, language, and reasoning. It was hypothesized that, to explain this unusual behaviour, there must be an overarching system that co-ordinates other cognitive resources.

Much of the experimental evidence for the neural structures involved in executive functions comes from laboratory tasks such as the Stroop task or the Wisconsin Card Sorting Task (WCST). In the Stroop task, for example, human subjects are asked to name the color that color words are printed in when the ink color and word meaning often conflict (for example, the word "RED" in green ink). Executive functions are needed to perform this task, as the relatively overlearned and automatic behaviour (word reading) has to be inhibited in favour of a less practiced task - naming the ink color. Recent functional neuroimaging studies have shown that two parts of the PFC, the anterior cingulate cortex (ACC) and the dorsolateral prefrontal cortex (DLPFC), are thought to be particularly important for performing this task.

Context-sensitivity of PFC neurons

Other evidence for the involvement of the PFC in executive functions comes from single-cell electrophysiology studies in non-human primates, such as the macaque monkey, which have shown that (in contrast to cells in the posterior brain) many PFC neurons are sensitive to a conjunction of a stimulus and a context. For example, PFC cells might respond to a green cue in a condition where that cue signals that a leftwards fast movement of the eyes and the head should be made, but not to a green cue in another experimental context. This is important, because the optimal deployment of executive functions is invariably context-dependent. To quote an example offered by Miller and Cohen, a US resident might have an overlearned response to look left when crossing the road. However, when the "context" indicates that he or she is in the UK, this response would have to be suppressed in favour of a different stimulus-response pairing (look right when crossing the road). This behavioural repertoire clearly requires a neural system that is able to integrate the stimulus (the road) with a context (US, UK) to cue a behaviour (look left, look right). Current evidence suggests that neurons in the PFC appear to represent precisely this sort of information. Other evidence from single-cell electrophysiology in monkeys implicates ventrolateral PFC (inferior prefrontal convexity) in the control of motor responses. For example, cells that increase their firing rate to NoGo signals as well as a signal that says "don't look there!" have been identified.

Attentional biasing in sensory regions

Electrophysiology and functional neuroimaging studies involving human subjects have been used to describe the neural mechanisms underlying attentional biasing. Most studies have looked for activation at the 'sites' of biasing, such as in the visual or auditory cortices. Early studies employed event-related potentials to reveal that electrical brain responses recorded over left and right visual cortex are enhanced when the subject is instructed to attend to the appropriate (contralateral) side of space.

The advent of bloodflow-based neuroimaging techniques such as functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) has more recently permitted the demonstration that neural activity in a number of sensory regions, including color-, motion-, and face-responsive regions of visual cortex, is enhanced when subjects are directed to attend to that dimension of a stimulus, suggestive of gain control in sensory neocortex. For example, in a typical study, Liu and coworkers presented subjects with arrays of dots moving to the left or right, presented in either red or green. Preceding each stimulus, an instruction cue indicated whether subjects should respond on the basis of the colour or the direction of the dots. Even though colour and motion were present in all stimulus arrays, fMRI activity in colour-sensitive regions (V4) was enhanced when subjects were instructed to attend to the colour, and activity in motion-sensitive regions was increased when subjects were cued to attend to the direction of motion. Several studies have also reported evidence for the biasing signal prior to stimulus onset, with the observation that regions of the frontal cortex tend to come active prior to the onset of an expected stimulus.

Connectivity between the PFC and sensory regions

Despite the growing currency of the 'biasing' model of executive functions, direct evidence for functional connectivity between the PFC and sensory regions when executive functions are used, is to date rather sparse. Indeed, the only direct evidence comes from studies in which a portion of frontal cortex is damaged, and a corresponding effect is observed far from the lesion site, in the responses of sensory neurons. However, few studies have explored whether this effect is specific to situations where executive functions are required. Other methods for measuring connectivity between distant brain regions, such as correlation in the fMRI response, have yielded indirect evidence that the frontal cortex and sensory regions communicate during a variety of processes thought to engage executive functions, such as working memory, but more research is required to establish how information flows between the PFC and the rest of the brain when executive functions are used. As an early step in this direction, an fMRI study on the flow of information processing during visuospatial reasoning has provided evidence for causal associations (inferred from the temporal order of activity) between sensory-related activity in occipital and parietal cortices and activity in posterior and anterior PFC. Such approaches can further elucidate the distribution of processing between executive functions in PFC and the rest of the brain.

Bilingualism and executive functions

A growing body of research demonstrates that bilinguals show advantages in executive functions, specifically inhibitory control and task switching. A possible explanation for this is that speaking two languages requires controlling one's attention and choosing the correct language to speak. Across development, bilingual infants, children, and elderly show a bilingual advantage when it comes to executive functioning. Interestingly, bimodal bilinguals, or people who speak one language and also know sign language, do not demonstrate this bilingual advantage in executive functioning tasks. This may be because one is not required to actively inhibit one language in order to speak the other. Bilingual individuals also seem to have an advantage in an area known as conflict processing, which occurs when there are multiple representations of one particular response (for example, a word in one language and its translation in the individual’s other language). Specifically, the lateral prefrontal cortex has been shown to be involved with conflict processing.

Future directions

Other important evidence for executive functions processes in the prefrontal cortex have been described. One widely cited review article emphasizes the role of the medial part of the PFC in situations where executive functions are likely to be engaged – for example, where it is important to detect errors, identify situations where stimulus conflict may arise, make decisions under uncertainty, or when a reduced probability of obtaining favourable performance outcomes is detected. This review, like many others, highlights interactions between medial and lateral PFC, whereby posterior medial frontal cortex signals the need for increased executive functions and sends this signal on to areas in dorsolateral prefrontal cortex that actually implement control. Yet there has been no compelling evidence at all that this view is correct, and, indeed, one article showed that patients with lateral PFC damage had reduced ERNs (a putative sign of dorsomedial monitoring/error-feedback) - suggesting, if anything, that the direction of flow of the control could be in the reverse direction. Another prominent theory emphasises that interactions along the perpendicular axis of the frontal cortex, arguing that a 'cascade' of interactions between anterior PFC, dorsolateral PFC, and premotor cortex guides behaviour in accordance with past context, present context, and current sensorimotor associations, respectively.

Advances in neuroimaging techniques have allowed studies of genetic links to executive functions, with the goal of using the imaging techniques as potential endophenotypes for discovering the genetic causes of executive function.

See also

• Cognitive neuropsychology
• Executive dysfunction
• Metacognition
• Nonverbal learning disorder
• Purkinje cell
• Self control
• Supervisory Attentional System (SAS)

External links

• The National Center for Learning Disabilities
• Notes on dysexecutive syndrome and the executive system