Thoughts Into Action Executive Function PDF

Summary

This document provides insights into the concept, history, and neural substrates of executive function. It explores how cognitive control, a component of executive function, plays a role in organizing complex tasks. The document also analyzes the computational representation and different aspects of executive function via various examples and studies.

Full Transcript

07 March 2024 12:54 Main Ideas Notes Notes Executive Function Executive function is an umbrella term referring to a set of cognitive functions that underlie the flexible and adaptive aspects of 'intelligent' behaviours Exec. Functions - transform current context and current thought into appropriate...

07 March 2024 12:54 Main Ideas Notes Notes Executive Function Executive function is an umbrella term referring to a set of cognitive functions that underlie the flexible and adaptive aspects of 'intelligent' behaviours Exec. Functions - transform current context and current thought into appropriate action History behind investigating executive function Families of executive function Studies focus on specific cognitive processes assumed to be important for EF Cognitive process had a general property of going beyond mere S -R organisation ○ Such studies rarely describe how these processes contributed to executive function Descriptive characterization of individual cognitive components did not provide an overall cognitive theory of executive func tion Retaining a stimulus in working memory Working memory is the ability to hold an item of information transiently in mind in the service of comprehension, thinking, a nd planning Delay-period activity Reorganising S-R pattern rules Control processes reconfigure mental resources for a change of task by requiring subjects to switch frequently among a small set of simple tasks Task-switching Organising complex chains of action Every sensorimotor action is embedded within a complex set of contexts, which vary across episodes, and have a hierarchical branching structure Classical neuropsychology ○ WM/Undifferentiated - "Central Executive" ○ MET: Complex behaviour ○ Phineas Gage ○ Limited viability of functional localisations Neurophysiology/Neuronal Coding ○ "Mixed Selectivity" tuning ○ High-dimensional spaces ○ Multiple demand system AI: Computation Representation ○ Hierarchy of abstraction ○ Hierarchy of control ○ Topography-like frontal connectivity Motivated Behaviour ○ Convergence of subcortical loops ○ Substance use ○ Stimulation evokes "urge" experiences Neuropsychology of frontal lobes Elusive in any single cognitive test, but strong involvement in everyday activities Organized vs. disorganized behaviour Complex tasks > simple tasks Refraining from a pre-potent response Eriksen-Flanker Task Cognitive control Post-error slowing: ○ Response conflict is managed by a (slow) control process ○ Following an error, this process becomes especially important Conflict adaptation Making sense of lists of executive functions list of “executive functions” seems arbitrary and disjointed But frontal (or frontoparietal) circuits are generally involved Factor analyse deficits of patients with frontal damage to identify minimum set of concepts/processes required Analysis based on TASKS, aiming to identify core PROCESSES ○ Results ▪ Energisation ▪ Task-setting ▪ Monitoring (no inhibition) Cognitive Theory Classical cognitive theories of executive function are often homuncular, and fail to explain the source of control signals Integrated intelligence from distributed brain activity Segment complex problem by analysis into a series of subgoals Integrate previous knowledge, current stimuli and current subgoal in focussed attention (problem then seems to solve itself a lmost automatically) Summary “Executive functions” is a category that arose historically as a postulate or theoretical construct to explain the deficits o f patients with frontal lesions Its conceptual history has been limited by the ontological difficulty of identifying and characterising individual cognitive functions within the EF family The general unifying concept of cognitive control risks being homuncular, unless carefully spelled out in cognitive theory We investigated a new approach from segment/integrate theories, inspired by ‘good old -fashioned AI’ The neural substrates of EF seem to lack the specialized neurons that characterise primary input/output areas. Complex/mixed codes predominate PSYC0032 The Brain in Action Page 1 Notes Matrix tests and fluid intelligence Segmentation of the problem Integration "Evidently, information is needed on arrow direction in each cell of the matrix. This direction information must be correctly bound to positions within each figure and in the matrix as a whole. The layout of left and right arrows must be related to an overall conception of the problem to be solved and what it means for the solution to ‘look right’." Multiple demand system Five frontal nodes of an interconnected multiple demand system These areas seem to be activated for almost every cognitive task, and are more activated the more the task involves segment/integrate processes Individual frontal and prefrontal neurons show "mixed selectivity tuning" Neurons do not have simple tuning to a single dimension Instead, neurons are tuned to novel but currently relevant conjunctions of input, knowledge and task. Neuron may respond best to some specific combinations of task and cue. High-dimensional neural representations encode information that is nonlinearly mixed in a conjunction format suitable for local ne ural processing. Challenges for segment/integrate theories Has the segment/integrate model been sufficiently freed from the focus and constraints of specific cognitive tasks? Are there any tasks in which the brain’s MD network is not activated? Is that a concern for falsifiability? Is the MD network sufficiently differentiated from other global workspaces and brain modes? Are mixed selectivity neurons just data-mining? Is it a problem that apparently complexifying high-dimensional neural solutions underlie an apparently simplifying cognitive process of segmentation? A graph of brain activity It is specifically about a Cognitive control is a se Conflict resolution is a s options. The different sections of ○ Task (Stroop, Flan ○ Condition (Congru incongruent (e.g., (reading the word ○ Block: Different bl The graph shows that th suggests that these brain Notes y related to cognitive control processes. study on task-specific neural processes underlying conflict resolution. t of mental processes that allow us to control our thoughts, emotions, and behavior in order to achieve goals. pecific aspect of cognitive control that refers to the ability to choose the most appropriate response when faced with competing f the image show brain activity (z-scored high gamma power) in different conditions: nker, Number): These refer to different cognitive tasks used in the study. uent, Incongruent):Whether the stimuli presented in each task were congruent (e.g., the word "RED" written in red ink) or the word "RED" written in blue ink). Incongruent stimuli require more conflict resolution because the automatic response d) conflicts with the correct response (saying the colour of the ink). locks of trials within each task. here was greater high gamma power in some brain regions during incongruent conditions compared to congruent conditions. This n regions were more active when participants had to resolve conflict during the tasks. Response times appear to be faster in cued conditions compared to uncued conditions for all three groups. Variation across tasks, but invariance within tasks Different stimuli within Stroop task: electrode shows sensitivity to inhibitory processing, irrespective of what the stimulus is and what motor response must be inhibited “the neural mechanisms underlying cognitive control generalize across different conditions within a task but are not task invariant” No “pure” processes for specific exec. functions like “inhibition” Instead, neurons may have functions that are nested within a task – could these population recordings reflect “mixed selectivity” neurons again PSYC0032 The Brain in Action Page 2

Use Quizgecko on...
Browser
Browser