What is cognitive science and what does it mean?

What is cognitive science and what does it mean?

Researchers in the field of cognitive sciences study human intelligence and behavior by focusing on how the nervous system displays, processes, and transmits information. The mental parts that are important for researchers in this field include language, perception, memory, attention, wisdom and emotions. To understand these mental parts, cognitive scientists draw on scientific disciplines such as linguistics, psychology, artificial intelligence, philosophy, neuroscience, and anthropology. Conventional analysis in the cognitive sciences extends to different levels of the organization, from learning and decision making to logic and planning, from neural circuits to the modular structure of the brain. The basic concept of cognitive science is that “the process of thinking can be best understood by knowing the representative structures in the human mind and the computational processes that operate on these structures.”

The cognitive sciences began in the 1950s as an intellectual movement often referred to as the cognitive revolution.

Fundamentals of Cognitive Science

Levels of analysis

The key principle of the cognitive sciences is that a full understanding of the mind or brain is not possible by studying one level alone. An example of this is the difficulty of forgetting a phone number and remembering it at another time. One approach to understanding this process is to study behavior through direct observation, or in other words, natural observation. A person can be given a phone number and asked to remember it after a period of time. In this way, the accuracy of the answer can be measured. Another approach to measuring cognitive ability is to study the fire of neurons (the activation of neurons by the stimulus reaching a threshold value) while a person is trying to remember a phone number. Neither of these two experiments alone can justify the process of remembering a person’s phone number. Even if we have the technology to map and identify all the neurons in the brain instantly and know when to fire each neuron, still knowing how to fire a certain arrangement of neurons leads to the observed behavior. It will be impossible. Therefore, it is imperative to understand how these two levels are related. “The new sciences of the mind must broaden their horizons to include both human experiences and the potential for inherent change in human experiences,” says the book Visualization of the Mind: Cognitive Sciences and Human Experience. . This can be done through the application leveling of this process. Studying a particular phenomenon from several different levels helps us better understand the processes that take place in the brain that trigger a particular behavior. The British neuroscientist David Marr described his three levels of analysis as follows:

  1. Computational theory of mind: Defining computational goals
  2. Displays and Algorithms: Provides an overview of input and output data and algorithms that convert input to output
  3.  Hardware implementation: How algorithms and data representations are physically realized

Interdisciplinary nature

Cognitive science is an interdisciplinary field in which researchers from different disciplines such as psychology, neuroscience, linguistics, philosophy of mind, computer science, anthropology, sociology and biology work. Cognitive scientists work together in the hope of understanding the mind and its interactions with the world around it, similar to what other sciences do. This field of science considers itself compatible with the physical sciences and uses the scientific method along with simulation or modeling and often compares the output of the models with different aspects of human cognition. Similar to psychology, there are doubts as to whether there is an integrated science called cognition. This has led some researchers to prefer to use the term cognitive science.

Many people who consider themselves behavioral scientists (but not all of them) look at the human mind with a functionalist approach. The functionalist approach refers to the view that mental states and processes must be described through their function. According to the theory of multiple feasibility, one of the reasons for the functionalist approach is that knowledge can even be attributed to non-human systems such as robots or computers.

The term cognitive science

The term “cognitive” is used in the “cognitive sciences” to mean “any kind of operation or mental structure that can be accurately studied.” This conceptualization is very broad and should not be confused with the word “cognitive”, which is used in some traditions of analytic philosophy and means “cognitive”, formal rules and semantics.

The first recorded meanings for the word “cognitive” in the Oxford English Dictionary describe it as “related to the act or process of knowing”. The first recorded entry, from 1586, indicates that the term was once used in discussions of Platonic theories of knowledge. However, many people in the field of cognitive science do not believe that their field is the study of something with certainty, similar to the knowledge that Plato sought.

Cognitive science range

Cognitive science is a broad scientific field and covers various topics in the field of cognition. It should be noted, however, that it has not always been equally important for any subject that may be related to the nature and operation of the mind. Among philosophers, classical activists in the field of recognizing the effects of social and cultural factors, emotions, self-awareness, recognizing the capacities of the animal mind, and adaptive and evolutionary psychology have underestimated or completely ignored them. However, with the decline of behaviorism, inner states such as emotions as well as consciousness and mental attention were re-accessed and used. For example, theories of bodily cognition or cognition depend on the current state of the environment as well as the role of the body in cognition. With the emerging emphasis on information processing, observable behavior was no longer considered the basis of psychological theories and was replaced by modeling or recording mental states.

The following are some of the issues that cognitive science deals with. This list is not exhaustive.

Artificial intelligence

Artificial intelligence (AI) involves the study of cognitive phenomena in machines. One of the practical goals of AI is to implement aspects of human intelligence in computers. Computers are also used as a tool to study cognitive phenomena. Computer modeling uses simulation to study how human intelligence is structured.

In this context, it is usually debated whether it is better to think of the mind as a huge array of small and weak elements (neurons) or as a set of higher level structures such as symbols, designs. , Programs and rules collided. The first approach uses connectivity to study the mind, while the second approach emphasizes symbolic computations. One way to look at this is to look at whether it is possible to accurately simulate the human brain in a computer without accurately simulating the neurons that make it up.


Paying attention to the meaning of choosing information is important. The human mind is bombarded with millions of stimuli and must have a way of choosing the information to be processed from among other information. Attention is sometimes thought of as a spotlight; This means that it can only focus on a specific set of information. Experiments that support this analogy include the dichotomous hearing test (cherry, 1957) and the Unwanted Blindness Research Series (Mac and Rock, 1998). In a dichotomous hearing test, test subjects are bombarded with two different sets of voice messages, each in one ear, and asked to focus on only one of these messages. At the end of the experiment, when people are asked about the content of messages they did not pay attention to, they could not say it.

Knowledge and language processing

The ability to learn and understand a language is an extremely complex process. Language is acquired in the first few years of life and can be mastered by all normal people. One of the drivers of theoretical linguistics is finding the nature that language must have in abstraction in order to be learned in its current order. Some of the main research questions that arise in examining how language is processed by the brain include the following: 1. To what extent is language knowledge innate or learnable? Why is learning a secondary language more difficult for adults than learning a mother tongue for infants? 3- How do people understand new sentences?

The study of language processing covers a wide range, from the study of speech patterns to the meaning of complete words and sentences. Linguistics often divides language processing into different categories called grammar, phonetics, phonology, monosyllabic, semantics, semantics, and application. Using each of these components and their interaction with each other, many aspects of language can be explored.

The study of language processing in the cognitive sciences is closely related to linguistics. Linguistics has traditionally been studied as part of the humanities, including the study of history, art, and literature. In the last fifty years, more and more researchers have studied the knowledge and use of language as a cognitive phenomenon. The main problem here is how language knowledge is created, how it is used, and exactly what it consists of. Linguists have found that although humans form sentences using ways controlled by very complex systems, they themselves are unaware of the laws that govern their speech. As a result, linguists must use indirect methods to determine these rules, if any.

Learning and growing

Learning and growth are processes through which we acquire knowledge and information over time. Babies are born with little or no knowledge (depending on how knowledge is defined), but they quickly learn the ability to use language, walk, and recognize people and objects. The purpose of research on learning and growth is to describe the mechanisms through which these processes take place.

One of the main questions in the field of cognitive development is to what extent some abilities are innate or learnable. This question is often asked in the form of nature. Essentialists emphasize that certain traits are inherent in an organism and are determined by genetic inheritance. Empiricists, on the other hand, emphasize that some abilities are learned through the environment. Although it is clear that both genetic and environmental inputs are needed for a child to grow normally, there is still much debate about how genetic information can guide cognitive development. For example, in the field of language acquisition, some (such as Steven Pinker) believe that genes should contain specific information, including the laws of global grammar, while others (such as Jeffrey Inman and colleagues in Rethinking Innateness) believe that Pinker’s claims They are not biologically realistic. They believe that genes determine the architecture of a learning system, but certain facts about how grammar works can only be learned through experimentation.


Memory allows us to store information and re-access it at another time. It is often thought that memory has two locations for storing information in the short and long term. Long-term memory allows us to store information for long periods of time (days, weeks and years). We do not yet know the practical limit of long-term memory capacity. Short-term memory allows us to store information on short time scales (seconds and minutes).

Memory is often categorized as news memory and process memory. News memory, which is itself subdivided into semantic memory and event memory, is for specific facts and knowledge, specific meanings, and specific experiences (for example, to answer the question of who was the first president of the United States? Or four days before breakfast. I ate?) Refers to memory. Process memory allows us to recall tasks and movement sequences (such as cycling) and is often referred to as knowledge or intangible memory.

Cognitive scientists study memory similarly to psychologists, but focus more on the effect of memory on cognitive processes and the interrelationship between memory and cognition. An example of this relationship is what mental processes must a person go through to recall a memory they have long forgotten? Or what is the difference between the cognitive process of recognizing (seeing signs of something before remembering it, or background memory) and remembering (re-reading a memory, for example to fill in blanks)?

Perception and action

Perception is the ability to obtain information through the senses and process them in a specific way. Sight and hearing are the two dominant senses that allow us to understand the environment. For example, some of the questions posed in the study of visual perception include: 1. How can we recognize objects? 2. Why do we perceive a continuous visual environment when we see only small parts of it at a given time? One of the tools available to assess visual perception is to examine how individuals process visual error. The image below is a necker cube that is an example of dual perception; This means that the cube can be seen in two different directions.

The study of tactile, olfactory and taste stimuli is also in the realm of perception.

Here the action is the output of a system. In humans, this happens through motor responses. Spatial planning and movement, complex speech and movement are all aspects of action.


Self-awareness means knowing whether something is an external object or something inside. Self-awareness helps the mind to be able to experience or feel independent.

research methods

Different methodologies are used in the study of cognitive sciences. Given that this field is highly interdisciplinary, research in this field is also related to research methods used in psychology, neuroscience, computer science, and systems theory.

Behavioral experiments

To describe what constitutes intelligent behavior, a researcher must examine the behavior itself. This type of research is closely related to cognitive psychology and psychophysics. By measuring behavioral responses to different stimuli, the researcher can gain a better understanding of how these stimuli are processed. Lewandowski and Stromtz (2009) examined a creative set of applications of behavioral measurement in psychology, including behavioral effects, behavioral observations, and behavioral choices. Behavioral effects are pieces of evidence that indicate that a behavior has occurred but that the person performing the behavior is not present at the scene (such as garbage dumped in a parking lot or meter readings). Behavioral observations involve directly observing the person performing the desired behavior (such as observing how close one person sits to another).

  • Reaction time: The time between creating the stimulus and responding appropriately to it can show the differences between the two cognitive processes and also reveal things about their nature. For example, if during a search, the reaction time changes in proportion to the number of elements being searched, it is clear that this cognitive search process uses sequential processing instead of parallel processing.
  • Psycho-physical answers: Psycho-physical experiments are an old psychological technique that is also used in cognitive psychology. These experiments often involve judging a physical property, such as how loud a sound is. The relationship of subjective scales between different individuals can show cognitive or sensory biases compared to actual physical measurements. Examples include:
  1. Judge similarity for colors, sounds, textures and…
  2. Difference thresholds for colors, sounds, textures and…
  • Eye tracking: This methodology is used to study different types of cognitive processes, especially visual perception and language processing. In this method, the fixed point of the eyes is related to the focus of a person’s attention. As a result, by monitoring the movement of the eyes, we can study what information is being processed at a given time. Eye tracking makes it possible to study cognitive processes on very short time scales. Eye movements represent the momentary decision-making process of a task and give us a better understanding of the ways in which these decisions are processed.

Brain imaging

Brain imaging involves analyzing the amount of activity inside the brain while performing various tasks. This helps us link the behavior and function of the brain to understand how information is processed. Different imaging methods have different temporal and spatial resolution. Brain imaging is mainly used in cognitive neuroscience.

  • Single-photon computed tomography (SPECT) and positron emission tomography (PET): SPECT and PET use radioactive isotopes that are injected into the bloodstream and absorbed by the brain. By looking at which parts of the brain absorb the radioactive isotope, we can see which parts of the brain are more active than the other. PET has the same spatial resolution as fMRI but its temporal resolution is very poor.
  • EEG: An EEG measures the electric fields produced by a large number of neurons in the cortex by placing a set of electrodes on the scalp. This method has a very high temporal resolution but its spatial resolution is relatively poor.
  • fMRI: This method measures the relative amount of oxygenated blood flowing to different parts of the brain. It is assumed that oxygenated blood in a particular area is more related to increased neuronal activity in that area. This allows specific functions to be localized within different parts of the brain. fMRI has moderate temporal and spatial resolution.
  • Optical Imaging: This method uses infrared transmitters and receivers to measure the amount of light reflected by the blood near different parts of the brain. Given that oxygenated and deoxygenated blood reflects light in varying amounts, we can study the parts that are more active (ie, those with more oxygenated blood). Optical imaging has a moderate temporal resolution but poor spatial resolution. This method also has the advantage that it is very safe and can be used to study the brain of infants.
  • MEG: This method measures the magnetic fields caused by the activity of the cerebral cortex. This method is similar to an electroencephalogram, but the difference is that MEG has a better spatial resolution because the magnetic fields measured in this method are as much as the electric fields measured in the electroencephalogram by the scalp, meninges, and The components are no longer blurred or weakened. MEG uses SQUID sensors to detect small magnetic fields.

Computational modeling

The need to create computational models requires a complete expression of the problem mathematically and logically. Computer models are used in simulation and experimental validation of general and specific properties of intelligence. Computational modeling can help us understand the function of a particular cognitive phenomenon. In cognitive modeling, there are two general approaches. The first approach focuses on the abstract mental functions of an intelligent mind and operates using signs and symptoms. The second approach focuses on the neural and participatory properties of the human brain and is called subsymbolic modeling.

  • Sign modeling evolved with the help of similar examples in computer science and using knowledge base systems technologies as well as philosophical perspectives (for example, good old artificial intelligence – GOFAI). These models were developed by the first cognitive scientists and later used in information engineering for expert systems. Since the early 1990s, these models have been generalized to systems science to be used to study applied models of human-like intelligence, such as personoids. In parallel, this model was used as the Soar environment (a kind of cognitive architecture). Recently, especially in the field of cognitive decision making, cognitive modeling has also expanded into a cognitive social approach. In this approach, organizational and social cognition are interconnected through a subconscious subconscious layer.
  • The following sign modeling includes connection-oriented models or neural networks . Connectionism is based on the idea that the mind or brain is made up of simple nodes and that the power of this system is largely due to the existence and manner of connections between simple nodes. Neural networks are the exact implementation of this approach. Some critics of this approach feel that although these models use biological reality as a demonstration of how they work, they lack explanatory power because complex systems of connections, even with simple rules, are extremely complex and often uninterpretable. They are more than the system they are supposed to model.
  • Other approaches that are becoming more popular include: 1- Using dynamic systems theory; and 2- Methods that adapt signal models and connection-oriented models (neural-signal integration or hybrid intelligent systems). ) Are. Bayesian models, derived from machine learning , are also becoming more popular.

All of the above approaches tend to be generalized in the form of integrated computational models of synthetic or abstract intelligence so that they can be used to explain and improve individual and socio-organizational decisions and arguments.

Neuro-biological methods

Research methods borrowed directly from neuroscience and neuropsychology can also help us understand aspects of intelligence. These methods allow us to understand how intelligent behavior is implemented in a physical system.

  • Single unit registration
  • Direct brain stimulation
  • Animal models
  • Autopsy studies

Main findings

Cognitive science has led to the development of models of cognitive bias and risk perception and has also played an important role in the development of behavioral economics, which is part of the economy. In addition, the cognitive sciences have given rise to new theories of the philosophy of mathematics, and many theories of artificial intelligence, persuasion, and coercion. The effect of this science on the philosophy of language and epistemology is not hidden from anyone. In addition, the cognitive sciences are one of the wings of modern linguistics. The fields of cognitive science have been instrumental in understanding the function of specific brain systems (or functional defects), from speech ability to auditory and visual perception processes. These sciences have advanced our understanding of how damage to certain parts of the brain also affects cognition, and have also helped us discover the roots and consequences of disorders such as dyslexia, anopia, and unilateral neglect. اند.

History of Cognitive Sciences

The cognitive sciences began in the 1950s as an intellectual movement often referred to as the cognitive revolution. Cognitive science has a long history dating back to the texts of ancient Greek philosophy (such as Menon Plato or Aristotle) ​​but also includes authors such as Descartes, David Hume, Kant, Baruch Spinoza, Nicolas Malbransch, Pierre Cabanis, Leibniz, and John Locke. Is. However, although these writers greatly contributed to the philosophical discovery of the mind, and this eventually led to the development of psychology, they came up with a set of basic tools and concepts different from those of cognitive scientists. they did.

The modern culture of the cognitive sciences can be traced back to early cybernetic scientists such as Walter McClock and Walter Pitts in the 1930s and 1940s. They tried to understand the organizing principles of the mind. McCullough and Pitts developed the first computational models inspired by the biological structure of neural networks, what are now called artificial neural networks.

Another innovation was the development of computational theory and digital computers in the 1940s and 1950s. Kurt Goodell, Alonzo Church, Alan Turing and John von Newman were key figures in this. The modern computer, or von Neumann machine, later played a major role in the cognitive sciences, both as a metaphor for the mind and as a tool for inquiry.

The first cognitive science experiment was conducted at an academic institution, MIT, School of Business. This work was created by Joseph Carl Leklider, who worked in the Department of Social Psychology and conducted experiments on computer memory as models of human cognition.

In 1959, Noam Chomsky published a scathing critique of BF Skinner’s book Verbal Behavior. At the time, Skinner’s behaviorist model dominated American psychology. Most psychologists focused on the functional relationship between the stimulus and the response, regardless of internal perceptions. To explain language, Chomsky argued, we need a theory such as reproductive grammar that takes into account not only internal perceptions but also the order in which they exist.

Terminology was first used by Christopher Langett Higgins in 1973 in his interpretation of the Light Hill report, which examined the state of artificial intelligence at the time. In the same decade, the journal Cognitive Science and the Cognitive Science Association were formed. The Cognitive Science Association convened in 1979 at the University of California, San Diego, and led to the emergence of international cognitive science. In 1972, Hampshire College introduced the first undergraduate program in cognitive science, led by Neil Stillings. In 1982, with the help of Professor Stillings, Wasser College became the first institution in the world to award undergraduate degrees in cognitive science to students. In 1986, the world’s first department of cognitive science was established at the University of California, San Diego.

In the 1970s and early 1980s, with increasing access to computers, research into artificial intelligence also expanded. Researchers like Marvin Minsky have written computer programs in languages ​​such as LISP to formally describe the steps human beings take, for example in decision-making or problem-solving, so that they can better understand the human mind. Understand. Their other motivation was the hope of building artificial minds. This approach is called Symbolic AI.

Gradually, the limitations of the AI ​​program became apparent. For example, listing human knowledge comprehensively in a way that could be used by a token computer program seemed unrealistic. In the late 1980s and 1990s, neural networks and connectivity emerged as research models. From this perspective, attributed to James McClelland and David Rommel Hart, the mind can be described as a set of complex connections, represented as a layered network. Critics argue that there are phenomena that are better described by sign models, and that connectionist models are often so complex that they have little explanatory power. Recently, signal and connection-oriented models have been combined, so that both types of expression can be used. Although both symptomatic and connection-oriented approaches have been useful in examining different hypotheses and approaches in order to understand aspects of lower-level cognition and function in the brain, neither is biologically close to reality. As a result, none of them are reliable in terms of neuroscience. Connectiveness has been computationally useful in discovering how to create cognition in the development of models and in the human brain, and provides a different approach to domain-specific or general approaches. For example, scientists such as Jeff Elman, Liz Bates, and Ant Karmilov-Smith have shown that brain networks arise from the dynamic interaction between them and environmental inputs. Connectiveness has been computationally useful in discovering how to create cognition in the development of models and in the human brain, and provides a different approach to domain-specific or general approaches. For example, scientists such as Jeff Elman, Liz Bates, and Ant Karmilov-Smith have shown that brain networks arise from the dynamic interaction between them and environmental inputs. Connectiveness has been computationally useful in discovering how to create cognition in the development of models and in the human brain, and provides a different approach to domain-specific or general approaches. For example, scientists such as Jeff Elman, Liz Bates, and Ant Karmilov-Smith have shown that brain networks arise from the dynamic interaction between them and environmental inputs.

The main researchers

name Year of Birth Year of service Service
Daniel Dennett 1942 1987 Outlook for Computing Systems
John Sarl 1932 1980 Chinese room
Jerry Fodor 1935 1968, 1975 Functionalism
David Chalmers 1966 1995 Dualism, the difficult issue of self-awareness
Douglas Hofstadter 1945 1979 Book of Godel, Asher, Bach
Marvin Minsky 1927 70s and early 80s Writing computer programs in languages ​​such as LISP to formally describe the steps a person takes to perform processes such as decision-making or problem-solving
Christopher Longt-Higgins 1923 1973 Invention of the word cognitive sciences
McCullough and Pitts 30s and 40s Development of the first artificial neural networks
Lyclider 1915 Establishment of MIT School of Management
Noam Chomsky 1928 1959 The publication of a critique of the book Verbal Behavior by Skinner, which began the current of epistemology as a rival to the dominant current of behaviorism.


Some of the most well-known names in the cognitive sciences are often the most controversial or have the most references. Within philosophy, some familiar names include Daniel Dennett, who writes from the perspective of computing systems, John Searle, who is known for his controversial china argument, and Jerry Fodor, who advocates functionalism.

Others include David Chalmers, a proponent of dualism and the problem of self-consciousness, and Douglas Hofstadter, best known for writing G گdel, Asher, and Bach, which questioned the nature of words and thoughts. .

In the realm of linguistics, Noam Chomsky and George Lykoff have been influential (both of whom have become significant political analysts). In artificial intelligence, Marvin Minsky, Herbert Simon and Ellen Newell are key figures.

Popular names in the field of psychology include George Miller, James McClelland, Philip Johnson-Lord and Steven Pinker. Anthropologists such as Dan Esperber, Edwin Hutchins, and Scott Ethern have collaborated on projects with cognitive and social psychologists, political scientists, and evolutionary biologists to provide general theories on the formation of culture, religion, and political affiliation.

Computational theories (with models and simulations) have also been developed by the likes of David Rummelhart, James McClelland, Philip Johnson-Lord, and others.

Other services have been provided by Marvin Minsky and Noam Chomsky.