What do you think?
Rate this book
ذهن: درآمدی بر علوم شناختی
Cognitive science approaches the study of mind and intelligence from an interdisciplinary perspective, working at the intersection of philosophy, psychology, artificial intelligence, neuroscience, linguistics, and anthropology. With Mind, Paul Thagard offers an introduction to this interdisciplinary field for readers who come to the subject with very different backgrounds. It is suitable for classroom use by students with interests ranging from computer science and engineering to psychology and philosophy.
Thagard's systematic descriptions and evaluations of the main theories of mental representation advanced by cognitive scientists allow students to see that there are many complementary approaches to the investigation of mind. The fundamental theoretical perspectives he describes include logic, rules, concepts, analogies, images, and connections (artificial neural networks). The discussion of these theories provides an integrated view of the different achievements of the various fields of cognitive science.
This second edition includes substantial revision and new material. Part I, which presents the different theoretical approaches, has been updated in light of recent work the field. Part II, which treats extensions to cognitive science, has been thoroughly revised, with new chapters added on brains, emotions, and consciousness. Other additions include a list of relevant Web sites at the end of each chapter and a glossary at the end of the book. As in the first edition, each chapter concludes with a summary and suggestions for further reading.
Thagard's systematic descriptions and evaluations of the main theories of mental representation advanced by cognitive scientists allow students to see that there are many complementary approaches to the investigation of mind. The fundamental theoretical perspectives he describes include logic, rules, concepts, analogies, images, and connections (artificial neural networks). The discussion of these theories provides an integrated view of the different achievements of the various fields of cognitive science.
This second edition includes substantial revision and new material. Part I, which presents the different theoretical approaches, has been updated in light of recent work the field. Part II, which treats extensions to cognitive science, has been thoroughly revised, with new chapters added on brains, emotions, and consciousness. Other additions include a list of relevant Web sites at the end of each chapter and a glossary at the end of the book. As in the first edition, each chapter concludes with a summary and suggestions for further reading.
344 pages, Paperback
First published June 1, 1996
52 people are currently reading
681 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 21 of 21 reviews
February 8, 2018
توضیح کتاب کاملا همون اسم کتابه
درآمدی بر علوم شناختی!
اگر هیچ ایده ای ندارید علوم شناختی چیه و یا دوست دارید بدونید ذهن چطور کار میکنه و تفکر به چه معناست کتاب مقدماتی و خوبی هستش
اگر زبانتون خوبه توصیه میکنم زبان اصلی بخونید. ترجمه ش درک یه سری مفاهیم رو سخت میکرد.
درآمدی بر علوم شناختی!
اگر هیچ ایده ای ندارید علوم شناختی چیه و یا دوست دارید بدونید ذهن چطور کار میکنه و تفکر به چه معناست کتاب مقدماتی و خوبی هستش
اگر زبانتون خوبه توصیه میکنم زبان اصلی بخونید. ترجمه ش درک یه سری مفاهیم رو سخت میکرد.
June 22, 2020
While this book helped me understand cognitive science as an academic discipline, it didn’t really help me understand the mind. Perhaps it’s simply too broad and too general to helpful to me. I think I’ll stick with more specialized books on neuroscience, neurophilosophy, and neuropsychology moving forward.
June 11, 2021
A nice little intro to cog science-y things. Not in depth. Neither talks much from a psychological perspective. But a nice overall computational overview of how things work (or atleast seem to).
April 7, 2017
Excellent textbook ^-^
June 19, 2024
This is my first time reading a textbook on Cognitive Science. It offers an introduction to the subject that is thoroughly organized. Although it can be useful to anybody who is attracted to this field, it is mainly directed at undergraduate Cognitive Science majors.
March 5, 2024
Amazon
Highschool reading
What is the difference between psychology and cognitive science? The difference is pedagogical as far as undergrad is concerned. However at the grad level the assumed background of Cog-sci researchers includes more much more mathematical maturity and familiarity with computer science.
To oversimplify, psychology is qualitative while cog-sci is quantitative.
CONS:
1 Everything is motivated by singular examples (Should be theory then illustrated with multiple examples)
2 Basically just tells you that to be a cogn researcher you're going to have to learn math/comp-sci
3 Subtle ideas are not pointed out
4 No idea who the target audience is
5 Way too short on such a deep subject. What could you possibly teach in such a short book?
Why is this book in my uni's bookstore? Why did I make the mistake of buying it? This book is about 230 pages. That should just about tell you everything you need to know about this book.
A subject as deep in theory and application as cognitive science only being allocated 230 pages!? No. This book doesn't explain anything, and doesn't even try.
This book doesn't teach anything about cog-sci.
I think the authors goal was just to make aware cog-sci students of the scientific fields they have ahead of them; if you want to be a grad level researcher then prepare to learn really advanced math. But also skip this text entirely.
It is an introduction to who did what and which paper is worth reading, so in that sense it is a good reference collection. I would give you an excerpt to illustrate the fact that most of this book leaves much unexplained, but whats the point?
So many damn examples, that an excerpt wouldn't even make sense unless I explained the chapter-long analogy he is working with.
So, Thagard describes 'concepts' as containing 'slots' which are basically propositions associated with the concept (like 'desk' is a slot associated with 'professor') and that these associations are stereotypical generalizations of an individuals empirical experience/observations.
In other words the association between concepts and their slots is one of a statistical natural (the stereotypical aspect) but is also multivalued (many concepts can be associated with many slots and vice versa).
Okay that's great, its certainly logically consistent.
The issue is that it is not formulated at a sufficient level of detail for rigorous analysis.
In fact the entire book is just vague and shallow in terms of scientific exposition.
For example, after reading the parts on these concepts and slots, the following questions are clearly unanswered:
a. How do you measure a slots association with a concept? Is the slot "PHD" in stronger association then the slot "desk" with the concept "professor"?
b. Can a given slot belong to multiple concepts? How do slots map to multiple concepts?
c. Does the brain store a finite number of slots per concept, or does it somehow afford a infinite number of slots per concept?
d. Can we identify a concept with just its slot description?
To me these are the first things I thought of when I was reading his chapter on concepts.
And because he didn't explain them, I just put the book down.
What can I do with slots if I don't know anything about their (mathematical) relationship to concepts?
You might think I just have a penchant for formalism, but in reality, formalism is what cogsci is all about. To say it bluntly, anything less than a formal (mathematical) model for a scientific theory is unacceptable.
Thagaard's book may convince you that this is indeed characteristic of cogsci as a science, but you will be no further educated in that endeavour beyond a superficial awareness of mere terminology.
gakusya
September 12, 2016
It seems MIT Press is the publisher to look for in Cognitive Science.
Quite enjoyable beginner’s text book. It covers some aspects that none of the ones I’ve read so far do, such as: emotion, society, and dynamic systems.
Proposes these criteria for evaluating Theories of Mental Representation:
1.Representational Power
2.Computational Power
1.Problem Solving
i. Planning
ii. Decision
iii. Explanation
2.Learning
3.Language
2.Psychological plausibility
3.Neurological plausibility
4.Practical applicability
1.Education
2.Design
3.Intelligent Systems
4.Mental Illness
Which is interesting, because 1&2 categorize capabilities, 3 & 4 plausibility and 5 applicability.
Pg. 39. Logic. Neurological plausibility.
“Brain scanning experiments are being used to determine whether people perform deductions using the left half of their brain, as suggested by mental logic view that deduction is formal and independent of the content. The alternative hypothesis is that people perform deductions using the right half of the brain, as suggested by the mental models view that deduction requires regions of the brain that involve spatial reasoning.
Pg. 46. Rules. Problem Solving.
Rules have very simple If-Then structure.
Heuristic are rules of thumb that contribute to satisfactory solutions without considering all possibilities.
Pg. 49. Rules. Learning.
In inductive generalization, rules are formed from examples; but rules can also be formed from other rules by a process that in the SOAR model is called chinking and in the ACT model is called composition.
SOAR. State Operator and Result. A computational theory of thinking.
ACT Adaptive Control of Thought. Another computational theory of thinking.
The more a rule gets used successfully, the more likely is to be used in the future.
Pg. 51. Rules. Language
Chomsky continues to maintain that every human is born with an innate universal grammar. Contrary to his initial beliefs about children acquiring the ability to use language abductively by forming hypothesis about what rules to apply to their individual language (1972), he currently holds that children learn a language automatically by merely recognizing which of a set of finite set of possibilities that language employs (Chomsky 1988)
Pg 67. Concepts. Learning.
Consider the example of human face consisting of two eyes, a nose and a mouth. Perhaps babies learn this concept from experience as they repeatedly encounter examples of faces. But there is experimental evidence that babies do not have to learn the typical structure of faces, but rather are born expecting faces to look a certain way.
Pg 89. Analogies. Psychological plausibility.
Metaphorical interpretation appears to be an obligatory process that accompanies literal processing, rather than an optional process that occurs after literal processing.
Analogies. Neurological Plausibility.
Boroojerdi (2001) found that the left prefrontal cortex is involved in analogical reasoning by determining that magnetic stimulation of that part of the brain speeds up solution times for solving analogical problems. This is consistent with recent findings that reasoning involving complex relations, which is crucial for analogical thinking, also involves the left prefrontal cortex (Christoff 2001)
Pg. 106. Images. Neurological plausibility.
The areas of the brain most immediately connected to the retina have a spatial organization that is structurally similar to that of the retina. Since these areas preserve some of the spatial structure of objects presented to the retina (I recall reading somewhere that exposure to say, a column, produces neural activity in an area with some shape of a column). Kosslyn (2001) review neurological studies of visual, auditory and motor imagery.
Images. Summary.
Pg. 118. Connections (networks). Problem Solving.
Parallel constrain satisfaction. Constrains can be satisfied in parallel by repeatedly passing activations among all the units (iterations), until after some number of cycles of activity all units have reached stable activation levels. This process is called relaxation, by analogy to a physical process that involve objects gradually achieving a stable shape or temperature. Achieving stability is called settling. Relaxing the network means adjusting the activation of all units based on the units to which they are connected until all units have stable high or low activations.
Pg. 124. Connections. Learning.
To bypass the impossibly long times required for some iterations – propagations McClelland et al (1995) advocate complementary learning systems that use both a slow-learning component for semantic as well as a fast learning one for object names and other information.
Pg. 192. Bodies in the world.
Lakoff and Johnson (1999) argue that human concepts are embodied in the sense that they are crucially shaped by our bodies and brains, specially by our sensory and motor systems. For example, our concepts of color are shaped in part by two aspects of our bodies: the color cones in our retinas that absorb light at different wavelengths, and the complex neural circuitry connected to those cones. Moreover, the basic concepts that we use to categorize the world are derived in part from the way that our visual and other sensory system detect the overall part-whole structure of the world. We can form visual images of elephant and chairs, but not of more abstract concepts such as animal and furniture.
*And yet, I’ve been thinking all along that even the most abstract concepts, including religion, must necessarily be build with elements of other concepts, all the way down to concepts that are senses related. To put it crudely, I think that God’s omnipotence and eternity can be distant grandchildren of seeing lighting or the ocean.
Pg. 195. Bodies, the world, dynamic systems.
Intentionality. The Chinese room.
“Unknown to you, the symbols are Chinese, and when you pass back the symbols you have looked up, you are providing sensible answers to questions that you received. Searle contents that is obvious that you are merely manipulating symbols you do not understand, so that a similarly a computer manipulating symbols is lacking in understanding.
Pg. 211 Societies.
Culture and Language.
People with different languages vary with respect to their ways of thinking about space, time, objects, color, shapes events and other minds. Boroditsky 2003. For example English speakers use front-back spatial terms to talk about time: childhood is behind you, etc. Mandarin Speakers use an alternative up-down spatial terms to talk about time.
Quite enjoyable beginner’s text book. It covers some aspects that none of the ones I’ve read so far do, such as: emotion, society, and dynamic systems.
Proposes these criteria for evaluating Theories of Mental Representation:
1.Representational Power
2.Computational Power
1.Problem Solving
i. Planning
ii. Decision
iii. Explanation
2.Learning
3.Language
2.Psychological plausibility
3.Neurological plausibility
4.Practical applicability
1.Education
2.Design
3.Intelligent Systems
4.Mental Illness
Which is interesting, because 1&2 categorize capabilities, 3 & 4 plausibility and 5 applicability.
Pg. 39. Logic. Neurological plausibility.
“Brain scanning experiments are being used to determine whether people perform deductions using the left half of their brain, as suggested by mental logic view that deduction is formal and independent of the content. The alternative hypothesis is that people perform deductions using the right half of the brain, as suggested by the mental models view that deduction requires regions of the brain that involve spatial reasoning.
Pg. 46. Rules. Problem Solving.
Rules have very simple If-Then structure.
Heuristic are rules of thumb that contribute to satisfactory solutions without considering all possibilities.
Pg. 49. Rules. Learning.
In inductive generalization, rules are formed from examples; but rules can also be formed from other rules by a process that in the SOAR model is called chinking and in the ACT model is called composition.
SOAR. State Operator and Result. A computational theory of thinking.
ACT Adaptive Control of Thought. Another computational theory of thinking.
The more a rule gets used successfully, the more likely is to be used in the future.
Pg. 51. Rules. Language
Chomsky continues to maintain that every human is born with an innate universal grammar. Contrary to his initial beliefs about children acquiring the ability to use language abductively by forming hypothesis about what rules to apply to their individual language (1972), he currently holds that children learn a language automatically by merely recognizing which of a set of finite set of possibilities that language employs (Chomsky 1988)
Pg 67. Concepts. Learning.
Consider the example of human face consisting of two eyes, a nose and a mouth. Perhaps babies learn this concept from experience as they repeatedly encounter examples of faces. But there is experimental evidence that babies do not have to learn the typical structure of faces, but rather are born expecting faces to look a certain way.
Pg 89. Analogies. Psychological plausibility.
Metaphorical interpretation appears to be an obligatory process that accompanies literal processing, rather than an optional process that occurs after literal processing.
Analogies. Neurological Plausibility.
Boroojerdi (2001) found that the left prefrontal cortex is involved in analogical reasoning by determining that magnetic stimulation of that part of the brain speeds up solution times for solving analogical problems. This is consistent with recent findings that reasoning involving complex relations, which is crucial for analogical thinking, also involves the left prefrontal cortex (Christoff 2001)
Pg. 106. Images. Neurological plausibility.
The areas of the brain most immediately connected to the retina have a spatial organization that is structurally similar to that of the retina. Since these areas preserve some of the spatial structure of objects presented to the retina (I recall reading somewhere that exposure to say, a column, produces neural activity in an area with some shape of a column). Kosslyn (2001) review neurological studies of visual, auditory and motor imagery.
Images. Summary.
Pg. 118. Connections (networks). Problem Solving.
Parallel constrain satisfaction. Constrains can be satisfied in parallel by repeatedly passing activations among all the units (iterations), until after some number of cycles of activity all units have reached stable activation levels. This process is called relaxation, by analogy to a physical process that involve objects gradually achieving a stable shape or temperature. Achieving stability is called settling. Relaxing the network means adjusting the activation of all units based on the units to which they are connected until all units have stable high or low activations.
Pg. 124. Connections. Learning.
To bypass the impossibly long times required for some iterations – propagations McClelland et al (1995) advocate complementary learning systems that use both a slow-learning component for semantic as well as a fast learning one for object names and other information.
Pg. 192. Bodies in the world.
Lakoff and Johnson (1999) argue that human concepts are embodied in the sense that they are crucially shaped by our bodies and brains, specially by our sensory and motor systems. For example, our concepts of color are shaped in part by two aspects of our bodies: the color cones in our retinas that absorb light at different wavelengths, and the complex neural circuitry connected to those cones. Moreover, the basic concepts that we use to categorize the world are derived in part from the way that our visual and other sensory system detect the overall part-whole structure of the world. We can form visual images of elephant and chairs, but not of more abstract concepts such as animal and furniture.
*And yet, I’ve been thinking all along that even the most abstract concepts, including religion, must necessarily be build with elements of other concepts, all the way down to concepts that are senses related. To put it crudely, I think that God’s omnipotence and eternity can be distant grandchildren of seeing lighting or the ocean.
Pg. 195. Bodies, the world, dynamic systems.
Intentionality. The Chinese room.
“Unknown to you, the symbols are Chinese, and when you pass back the symbols you have looked up, you are providing sensible answers to questions that you received. Searle contents that is obvious that you are merely manipulating symbols you do not understand, so that a similarly a computer manipulating symbols is lacking in understanding.
Pg. 211 Societies.
Culture and Language.
People with different languages vary with respect to their ways of thinking about space, time, objects, color, shapes events and other minds. Boroditsky 2003. For example English speakers use front-back spatial terms to talk about time: childhood is behind you, etc. Mandarin Speakers use an alternative up-down spatial terms to talk about time.
January 14, 2023
This was really cool from a computer science / AI perspective, because - surprise, surprise - it's outdated (1996)! Because it was written more than a couple decades ago, the book is charmingly blind to any developments in AI that happened after 2002, so we are left with the sequence of events that led right up to the present-day deep learning boom.
It was fun to learn how the cognitive science and AI fields actually weren't too different back in the day, and that much of our machine learning tools today actually grew out of cognitive science labs, not computer science labs. I especially appreciated that the author covered the limitations of computational models of the mind in great detail - in fact, the coverage of limitations takes up half the book!
Worth giving a read if you're interested in learning about the history of AI from someone who has been in the field themselves.
It was fun to learn how the cognitive science and AI fields actually weren't too different back in the day, and that much of our machine learning tools today actually grew out of cognitive science labs, not computer science labs. I especially appreciated that the author covered the limitations of computational models of the mind in great detail - in fact, the coverage of limitations takes up half the book!
Worth giving a read if you're interested in learning about the history of AI from someone who has been in the field themselves.
March 29, 2025
It deserves its name. With Thagard's basic explanations, I believe this book provides a brief first look at cognitive science. I followed the discussions in Cogist while reading this book. Experts in the field found the book limited, but I think it is good enough for a start.
While reading, I thought that people who study cognitive science stay in one category too much. I think that while the field was first thought and worked on the architecture of consciousness, today, developments are achieved as a result of the overlapping of separate studies in much more specialized fields (as much as possible). I may have given some baseless information, like gossip, but I hope that I will return to this review as I read more and update my current thoughts.
While reading, I thought that people who study cognitive science stay in one category too much. I think that while the field was first thought and worked on the architecture of consciousness, today, developments are achieved as a result of the overlapping of separate studies in much more specialized fields (as much as possible). I may have given some baseless information, like gossip, but I hope that I will return to this review as I read more and update my current thoughts.
July 29, 2018
I think the implementation of cognitive science is much more fun than the philosophical debates haha. The textbook is kind of dull to read for me. However, this book prompts me to contemplate what is the essence of human beings. If in the future a robot can simulate the full functionality of human intelligence, including the brain activities of consciousness and emotions, are they considered conscious? What is still left to make us different?
April 16, 2021
The book takes an interdisciplinary perspective and is a vital read for anyone trying to build concepts of cognitive science. It follows somewhat a textbook approach due to which it seems quite lengthy but is an important book that incorporates necessary frameworks of this discipline.
January 7, 2023
This was required reading for a course I took at University. I found it challenging to digest at times, but none the less worthwhile for anyone who is interested in the intersection between philosophy and cognition.
February 21, 2024
این کتاب رو برای کنکور کارشناسی ارشد علوم شناختی که چند روز دیگه دارم خوندم. به عنوان مرور ادبیات حوزه بد نبود. ساده و روان بود. هرچند خیلی کمدتر از این بود که بخوام برای آشنایی با این حوزه به کسی پیشنهاد کنم.
November 4, 2019
Es un buen libro para iniciarse en el tema. Funciona como un manual para la materia que dar Thagard. Me hubiera gustado que de más ejemplos y mejores explicaciones de los modelos de los que habla.
February 6, 2020
This book is VERY introductory. I'm not sure whether anybody in 21th century needs such introduction.
September 29, 2024
This book is required reading for an Intro to Cog Sci course in grad school. I like its structure and organization. It is very useful for readers who have zero background in cognitive science, and it gives a very helpful summary of what the field is all about, leaving you wanting to learn more.
June 26, 2021
Very academic.
February 16, 2013
Thagard writes an easy to understand introduction to cognitive science, which is exactly what the book's meant to be. The writing style is generally fluid and the material does not generally go beyond what a layman should understand. There are however, certain parts that can be further elaborated upon in order to improve a reader's understanding of the concepts (ex. the differences between rules and logic and mental representations). Thagard's own computational representational understanding of mind (CRUM) is well-developed, but some insight into alternative models would also be appreciated for the sake of a more well-rounded understanding of the field in general.
July 18, 2011
With exceptional clarity, Thagard gives a concise broad overview of research into strong artificial intelligence. Computer scientists will want something with more substance.
April 10, 2017
A well-rounded introduction to cognitive science. I enjoyed the breakdown between the different aspects of cognitive science and their respective coverage.
February 8, 2016
A great introduction to cognitive science.
Displaying 1 - 21 of 21 reviews