What does cognitive psychology entail? - Chapter 14

There are three issues at the heart of this chapter. First, Hobbes was busy with analysing the properties and processes that underlie human reason. Hobbes was concerned with how the mind receives and acquires knowledge. His answer led to a second point. He suggested that mathematical calculations and logical reasoning are the same. Thirdly, since the use of Indo-Arabic numbers, mathematics had become a routine process through it apply a few simple rules.

How did artificial intelligence develop?

Blaise Pascal's machine (1623-1662) was the first in a series of theoretical and technological innovations to create a mechanical likeness of the human mind. The Pascaline stands known as one of the first machines specially built for artificial intelligence (AI). After the "Pascaline", Leibniz came up with a new mechanism, which he called the "stepped cylinder". This could also carry out multiplications and divisions.

Leibniz began by constructing a new universal language for philosophy, the signs of which would function in the same way as the mathematical symbols. He imagined that different words would contain a different word, and that there was a logical hierarchy. An example is: the concept of "human" is included in the concept of "animal", and the concept "animal" is back in the concept of a "living thing". When the new universal language would be complete, all concepts would be given a rating that reflects the exact degree of inclusiveness in relation to each other. People would go out with the help of the new language different languages ​​and cultures can not only communicate with each other, but also be able to calculate solutions for the problems that divided them. However, the dream of Leibniz didn’t ever come true. Logical statements could not be reduced to arithmetic. Classic logic and traditional arithmetic was seen as an example of general symbolic logic.  Leibniz also came up with the idea of ​​binary arithmetic - the representation of all figures through of ones and zeros.

What did Charles Babbage do?

Charles Babbage (1792-1871) designed a mechanical calculator that was very accurate in generating  arithmetic tables and calculate polynomial functions. This machine he called the "difference engine" because it would use the "method of difference" to calculate things. He also came up with the analytical machine (or the "programmable computer"), with which he practically could perform all kinds of calculations. The machine consisted of five main components: An input system for the data and instructions. "The mill" to be able to carry out the calculations. This part was similar to the "difference engine". A controller to receive instructions from the input and check whether the "mill" actually performed the prescribed calculations in the right order. A memory to store the original data and the results and finally an output system. Today's computers still contain these five components. George Boole (1815- 1864) came with the symbolic logic, which provided a universal language through which it became possible to translate traditional logical problems into formal, mathematical terminology ("Boolean algebra").

What was the turing machine?

Alan Turing (1912-1954) came with the Turing machine, with which he could calculate and manipulate figures. With calculatable numbers he meant a series of numbers or symbols that could be manipulated in the Boolean way according to a collection of specifiable and consistent rules. 

Turing also came up with the Turing Test, a game that could determine whether or not one computer program could successfully fool human interrogators by making them think they are talking to a person. The test assesses the intelligence of a machine in its ability to interact with people during a complex task, or whenever else intelligent behaviour is needed.

Claude Shannon (1916-2001) showed that the notation in binary codes (i.e. the combination of ones and zeros) could not only be used to represent normal arithmetic, but also the more generalized problems of Boolean algebra and symbolic logic. People  could, by using the binary code, manage to display dichotomous logical conditions, such as "true / false" and "or". This could be displayed using mechanical sequences of circuits that could either be opened or be closed. The "off" or "on" setting of the circuit could be used to set the display onto patterns of ones and zeroes.

The question of machines as a mind became a prominent question. Turing wrote a paper about this: Computing Machinery and Intelligence. He suggested a situation involving external observer communicating in an impersonal way, such as typed messages, or with real people in one separate room or with a computer programmed to respond to the messages. The task of the observer is to determine, by asking questions and receiving answers, whether the respondent is a person or a machine. This is the Turing Test, and can manage to determine the intelligence of a machine. The goal of scientists was to design machines and software that were just as good or even better than people in performing intellectual tasks or judgements. Turing sadly committed suicide.

Shannon also remained influential and developed the information theory. According to this picture, every communication or signal are analysed in terms of a fundamental unit Shannon called the bit: the amount of information that can be transferred by the open or closed status of a single binary switch (1 or 0).

What did Alan Newell (1927-1992) and Herbert Simon (1916-2001) do?

Newell and Simon came up with the "thinking machine" which they called "Logic Theorist (or" LT "). Their program contained five logical basic axioms and three operations to enter certain characters to transform other new, logical and consistent characters. They started with a theory, and the LT caused it to be reduced to axioms. Newell and Simon did note that the LT was limited compared to a human thinker. To begin with, the machine tried to run all possible transformations on all possible signs, without pre-selection. Such a systematic search through all possibilities ("algorithm"), is only suitable if there is a limited number possibilities. Heuristics are the opposite of algorithmic strategies. Here Limits the search space that is explored for solutions. Eventually the LT was replaced by the General Problem Solver (GPS), a general machine who is, theoretically, able to solve any problem. The GPS used the "means-end analysis ", in which the desired solution for a problem is compared with the current state and where the difference between two is assessed. 

The GPS was able to recognize twelve kinds of differences between two states and was programmed with certain specific ones operations. Newell and Simon did not assume that their computer would function exactly like a human brain does. George Miller, Eugene Galanter and Karl Pribram were inspired by the GPS and wrote a book in which they presented the TOTE unit as a new central concept in the analysis of thinking and reasoning. The letters represent the order test-operate-test-exit. Their publication 'Plans and the Structure of Behavior' was a significant and symbolic event in the American psychological transition from a strong behaviourist orientation to a stronger focus on inner cognitive processes. Despite the early success of the GPS, this success proved to be limited to solving 'closed' logical problems that you could express in formal symbols and where there were correct solutions for. Other types of problems, such as learning to navigate in a new area, fell outside the capacity of the program. The GPS could only detect differences, just like it could only use operations that had been introduced by people. Formulating the problems, the specification of the search areas and the planning of specific strategies had to continue people are being done. Some AI scientists took a connectionist-computing model (i.e. the "parallel distributed processing "or" PDP "), which works differently than the previous models. The LT and GPS worked by performing a series of actions according to a specific order on one specified set of symbols, where both the actions and the symbols are stored in the memory. These programs are described as serial and symbolic in their own process strategy. At PDP, several actions take place simultaneously, and they run according to a pattern of activity throughout the system, rather than as symbols in predefined locations. Some PDPs have shown that they are important aspects of the logical problem-solving ability to simulate something that the serialist programs cannot. Margaret Boden distinguished between "improbabilist" and "impossibilist creativity". At the first being use well-known ideas or components according to certain rules. An example is repeating of a previously used sentence or word according to our Dutch grammar. With "impossibilist creativity" the rules are changeable.

What is the thought experiment of John Searle?

John Searle (b.1932) came up with a thought experiment called the Chinese Room, which describes that when someone is in a room where a language is spoken that he or she does not understand, and where papers with the correct answer in Chinese are being passed around, one would respond as a computer would. He compared the situation in the Chinese room with that of a computer that is programmed to read and respond according to a formal system of symbols: both react mechanically and without real understanding. Searle distinguished between a weak AI and a strong AI. In a weak AI, computer processes can be useful models for significant aspects of human cognitive processes that are seen, and the computer becomes one useful tool for testing concrete hypotheses about how the mind could to work. Searle was fine with this, but with strong AI he did not agree. This is the assumption that a computer program can possibly be the mind and therefore cannot be distinguished from the real spirit in all functional aspects. The preference of Searle is based on the thought experiment, or the introspective consideration of his own conscious subjective state.

Who was George A. Miller (1920-2012)?

Miller carried out research in the field of speech and communication. Miller got interested in the information theory from Shannon and used this in his research. He examined how and under which conditions, meaningful speech could be recognized if there were different intensities and types of background noise. He studied the phonemes of English. He got a good reputation as a psycholinguist. He invented the magic number seven, plus or minus two. This meant that people in general can follow and correctly remember an order of seven items. This also applied if the items were more complex. This he explained by chuncking: one process in which people organize simple stimuli in higher-order concepts.

Noam Chomsky was involved in syntax, the logical relationship between words that are led by the grammar of the language. According to Chomsky, there was an innate universal grammar in the human brain. Chomsky and Miller became friends. Chomsky was critical of the behaviourist analysis of language. Miller became convinced of the nativist argument. Miller worked with his older Harvard colleague Jerome S. Bruner (1915-2016). He did both his bachelor's and master's in experimental psychology at the Duke and followed a PhD at Harvard. He worked with students and colleagues on a research that became known as new perspective on perception. These studies showed how non-objective factors systematically become a process of perception. Bruner also conducted research into concept formations. He called this entire research program the 'Cognition Project'. Bruner and Miller together founded the Harvard Centre for Cognitive Studies. Some see this as the start of a cognitive revolution in psychology. British psychologists have in fact always been less behaviourist than American. The best-known British experimentalist was Sir Frederic Bartlett (1886-1969), and his approach was cognitive. He did research on memory and how this is formed by cultural and social predispositions, as well as schemata. People remember stories about time less well and adapt it more to their own cultural background.

What did Ulric Neisser do for cognitive psychology?

Ulric Neisser (b.1928) came up with the "flash bulb memory", a vivid memory about where someone was and what someone was doing when a certain impressive event took place. Later he found out that memory is generally not a perfect photo replica of an experience, but rather that there is a construction based on an experience, adapted and processed by emotional and mental factors. Oliver Selfridge (1926-2008) came with the "Pandemonium", a program that consists of small mini-programs (or "Demons" as he himself called them). These demons worked simultaneously, but they independently of each other worked on different aspects of a particular problem. Created around 1960, Neisser had a mostly general idea. He was of the opinion that the perception, the attention span, the visual search and computer pattern recognition were the input while finding solutions to problems and memory were the output. Everything in between was a different kind of processing. Also Neisser stressed that the mind processes only the sensory input. This was a modern formulation of the earlier assumption of the mind as an active, transformational mediator. He summarized this his book ‘Cognitive Psychology’. According to him, the definition of cognition is: all processes through which sensory input is transformed, reduced, elaborated, stored, restored and used. After the publication of his book, cognitive psychology was quickly accepted as a sub-discipline in academic psychology.

ExamTips

• Try to understand how our modern technology was formed out of machines like the GPS and the Turing Test. What are the psychological implications of this advancement.
• Why are we so fascinated with technology outsmarting us?